Keywords: Optimal control, Nonlinear system theory
Abstract: The dissipativity notion for open dynamic systems conceived by Jan Willems as well as the optimal control twin breakthroughs, i.e. the maximum principle by Lev Pontryagin et al. and Richard Bellman’s dynamic programming, are fundamental pillars of systems and control. At first glance, it may appear as if dissipation inequalities are only loosely related to optimal control. Yet, in this talk we explore the fundamental relations between both through the turnpike phenomenon in optimal control.

The turnpike phenomenon refers to situations where for different initial conditions and varying horizons, the optimal state and input trajectories spend increasing amounts of time close to the optimal steady state. The first observations of such behavior can be traced back to macroeconomics and the works of John von Neumann and Frank P. Ramsey in the 1930s and 1920s. Here, we discuss the equivalence of strict dissipativity and turnpike properties of optimal control problems.

Closing the loop with MPC, we explore how dissipativity helps to analyze the properties of receding-horizon approximations to infinite-horizon problems. Considering port-Hamiltonian structures of thermodynamic systems, we show how to exploit physics in optimal control problems with respect to energy, entropy, and exergy. This leads to turnpikes on the manifold of thermodynamic equilibria. Exploring dissipativity in stochastic optimal control then brings forth turnpikes in the Wasserstein metric.

At last, we enter the turnpike once more to analyze the training of neural networks through the lens of optimal control. Specifically, we discuss the dissipativity properties of cross-entropy loss functions – thus we conclude by leveraging optimal control and dissipativity concepts for deep learning.

Keywords: Identification, Modeling, Network analysis and control
Abstract: This paper addresses the problem of consistently estimating a continuous-time (CT) diffusively coupled network (DCN) to identify physical components in a physical network. We develop a three-step frequency-domain identification method for linear CT DCNs that allows to accurately recover all the physical component values of the network while exploiting the particular symmetric structure in a DCN model. This method uses the estimated noise covariance as a non-parametric noise model to minimize variance of the parameter estimates, obviating the need to select a parametric noise model. The method is illustrated with an application from In-Circuit Testing of printed circuit boards. Experimental results highlight the method's ability to consistently estimate component values in a complex network with only a single excitation.

Keywords: Identification, Nonlinear system identification, Energy systems
Abstract: Accurate identification of lithium-ion (Li-ion) battery parameters is essential for managing and predicting battery behavior. However, existing discrete-time methods hinder the estimation of physical parameters and face the fast-slow dynamics problem of the battery. In this paper, we develop a continuous-time approach that enables the estimation of battery parameters directly from sampled data. This method avoids discretization errors in converting continuous-time models into discrete-time ones. Moreover, the developed method is capable of jointly identifying the open-circuit voltage (OCV) and the state of charge (SOC) relation of batteries without utilizing offline OCV tests. By modeling the OCV-SOC curve as a cubic B-spline, we represent the piecewise nonlinearity of the OCV curve with high fidelity, facilitating its estimation. By solving a rank and L1 regularized least squares problem, we identify battery parameters and the OCV-SOC relation directly from the battery's dynamic data. Simulated and real-life data validate the effectiveness of the developed method.

Keywords: Identification
Abstract: The work focuses on direct data-driven controller design using input-output measurements corrupted by bounded additive noise. The main goal is to tackle the data-driven model reference control problem. To this end, we employ a set-membership framework and define the feasible controller parameter set, i.e., the set of parameters consistent with the noise bounds and the model-matching condition. We determine the controller parameters as the Chebyshev center of this set. We also provide a data-driven condition that is sufficient for establishing stability. This condition is also robust against the presence of bounded noise. The approach utilizes polynomial optimization and achieves global optimality via semidefinite relaxation techniques. A simulation example demonstrates the effectiveness of the proposed approach.

Keywords: Identification, Machine learning, Optimization
Abstract: Accurate estimation of financial portfolio covariance matrix is crucial for effective risk management and portfolio optimization. Traditional methods, such as Rolling Window, Exponentially Weighted Moving Average (EWMA), and Multivariate GARCH (MGARCH), often struggle in its prediction due to the harsh predictability of the dynamic relationships among financial assets data. This paper introduces a hybrid predictor that linearly combines traditional covariance prediction methods with an Expanding Window (EW) model, exploiting a time-varying weighting mechanism to enhance prediction accuracy. The proposed method’s efficacy is validated on actual time-series data from 2010 to 2024 of 9 major S&P500 stocks. Results show that the proposed strategy significantly improves covariance estimation, particularly in the end of financial quarters. The higher covariance prediction quality directly reflects into more accurate risk-return trade-offs, as a consequence of portfolio optimization when using predicted covariance as portfolio risk proxy. These results can ultimately lead to sharper investment decisions.

Keywords: Identification, Manufacturing processes, Linear systems
Abstract: Some industrial processes cannot have their control loop opened for classic direct system identification, therefore closed-loop system identification is required, thus leading to issues with cross-correlated signals. Instrumental variables can be used to estimate noise-free control and output signals. Typically, a filter is designed to limit high-frequency noise during the identification process. However, an optimal filter can minimize noise without reducing valuable information from the input and output signals. When the output is tainted with white noise, the simplified refined instrumental variable method for closed-loop systems (CLSRIV) can be used. However, in the context of machining with Computer Numerical Control machine tools, the closed-loop of the spindle angular speed is more complex. Indeed, this loop includes a parallel proportional-integral controller, where a prefiltered reference is only applied to the integral component. The aim of this paper is to explore various closed-loop system identification methods in a white noise context by considering such specific prefiltering and parallel controller. A Monte-Carlo simulation is then proposed to analyze the statistical properties of the proposed methods.

Keywords: Identification, Stochastic systems, Signal processing
Abstract: We propose a recursive Bayesian estimation procedure for multivariate autoregressive models with exogenous inputs based on message passing in a factor graph. Unlike recursive least-squares, our method produces full posterior distributions for both the autoregressive coefficients and noise precision. The uncertainties regarding these estimates propagate into the uncertainties on predictions for future system outputs, and support online model evidence calculations. We demonstrate convergence empirically on a synthetic autoregressive system and competitive performance on a double mass-spring-damper system.

Keywords: Observers for nonlinear systems, Lyapunov methods
Abstract: This paper investigates the problem of observability and state reconstruction for nonlinear systems subject to unknown disturbances. We propose definitions of practical observability in this context and two kinds of practical observers. Under certain assumptions, we use the Lyapunov stability analysis method to show how to build an observer so that the estimation error converges in a "practical" sense. Examples illustrate the concepts and the results.

Keywords: Observers for nonlinear systems, Machine learning, Neural networks
Abstract: Contraction analysis offers, through elegant mathematical developments, a unified way of designing observers for a general class of nonlinear systems, where the observer correction term is obtained by solving an infinite dimensional inequality that guarantees global exponential convergence. However, solving the matrix partial differential inequality involved in contraction analysis design is both analytically and numerically challenging and represents a long-lasting challenge that prevented its wide use. Therefore, the present paper proposes a novel approach that relies on an unsupervised Physics Informed Neural Network (PINN) to design the observer's correction term by enforcing the partial differential inequality in the loss function. The performance of the proposed PINN-based nonlinear observer is assessed in numerical simulation as well as its robustness to measurement noise and neural network approximation error.

Keywords: Observers for nonlinear systems, Nonlinear system theory, Uncertain systems
Abstract: This work introduces a variable structure velocity observer formulation for a broad class of second order nonlinear engineering systems. The proposed observer formulation incorporates a novel error signal that performs similarly to a sliding-mode observer for larger values of the position observation error and acts as a high-gain term when the position observation error is small. The stability of the observer formulation is validated through a novel Lyapunov-like argument, ensuring the practical asymptotic convergence of the observation error. %uniform ultimate boundedness of the observation error. Specifically, the velocity observation error is guaranteed to converge to a small region near the origin, that can be adjusted to be arbitrarily small by the observation gains. The effectiveness of the observer formulation is examined through numerical simulations on the New England 39-bus power system, a testing platform comprising 39 buses and 10 generators.

Keywords: Observers for nonlinear systems, Switched systems, LMI's/BMI's/SOS's
Abstract: This work proposes an unknown input observer, a classical observer, and an observer-based controller for switched linear systems with non-detectable and non-stabilizable modes. Sufficient conditions for observer synthesis are formulated as LMIs. The results, extended to observer-based control, rely on a switched Lyapunov function and average dwell time for system convergence. Two examples illustrate the methodology: an academic case for comparison with existing approaches and a buck converter, a benchmark for switched systems.

Keywords: Observers for nonlinear systems
Abstract: This manuscript introduces a two-step interval estimation approach for continuous-time Lipschitz nonlinear systems, utilizing robust observer design and ellipsoidal analysis. First, the peak-to-peak performance index is adopted to design a novel robust observer with more freedom, attenuating the influence of the uncertainties. Subsequently, interval estimation is achieved by ellipsoidal analysis of an auxiliary error dynamic system, further enhancing the estimation accuracy. Finally, we verify the merit and effectiveness of our approach through comparative numerical simulations.

Keywords: Output feedback, Observers for nonlinear systems, Uncertain systems
Abstract: This paper presents an optimal dynamic control framework for bounded Jacobian nonlinear discrete-time (DT) systems with nonlinear observations affected by both state and process noise. Rather than directly stabilizing the uncertain system, we focus on stabilizing an interval observer in a higher-dimensional space, whose states bound the true system states. Our nonlinear dynamic control method introduces added flexibility over traditional static and linear approaches, effectively compensating for system nonlinearities and enabling potentially tighter closed-loop intervals. Additionally, we establish a separation principle that allows for the design of observer and control gains. We further derive tractable matrix inequalities to ensure system stability in the closed-loop configuration. The simulation results show that the proposed dynamic control approach significantly outperforms a static counterpart method.

Keywords: Uncertain systems, Adaptive control, Constrained control
Abstract: This paper introduces a novel set-theoretic model reference adaptive control framework with bandwidth regulation, which has the capability to provide strict closed-loop performance guarantees without inducing high-frequency oscillations. While standard model reference adaptive control approaches aim to reduce the error between an uncertain dynamical system and a reference model, they cannot impose a user-defined bound on this error---a critical requirement for safety-critical applications. Set-theoretic model reference adaptive control employs generalized Lyapunov barrier functions to ensure the error strictly remains within a user-defined bound. However, as the error approaches this bound, high-gain learning rates may induce high-frequency oscillations that potentially leads to system instability in practice. The key feature of the proposed framework is an adaptive bandwidth regulation mechanism, which adjusts the error dynamics to prevent such oscillations. The presented system-theoretical analysis confirms the stability of the proposed framework and illustrative numerical examples demonstrate its effectiveness in maintaining error bounds while ensuring smooth state and control histories.

Keywords: Uncertain systems, Robust control, Feedback linearization
Abstract: We study trajectory tracking for flat nonlinear systems with unmatched uncertainties using the model-following control (MFC) architecture. We apply state feedback linearisation control for the process and propose a simplified implementation of the model control loop which results in a simple model in Brunovský-form that represents the nominal feedback linearised dynamics of the nonlinear process. To compensate possibly unmatched model uncertainties, we employ Lyapunov redesign with numeric derivatives of the output. It turns out that for a special initialisation of the model, the MFC reduces to a single-loop control design. We illustrate our results by a numerical example.

Keywords: Robust adaptive control, Uncertain systems, Constrained control
Abstract: We study the trajectory-tracking problem for systems with uncertain nonlinear dynamics. We focus on the prescribed performance control problem, which aims to establish predefined transient- and steady-state system specifications by forcing the evolution of the tracking error inside predefined functions of time, called performance functions. A typical problem in related works is the difficulty of achieving asymptotic tracking; arbitrarily small steady-state errors can typically be achieved via discontinuous controllers or performance functions that converge to arbitrarily small values, both of which induce control chattering. We develop a novel control algorithm that solves the trajectory-tracking problem under prescribed performance specifications for control-affine systems with unknown nonlinear terms. The algorithm relies on an innovative integration of prescribed performance and boundary layer control and guarantees the finite-time convergence of the tracking error to a predefined and arbitrarily small set, independent of the performance functions, the system dynamics, or the control gains’ selection. Comparative simulation studies demonstrate that the proposed algorithm outperforms standard prescribed performance algorithms in the sense that it achieves lower steady-state errors without chattering.

Keywords: Uncertain systems, Linear time-varying systems, Computational methods
Abstract: We consider a two-dimensional SISO LTI system closed by uncertain linear feedback. The feedback gain is time-varying, bounded, and has a bounded derivative (both bounds are known). We investigate the asymptotic stability of this system under all admissible behaviors of the gain. Note that the situation is similar to the classical absolute stability problem of Lurie–Aizerman with two differences: linearity and derivative constraint. Our method of analysis is therefore inspired by the variational ideas of Pyatnitskii, Barabanov, Margaliot, and others developed for the absolute stability problem. We derive the Hamilton–Jacobi–Bellman equation for a function describing the “most unstable” of the possible portraits of the closed-loop system. A numerical method is proposed for solving the equation. Based on the solution, sufficient conditions are formulated for the asymptotic stability and instability. The method is applied to an equation arising from the analysis of a power electronics synchronization circuit.

Keywords: Uncertain systems, Robust adaptive control
Abstract: In this paper, we consider the problem of finding the shortest path in a graph when there is aleatoric uncertainty about the presence and/or cost of certain edges. We investigate hybrid path planning, in which an agent observes the uncertain information as it traverses the graph and may adapt to the new information. We model this problem as a robust partially observable Markov decision process (robust POMDP) and identify an information state for dynamic programming decomposition. We propose a series of pruning steps, which truncate the state space based on the relationship between the cost and the uncertainty. We then show how to adapt the Neural Monte Carlo Tree Search (neural MCTS) algorithm to obtain an approximate solution. Finally, we present a numerical study to illustrate the effectiveness of the proposed approach.

Keywords: Uncertain systems, Observers for nonlinear systems, Lyapunov methods
Abstract: In this paper, the problem of designing a robust unknown input observer (RUIO) for quadratically inner bounded (QIB) one-sided Lipschitz (OSL) nonlinear systems considering noise and unknown inputs is addressed. The RUIO is formulated as a convex optimization problem where parameter uncertainty and noise are considered. Sufficient conditions for observer gain synthesis are shown to be equivalent to solving a finite sets of Linear Matrix Inequalities (LMIs) and Linear Matrix Equalities (LMEs). Two illustrative examples (a bioreactor for biomass production from substrate consumption and a robot manipulator) are used to test the effectiveness of the approach in the presence of unknown inputs, parameter uncertainty and measurement noise.

Keywords: Game theoretical methods, Modeling, Manufacturing processes
Abstract: Considering a supply chain (SC) with partial vertical integration, we attempt to seek answers to several questions related to the cooperation-competition based friction, abundant in such networks. Such an SC can represent a supplier with an in-house production unit that attempts to control an out-house production unit via the said friction. The two production units can have different sets of loyal customer-bases and the aim of the manufacturer supplier-duo would be to get the best out of the two customer bases. Our analysis shows that under certain market conditions, an optimal strategy might be to allow both units to earn positive profits—particularly when they hold similar market power and when customer loyalty is high. In cases of weaker customer loyalty, however, the optimal approach may involve pressurizing the out-house unit to operate at minimal profits. Even more intriguing is the scenario where the out-house unit has a greater market power and customer loyalty remains strong; here, it may be optimal for the in-house unit to operate at a loss just enough to dismantle the downstream monopoly.

Keywords: Game theoretical methods, Optimization algorithms, Variational methods
Abstract: We consider generalized Nash equilibrium (GNE) problems in games with strongly monotone pseudo-gradients and jointly linear coupling constraints. We establish the convergence rate of a payoff-based approach intended to learn a variational GNE (v-GNE) in such games. While convergent algorithms have recently been proposed in this setting given full or partial information of the gradients, rate of convergence in the payoff-based information setting has been an open problem. Leveraging properties of a game extended from the original one by a dual player, we establish a convergence rate of O(frac{1}{t^{4/7}}) to a v-GNE of the game.

Keywords: Game theoretical methods, Stochastic control, Optimal control
Abstract: We analyze best response dynamics for finding a Nash equilibrium of an infinite horizon zero-sum stochastic linear quadratic dynamic game (LQDG) with partial and asymmetric information. We derive explicit expressions for each player's best response within the class of pure linear dynamic output feedback control strategies where the internal state dimension of each control strategy is an integer multiple of the system state dimension. With each best response, the players form increasingly higher-order belief states, leading to infinite-dimensional internal states. However, we observe in extensive numerical experiments that the game's value converges after just a few iterations, suggesting that strategies associated with increasingly higher-order belief states eventually provide no benefit. To help explain this convergence, our numerical analysis reveals rapid decay of the controllability and observability Gramian eigenvalues and Hankel singular values in higher-order belief dynamics, indicating that the higher-order belief dynamics become increasingly difficult for both players to control and observe. Consequently, the higher-order belief dynamics can be closely approximated by low-order belief dynamics with bounded error, and thus feedback strategies with limited internal state dimension can closely approximate a Nash equilibrium.

Keywords: Game theoretical methods, Agents and autonomous systems, Optimal control
Abstract: An open problem in linear quadratic (LQ) games has been characterizing the Nash equilibria. This problem has renewed relevance given the surge of work on understanding the convergence of learning algorithms in dynamic games. This paper investigates scalar discrete-time infinite-horizon LQ games with two agents. Even in this arguably simple setting, there are no results for finding textit{all} Nash equilibria. By analyzing the best response map, we formulate a polynomial system of equations characterizing the linear feedback Nash equilibria. This enables us to bring in tools from algebraic geometry, particularly the Gröbner basis, to study the roots of this polynomial system. Consequently, we can not only compute all Nash equilibria numerically, but we can also characterize their number with explicit conditions. For instance, we prove that the LQ games under consideration admit at most three Nash equilibria. We further provide sufficient conditions for the existence of at most two Nash equilibria and sufficient conditions for the uniqueness of the Nash equilibrium. Our numerical experiments demonstrate the tightness of our bounds and showcase the increased complexity in settings with more than two agents.

Keywords: Game theoretical methods, Optimization, Modeling
Abstract: Resource allocation plays an important role in the offense-defense game process. This problem is complex, especially involving multiple stages, attackers, and defense measures. To address the above-mentioned problem, a game theory-reinforcement learning model is used to determine the optimal resource allocation for both players. In each stage, the defender can deploy false targets (FTs) and strengthen the target, whereas the attacker can choose to identify the FTs or attack the targets randomly. Each player aims to maximize their utility by choosing appropriate actions. In specific, the Q-learning method is adopted in the game theoretic model, exploring the best resource allocation strategy over the entire planning horizon. Comparative studies demonstrate the effectiveness and feasibility of the proposed model. The model can provide certain support for the resource allocation problem.

Keywords: Game theoretical methods
Abstract: We study model predictive control (MPC) schemes for non-cooperative dynamic games regarding stabilization. The dynamic games are modelled as generalized Nash equilibrium problems (GNEPs), in which a shared constraint is given as a jointly controlled time-discrete (linear) dynamics. Furthermore, the players’ objectives areinterdependent. For the stability analysis, we introduce the class of α-quasi-GNEPs, which approximate the original games. Basing the MPC scheme on these instead, we derive conditions on the players’ objectives which guarantee asymptotic stability in the presence of stabilizing constraints in the form of a Lyapunov terminal cost. Furthermore, we show that a suitable Lyapunov terminal cost can be obtained from a non-game-based MPC scheme. This non-game-based MPC scheme relies on a classical optimal control problem for the aggregated cost. Hence, known results for determining the Lyapunov cost can be applied and carried over to the game-based setting. The theoretical results are complemented by numerical experiments.

Keywords: Energy systems, Machine learning, Neural networks
Abstract: Accurate estimation of battery State-of-Health (SOH) is crucial in the battery monitoring and management process. Several methods have been proposed to model and estimate battery aging dynamics, either formally, model-based or data-driven. One key challenge in SOH modeling is the generality of the SOH modeling approach, which requires consideration of inherent dependencies among the various multidisciplinary stress factors involved. In this paper, we present an end-to-end self-supervised approach based on Convolutional Auto-Encoders (CAEs) for learning informative intermediate features from battery measurable properties such as voltage, current and temperature. We then employ the learned features to estimate the change in battery SOH by a light-weight feed-forward neural network. The learned features represent essential information in battery dynamics and surpass the human-engineered features in terms of correlation with the target SOH characteristic. Utilizing these representative features, our SOH estimation model yields 58.7% and 45.0% average performance improvement on two large battery datasets compared to the state-of-the-art machine learning methods.

Keywords: Energy systems, Stochastic control, Uncertain systems
Abstract: As more and more energy is generated from intermittent sources, such as wind and solar, it becomes increasingly difficult to balance energy supply and demand. Flexible energy consumption, also referred to as demand response, is a cost-effective way to mitigate this growing challenge. In this paper, we propose an approach that generates optimal price signals for indirect control of the flexible energy consumption. With a stochastic and dynamic demand-response model, called the flexibility function, we generate the price signals by solving a stochastic open-loop optimal control problem (OCP). Using numerical examples, we demonstrate the improved performance of the generated price signals compared to those obtained by solving a deterministic OCP. Furthermore, we compare with the performance of an optimal stationary feedback law obtained by solving the Hamilton-Jacobi-Bellman equation. Based on the numerical results, we conclude that the performance is improved by accounting for uncertainty when generating the price signals. Furthermore, the improvement increases with the level of uncertainty in the demand response model.

Keywords: Energy systems, Emerging control applications, Optimal control
Abstract: The problem of optimizing the power output of a class of Airborne Wind Energy Systems (AWES), named fly-gen, is considered. Fly-gen AWES, called windplanes in this paper, harvest wind power by means of an autonomous tethered aircraft that carries out periodic trajectories roughly perpendicular to the wind flow (crosswind conditions), using on-board turbines and converters and an electric tether to transfer power to the ground. The amount of generated power and its variability strongly depend on the flown trajectory, whose optimization is a highly nonlinear and non-convex problem. Differently from most of the existing literature on the topic, this problem is here addressed from a multi-objective perspective, where both the average power and its variability are considered. Through a recently-proposed pseudo-spectral decomposition of the states and inputs, a rather small-scale nonlinear program is derived to obtain a periodic orbit that maximizes the average power under a constraint on its variability. Then, a series of such programs is formulated and solved to approximate the Pareto front of the problem. Finally, the latter is exploited to analyze the possible trade-offs. The main finding of this work is that, contrary to what postulated so far in the scientific community, it is possible to operate the windplane with minimal power fluctuations (10% of the average) with a very small reduction of mean power, of the order of 5% with respect to the maximum achievable. Additional considerations regarding the sensitivity of the optimal trajectories to various factors are presented, too. These results pave the way for a completely novel way of optimizing and controlling windplanes.

Keywords: Energy systems, Observers for nonlinear systems, Lyapunov methods
Abstract: Ensuring the efficient and safe operation of lithium-ion batteries requires accurate knowledge of its internal states. However, these internal battery variables cannot be measured in general and must therefore be estimated. In this context, we present an observer design that robustly estimates the state of charge of a lithium-ion cell. This design is appealing for several reasons. First, it is based on a reduced-order physicsbased model from the literature, which has the distinctive feature to combine the advantages of equivalent circuit models in terms of simplicity and low computational effort, and of single particle electrochemical models, which more faithfully describe the internal dynamics of the battery. Second, the proposed observer is very simple to design and its robust convergence towards the battery internal states is systematically guaranteed based on Lyapunov arguments. Simulation results demonstrate the effectiveness of the designed observer in estimating the state of charge of the battery given data generated by a higher-fidelity model to emulate an experimental setting.

Keywords: Energy systems, Optimization algorithms, Optimization
Abstract: This paper presents a Hop Optimized Multi-Exchange Routing and Scheduling (HOMERS) algorithm designed to solve concurrent Peer-to-Peer (P2P) exchanges in the Energy Internet. HOMERS formulates multiple P2P exchanges as a single-step mixed-integer nonlinear programming optimization problem and effectively routes and schedules them. Using graph theory, HOMERS ensures power distribution from each source to its corresponding destination with required equivalent multi-path solutions, without any power flow conflicts. Performance evaluations of HOMER were conducted on randomly generated dense topologies with community sizes ranging from 10 to 400 nodes and were compared against Dijkstra's algorithm modified to prevent power flow violations. The results in a 400-node network show that HOMERS achieved a 68.29% reduction in the mean number of hops, which translates to a decrease in the total distribution loss. Though Dijkstra was computationally faster in routing for single-path solutions, HOMERS performed better in providing equivalent multi-path solutions for congestion reduction among 400 nodes with a 65.27% reduction in the mean runtime. Thus, the algorithm is scalable, effectively manages congestion with multi-path solutions, and also routes for feasible solutions under heterogeneous and limited distribution capacity constraints.

Keywords: Energy systems, Robust control, Constrained control
Abstract: The proper disposal and repurposing of end-of-life electric vehicle batteries are critical for maximizing their environmental benefits. This study introduces a robust model predictive control (MPC) framework designed to optimize the battery discharging process during pre-treatment, ensuring both efficiency and safety. The proposed method explicitly incorporates temperature constraints to prevent overheating and potential hazards. By leveraging a control-oriented equivalent circuit model integrated with thermal dynamics, the MPC algorithm dynamically adjusts the discharging profile to maintain safe operating temperatures. Additionally, the robust controller is designed to account for model mismatches between the nonlinear battery dynamics and the linearized model, ensuring reliable performance under varying conditions. The effectiveness of this approach is demonstrated through simulations comparing the robust MPC method with conventional discharging strategies, including constant current-constant voltage (CC-CV) and constant current-constant temperature (CC-CT) methods. Results indicate that the robust MPC framework significantly reduces discharging time while adhering to safety constraints, offering a promising solution for the recycling and second-life applications of lithium-ion batteries.

Keywords: Control education, Computer aided learning, Control courses and labs
Abstract: This study examines the effectiveness of LU-PZE https://lu-pze.github.io - a web-based interactive tool designed to help students visualize and interact with fundamental concepts in automatic control. A total of 200 students enrolled in an introductory course on automatic control had the option to use LU-PZE as an additional resource for learning. LU-PZE offers users randomized quizzes, structured assignments, and real-time visualizations of theoretical concepts. An analysis of student usage shows that LU-PZE successfully increased student engagement and improved their understanding of automatic control principles. Student feedback also reveals that the students appreciate interactive tools for learning. These findings suggest that LU-PZE can be a valuable tool for teachers to promote active learning and enhance student outcomes in introductory courses in automatic control.

Keywords: Control education, Computer aided learning, Predictive control for nonlinear systems
Abstract: This paper presents a comprehensive framework aimed at enhancing education in modeling, optimal control, and nonlinear Model Predictive Control (MPC) through a practical greenhouse climate control model. The framework includes a detailed mathematical model of lettuce growth and greenhouse environment, which are influenced by real-time external weather conditions obtained via an application programming interface (API). Using this data, the MPC-based approach dynamically adjusts greenhouse conditions, optimizing plant growth and energy consumption and minimizing the social cost of CO2. The presented results demonstrate the effectiveness of this approach in balancing energy use with crop yield and reducing CO2 emissions, contributing to economic efficiency and environmental sustainability. The framework also provides a valuable resource for making control systems education more engaging and effective. The main aim is to provide students with a hands-on platform to understand the principles of modeling, the power of MPC and the trade-offs between profitability and sustainability in agricultural systems. The framework gives students a hands-on experience, helping them to understand the control theory better, connecting it to the practical implementation, and developing their problem-solving skills. It can be accessed at https://ecompc4greenhouse.streamlit.app.

Keywords: Emerging control applications, Markov processes, Game theoretical methods
Abstract: This paper presents an incentivizing control of resource cycling problems for surveillance that enables autonomous frontline patrollers to achieve a socially optimal real‑time allocation under moral hazard conditions. We model the resource cycling problem as a discrete‑time Markov jump linear system that captures transient transition probabilities while recasting the surveillance routing problem in the steady state. To reflect the individual decision-making of the patrollers to optimize the surveillance performance for urban security, we propose that the solution to the resource cycling problem is reduced to the continuous-time theoretically-optimal incentivizing control solution of the long-term time-average revenue case, which is inspired by contract theory, using the forward-difference-type Euler approximation. The proposed incentivizing control is proven to be a strategy‑proof mechanism, and its effectiveness is verified through numerical examples.

Keywords: Modeling
Abstract: Cyber-Physical Systems (CPS) leverage Digital Twins (DT) to enable real-time plant monitoring, control and management. This paper explores the critical problem of building DTs that integrate continuous-time models made of Differential and Algebraic Equation (DAE) systems and event-based models made of Discrete Event Systems (DEVS). We address the complexities of this integration, and particularly the challenges posed when a computationally efficient integration of DAE models must co-exist with the occurrence of events from DEVS ones. We propose a framework that employs Object-Oriented Modelling (OOM), and in particular the Modelica language, facilitating the efficient assembly of multi-physics systems while maintaining high performance. We demonstrate our approach through a case study, highlighting the potential for scalable and effective simulations within the CPS landscape.

Keywords: Transportation systems, Emerging control applications, Modeling
Abstract: Large crowds at events disrupt city transportation networks, and incentivizing public transit to mitigate these impacts remains challenging. This paper presents a method to design and assess public transit incentives for large events. This is done by first developing a traveller mobility model that describes how information from routing apps data, public transit fare and frequency guides the choice of travellers between two modes and updates the traffic state based on their mode choice. Traveller’s wait time, departure time, public transit capacity, road capacity, road congestion, and parking capacity of private vehicles at destination are all considered when updating the number of users who completed their trip. The influence of public transit fare and frequency on the population’s overall travel time, modal split and parking overcapacity is studied, along with the influence of the ratio of travellers informed with routing apps. Finally, incentives are selected by including operational cost and user satisfaction in an optimization approach.

Keywords: Process control, Adaptive control
Abstract: Modern process plants increasingly rely on Digital Twins (DTs) for optimal management, and often incorporate autotuning controllers to enhance performance. Since autotuning captures snapshots of plant operation, a key question arises: can autotuners act as probes to ensure alignment between the Digital Twin and its physical counterpart? Additionally, can they help detect and possibly quantify discrepancies between the two, and enhance condition monitoring? This paper explores these possibilities, formulating suggestions for a new generation of ``DT-oriented'' autotuners.

Keywords: Robotics, Autonomous robots, Robust adaptive control
Abstract: Deep reinforcement learning (DRL) is emerging as a promising method for adaptive robotic motion and complex task automation, effectively addressing the limitations of traditional control methods. However, ensuring safety throughout both the learning process and policy deployment remains a key challenge due to the risky exploration inherent in DRL, as well as the discrete nature of actions taken at intervals. These discontinuities, despite being part of a continuous action space, can lead to abrupt changes between successive actions, causing instability and unsafe intermediate states. To address these challenges, this paper proposes an integrated framework that combines DRL with a jerk-bounded trajectory generator (JBTG) and a robust low-level control strategy, significantly enhancing the safety, stability, and reliability of robotic manipulators. The low-level controller ensures the precise execution of DRL-generated commands, while the JBTG refines these motions to produce smooth, continuous trajectories that prevent abrupt or unsafe actions. The framework also includes pre-calculated safe velocity zones for smooth braking, preventing joint limit violations and ensuring compliance with kinematic constraints. This approach not only guarantees the robustness and safety of the robotic system but also optimizes motion control, making it suitable for practical applications. The effectiveness of the proposed framework is demonstrated through its application to a highly complex heavy-duty manipulator.

Keywords: Robotics, Hybrid systems
Abstract: This paper discusses a smoothing method for a signed distance function (SDF) between polytopes that is applicable in direct optimal control methods for contact-rich systems. It is well known that the Euclidean or similar SDF are nondifferentiable with respect to degrees of freedom for rigid bodies that do not have sufficiently smooth surfaces. We consider a growth distance for polytopes modeled as a parametric linear program and give a thorough proof that smoothing this SDF by solving perturbed optimality conditions results in well-defined functions. In previous work, exact optimality conditions of the SDF were embedded directly into an optimal control problem (OCP). In contrast, smoothing the SDF allows for external evaluation and thereby smaller instances of OCP, distributing the computational effort more equally among function evaluations and the load on numerical solvers. We discuss an implementation of the smooth SDF in the CasADi toolbox and compare the performance to the equivalent direct embedding into the OCP. To this end, a planar peg-in-hole task is considered for which an ensemble of trajectories that share the same control input is optimized simultaneously.

Keywords: Robotics, Robust control, Nonlinear system identification
Abstract: Mobile robot navigation can be challenged by system uncertainty. For example, ground friction may vary abruptly causing slipping, and noisy sensor data can lead to inaccurate feedback control. Traditional model-based methods may be limited when considering such variations, making them fragile to varying types of uncertainty. One way to address this is by leveraging learned prediction models by means of the Koopman operator into nonlinear model predictive control (NMPC). This paper describes the formulation of, and provides the solution to, an NMPC problem using a lifted bilinear model that can accurately predict affine input systems with stochastic perturbations. System constraints are defined in the Koopman space, while the optimization problem is solved in the state space to reduce computational complexity. Training data to estimate the Koopman operator for the system are given via randomized control inputs. The output of the developed method enables closed-loop navigation control over environments populated with obstacles. The effectiveness of the proposed method has been tested through numerical simulations using a wheeled robot with additive stochastic velocity perturbations, Gazebo simulations with a realistic digital twin robot, and physical hardware experiments without knowledge of the true dynamics.

Keywords: Robotics, Safety critical systems, Robust adaptive control
Abstract: Robotic grasping requires safe force interaction to prevent a grasped object from being damaged or slipping out of the hand. In this vein, this paper proposes an integrated framework for grasping with formal safety guarantees based on Control Barrier Functions. We first design contact force and force closure constraints, which are enforced by a safety filter to accomplish safe grasping with finger force control. For sensory feedback, we develop a technique to estimate contact point, force, and torque from tactile sensors at each finger. We verify the framework with various safety filters in a numerical simulation under a two-finger grasping scenario. We then experimentally validate the framework by grasping multiple objects, including fragile lab glassware, in a real robotic setup, showing that safe grasping can be successfully achieved in the real world. We evaluate the performance of each safety filter in the context of safety violation and conservatism, and find that disturbance observer-based control barrier functions provide superior performance for safety guarantees with minimum conservatism.

Keywords: Robotics, Sliding mode control
Abstract: In this paper, an estimation scheme based on Adaptive Integral Sliding Mode Unknown Input Observers (AISM-UIOs) is proposed to provide an estimate of the cartesian force applied by the patients on the robot end-effector during rehabilitation exercises. In particular, the robot joints are decoupled leveraging the inverse dynamics control, and the acceleration perturbation caused by the external force is estimated using the observers. Such an estimate is then employed to reconstruct the original force applied by the patient. The proposal is analyzed from the theoretical point of view and assessed both in simulation and experimentally on a Franka Emika Panda robot manipulator.

Keywords: Robotics, Uncertain systems, Optimization
Abstract: Singularities, manifesting as special configuration states, deteriorate robot performance and may even lead to a loss of control over the system. This paper addresses the kinematic singularity concerns in robotic systems with model mismatch and actuator constraints through control barrier functions. We propose a learning-based control strategy to prevent robots entering singularity regions. More precisely, we leverage Gaussian process regression to learn the unknown model mismatch, where the prediction error is restricted by a deterministic bound. Moreover, we offer the criteria for parameter selection to ensure the feasibility of CBFs subject to actuator constraints. The proposed approach is validated by high-fidelity simulations on a 2 degrees-of-freedom planar robot.

Keywords: Robust control, Uncertain systems, Electrical machine control
Abstract: One of the major challenges in the reliable operation of power converters are the harmonics prop- agated along the electromechanical system. Limited knowledge of the load, whose mechanical characteristics possibly change over time, demand a robust approach to optimally operate the converter. In this work, we present a frequency domain control solution to address oscillation propagation along the drive, improve its performance, reduce the chance of triggered faults and shaft damage.

Keywords: Power electronics, Predictive control for nonlinear systems, Constrained control
Abstract: With the increasing amount of energy resources connected to the electric powered grid via power electronics converters, commonly referred to as inverted-based resources (IBR), the inertia and robustness of power systems have been greatly impacted. Grid-forming control arose as a solution to these issues by operating the IBRs as voltage sources. However, limiting the amplitude of the converter current became an issue. In this context, the concept of a constrained control strategy capable of limiting the amplitude of the system variables gained relevance. As a solution, this work proposes a nonlinear model predictive control strategy, that is categorized neither as grid-forming nor grid-following. It is capable of controlling the power flow through the converter to the grid, without a phase-locked-loop, allowing constraint satisfaction even in the condition of unknown grid impedance. The control strategy was tested through simulation of nominal operation and during symmetric and asymmetric faults. The results proved its effectiveness in achieving these objectives even in cases in which the grid behavior cannot be predicted.

Keywords: Adaptive control, Electrical power systems, Optimal control
Abstract: Power electronic converters are becoming the main components of modern power systems due to the increasing integration of renewable energy sources. However, power converters may become unstable when interacting with the complex and time-varying power grid. In this paper, we propose an adaptive data-driven control method to stabilize power converters by using only online input-output data. Our contributions are threefold. First, we reformulate the output-feedback control problem as a state-feedback linear quadratic regulator (LQR) problem with a controllable non-minimal state, which can be constructed from past input-output signals. Second, we propose a data-enabled policy optimization (DeePO) method for this non-minimal realization to achieve efficient output-feedback adaptive control. Third, we use high-fidelity simulations to verify that the output-feedback DeePO can effectively stabilize grid-connected power converters and quickly adapt to the changes in the power grid.

Keywords: Modeling, Machine learning
Abstract: Precisely forecasting the energy consumption of electric vehicles not only alleviates the anxiety associated with driving range but also serves as the foundation for progressive advancements, including optimizing charging strategy and energy utilization. The main challenge lies in the inaccuracy of current methods, whether they are empirical models, physics-based models, or data-driven models. Based on newly constructed and engineered physics-informed features, this paper introduces a machine learning-based prediction framework, employing a synergy of offline global models and vehicle-based online adaptation. This combination aims to elevate accuracy in point predictions and also provide valuable information on prediction uncertainties. The developed framework is trained and extensively tested using data from a fleet of real-world electric vehicles. The leading global model, quantile regression neural network (QRNN), demonstrates an average error of 6.30%. Subsequent online adaptation results in a notable reduction to 5.04%, with both surpassing the performance of existing models significantly. Concurrently, the online QRNN exhibits a strong capability in enhancing the coverage probability and decreasing the average width of prediction intervals.

Keywords: Energy systems, Hybrid systems, Distributed control
Abstract: In this paper, we present a control-oriented benchmark of a network of dynamical systems representing an abstraction of the European Economic Area (EEA) electricity network. In the network each node represents a country of the EEA as an equivalent electrical area with specific generation and load features. The benchmark has been developed to provide the research community with a tool to assess non-centralized control strategies over a standardized case study. The Load Frequency Control (LFC) problem in the presence of renewable energy sources is considered, where the objective is to maintain a nominal operating frequency of the electricity network despite the presence of variations in the load request, and renewable energy production. A hybrid implementation of Energy Storage Systems (ESSs) with different operating modes is considered in the network to support energy generation. We test the features of the system through control simulations with centralized Model Predictive Control (MPC), and a Distributed MPC (DMPC) based on the Alternating Direction Method of Multipliers (ADMM). The benchmark is provided together with a long-term access repository containing both the data, and the scripts to access and process the data.

Keywords: Traffic control, Sliding mode control, Constrained control
Abstract: In this paper, a novel ramp metering strategy with constraints compliance is introduced for a road traffic system with uncertain demand. The uncertain part of the flow entering the on-ramps is estimated relying on Integral Sliding Mode based Unknown Input Observers. The estimation is then employed in a Quadratic Programming (QP) problem, in which constraints on queue lengths at the on-ramps are included via suitable Control Barrier Functions (CBFs). This enhances classical ramp metering strategies by guaranteeing that the constraints on queue lengths are always satisfied. The proposal is theoretically analyzed and assessed in simulation relying on the well-known Cell Transmission Model (CTM) with capacity drop.

Keywords: Traffic control, Optimization, Distributed control
Abstract: Vehicular demand management strategies emerge as a potential solution to traffic congestion. These strategies aim to regulate the inflow of vehicles across specific regions of a vehicular transportation network. This work investigates the stability and optimality of vehicular demand management schemes in regional traffic networks, considering inter-boundary flow constraints and triangular macroscopic fundamental diagram relationships between densities and flows. We first formulate an optimization problem that aims to maximize the total vehicular throughput at steady-state. Due to the triangular macroscopic fundamental diagram relationships and inter-boundary flow constraints, the optimization problem is nonconvex. We tackle this challenge by reformulating the problem as a Mixed Integer Linear Program that can be solved with standard mathematical programming solvers to determine the optimal operating set-points. Nonetheless, it has been demonstrated that operating at maximum throughput set-points, particularly near local critical density points, may lead to instability and gridlock. To address this issue, we propose a decentralized proportional vehicular demand management controller, accompanied by proper local design conditions, such that stability is guaranteed. The effectiveness and practicality of the proposed approach are demonstrated through numerical simulations in a six-region traffic network system.

Keywords: Traffic control, Modeling, Optimization
Abstract: This work presents a novel optimization-based control framework for managing traffic flow in urban networks with mixed autonomy, where Connected and Automated Vehicles (CAVs) and Human-Driven Vehicles (HDVs) coexist. The proposed approach extends the store-and-forward model by incorporating a dynamic saturation flow rate that reflects the level of autonomy in vehicle queues. The problem is formulated as a Non-convex Quadratic Program (NQP), capturing the dynamics of queue lengths, spillback effects, green time allocation, CAV routing, and variable saturation flow rates. To solve the NQP efficiently, we reformulate bilinear terms using under- and over estimators, transforming the non-convex problem into a series of convex subproblems—specifically, a Mixed-Integer Quadratic Program (MIQP)—which is further converted into a Mixed-Integer Linear Program (MILP) by linearizing quadratic terms in the objective function. This approach significantly reduces computational complexity while enabling potential real-time implementation. Numerical simulations on a grid network demonstrate the effectiveness and efficiency of the proposed methodology.

Keywords: Cooperative control, Traffic control, Transportation systems
Abstract: This paper addresses lane-free traffic of connected autonomous vehicles (CAVs) under a mixed-autonomy condition on freeways. It introduces an integrated control method that combines road internal boundary control (IBC) and active lane-changing control (LCC) for CAVs. To cope with the coexistence of CAVs and human-driven vehicles (HVs), a dedicated bi-directional zone is established for CAVs in the center of the considered freeway, which is operated in the lane-free mode in either direction. IBC is considered within the dedicated zone of CAVs. The rest of space in either direction is shared by CAVs and HVs, and an LCC strategy is applied to regulate the lane-changing flow of CAVs between the dedicated zone and mixed area in either direction. A model predictive control (MPC) framework is adopted to coordinate IBC and LCC of CAVs, for the sake of optimizing road space utilization and traffic flow efficiency. The study results demonstrate that the proposed control method is able to achieve the control goal.

Keywords: Optimal control, Traffic control, Transportation systems
Abstract: This paper presents an event-triggered model predictive control (MPC) scheme for diverse automated vehicle entities populating a lane-free traffic environment with vehicle nudging. The vehicle entities comprise, beyond individual vehicles, 1-D snake-like interruptible vehicle platoons and 2-D flexible-shape vehicle flocks. Each entity is driven independently by use of a proper movement strategy for one or multiple automated vehicles deriving from a generic single-vehicle or joint optimal control problem, respectively. A two-dimensional double-integrator model is considered for the longitudinal and lateral movements of each vehicle, considering constant and state-dependent bounds on control inputs, including road boundary constraints. A multi-objective function, comprising various weighted sub-objectives, is designed for all vehicles of each entity, considering minimization of fuel consumption, intra-entity and inter-entity collision avoidance, entity-specific desired speed, prevention of infeasible maneuvers and, for multi-vehicle entities, operation of a flexible platoon or deformable flock. A computationally efficient feasible direction algorithm is called, on an event-triggered basis, to compute in real time the numerical solution of each entity’s optimal control problem for finite time-horizons within an MPC framework. Preliminary, but challenging scenarios are examined on a straight lane-free motorway stretch, producing promising results for further exploration.

Keywords: Delay systems, Traffic control, Automotive
Abstract: We develop a predictor-based cooperative adaptive cruise control (CACC) design for platoons with heterogeneous vehicles, whose dynamics are described by a third-order linear system subject to actuators delays, which are {em distinct} for each individual vehicle. The design achieves individual vehicle stability, string stability, and zero, steady-state speed/spacing tracking errors, relying on a nominal, constant time headway (CTH)-type CACC design that achieves these specifications when all actuators delays are zero. This is achieved owing to the delay-compensating mechanism, of the CACC law introduced, for long delays values and despite the fact that each vehicle’s dynamics are subject to different input delays, which makes the available predictor-feedback CACC designs inapplicable. The proofs of individual vehicle stability, string stability, and regulation rely on employment of an input-output approach on the frequency domain. In addition, we present consistent simulation results that validate the performance of our design in realistic scenarios.

Keywords: Predictive control for nonlinear systems, Autonomous systems
Abstract: In this paper, we establish a connection between model predictive control (MPC) techniques and Control Barrier Functions (CBFs). Recognizing the similarity between CBFs and Control Lyapunov Functions (CLFs), we propose a MPC formulation that ensures invariance and safety without relying on explicit stability conditions. The value function of our proposed MPC is a CBF, which we refer to as the Predictive Control Barrier Function (PCBF), similar to traditional MPC formulations which encode stability by having value functions as CLFs. Our formulation is simpler than previous PCBF approaches and is based on weaker assumptions while proving a similar theorem that guarantees safety recovery. Notably, our MPC formulation does not require the value function to be strictly decreasing to ensure convergence to a safe invariant set. Numerical examples demonstrate the effectiveness of our approach in guaranteeing safety and constructing non-conservative CBFs.

Keywords: Predictive control for nonlinear systems, Distributed control, Large-scale systems
Abstract: A key motivation in the development of Distributed Model Predictive Control (DMPC) is to accelerate centralized Model Predictive Control (MPC) for large-scale systems. DMPC has the prospect of scaling well by parallelizing computations among subsystems. However, communication delays may deteriorate the performance of decentralized optimization, if excessively many iterations are required per control step. Moreover, centralized solvers often exhibit faster asymptotic convergence rates and, by parallelizing costly linear algebra operations, they can also benefit from modern multicore computing architectures. On this canvas, we study the computational performance of cooperative DMPC for linear and nonlinear systems. To this end, we apply a tailored decentralized real-time iteration scheme to frequency control for power systems. DMPC scales well for the considered linear and nonlinear benchmarks, as the iteration number does not depend on the number of subsystems. Comparisons with multi-threaded centralized solvers demonstrate competitive performance of the proposed decentralized optimization algorithms.

Keywords: Predictive control for nonlinear systems, Energy systems, Emerging control applications
Abstract: Wave energy is one of the untapped renewable energy sources, requiring further development of wave energy converters (WECs) to become competitive with wind and solar energy. A significant challenge for WEC development is the high levelized cost of energy (LCoE) associated with traditional heaving or diffraction-based devices. However, analytical and experimental evaluation of lift-based cyclorotor WECs indicate that these devices can achieve superior power absorption when optimised using advanced control techniques, potentially increasing power production by several times, compared to uncontrolled scenarios. This work presents the first implementation of Nonlinear Model Predictive Control (NMPC) for a cyclorotor WEC. The control strategy relies on the separation principle, assuming accurate wave prediction over the control horizon for panchromatic waves. A comparison of various pitch and/or velocity control strategies is conducted for different irregular sea states. The results, obtained by simulations, confirm and exceed the capability, previously predicted by the theoretical optimal control solution, of a cyclorotor WEC to absorb up to 70 % of wave energy.

Keywords: Predictive control for nonlinear systems, Machine learning, Autonomous systems
Abstract: In this paper, we present a learning-based nonlinear model predictive controller (NMPC) using an original reinforcement learning (RL) method to learn the optimal weights of the NMPC scheme, for which two methods are proposed. Firstly, the controller is used as the current action-value function of a deep Expected Sarsa where the subsequent action-value function, usually obtained with a secondary NMPC, is approximated with a neural network (NN). With respect to existing methods, we add to the NN's input the current value of the NMPC's learned parameters so that the network is able to approximate the action-value function and stabilize the learning performance. Additionally, with the use of the NN, the real-time computational burden is approximately halved without affecting the closed-loop performance. Secondly, we combine gradient temporal difference methods with a parametrized NMPC as a function approximator of the Expected Sarsa RL method to overcome the potential parameters' divergence and instability issues when nonlinearities are present in the function approximation. The simulation results show that the proposed approach converges to a locally optimal solution without instability problems.

Keywords: Predictive control for nonlinear systems, Machine learning, Computational methods
Abstract: This paper considers the design of nonlinear data-enabled predictive control (DeePC) using kernel functions. Compared with existing methods that use kernels to parameterize multi-step predictors for nonlinear DeePC, we adopt a novel, operator-based approach. More specifically, we employ a universal product kernel parameterization of nonlinear systems operators as a prediction mechanism for nonlinear DeePC. We show that by using a product reproducing kernel Hilbert space (RKHS) to learn the system trajectories, big data sets can be handled effectively to construct the corresponding product Gram matrix. Moreover, we show that the structure of the adopted product RKHS representation allows for a computationally efficient DeePC formulation. Compared to existing methods, our approach achieves a factor 30 reduction in online computation time for the same data size. This allows for the use of much larger data sets and enhanced control performance.

Keywords: Predictive control for nonlinear systems, Optimal control, Chemical process control
Abstract: This paper presents a nonlinear model predictive control (NMPC) algorithm for systems modeled by semi-explicit stochastic differential-algebraic equations (DAEs) of index 1. The NMPC combines a continuous-discrete extended Kalman filter (CD-EKF) with an optimal control problem (OCP) for setpoint tracking. We discretize the OCP using direct multiple shooting. We apply an explicit singly diagonal implicit Runge- Kutta (ESDIRK) integration scheme to solve systems of DAEs, both for the one-step prediction in the CD-EKF and in each shooting interval of the discretized OCP. The ESDIRK method uses iterated internal numerical differentiation to compute precise integrator sensitivities, which are used to provide accurate gradient and constraint Jacobian information of the OCPs, as well as to efficiently compute the estimation covariance in the CD-EKF. Subsequently, we present a simulation case study where we apply the NMPC to a simple alkaline electrolyzer stack model. We use the NMPC to track a time-varying setpoint for the stack temperature subject to input bound constraints.

Keywords: Emerging control theory, Emerging control applications
Abstract: This event is meant to provide a moment for women researchers in control attending ECC 2025 to share their insights, experiences, and visions for the future of the field.

Speakers:

Zoe Doulgeri (Aristotle University of Thessaloniki, Greece) and Maria Laura Delle Monache (University of California, Berkeley)

with the participation of:

Elena Valcher (University of Padova, EUCA President), Christophe Prieur (Gipsa-lab, CNRS-Univ. Grenoble Alpes, EUCA Vice-President) and Antonella Ferrara (University of Pavia, EUCA CEB Chair)

Keywords: Adaptive control
Abstract: In control engineering, information and sensing technologies are essential components of real-time measurement systems, providing variable resolution quantitative measurements in control systems, such as in electric power grid systems, intelligent transportation systems, vehicle coordination systems, and offshore mechatronics. Typically, these control systems rely on communication channels. An important aspect is to use quantization schemes with sufficient precision while maintaining low communication rates.This talk explores the challenges and advancements in quantized control, focusing on the theoretical foundations of stability analysis and the interplay between quantization and control system design. Specifically, it addresses the adaptive control of dynamic systems in the presence of uncertainties, nonlinear dynamics, and quantization. The discussion highlights how adaptive design and Lyapunov-based analysis can be applied systematically to mitigate, offering a unified and systematic framework for tackling these challenges.

Keywords: Safety critical systems
Abstract: Intelligent, autonomous systems are increasingly prevalent in today's society and will continue to proliferate. Many of these systems are critical, necessitating rigorous safety guarantees. However, designing and verifying safe controllers remains challenging, even for systems with linear dynamics. This talk will focus on the design of safe control systems using a Control Barrier Function (CBF) approach, which ensure forward invariance of safe sets. I will first present a new convex design method for linear systems, enabling the co-design of the controller and the barrier function. I will then extend the analysis to more complex scenarios, including systems with nonlinear dynamics, constraints, high relative degree, and model uncertainty. Finally, I will consider the case of multi-agent systems, proposing a novel distributed control algorithm with parallel computation for enhanced safety. Recognizing that computational limitations may necessitate early termination, I will present a probabilistic result for guaranteeing safety. This work is a collaborative effort with Dr. Han Wang and Professors Kostas Margellos and Claudio de Persis.

Keywords: Identification for control, Filtering, Signal processing
Abstract: This paper introduces a new digital filter design approach for separating signal mixtures into periodic and aperiodic components using a regularized least-squares framework. The framework models the aperiodic component using a geometric polynomial, while the periodic component is represented through trigonometric functions. By incorporating a regularization term within the least-squares formulation, the method effectively preserves signal fidelity and prevents overfitting. The digital filter coefficients are obtained by solving the regularized least-squares problem. Additionally, this approach allows for the calculation of confidence intervals for the extracted signal components, providing a measure for the reliability of the results. A comprehensive analysis of the model parameters is carried out, and the method is validated using both real-world measurement data from an industrial process and synthetic datasets.

Keywords: Identification for control, Robust control, LMI's/BMI's/SOS's
Abstract: We tackle the problem of providing closed-loop stability guarantees with a scalable data-driven design. We combine virtual reference feedback tuning with dissipativity constraints on the controller for closed-loop stability. The constraints are formulated as a set of linear inequalities in the frequency domain. This leads to a convex problem that is scalable with respect to the length of the data and the complexity of the controller. An extension of virtual reference feedback tuning to include disturbance dynamics is also discussed. The proposed data-driven control design is illustrated by a soft gripper impedance control example.

Keywords: Identification for control
Abstract: The selection of an appropriate reference model represents a critical decision in direct data-driven model-reference control. In this framework, we introduce a novel approach to estimate the closed-loop behavior of an unknown system in feedback with a given controller using only open-loop data already available. Such an estimation method is then used to build an automatic tuning scheme for model-reference control techniques. The proposed autotuning strategy allows one to select the reference model that best balances user-defined goals and the inevitable limitations of the selected direct control technique, toward the controller’s refinement. We illustrate the effectiveness of our approach through extensive numerical simulations for automatically tuning a VRFT-based controller.

Keywords: Linear parameter-varying systems, Identification for control, Aerospace
Abstract: This paper presents the application of a recently proposed data-driven Linear Parameter Varying (LPV) controller design method to the Quanser Aero. The employed approach is based on Quadratic Matrix Inequalities (QMIs) and the strict matrix S-lemma. The existing approach is tweaked with the aim of forcing a higher control aggressiveness. Simulation and experimental results involving stabilization and multistep-reference tracking are used to illustrate the validity and performance of the design.

Keywords: Identification for hybrid systems, Energy systems, Observers for nonlinear systems
Abstract: Aquifer thermal energy storages (ATES) represent groundwater saturated aquifers that store thermal energy in the form of heated or cooled groundwater. Combining two ATES, one can harness excess thermal energy from summer (heat) and winter (cold) to support the building's heating, ventilation, and air conditioning (HVAC) technology. In general, a dynamic operation of ATES throughout the year is beneficial to avoid using fossil fuel-based HVAC technology and maximize the ``green use'' of ATES. Model predictive control (MPC) with an appropriate system model may become a crucial control approach for ATES systems. Consequently, the MPC model should reflect spatial temperature profiles around ATES' boreholes to predict extracted groundwater temperatures accurately. However, meaningful predictions require the estimation of the current state of the system, as measurements are usually only at the borehole of the ATES. In control, this is often realized by model-based observers. Still, observing the state of an ATES system is non-trivial, since the model is typically hybrid. We show how to exploit the specific structure of the hybrid ATES model and design an easy-to-solve moving horizon estimator based on a quadratic program.

Keywords: Identification for control, Robust control
Abstract: In this paper, we present a new parametrization to perform direct data-driven analysis and controller synthesis for the error-in-variables case. To achieve this, we employ the Sherman-Morrison-Woodbury formula to transform the problem into a linear fractional transformation with unknown measurement errors and disturbances as uncertainties. For bounded uncertainties, we apply robust control techniques to derive a guaranteed upper bound on the H2-norm of the unknown true system. To this end, a single semidefinite program needs to be solved, with complexity that is independent of the amount of data. Furthermore, we exploit the signal-to-noise ratio to provide a data-dependent condition, that characterizes whether the proposed parametrization can be employed. The modular formulation allows to extend this framework to controller synthesis with different performance criteria, input-output settings, and various system properties. Finally, we validate the proposed approach through a numerical example.

Keywords: Reduced order modeling, Computational methods, Large-scale systems
Abstract: In this paper, a method that constructs bilinear systems from input-output data in the time-domain is presented. A nuclear norm heuristic is used to tackle the case of noisy data and to build a linear model. This mitigates the challenge of calculating the Markov parameters through noisy data for linear identification. Then, if needed, the order of the linear model can be reduced by using model reduction techniques. Finally, a least squares problem and a nonlinear optimization step are solved to add the nonlinear (bilinear) term and enrich the structure of the learned system. The applicability of the method is demonstrated on an industrial dataset of a heat exchanger experiment.

Keywords: Reduced order modeling, Identification for control, Large-scale systems
Abstract: We present an extension of optimal mode decomposition (OMD) for autonomous systems to systems with controls. The extension is developed along the same lines as the extension of dynamic mode decomposition (DMD) to DMD with control (DMDc).

DMD identifies a linear dynamic system from high-dimensional snapshot data. DMD is often combined with a subsequent reduction by a projection to a truncated basis for the space spanned by the snapshots. In OMD, the identification and reduction are essentially integrated into a single optimization step, thus avoiding the somewhat adhoc decoupled, a posteriori reduction that is necessary if DMD is to be used for model reduction. DMD was devised for autonomous systems and later extended to DMD for systems with control inputs (DMDc). We present the analogous extension of OMD to OMDc, i.e. OMD for systems with control inputs. We illustrate the proposed method with an application to coupled diffusion-equations that model the drying of a wood chip. Reduced models of this type are required for the efficient simulation of industrial drying processes.

Keywords: Model/Controller reduction, H2/H-infinity methods, Optimization
Abstract: This article addresses the problem of mathcal{H}_{infty} model order reduction for continuous-time linear time-invariant (LTI) systems over finite-frequency intervals. The goal is to compute a reduced-order model such that the error characterised by mathcal{H}_{infty}-norm between the original system and reduced-order model is minimised within a specified frequency interval. Utilising the generalised Kalman-Yakubovich-Popov lemma, the problem is formulated as an optimisation problem constrained by nonlinear matrix inequalities. By employing a novel application of the projection lemma, the problem is reformulated into an optimisation problem with linear matrix inequality constraints. An iterative algorithm is then given to compute the solution of the finite-frequency model order reduction problem. In case studies, the proposed approach is shown to outperform existing results in the literature.

Keywords: Model/Controller reduction, Reduced order modeling, Linear systems
Abstract: Input redundant systems have more inputs than needed to control the output and offer advantages such as fault tolerance and improved performance. However, identifying these redundant inputs in high-dimensional systems can be computationally intensive. This article proposes an approach that uses model reduction by balanced truncation to simplify input redundancy analysis. By reducing the system order, we identify “invisible inputs” for the reduced-order model and design a reduced order state feedback mechanism to decouple them from the “visible inputs.” Then, error bounds from balanced truncation provide guarantees for applying this decoupling to the original system, as a feedforward input. The approach offers significant computational savings, as demonstrated in numerical examples.

Keywords: Model/Controller reduction, Neural networks
Abstract: This paper studies a multi-task control problem where multiple linear systems are to be regulated by a single non-linear controller. In particular, motivated by recent advances in multi-task learning and the design of brain-inspired architectures, we consider a neural controller with (smooth) ReLU activation function. The parameters of the controller are a connectivity matrix and a bias vector: although both parameters can be designed, the connectivity matrix is constant while the bias vector can be varied and is used to adapt the controller across different control tasks. The bias vector determines the equilibrium of the neural controller and, consequently, of its linearized dynamics. Our multi-task control strategy consists of designing the connectivity matrix and a set of bias vectors in a way that the linearized dynamics of the neural controller for the different bias vectors provide a good approximation of a set of desired controllers. We show that, by properly choosing the bias vector, the linearized dynamics of the neural controller can replicate the dynamics of any single, linear controller. Further, we design gradient-based algorithms to train the parameters of the neural controller, and we provide upper and lower bounds for the performance of our neural controller. Finally, we validate our results using different numerical examples.

Keywords: Robotics, Autonomous robots, Reduced order modeling
Abstract: This paper presents a framework for the planning and control of wheeled mobile manipulators, integrating a novel reduced-order dynamics model that can be used during motion planning to enhance safety and prevent tip-overs. Leveraging centroidal momentum dynamics, the model captures key forces at the center of mass, enabling efficient, dynamically feasible trajectory generation that considers both manipulator motion and ground interaction forces. Unlike traditional methods that rely on conservative planning or separate reactive tip-over prevention mechanisms, our approach incorporates stability considerations directly into the motion planning phase. By embedding the zero moment point criterion within the model, our framework ensures tip-over prevention even during fast and payload-intensive tasks. Simulations demonstrate the effectiveness of this approach, achieving stable and efficient task execution across various scenarios.

Keywords: Nonlinear system theory, Output regulation, LMI's/BMI's/SOS's
Abstract: This work addresses data-driven output regulation of nonlinear systems with unknown system parameters. Using incremental passivity, output regulation is achieved by interconnecting an incrementally passive internal model and an incrementally passive closed-loop system of the plant. The controller that makes the closed-loop system incrementally passive is characterized by linear matrix inequalities using finite offline data. A numerical example verifies the proposed data-driven regulator design.

Keywords: Behavioural systems, Emerging control theory, Nonlinear system theory
Abstract: This paper considers the receding horizon control design of dynamical systems using their trajectories, which would contribute towards the emerging big data-centric control paradigm to utilize the ever-increasingly rich process data available in modern industries. The development is in the behavioral framework, in which dynamical systems are defined by their behaviors, i.e., the set of all admissible trajectories of the systems, and interconnections are interpreted as constraints on the choice of trajectories. This work takes a set-theoretic viewpoint, and the results are completely representation-free. The interconnection, control performance specification, and controller constraints are treated as systems. Such a viewpoint leads to the development of necessary and sufficient conditions for the existence of implementable controlled behavior along with the corresponding controller behavior. These results provide a generic framework for trajectory-based receding horizon control design of general systems.

Keywords: Nonlinear system theory, Network analysis and control, Energy systems
Abstract: Motivated by theories and evidence from the social psychology literature, we propose a novel continuous-time mathematical model that captures the evolution of motivation and behavior of energy consumers in a social network. In our model, consumers are connected to the energy grid and their energy demand (which we shall refer to as their behavior) is affected by their personal motivation on reducing (or increasing) their energy consumption. The motivation-behavior dynamics of each consumer is modeled using a second-order continuous-time bilinear differential equation. Each consumer has the ability to share their motivation and observe behavior of other consumers on a social network. Using the information gathered from their peers and their personal bias about the energy consumption behavior, consumers have the ability to revise their own motivation about energy consumption and, ultimately, their behavior. Moreover, we incorporate into the model an external control action that captures the implementation of external behavioral interventions that influence the weight each consumer assigns to their own bias. Then, we use the propose framework to shed light onto the collective motivation-behavior dynamics of all the consumers, establishing conditions for existence of equilibrium points, characterizing them, and performing a sensitivity analysis of such equilibria with respect to variations in the steady-state interventions provided to each consumer.

Keywords: Nonlinear system identification, Hybrid systems, Power electronics
Abstract: Digital twins for power electronics require accurate power losses whose direct measurements are often impractical or impossible in real-world applications. This paper presents a novel hybrid framework that combines physics-based thermal modeling with data-driven techniques to identify and correct power losses accurately using only temperature measurements. Our approach leverages a cascaded architecture where a neural network learns to correct the outputs of a nominal power loss model by backpropagating through a reduced-order thermal model. We explore two neural architectures, a bootstrapped feedforward network, and a recurrent neural network, demonstrating that the bootstrapped feedforward approach achieves superior performance while maintaining computational efficiency for real-time applications. Between the interconnection, we included normalization strategies and physics-guided training loss functions to preserve stability and ensure physical consistency. Experimental results show that our hybrid model reduces both temperature estimation errors (from 7.2±6.8°C to 0.3±0.3°C) and power loss prediction errors (from 5.4±6.6W to 0.2±0.3W) compared to traditional physics-based approaches, even in the presence of thermal model uncertainties. This methodology allows us to accurately estimate power losses without direct measurements, making it particularly helpful for real-time industrial applications where sensor placement is hindered by cost and physical limitations.

Keywords: Nonlinear system identification, Predictive control for nonlinear systems, Biological systems
Abstract: This work investigates the benefits of implementing a systematic approach to social isolation policies during epidemics. We develop a mixed integer data-driven model predictive control (MPC) scheme based on an SIHRD model which is identified from available data. The case of the spread of the SARS-CoV-2 virus (also known as COVID-19) in Mauritius is used as a reference point with data obtained during the period December 2021 to May 2022. The isolation scheme is designed with the control decision variable taking a finite set of values corresponding to the desired level of isolation. The control input is further restricted to shifting between levels only after a minimum amount of time. The simulation results validate our design, showing that the need for hospitalisation remains within the capacity of the health centres, with the number of deaths considerably reduced by raising the level of isolation for short periods of time with negligible social and economic impact. We also show that the introduction of additional isolation levels results in a smoother containment approach with a considerably reduced hospitalisation burden.

Keywords: Nonlinear system theory, Stability of nonlinear systems, Lyapunov methods
Abstract: A class of homogeneous control laws is introduced for the finite-time or nearly fixed-time stabilization of linear MIMO systems. This analysis utilizes an explicitly defined Lyapunov function, which distinguishes it from similar studies. The approach provides a parameter tuning procedure formalized in the form of linear matrix inequalities. Robust properties of the proposed control is provided as well. The theoretical findings are validated through numerical examples.

Keywords: Game theoretical methods, Optimization algorithms, Variational methods
Abstract: We design a computational approach to find equilibria in a class of Nash games possessing a hierarchical structure. By using tools from mixed-integer optimization and the characterization of variational equilibria in terms of the Karush--Kuhn--Tucker conditions, we propose a mixed-integer game formulation for solving this challenging class of problems. Besides providing an equivalent reformulation, we design a proximal Gauss--Seidel method with global convergence guarantees in case the game enjoys a potential structure. We finally corroborate the numerical performance of the algorithm on a novel instance of the ride-hail market problem.

Keywords: Intelligent systems, Game theoretical methods, Robust control
Abstract: In this paper, a learning-based robust evasion method is proposed to solve the problem of evading strategies from a scalable number of pursuers. The method aims to provide more escape possibilities while overcoming the issue of sparse reward and local optima in unknown environments. The approximation of Q-functions via Gaussian process allows for an accurate online update of parameters and significantly enhances the training efficiency when dealing with high-dimensional data. Simulation and experimental results on several escape tasks for robots demonstrate the effectiveness and robustness of the method. The evading strategies can be scaled to pursuit-evasion problems involving multiple pursuers and evaders, thus providing a framework for multiplayer pursuit-evasion games.

Keywords: Mechatronics, Adaptive control, Optimization algorithms
Abstract: Electromechanical switching devices such as relays, solenoid valves, and contactors offer several technical and economic advantages that make them widely used in industry. However, uncontrolled operations result in undesirable impact-related phenomena at the end of the stroke. As a solution, different soft-landing controls have been proposed. Among them, feedforward control with iterative techniques that adapt its parameters is a solution when real-time feedback is not available. However, these techniques typically require a large number of operations to converge or are computationally intensive, which limits a real implementation. In this paper, we present a new algorithm for the iterative adaptation that is able to eventually adapt the search coordinate system and to reduce the search dimensional size in order to accelerate convergence. Moreover, it automatically toggles between a derivative-free and a gradient-based method to balance exploration and exploitation. To demonstrate the high potential of the proposal, each novel part of the algorithm is compared with a state-of-the-art approach via simulation.

Keywords: Optimization, Robust control, Automotive
Abstract: The increasing penetration of electric vehicles (EVs) in the mobility market and their significant impact on the power grids calls for new and ``smart'' approaches to the management of the charging process for these vehicles, aimed at maximizing the efficiency while respecting power budgets and minimizing costs. In this paper, we propose a robust model for optimal scheduling the charge power at sockets for a large EVs parking facility. We start by developing a nominal linear programming (LP) model for the smart charging problem and then observe that, in practice, key quantities such as electricity prices and the vehicles' energy demand are subject to uncertainty. We hence formulate a robust LP version of the problem, which provides charging plans that are resilient to uncertainties. The effectiveness of such model is analyzed by means of a-posteriori evaluations, where we test the candidate plan against scenarios and realizations of the uncertain data, using performance metrics such as the {em regret} that allow for a fair comparison between different solutions (e.g., robust and nominal). The robust optimization model indeed handles uncertainties without drastically compromising performance, and offers a promising approach when deployed in real-time by means of a receding-horizon scheme.

Keywords: Optimization algorithms, Manufacturing processes, Mechatronics
Abstract: The combination of highly dynamic systems with a limited work envelope, with a less dynamic system which has a larger working envelope, promises to combine the advantages of both systems while eliminating the disadvantages. Due to the created kinematic redundancy it is not trivial to generate trajectories that make use of both systems, while considering limits and allowing for parallel motion. There are several approaches to creating trajectories for such systems. One of which is to formulate an optimisation problem that directly considers the stroke and dynamic limits of each axis to plan trajectories. However, the dimension of the resulting optimisation problems can quickly make this approach infeasible, as they result in long computation times. The objective of this contribution is to develop an optimisation based approach for trajectory planning of redundant systems, which scales well with the problem size. Furthermore, the approach aims at avoiding tunable parameters which can not be physically linked to the machine or the process. The primary method of achieving that objective is the incorporation of b-spline functions into the optimisation problem. These represent the position, velocity, acceleration and jerk of each axis over time. By only having to solve the optimisation problem for the b-spline coefficients, the dimension of the problem is reduced without sacrificing temporal resolution. Constraints are formulated to ensure that axis bounds are respected and an objective function is designed to minimise the jerk of the base axes. In the presented non-redundant test case, the optimisation approach generated valid trajectories comparable to those generated by the TwinCAT CNC kernel, confirming the plausibility of the approach. For the redundant system, the algorithm effectively resolved the kinematic redundancy by generating synchronised trajectories that respected the dynamic constraints of both systems.

Keywords: Quantum information and control, Predictive control for nonlinear systems, Optimization
Abstract: We propose a quantum computing approach for the solution of NMPC optimization problems. Assuming the availability of an efficient quantum computer, the approach can considerably decrease the computational time and/or enhance the solution quality compared to classical algorithms. The approach is illustrated by means of a simulation example, concerned with automated driving. The results show that the approach can substantially reduce the computational cost of optimization-based methods.

Keywords: Aerospace, Stochastic filtering, Emerging control applications
Abstract: Magnetic anomaly navigation (MagNav) is a navi- gation method using the magnetic signature of the Earth’s crust as a position reference. It has recently gained interest within the navigation community. The signal’s global availability at any time without dependence on external transmitters makes it attractive for the development of more resilient airborne navigation systems. Increasing risk of global navigation satellite systems (GNSS) unavailability combined with technological ad- vances in high-performing quantum magnetometers currently increases the interest in this method. Here we present a novel magnetic anomaly navigation algorithm based on an error-state Kalman filter on the SE2(3) Lie Group of extended poses. In state-of-the-art algorithms for aviation, the growing error of a position provided by an inertial system is corrected with a feedforward filter. In contrast, our algorithm is based on feedback corrections of the inertial system. We demonstrate the performance of our navigation algorithm on simulated data and show that it can successfully compensate the posi- tion drift of both tactical and navigation grade IMUs using magnetic anomaly maps with sufficient spatial variability. Our work paves the way towards more robust implementations of alternative aircraft navigation methods in case of GNSS-denial.

Keywords: Aerospace, Robotics, Robust control
Abstract: Intercepting dynamic objects in uncertain environments involves a significant unresolved challenge in modern robotic systems. Current control approaches rely solely on estimated information, and results lack guarantees of robustness and feasibility. In this work, we introduce a novel method to tackle the interception of targets whose motion is affected by known and bounded uncertainty. Our approach introduces new techniques of reachability analysis for rigid bodies, leveraged to guarantee feasibility of interception under uncertain conditions. We then propose a Reachability-Guaranteed Optimal Control Problem, ensuring robustness and guaranteed reachability to a target set of configurations. We demonstrate the methodology in the case study of an interception maneuver of a tumbling target in space.

Keywords: Aerospace, Autonomous robots, Predictive control for nonlinear systems
Abstract: This paper introduces a novel quasi Linear Pa- rameter Varying (qLPV) model to describe the quadrotor dynamics and constraints, overcoming the limitations of ex- isting LPV models that rely on linearization, approximation, and mainly limited to attitude dynamics. Alternatively, the proposed LPV model captures both position and attitude dynamics within one controller without involving cascaded structures. In addition, the nonlinear constraints associated with unit quaternion representation are reformulated as Q- LPV constraints, enabling their integration into the control and estimation processes. The efficiency of the model for control and estimation is demonstrated through the application of Model Predictive Control (MPC) and Moving Horizon Estimation (MHE), achieving comparable tracking accuracy to nonlinear MPC (NMPC) while reducing execution time by a factor of nine. This contribution offers a computationally efficient and real-time solution for quadrotor control and estimation, filling an important gap in the existing literature.

Keywords: Aerospace, Optimal control
Abstract: The dynamical parameter continuation method is considered. It is implemented for the planar non-linear guidance problem with the stationary target. In this approach, the system dynamics is parameterized in such a way that if the parameter is equal to zero, the dynamics is the quadratic-kinematic approximation of the original system, whereas if the parameter is equal to one, it is the original one. We propose an improved formalization of the method that allows convenient handling of the boundary conditions, and derive a numerical procedure. It is shown that the proposed method converges from the wider set of initial and terminal conditions than the classical parameter continuation method.

Keywords: Aerospace, UAV's, Fault detection and identification
Abstract: The Global Positioning System (GPS) spoofing attack detection problem for advanced air mobility is addressed using the model-based nullspace-based residual filter design paradigm. The nullspace-based approach enables numerically stable filter designs and results in residual filters of minimal order. This ensures optimal utilization of the given system structure, minimizing model dependence and thereby enhancing robustness against model uncertainties. The general problem of undetectable constant GPS spoofing attacks in typical aerial system setups is overcome by integrating a data-driven technique that explicitly utilizes topographic data of the underlying urban environment. The approach relies on the idea that altitude measured by a radar altimeter in an urban environment together with the topographic data at the current GPS position must correlate with the altitude provided by the static pressure system. A lack of correlation is an indicator of GPS spoofing attacks. The proposed technique is verified using a modern urban air mobility simulation model.

Keywords: Aerospace, UAV's, Sensor and signal fusion
Abstract: A novel approach for in-flight wind estimation on small unmanned aerial systems (UAS) using the nullspace-based estimator design technique is presented. With this mode-based design technique, three individual wind estimators are derived, one for each wind velocity component. The nullspace-based approach allows synthesizing estimators such that their estimates are decoupled from each other, from any control inputs, and any modeled external disturbances. The resulting estimators have minimal dynamical system order, which is one of the major advantages of the proposed approach. Due to this minimal order, the dependence on the underlying design model is decreased, thereby increasing robustness against model uncertainties. The applicability of the method is demonstrated on a wind estimator design for TU Dresden's research UAS, the Urban Condor, in two different sensor configurations. The first configuration makes use of the full sensor capabilities of Urban Condor, including GPS, an inertial measurement unit (IMU), and an air data system. The second configuration considers the more realistic scenario for commercial UAS by only using GPS and the IMU. The developed estimators are verified in a high-fidelity nonlinear simulation environment.

Keywords: Modeling, Process control, Large-scale systems
Abstract: Machine learning holds a significant potential for pattern recognition in large datasets. However, access to the necessary data is not always possible, particularly for physical systems. To address this, physics-informed neural networks (PINNs) have emerged as an innovative approach, incorporating physical principles into neural networks to enhance the model quality and data efficiency. Despite this, the performance of PINNs can be challenged by the complexity of the underlying physics. This paper introduces a novel framework aimed at improving both the accuracy and training efficiency of PINNs, particularly for large or complex systems. The proposed method outperforms traditional approaches, as demonstrated through various simulations presented in this paper.

Keywords: Modeling, Process control
Abstract: We present a dynamic model to characterise the superheating phase in electric arc steelmaking, aiming to capture the key dynamics through a first-principles approach. This approach ensures that the model parameters are physically meaningful, while maintaining computational efficiency suitable for future implementation in model predictive control. The paper details the process being modeled, elaborates on the underlying assumptions and their physical justifications, and validates the model through comparison with experimental data from a full-scale production plant. The obtained result indicate that the presented model is suitable for online use in control applications, a topic that will be further explored in the sequel of the research.

Keywords: Process control, Chemical process control, Optimization
Abstract: Achieving the Paris Agreement's goal necessitates not only reducing carbon dioxide emissions to net zero but also actively removing CO_2 from the atmosphere. Direct Air Capture (DAC) emerges as a pivotal technology in this effort, offering a reliable, flexible, and scalable solution for negative emissions. However, DAC performance is highly sensitive to environmental factors such as temperature and humidity. Consequently, it is vital to develop dynamic control and optimization mechanisms that can enhance the cost-efficiency of DAC. Due to the complexity and lack of a comprehensive model for DAC systems, the need for expert knowledge for modeling, and high computational costs, traditional model-based methods are not feasible. Therefore, we suggest a model-free, data-driven optimization technique based on Bayesian optimization to enhance the productivity and cost-effectiveness of DAC.

Keywords: Process control, Manufacturing processes, Identification
Abstract: This paper addresses the design of temperature controllers for shrink tunnels, which are machines composed of an industrial oven and a conveyor belt employed for packaging purposes. Specifically, the present work focuses on the rejection of disturbances arising from two sources: (i) the flow of products inserted into the oven by the conveyor belt, which causes abrupt temperature drops, and (ii) the varying grid voltage, which perturbs the heat generated by the heat resistors commanded by the regulator. To that end, a recent control architecture is extended with the addition of (i) feedforward compensators and (ii) a suitable rescaling strategy for the control actions, effectively addressing the previously mentioned disturbances. The performances of the proposed strategy are validated on an experimental workbench, showing improved disturbance rejection capabilities compared to the baseline approach.

Keywords: Machine learning, Control of metal processing, Manufacturing processes
Abstract: Additive Manufacturing (AM) technology is one of the most promising processes for the production of complex shape parts. A number of issues still remain unsolved, among which the fundamental one is the definition of a model able to predict the mechanical behavior of additively manufactured components starting from the knowledge of the alloy microstructure and the process-induced defects. Ongoing research aims at optimizing the use of AM technologies in several industrial fields, like medical, aerospace and mechanical. This goal requires in-depth characterization of the alloy microstructure and of the metallurgical defects in terms of morphology and distribution, the determination of mechanical properties of AM-produced specimens (like fatigue of materials, fatigue damage, fracture toughness,...), and the development of a predictive model based on artificial intelligence (AI) algorithms. The first step requires the ability of classifying images obtained from specimens produced with different process parameters and, consequently, presenting various mechanical properties. In this paper, this goal is pursued by means of a totally automatized procedure based on advanced methods of machine learning (ML); the first results, obtained on real specimens fabricated using Electron Beam Powder Bed Fusion (EB-PBF), are promising, showing the classifier ability of obtaining satisfactory results after a training on limited number of images.

Keywords: Modeling, Differential algebraic systems, Chemical process control
Abstract: We present a 2D model for a grate belt cooler in the pyro-section of a cement plant. The model is formulated as an index-1 differential-algebraic equation (DAE) model based on first engineering principles. The model systematically integrates thermo-physical aspects, transport phenomena, reaction kinetics, mass and energy balances, and algebraic volume and energy relations. The model is used for dynamic simulation of the cooler and the paper provides dynamic and steady-state simulation results matching the expected behavior. The cooler model is one part of a full pyro-section model for dynamical simulations. The model can serve as a basis for the design of optimization and control systems towards improving energy efficiency and CO2 emission.

Keywords: Robotics
Abstract: This paper addresses the time-optimal path-tracking problem for redundant manipulators. By integrating path-velocity control into existing task-space robot controllers, the task-space motion can be dynamically scaled to satisfy the torque constraint under both kinematic and dynamic uncertainties. Numerical simulations and experiments demonstrate that trajectory feasibility and path-tracking accuracy of the task-space controllers can be significantly improved by integrating path-velocity control. In addition, the nullspace motion of redundant manipulators can be exploited to further improve the performance by tracking the approximate time-optimal joint trajectory associated with the tasks in nullspace.

Keywords: Robotics
Abstract: In this paper a trajectory tracking controller is designed for planar robotic cutting tasks. Based on a global model for the knife edge motion control problem the designed control input is shown to achieve convergence of the direction vector to the position error direction in finite time which then ensures the convergence of the position and direction tracking errors to zero as shown using Lyapunov arguments. The controller drives the system on the desired path reducing the distance traveled outside of it. The proposed approach is validated through simulations and compared to an existing method in the literature. Experimental results in a robotic cutting task further validate and demonstrate its performance.

Keywords: Robotics, UAV's
Abstract: This article presents a novel edge-based architecture to dynamically allocate resources to edge-offloaded controllers for multi-agent systems. The proposed controllers are designed to generate collision-free trajectories to track the desired reference positions. The computational complexity of the controllers' problem is estimated by a second-order polynomial regression model, while the Least Squares (LS) minimization technique is employed for the coefficients' estimation. The covariance matrix plays an essential role in assessing the confidence in the parameter estimates and in investigating correlations among parameters. Through this curve fitting process, we can dynamically estimate the complexity of the controllers' problem as conditions change, enabling effective and responsive resource allocation. Furthermore, a novel control law is designed to control the dynamic resource allocation, based on the measured communication and processing time delays. This approach allows us to control the controllers' response time, thereby ensuring the closed-loop system's stability. The overall architecture is enabled through a Kubernetes (k8s) cluster and is experimentally evaluated.

Keywords: Robotics, Sliding mode control
Abstract: In this work, a two-loop impedance control using the sliding mode control technique is presented. In the outer loop, a sliding mode control algorithm is proposed to impose the desired interaction behaviour on the system. Then, in the inner loop, another sliding mode control is proposed to control the motors of the elastic joints. Then, the presented sliding mode impedance controller is simulated to verify its performance.

Keywords: Robotics, Switched systems, System reconfiguration
Abstract: A standard motion control with feedback of the output displacement cannot handle unforeseen contact with environment without penetrating into the soft, i.e. viscoelastic, materials or even damaging the fragile materials. Robotics and mechatronics with tactile and haptic capabilities, and in particular medical robotics for example, place special demands on the advanced motion control systems that should enable the safe and harmless contact transitions. This paper shows how the basic principles of loop shaping can be easily used to handle sufficiently stiff motion control in such a way that it is extended by sensor-free dynamic reconfiguration upon contact with the environment. A thereupon based hybrid control scheme is proposed. A remarkable feature of the developed approach is that no measurement of the contact force is required and the input signal and the measured output displacement are the only quantities used for design and operation. Experiments on 1-DOF actuator are shown, where the moving tool comes into contact with grapes that are soft and simultaneously penetrable.

Keywords: Robotics, Modeling, Optimization
Abstract: Inverse dynamics models are crucial in robotics control. Traditional models, built on physical principles, often require precise system parameters, which can be difficult to obtain. Black-box models represent a valid alternative, however they usually lack physical plausibility, which prevents their usage in combination with standard control approaches. This paper focuses on black-box inverse dynamics identification with Gaussian Processes Regression (GPR), with a focus on ensuring physical consistency, in particular the positive definiteness of the inertia matrix. Our approach is based on the integration of positivity constraints into the standard empirical risk minimization problem. First, we propose an estimation method to infer the inertia matrix of a black-box model. Second, we introduce a constrained optimization strategy to modify the standard GPR solution by promoting inertia positiveness. Experimental validation on a 7 degrees-of-freedom robot highlights the benefits of the proposed method. Results show a minimal loss in estimation accuracy, but large benefits in terms of physical consistency, especially compared to unconstrained models that may yield non-physical behaviors.

Keywords: Traffic control, Modeling, Transportation systems
Abstract: As more and more Connected and Automated Vehicles enter public roads, Lagrangian traffic control strategies that leverage their presence to improve traffic flow are being proposed. As a result, the disparity in driving behaviours between different classes of vehicles is likely to continue increasing. Although numerous multi-class traffic models have been proposed, few adequately capture the complex interactions between vehicles with significantly different driving behaviour. In this work we study the simplest case of two-class traffic, with vehicle classes differentiated only by their desired speeds. The dynamics of this model are analysed in detail through solving the state discontinuity problems, where class traffic densities change, and interface discontinuity problems, where the reference speed of one class changes. The resulting traffic density profiles are compared with those arising from a corresponding microscopic model, incorporating both the car-following and lane-changing behaviour of vehicles of different classes. We show that the studied macroscopic model is congruent with an appropriately calibrated microscopic model.

Keywords: Traffic control, Modeling
Abstract: In this paper, using a continuation method cite{NikitinPaper}, we provide a simple procedure to derive macroscopic models from second order ordinary differential equations modeling individual vehicles moving on a ring. In particular, the result is applied to classical traffic models such as the optimal velocity follow the leader (OV-FTL) models, the intelligent driver model (IDM) or delayed driver models and we are able to modelize the macroscopic behavior of the drivers through a 2times 2 partial differential equations whose variables are the density of the traffic and the speed of the vehicles. The paper concludes with open questions regarding stability and microscopic control design.

Keywords: Transportation systems, Optimization
Abstract: The rapid deployment of electric and automated buses in cities holds great advantages for both environmental impact and safety reasons. However, the transition to these efficient propulsion systems requires a rethinking of the common control policies adopted to deal with non electric buses in urban environments. Indeed, in addition to the need of controlling the speed of the bus to meet the timetable, decisions regarding battery charging must be taken. The need to pursue multiple objectives, such as meeting the time schedule and preserving a certain level of energy in the bus battery, typically leads to the formalization and solution of multi-objective control problems such as the one proposed in this paper. However, solving a multi-objective problem can be challenging or impractical in the real-world implementations. This is why the aim of this paper is to identify explicit solutions that can be applied when specific operating conditions occur, thus alleviating the runtime computational load. A control algorithm that integrates the application of the explicit solutions with the solutions obtained by solving the multi-objective control problem is then proposed. The paper concludes with an application to a real case study to demonstrate the effectiveness of the proposed control scheme.

Keywords: Cooperative autonomous systems, Transportation systems, Modeling
Abstract: Ride-pooling services, such as UberPool and Lyft Shared Saver, enable a single vehicle to serve multiple customers within one shared trip. Efficient path-planning algorithms are crucial for improving the performance of such systems. For partially occupied vehicles with available capacity, we introduce a novel routing algorithm designed to maximize the likelihood of picking up additional passengers while serving the current passengers to their destination. Unlike traditional methods that group passengers and vehicles based on predefined time windows, our algorithm allows for immediate responses to passenger requests. Our approach optimizes travel time while dynamically considering passenger demand and coordinating with other vehicles. Formulated as an integer linear programming (ILP) problem, our method is computationally efficient and suitable for real-time applications. Simulation results demonstrate that our proposed method can significantly enhance service quality.

Keywords: Stability of nonlinear systems, Traffic control, Emerging control applications
Abstract: Building on previous theoretical findings, we conduct a numerical study on the boundary stabilization of Generic Second Order Macroscopic traffic models in Lagrangian coordinates. These consist in 2x2 nonlinear hyperbolic systems of balance equations with a relaxation-type source term, which is not required to satisfy the "sub-characteristic" stability condition. In the "super-characteristic" regime, equilibria are unstable and small disturbances may lead to the formation of large oscillations, modeling the emergence and persistence of stop-and-go waves. In this setting, we test various boundary control strategies at the right boundary, aiming to stabilize the system while respecting the physical and safety constraints associated with controlling a single vehicle's speed to stabilize traffic flow on a ring road.

Keywords: Optimization, Traffic control, Transportation systems
Abstract: This paper analyzes the differences between Nash and Stackelberg equilibria of Autonomous Mobility-on-Demand (AMoD) systems in mixed traffic conditions, whereby self- driving robotaxis provide on-demand mobility, possibly pooling users together, while sharing the road with selfish private cars. In particular, we first introduce the optimal fleet routing problem in mixed traffic conditions, considering a car-road network where also private, selfish vehicles are present. Second, we model the interactions between the centrally controlled AMoD fleet and the private cars with two equilibrium formulations: the first corresponds to a Nash equilibrium, where each agent (the fleet and the private users) reacts to the other agent’s action until convergence is reached. For the second approach, corresponding to a Stackelberg equilibrium, the leader (the fleet) can predict the best response of the follower (private users) and plan the strategy accordingly. The results of a case study of Sioux Falls, USA, indicate that the two equilibria are very similar in terms of the fleet’s objective function, suggesting that even if the operator can predict the best response of the private users, no benefit arises.

Keywords: Predictive control for nonlinear systems, Optimal control, Machine learning
Abstract: Model predictive control (MPC) is a powerful control strategy capable of handling constrained nonlinear systems. However, the need for the repeated online solution of optimization problems can be a significant challenge for its real-time application. To address this issue we employ neural network approximations of the MPC control law, which can be swiftly evaluated. Typical supervised training strategies require the solution of large amounts of optimization problems for the generation of training data. To avoid this requirement, we propose an unsupervised closed-loop primal-dual learning strategy, which ensures training data from relevant areas of the state-space without sampling of optimizer solutions. Furthermore, primal-dual learning reduces the importance of manually tuned penalty parameters by using learned Lagrange multipliers. The performance of the proposed strategy is investigated in the context of crosswind kite control, showing comparable performance to nominal MPC while significantly reducing the computational complexity.

Keywords: Predictive control for nonlinear systems, Optimal control, Uncertain systems
Abstract: This paper proposes a novel approach to formulate time-optimal point-to-point motion planning and control under uncertainty. The approach defines a robustified two-stage Optimal Control Problem (OCP), in which stage 1, with a fixed time grid, is seamlessly stitched with stage 2, which features a variable time grid. Stage 1 optimizes not only the nominal trajectory, but also feedback gains and corresponding state covariances, which robustify constraints in both stages. The outcome is a minimized uncertainty in stage 1 and a minimized total motion time for stage 2, both contributing to the time optimality and safety of the total motion. A timely replanning strategy is employed to handle changes in constraints and maintain feasibility, while a tailored iterative algorithm is proposed for efficient, real-time OCP execution.

Keywords: Predictive control for nonlinear systems, Optimization algorithms, Optimization
Abstract: Model mismatch and process noise are two frequently occurring phenomena that can drastically affect the performance of model predictive control (MPC) in practical applications. We propose a principled way to tune the cost function and the constraints of linear MPC schemes to improve the closed-loop performance and robust constraint satisfaction on uncertain nonlinear dynamics with additive noise. The tuning is performed using a novel MPC tuning algorithm based on backpropagation developed in our earlier work. Using the scenario approach, we provide probabilistic bounds on the likelihood of closed-loop constraint violation over a finite horizon. We showcase the effectiveness of the proposed method on linear and nonlinear simulation examples.

Keywords: Predictive control for nonlinear systems, Autonomous systems, UAV's
Abstract: This paper presents the use of robust model predictive control for the design of an intent-aware collision avoidance system for multi-agent aircraft engaged in horizontal maneuvering scenarios. We assume that information from other agents is accessible in the form of waypoints or destinations. Consequently, we consider that other agents follow their optimal Dubin's path--a trajectory that connects their current state to their intended state--while accounting for potential uncertainties. We propose using scenario tree model predictive control as a robust approach that demonstrates computational efficiency. We demonstrate that the proposed method can easily integrate intent information and offer a robust scheme that handles different uncertainties. The method is illustrated through simulation results.

Keywords: Predictive control for nonlinear systems, Constrained control, Autonomous robots
Abstract: Motivated by the application of using model predictive control (MPC) for motion planning of autonomous mobile robots, a form of output tracking MPC for non- holonomic systems and with non-convex constraints is studied. Although the advantages of using MPC for motion planning have been demonstrated in several papers, in most of the available fundamental literature on output tracking MPC it is assumed, often implicitly, that the model is holonomic and generally the state or output constraints must be convex. Thus, in application-oriented publications, empirical results dominate and the topic of proving completeness, in particular under which assumptions the target is always reached, has received comparatively little attention. To address this gap, we present a novel MPC formulation that guarantees convergence to the desired target under realistic assumptions, which can be verified in relevant real-world scenarios.

Keywords: Predictive control for nonlinear systems, Optimal control, Autonomous systems
Abstract: This work proposes a finite-horizon optimal control approach for set-point tracking using an Explicit Nonlinear Model Predictive Control (ENMPC) framework with integrated obstacle avoidance. A partitioning method is employed to evaluate the feasibility domain in the presence of obstacles. An approximate nonlinear multiparametric programming algorithm provides solutions as piecewise linear functions of the parameters, enabling efficient online implementation via piecewise linear state feedback without real-time optimization. Obstacle avoidance is ensured through control barrier functions and artificial variables, maintaining feasibility under varying references. A numerical example demonstrates the method on a differential mobile robot that tracks a reference while avoiding obstacles in its admissible space.

Keywords: Manufacturing processes
Abstract: One of the important skills to develop when researchers decide to take the commercialization route is to finance the development of their product. This requires long-term planning and considering some aspects that might appear foreign to the founder, especially from an academic background. While research funding and support from research labs can greatly help in the early phases, it soon becomes insufficient. There are different options for the way afterwards, which require considering not only of the technological or scientific value of the product, but also commercial potential. In this presentation, we will address this aspect, drawing on experience in working in startups, and evaluating startup funding applications, and will outline what makes a convincing application. As an example, I will draw from personal experience on how to transition scientific research to more applied research with a commercialization objective in manufacturing.

Keywords: Traffic control
Abstract: This talk traces the technical and organizational journey behind Unjam, a Swiss project developing the Unjam Traffic Gym — a control-oriented digital twin of Zurich’s traffic network built on closed-loop SUMO microsimulations. Beyond the control layer, we share lessons and anecdotes from applying traffic control in the wild: pitching ideas and prototyping within the Talent Kick accelerator, securing innovation-oriented funding, setting up legal and technical infrastructure (e.g., IP management, entity formation, NDA coordination), and building custom data pipelines to fetch and synchronize heterogeneous traffic signals — all in close collaboration with the Zurich startup ecosystem.

We reflect on how control theory informs both system design and broader architectural choices, offering robustness, modularity, and insight into complex urban dynamics. The talk culminates in a data-driven control framework that combines model-free optimization with interpretable, structure-aware policies, designed for real-world deployment and demonstrated within the Unjam Traffic Gym. Finally, we outline key open challenges, including online state estimation, distributed coordination, integration with legacy infrastructure, and scalability in large urban networks.

Keywords: Energy systems
Abstract: In today’s fast-evolving technological landscape, bridging the gap between academic breakthroughs and industrial application is key to fostering sustainable innovation. In my talk, I will share how, as the coordinator for knowledge and technology transfer at NCCR Automation, I have championed initiatives that nurture entrepreneurship and catalyze start-up creation. Through the Bench2Biz workshop and our Moonshot Initiatives — large-scale living labs that bring together young researchers, industry leaders, citizens, and policy makers—we have built a dynamic ecosystem where ideas quickly transform into market-ready solutions.

I will discuss the strategies and best practices that underpin these initiatives, highlighting both the successes and the challenges encountered along the journey from laboratory research to commercial reality. Attendees will learn how multi-stakeholder engagement and a collaborative, hands-on approach can unlock new avenues for technology transfer and create fertile ground for start-up growth. Whether you are an academic, an industry practitioner, or a policy maker, this talk offers valuable insights into developing robust partnerships that accelerate innovation and economic impact. Join me as we explore how aligning research with practical industry needs not only drives technological advancement but also inspires the next generation of entrepreneurial leaders.

Keywords: Autonomous systems
Abstract: In this first interactive session, we will discuss entrepreneurial ideas with the audience. We will form groups with the audience. Each group will discuss potential ideas for commercialization based on their research. One idea per group will be selected, and the group will prepare a pitch, using a dedicated tool. Finally, the pitch will be presented in front of all participants.

Keywords: Autonomous systems
Abstract: The second interactive session consists of a Q&A session where we will collect relevant questions, and will gauge interest for mentorship in the control filed and community. This will be useful input for the Task Force Entreprenurship and Startups of the IFAC Industry Committee (IFAC IC), which is organizing the workshop. The results will be communicated via the IFAC IC.

Keywords: Iterative learning control, Lyapunov methods, LMI's/BMI's/SOS's
Abstract: The paper deals with the problem of designing a novel iterative learning control (ILC) strategy for a class of input/state constrained dynamic systems. In particular, the proposed approach boils down to designing a input/state ILC for a virtual deterministic reference system. Based on the obtained control law as well as the reference system state, the final control strategy for the actual disturbed system is obtained. As a result, the actual system tracks the reference one with a desired performance taking into account the unappealing disturbance effect. The final part of the paper presents comprehensive simulation results, which clearly exhibit the effectives of the proposed approach.

Keywords: Iterative learning control, Model validation, Fault detection and identification
Abstract: Model-based algorithms are deeply rooted in modern control and systems theory. However, they usually come with a critical assumption - access to an accurate model of the system. In practice, models are neither perfect nor time-invariant. Even precisely tuned estimates of unknown parameters will deteriorate over time. We propose to combine statistical tests with dedicated parameter filters that track unknown system parameters from state data. These filters yield point estimates of the unknown parameters and, further, an inherent notion of uncertainty. This allows us to detect changes in the dynamics. When the point estimate leaves the confidence region, we trigger active learning experiments. Thus, models are only updated when necessary and statistically significant while ensuring guaranteed improvement, which we call event-triggered learning. We validate the proposed method in numerical simulations of a DC motor in combination with model predictive control.

Keywords: Iterative learning control
Abstract: This paper addresses the problem of designing the Iterative Learning Control for a specific subclass of 2D systems. It is also assumed that the system is constructed with elements with nonlinear characteristics which puts some additive requirements on the designed control performance. The designed ILC scheme is defined in terms of Differential Linear Repetitive Process and then the stability / stabilization theory (using an appropriate LMI conditions) for that class of systems is adapted in order to determine the controllers. The proposed design is then illustrated by the simulation example.

Keywords: Iterative learning control, Biological systems, Optimization algorithms
Abstract: For control problems that repeat with resets, such as batch processing and robotic trajectory tracking, iterative learning control is an established high-performance scheme. Fast learning is achievable with model-based schemes when the dynamics of the system are known to be described accurately by a prior model, but such accurate models are often difficult to obtain. The field of sensorimotor control studies the motion control systems of humans and other animals, which appear to quickly achieve accurate trajectory tracking without detailed prior knowledge. In this paper, we present a novel design that iteratively optimises the parameters of a system model, with a modular structure inspired by the learning behaviour of these sensorimotor control systems. The ‘modules’ used are a generalisation of pre-defined orthonormal basis functions, and the parameters of these modules are learnt using an alternating direction method of multipliers approach which is proven to converge under relatively mild assumptions. This establishes a framework suitable for the modelling of complex system dynamics while enforcing desirable model properties. We also discuss how this approach may be successfully applied when the reference changes on each trial. Generalising learnt skill in this way is a current challenge in iterative learning control design, and is a key benefit of the modular structure of sensorimotor- inspired schemes.

Keywords: Predictive control for nonlinear systems, Hybrid systems, Iterative learning control
Abstract: Improving the predictive accuracy of a dynamics model is crucial to obtaining good control performance and safety from Model Predictive Controllers (MPC). One approach involves learning unmodelled (residual) dynamics, in addition to nominal models derived from first principles. Varying residual models across an environment manifest as modes of a piecewise residual (PWR) model that requires a) identifying how modes are distributed across the environment and b) solving a computationally intensive Mixed Integer Nonlinear (MINLP) problem for control. We develop an iterative mapping algorithm capable of predicting time-varying mode distributions. We then develop and solve two tractable approximations of the MINLP to combine with the predictor in closed-loop to solve the overall control problem. In simulation, we first demonstrate how the approximations improve performance by 4-18% in comparison to the MINLP while achieving significantly lower computation times (upto 250x faster). We then demonstrate how the proposed mapping algorithm incrementally improves controller performance (upto 3x) over multiple iterations of a trajectory tracking control task even when the mode distributions change over time.

Keywords: Agents and autonomous systems, Autonomous robots
Abstract: In harsh industrial environments such as those affected by potholes or oil pollution, autonomous mobile robots (AMRs) often struggle to precisely park at designated stations, thereby hindering their ability to perform autonomous operations effectively. To solve the problem, this paper proposes a trajectory re-planning method for multiple accurate parking. The proposed method leverages an optimized Bi-RRT trajectory planning scheme, involving the angle constraints, the APF-assisted heuristic sampling method and the adaptive step length modification strategy. Finally, the proposed method is experimentally verified in a production line scenario. The results verify that the proposed method exhibits advantages in terms of improving the quality of trajectory planning, thereby ensuring accurate parking of AMRs in industrial scenarios.

Keywords: Robust adaptive control, Biomedical systems, Modeling
Abstract: Intravenous administration of medication for newborns and premature infants presents significant challenges due to the critical nature of dosage accuracy, particularly under settings including extremely low flow rates and fluids of different viscosities. Errors in dosing at such conditions can have serious consequences, including toxicity or even death. Classical delivery systems, such as syringe pumps, generally fail to offer reliable accuracy under these challenging settings. This work introduces a novel infusion system designed primarily for neonatal intensive care units, employing a pressure regulator and a robust controller based on sliding mode to overcome the previously mentioned limitations. The novel approach facilitates drug administration with good performances in spite of the presence of external perturbations. Comparative experiments show that the proposed system significantly reduces dosage errors which yields an improvement in terms of safety and efficacy of neonatal care units.

Keywords: Robust adaptive control, Process control, Machine learning
Abstract: The application of reinforcement learning (RL) to industrial processes involves significant challenges, particularly due to the safety risks of modeling errors and uncertainties. This paper presents a tube-based (RL) framework that integrates the robustness of the tube model predictive control through a dual-agent approach. The first agent operates in a simulated environment to control the process based on an idealized model, while the second agent adjusts for discrepancies between the simulated model and actual process dynamics. To approximate real-world conditions in the absence of direct process access, the simulation model is perturbed with parametric uncertainties and Gaussian disturbances. While this paper focuses on parametric uncertainties, the framework can flexibly address broader types of model uncertainties. A case study highlights the robust and adaptive performance of the proposed approach, showing that the tube-based (RL) method achieves stable control in uncertain environments.

Keywords: Robust adaptive control, Fault accommodation, Optimal control
Abstract: In order to solve continuous-time optimal control problems, direct methods transcribe the infinite-dimensional problem to a nonlinear program (NLP) using numerical integration methods. In cases where the integration error can be manipulated by the chosen control trajectory, the transcription might produce spurious local NLP solutions as a by-product. While often this issue can be addressed by increasing the accuracy of the integration method, this is not always computationally acceptable, e.g., in the case of embedded optimization. Therefore, alternatively, we propose to estimate the integration error using established embedded Runge-Kutta methods and to regularize this estimate in the NLP cost function, using generalized norms. While this regularization is effective at eliminating spurious solutions, it inherently comes with a loss of optimality of valid solutions. The regularization can be tuned to minimize this loss, using a single parameter that can be intuitively interpreted as the maximum allowable estimated local integration error. In a numerical example based on a system with stiff dynamics, we show how this methodology enables the use of a computationally cheap explicit integration method, achieving a speedup of a factor of 3 compared to an otherwise more suitable implicit method, with a loss of optimality of only 3%.

Keywords: Robust adaptive control, Optimal control
Abstract: This paper proposes a robust adaptive nonlinear optimal control strategy for mechanical systems that employs the weighted Sobolev space to achieve trajectory tracking with enhanced transient performance, even in the presence of parametric uncertainties and exogenous disturbances. The proposed adaptive W-infinity control law utilizes the Dynamic Regressor Extension and Mixing (DREM) estimator to estimate parameter values without relying on persistence of excitation (PE) conditions. Additionally, we use the concept of the Euler-Lagrange first integral to eliminate the need for acceleration measurements, which simplifies the application of DREM-based estimators. Numerical results involving a quadrotor unmanned aerial vehicle (UAV) are presented to corroborate the theoretical findings, demonstrating improved performance compared to a non-adaptive counterpart controller.

Keywords: Robust adaptive control, Predictive control for linear systems, Uncertain systems
Abstract: Robust MPC schemes address the presence of uncertainties in the model to achieve robust constraint satisfaction and stability of the closed-loop. A significant improvement of the performance of robust MPC schemes is possible by utilizing real-time measurement data from the controlled system to adapt the model uncertainty that is used in the controller online. We consider a linear system that is affected by both additive and multiplicative uncertainties. The true parametric uncertainty of the system is assumed to be contained in a convex polytopic set and time-invariant. In addition, it is considered that not all states are measured and that the measurements are affected by noise. We propose an adaptive robust MPC scheme that can handle uncertainties and can systematically update the limits of the unknown parameters in the system at every time step. The unknown but constant parameters are estimated using guaranteed parameter estimation. Recursive feasibility and robust stabilizing capabilities of the proposed approach are proven. Two examples are presented to show the benefits of the proposed adaptive scheme.

Keywords: Robust adaptive control, Predictive control for nonlinear systems, Optimization algorithms
Abstract: A computationally efficient nonlinear Model Predictive Control (NMPC) algorithm is proposed for safe learning-based control with a system model represented by an incompletely known affine combination of basis functions and subject to additive set-bounded disturbances. The proposed algorithm employs successive linearization around predicted trajectories and accounts for the uncertain components of future states due to linearization, modelling errors and disturbances using ellipsoidal sets centered on the predicted nominal state trajectory. An ellipsoidal tube-based approach ensures satisfaction of constraints on control variables and model states. Feasibility is ensured using local bounds on linearization errors and a procedure based on a backtracking line search. We combine the approach with a set membership parameter estimation strategy in numerical simulations. We show that the ellipsoidal embedding of the predicted uncertainty scales favourably with the problem size. The resulting algorithm is recursively feasible and provides closed-loop stability and performance guarantees.

Keywords: Sensor and signal fusion
Abstract: Living beings are rarely passive observers. Instead, they are characterized by a ”Perception-Action Coupling”, i.e. they perceive in order to move and they move in order to perceive, which constitutes an essential loop for learning. Drawing inspiration from nature, this work proposes a novel method for minimizing the uncertainty of the information gathered when observing a Repetitive Dynamic Event (RDE), considering a sensor that can change its position and orientation. The method considers an Unscented Kalman Filter (UKF) for fusing the sensor’s measurements with the evolution of the current dynamical model of the observed RDE. The proposed method aims at finding the optimum UKF parameters and pose of the sensor, for maximizing the quality of the estimation. Simulation and experimental evaluations show the improvement achieved by the proposed method, as compared to using a non optimized UKF, in terms of the quality of acquired information of a dynamic event. The experiments are conducted using a UR10e robotic manipulator with an eye-in-hand ZED 2 RGB-D camera.

Keywords: Sensor and signal fusion, Manufacturing processes, Emerging control applications
Abstract: Many autonomous systems are increasingly adopting Neural Networks (NNs) based perception in vision-in-the-loop (VIL) control systems. In many industrial applications, the features (shape, size and texture) of the object of interest varies, which imposes robustness requirements on the perception algorithm. Further, performance of the VIL system imposes strict latency requirements. Using NNs in VIL system poses two challenges. First, the NN models should be lightweight resulting in a low closed-loop latency. Second, availability of representative training data for ensuring robustness of the lightweight NN models. Collecting such training data is expensive and often, infeasible in many industrial systems. In this work we propose an approach for training the NNs used for VIL applications using digital twins (DT). The DT is used for automatically generating and labelling training data representing various features like object shapes and directional lighting. Starting from a lightweight NN base model, our proposed approach fine-tunes or retrains the model using DT-generated training data achieving desired performance and robustness on a different target operating condition. The approach is validated considering a VIL semiconductor motion stage system with square and rectangular dies of dimension of (0.5cm x 0.5 cm) and (0.5cm x 1 cm) respectively. The VIL system limits the positioning error in the range of 2% compared to 12% positioning error with no vision feedback.

Keywords: Sensor and signal fusion, Fault accommodation
Abstract: In many applications, attitude estimation algorithms rely on Magnetic field, Angular Rate, and Gravity measurements from a triad of sensors known as MARG sensors. Attitude estimation of rigid bodies is crucial for navigation systems which can be vulnerable to cyber-attacks. Several works in the literature focus on secure state estimation against randomly occurring false data injection (FDI) attacks on output measurements for both linear and nonlinear systems. However, no studies address this problem when the state belongs to the special orthogonal group SO(3), which provides a structured mathematical framework for attitude representation. Given the importance of SO(3) for attitude estimation, this paper proposes a secure MARG sensor-based attitude estimation on SO(3), subject to randomly occurring FDI attacks. A novel Kalman gain of the invariant extended Kalman filter (IEKF), which is typically used for attitude estimation under noisy measurements, is designed. The objective of the proposed design is to minimize the upper bound of the estimation error covariance matrix. The proposed algorithm is validated through Monte Carlo simulations.

Keywords: Observers for nonlinear systems, Sensor and signal fusion, Filtering
Abstract: The manifold extended Kalman filter (Manifold EKF) has found extensive application for attitude determination. Magnetometers employed as sensors for such attitude determination are easily prone to disturbances by their sensitivity to calibration and external magnetic fields. The TRIAD algorithm is well-known as a sub-optimal attitude estimator. In this article, we incorporate this sub-optimal feature of the TRIAD in mitigating the influence of the magnetometer reading in the pitch and roll axis determination in the Manifold EKF algorithm. We substantiate our results with experiments.

Keywords: Large-scale systems, Distributed estimation over sensor nets, Statistical learning
Abstract: In this paper, we focus on learning high-dimensional linear dynamical systems. To learn such systems, existing algorithms require either strong sensing capability or additional assumption on system observability. We propose an algorithm that learns the system model with a few sensors and without observability assumption. This is accomplished by integrating information from multiple data trajectories, each observing a possibly different set of coordinates of the high-dimensional states. Theoretical analysis of the algorithm is provided, showing that the system model can be learned accurately as long as every coordinate is observed in at least one trajectory. This approach significantly reduces the required number of sensors for data collection.

Keywords: Observers for nonlinear systems, Robotics
Abstract: This paper presents a robust moving horizon estimation (MHE) approach with provable estimation error bounds for solving the simultaneous localization and mapping (SLAM) problem. We derive sufficient conditions to guarantee robust stability in ego-state estimates and bounded errors in landmark position estimates, even under limited landmark visibility which directly affects overall system detectability. This is achieved by decoupling the MHE updates for the ego-state and landmark positions, enabling individual landmark updates only when the required detectability conditions are met. The decoupled MHE structure also allows for parallelization of landmark updates, improving computational efficiency. We discuss the key assumptions, including ego-state detectability and Lipschitz continuity of the landmark measurement model, with respect to typical SLAM sensor configurations, and introduce a streamlined method for the range measurement model. Simulation results validate the considered method, highlighting its efficacy and robustness to noise.

Keywords: Agents and autonomous systems
Abstract: The cohesive motion of a group of vehicles remains stable when all agents achieve velocity consensus and maintain a constant desired speed. However, when the reference velocity changes over time, the formation tracking controllers may struggle to ensure collision-free motion during transitional phases. In this paper, the simultaneous formation control and tracking task is formulated as a constrained optimization problem where the control input actions are restricted within a defined search space. The control objective is to ensure a desired formation geometry while tracking a time-varying reference trajectory by satisfying inter-agent separation conditions for collision avoidance and adhering to input saturation limits that define the allowable control space. Due to the sensor limitations of the deployed agents, the inter-agent distances are estimated using only relative bearing and velocity measurements. The stability and tracking convergence of the proposed distributed control and observer designs are analyzed and validated through simulations in a multi-vehicle deployment scenario. The results show that the control inputs stay within the feasible limits and effectively prevent inter-agent collisions while achieving the control objectives.

Keywords: Game theoretical methods, Markov processes, Intelligent systems
Abstract: The moving target defense (MTD) is a defense paradigm in repeated security games within the realm of cybersecurity. The defender’s strong Stackelberg equilibrium (SSE) strategy is optimal for deployment as an MTD strategy, assuming that the attacker adopts a best-response strategy after observing the defender’s actions. However, computing SSE strategies in repeated games is complex due to the non-convexity of players' expected utilities, especially when dealing with multiple targets in large-scale problems. Thus, we propose to use the zero-determinant (ZD) strategy as an MTD strategy to reduce computational complexity. To this end, we identify the conditions for the existence of the ZD strategy. Also, we provide a novel approach to construct the required ZD strategy.

Keywords: Nonlinear system theory, Algebraic/geometric methods, Large-scale systems
Abstract: Multilinear dynamical systems (MLDSs) are crucial for modeling multidimensional data across various fields, including neuroscience, social sciences, and signal processing, as they effectively capture higher-order interactions and complex tensorial structures beyond the scope of linear systems. However, analyzing and controlling MLDSs poses significant challenges due to their inherently high-dimensional, multilinear characteristics. In this article, we examine the controllability of MLDSs where the systems' evolution is governed by a multilinear operator based on Tucker decomposition. Notably, we extend the classical Kalman's rank condition to MLDSs by leveraging the factor matrices in the Tucker decomposition. Additionally, we explore the weak and strong controllability of MLDSs in the presence of stochastic noise. Numerical examples are provided to demonstrate the effectiveness and efficiency of our proposed framework.

Keywords: Observers for nonlinear systems, Control over networks, Genetic regulatory systems
Abstract: A cornerstone of understanding complex systems is observability: the capacity of data to encode a system’s state. Hypergraphs are an effective way to represent higher order data that captures group interactions within these systems. In this paper, we present hypergraph based algorithms that determine the observability properties of systems whose group interactions are governed by polynomial dynamics. Our proposed framework evaluates nonlinear observability criteria for polynomial dynamics using hypergraph walks, bypassing traditional matrix and tensor-based approaches. The improved efficiency of the proposed methods enables the analysis of hypergraph systems that are several orders of magnitude larger than previously possible. Finally, we use these methods to address the sensor and biomarker selection problem in a hypergraph model of the cell cycle, which contains hundreds of genes and thousands of group interactions.

Keywords: Distributed estimation over sensor nets, Optimization algorithms, Observers for linear systems
Abstract: We introduce a novel dynamic data fusion method for distributed state estimation in multi-agent systems. We consider a scenario where system outputs are measured by sensors distributed across multiple agents. We assume that the local measurement may lack detectability while the joint measurements ensure the detectability of the system. Each node is equipped with an observer that exchanges state estimates with neighboring nodes over a communication network, with the objective of estimating the entire system state locally. Our approach first reconstructs the locally observable portion of the state at each agent and then reformulates the distributed state estimation problem as a distributed online optimization task. By iteratively solving this optimization problem, the proposed scheme enables each agent to estimate the full system state within a bounded error, which depends on the system dynamics and the number of iterations per time step. Simulation results show the effectiveness of the proposed dynamic data fusion method.

Keywords: Safety critical systems
Abstract: This paper studies the safety-critical control of a class of underactuated systems, with a focus on a wheeled inverted pendulum as a case study. A safety-critical controller is desired to simultaneously keep the pendulum upright, restrict the position of the wheel within a desired range, and ensure that the rotational angular velocity of the wheel tracks a nominal reference signal as closely as possible. We first reformulate the safety constraints as saturated constraints on the state variables of a cascade connection of an integrator and a feedforward system, thereby mitigating conflicts among the safety constraints. Then, to ensure the satisfaction of the constraints, we propose a safety-critical controller that consists of a safety-critical virtual controller based on control barrier functions (CBFs) and a tracking controller designed using forwarding methods. Nonlinear small-gain techniques are employed to ensure the stable integration of the two controllers and the achievement of the safety-critical control objectives.

Keywords: Maritime, Optimal control, Differential algebraic systems
Abstract: Tugboats are deployed to support underactuated vessels, such as container ships or unactuated offshore platforms during a manoeuvre. This requires precise manoeuvring and a high degree of coordination between the individual vessels involved to avoid collisions and to ensure the safety of crew members. Based on this and the property that tugboats are typically fully actuated, the interest in autonomous tugboats has increased. Therefore, this paper proposes a method for coordinating tugboats in a multi-vessel compound system, together with subsequent manoeuvre planning. To this end, direct contact and tow ropes are considered and defined as motion constraints. The mathematical models of the individual vessels are coupled by applying a Lagrangian approach from the theory of multi-body dynamics, resulting in the structure of a differential-algebraic equation (DAE). Through an index reduction, this DAE-system can be converted into a system consisting of ordinary differential equations (ODE) only, which enables a solution by means of standard numerical solvers. Based on this approach, an optimal control problem is defined for a berthing and an unberthing manoeuvre, which is successfully evaluated by simulations.

Keywords: Maritime, Fault diagnosis, Fault tolerant systems
Abstract: Fault diagnosis for dynamic positioning vessels is vital for maritime safety and operational efficiency. The reliability of the dynamic positioning system is crucial for the safe and precise operations for which this system is used. However, over time, these systems degrade, compromising accuracy and performance. This issue has been investigated by several authors over the past years. After reviewing the works published in the literature, this article aims to provide an overview of fault detection and fault-tolerant control for dynamic positioning systems, with a focus on thrusters faults, and to explore potential future research directions.

Keywords: Maritime, Modeling
Abstract: This paper presents a numerical model for marine vehicles, developed primarily to perform a validation of automatic Dynamic Positioning (DP) verification methods. The primary focus of this work is to develop and test a DP system that can be applied across a specific real vessel. It is equipped with five thrusters, including two bow tunnel thrusters, a single fixed thruster positioned along the center line of the vessel responsible for forward and reverse propulsion, and two azimuth thrusters for enhanced maneuverability. The DP system plays a critical role in managing the actuators to achieve optimal positioning performance, whether the vessel is stationary or executing path-following operations under moderate weather conditions. To verify the control and thrust allocation logic, a comprehensive numerical model has been developed using MATLAB/Simulink™ software. This model integrates multiple subsystems, including ship dynamics, propulsion systems, thrust allocation, control systems, and environmental disturbances, all linked through suitable mathematical models to accurately capture their interactions and effects on the vessel’s performance. Eventually, simulation results are shown to evaluate the vessel model’s coherence and performance of the overall DP system.

Keywords: Maritime, Switched systems, Predictive control for nonlinear systems
Abstract: This work develops a control method for an inland vessel with three rudder propellers for low speed maneuvers like docking, station keeping, and lock entry. The rudder propellers’ thrust, rotation, and clutch state are adjusted to control the vessel position and orientation with a focus on clutching processes beyond classical control methods. The aim is a robust, adaptive approach that simplifies operation, aids automation, and saves fuel. Two control concepts are proposed: a hybrid NMPC using a mixed integer nonlinear program to handle clutching and a second NMPC approach approximating the clutch with a sigmoid function. Simulations with the nominal ship dynamics show that the computational requirements of the mixed integer approach are too high. To meet real-time requirements, a faster, suboptimal nonlinear controller for the approximation approach is developed and its resilience to model uncertainties and disturbances is demonstrated in software-in-the-loop simulations.

Keywords: Energy systems, Maritime
Abstract: Slaloming Tethered RiverinE And Marine Surface (STREAMS) Wings are a novel technology where a vertical underwater wing, rudder, and marine turbine are suspended beneath a tethered, streamlined surface hull in a water current. The STREAMS wing is controlled to slalom back and forth across the current at velocities higher than the current itself to generate electric power. The motion of the system is commanded by the angular position of the underwater wing and rudder to generate sufficient hydrodynamic loading, enabling the system to slalom across the current velocity field. The motion induced on the system drives a turbine mounted on board from which energy is collected. To ensure adequate energy production, the control signals for the wing and rudder must be appropriately commanded. Therefore, a Lyapunov-based stability analysis is conducted to assess input-output stability (IOS) and to provide a baseline backstepping closed-loop framework for controlling the STREAMS heading angle. Simulation results are included as an additional assessment tool to evaluate the system’s behavior and characteristics.

Keywords: Fault detection and identification, Machine learning, Maritime
Abstract: Catastrophic failures totally destroying or compromising components of marine engines are sudden and often unpredictable. Due to their abruptness, they are highly dangerous for navigation, crew, and passengers. Then, a big effort is required to maintain the ships healthy. Despite this, most of the works on maintenance focus on predicting the decay law of components, namely the gradual degradation of the performance. No care was paid to sudden, quick, and anomalous events leading to catastrophic failures. We propose here a preliminary study to address the lack of catastrophic failure early detection methods: we evaluate the second derivative, not just the rate, of the difference between the real sensor measurements and expected values of engine variables. Expectations are obtained by a Decision Tree, which implements a specific Machine Learning algorithm. The second derivative of the computed error gives more information on catastrophic failures than the error, because, more than the rate, it can be a symptom that something is suddenly happening inside the component. Also, information is obtained before the sensor measurements reach the thresholds that produce warning alarms. In this way, the ship operators can be notified in advance to shut down the engine. Simulation shows the effectiveness of the method. In fact, the second derivative, with its peaks and high frequency, makes it possible to recognize in advance some abrupt failure is occurring in the engine. In this way, an operator could shut down the engine in suitable time to avoid a big damage.

Keywords: Reduced order modeling, Model/Controller reduction, Linear systems
Abstract: This article introduces a novel model reduction method that preserves dissipativity properties while achieving moment matching at user-specified interpolation points. Building on the time-domain moment matching framework, our approach leverages the flexibility of free parameters in reduced-order models to ensure the preservation of (Q,S,R)-dissipativity, independent of the state dimension of the reduced model or the location of the interpolation points. By preserving dissipativity, this method enables more reliable and efficient reduced-order models for large-scale systems, while maintaining key physical and stability properties. Numerical examples are presented to illustrate the effectiveness and applicability of the proposed approach.

Keywords: Identification, Reduced order modeling, Linear systems
Abstract: First-principle models often fail to accurately capture system dynamics due to modeling simplifications and parameter uncertainties. This article introduces a data-driven technique for linear systems, enhancing baseline first-principle state-space models with black-box models obtained from experimental steady-state data. The proposed method parameterises models that achieve moment matching by integrating a known baseline model with a black-box component. Tools are provided to enforce a known interconnection structure or other physical knowledge. A mass-spring-damper system demonstrates the effectiveness of the technique.

Keywords: Large-scale systems, Reduced order modeling, Computational methods
Abstract: Simulations and data analysis in structural dynamics are challenged by large multi-dimensional data. Apart from two or three spatial directions, the problem coordinates include the dimension in frequency actuation and, possibly, dimensions to model uncertainties. If stored as a multidimensional array, these data quickly exceeds all storage capacities so that efficient approximative representations are needed for the data handling and, respectively, for simulations as a reduced-order model. Although the solution exhibits wave patterns, it is smooth and hence, well accessible to tensorized proper orthogonal decomposition (POD) approaches.

The frequency dimension, however, shows a number of characteristic poles which should be preserved so that the least-squares-based approach of POD may not well suited. Therefore, in this work, we call on recently developed methods for rational interpolation of matrix-valued functions, extend them to higher-dimensional arrays to approximately represent one specific mode of these tensors, and investigate the interplay with higher-order singular value decomposition of the remaining modes. We illustrate the findings for tensorized data and simulations of a two-dimensional plate under excitation.

Keywords: Model/Controller reduction, Nonlinear system theory
Abstract: Differential balancing theory for nonlinear model reduction relies on differential controllability and observability functions. In this paper, we further investigate them from two different perspectives. First, we establish novel connections between these differential energy functions and their incremental counterparts by assuming the existence of the corresponding optimal state feedback for each controllability function. Specifically, an upper bound on the incremental controllability/observability function is provided by the corresponding differential energy function. Conversely, an upper bound on the differential controllability function can be estimated from the incremental controllability function. Furthermore, the differential observability function can be constructed from the incremental observability function. Second, we explore the positive definiteness of the differential controllability/observability function in the context of controllability/observability and stability.

Keywords: Switched systems, Linear time-varying systems, Reduced order modeling
Abstract: Based on our recently established solution characterization of switched singular descriptor systems in discrete time, we propose a time-varying balanced truncation method. For that we consider the switched system on a finite time interval and define corresponding time-varying reachability and observability Gramians. We then show that these capture essential quantitative information about reachable and observable state directions. Based on these Gramians we formulate a time-varying balanced truncation method resulting in a fully-time varying linear system with possible varying state dimensions. We illustrate this method with a small dynamic Leontief model, where we can reduce the size to one third without altering the input-output behavior significantly. We also show that the method is suitable for a medium size random descriptor system (100 x 100) resulting in a time-varying system of less then a tenth of the size where the outputs of the original and reduced system are indistinguishable.

Keywords: Reduced order modeling, Model/Controller reduction, Stability of linear systems
Abstract: A model reduction-based controller design approach is presented, addressing the challenge that a lower-dimensional model, though open-loop stable, may not guarantee closed-loop stability—an essential criterion in controller design. To overcome this, a novel error minimization approach is presented for finding reduced-dimensional models of stable linear time-invariant higher-order systems, ensuring both open-loop and closed-loop stability. Open-loop stability is preserved using Lyapunov's second method, while closed-loop stability is enforced by incorporating gain margin and phase crossover frequency constraints into the optimization process. An initial reduced-order model is obtained without relative stability (RS) constraints, and a PID controller (PIDC) is derived. For comparison, PIDC is applied to both the higher-dimensional system (HDS) and the modified reduced model derived, that includes stability margin constraints. The effectiveness of the approach is validated via an example.

Keywords: Robust adaptive control, Robotics, Output feedback
Abstract: In this paper, we propose an adaptive performance control (APC) framework for output-feedback control of robot manipulators with input constraints, focusing on achieving joint position tracking with adaptive performance attributes. Building on the APC methodology, we introduce a singlefunnel control approach that dynamically adjusts performance constraints to accommodate the actuator limitations inherent in robotic systems. Moreover, a robust tracking differentiator with predefined convergence characteristics is developed to estimate the unmeasured velocity errors. The effectiveness of the proposed control scheme is validated through simulation results on a 2-DoF planar manipulator and experiments on a 3-fingered 12 degree-of-freedom robotic hand tasked with performing a musical piece on a keyboard.

Keywords: Feedback linearization, Linear parameter-varying systems, Robotics
Abstract: Feedback linearization is a valuable tool for designing controllers for vehicles with nonholonomic constraints. However, actuator dynamics and delays – particularly in steering – can limit its effectiveness. This paper presents an extension to existing feedback linearization approaches that incorporates actuator dynamics into the linearization law, and compensates for delays and non-idealities with an advanced gain-scheduling control action. This results in a linear time-invariant, decoupled system, enabling trajectory tracking controller design using linear control methods. Simulation results show the effectiveness of the proposed solution within a trajectory tracking framework.

Keywords: Servo control, Robust control, Robotics
Abstract: We propose a framework enabling mobile manipulators to reliably complete pick-and-place tasks for assembling structures from construction blocks. The picking uses an eye-in-hand visual servoing controller for object tracking with Control Barrier Functions (CBFs) to ensure fiducial markers in the blocks remain visible. An additional robot with an eye-to-hand setup ensures precise placement, critical for structural stability. We integrate human-in-the-loop capabilities for flexibility and fault correction and analyze robustness to camera pose errors, proposing adapted barrier functions to handle them. Lastly, experiments validate the framework on 6-DoF mobile arms.

Keywords: Sliding mode control, Robotics, Mechatronics
Abstract: Many Sliding Mode Control (SMC) algorithms are developed and tested on robotic manipulators based on the premise that finite-time convergence can be achieved in practical systems. However, this idealistic approach only works theoretically, as parasitic dynamics (unmodeled dynamics) in real systems lead to non-vanishing oscillations instead of convergence to an equilibrium point. In this work, we argue this fundamental point for the case of robotic manipulators. Two SMC techniques—conventional and Terminal SMC—are analyzed using Describing Function analysis and the Locus of Perturbed Relay Systems on a SOPDT model and a two-link manipulator. Findings confirm that chattering, caused by parasitic dynamics, challenges the feasibility of SMC’s finite-time convergence in real-world applications.

Keywords: Safety critical systems, Robotics, Robust control
Abstract: Set invariance techniques such as control barrier functions (CBFs) can be used to enforce time-varying constraints such as keeping a safe distance from dynamic objects. However, existing methods for enforcing time-varying constraints often overlook model uncertainties. To address this issue, this paper proposes a CBFs-based robust adaptive controller design endowing time-varying constraints while considering parametric uncertainty and additive disturbances. To this end, we first leverage Robust adaptive Control Barrier Functions (RaCBFs) to handle model uncertainty, along with the concept of Input-to-State Safety (ISSf) to ensure robustness towards input disturbances. Furthermore, to alleviate the inherent conservatism in robustness, we also incorporate a set membership identification scheme. We demonstrate the proposed method on robotic surface treatment that requires time-varying force bounds to ensure uniform quality, in numerical simulation and real robotic setup, showing that the quality is formally guaranteed within an acceptable range.

Keywords: Robotics
Abstract: This paper proposes a visibility-based planner for autonomous inspection tasks in outdoor ultra-high voltage centers, using a mobile robot. Regular inspection of components in such installations is essential to prevent faults and avoid power blackouts. The proposed system enables users to specify particular components for inspection. Based on this input and the environment’s map data, our algorithm suggests the most efficient inspection path, identifying the most traversable route across the selected components while maximizing optical and infrared data capture. We decompose the task into smaller sub-problems, addressing each individually. First, we perform a visibility-based waypoint selection and geometric path planning task, both executed offline as pre-processing steps. Using this data along with the user’s input, we then frame the problem as a Generalized Traveling Salesman Problem and solve it using the GLNS solver, which demonstrates superior performance and suitability for our problem type compared to other state-of-the-art methods. Experimental benchmarks indicate that our system consistently identifies optimal solutions tailored to our specific inspection use case.

Keywords: Transportation systems, Modeling, Network analysis and control
Abstract: This paper proposes a bi-level optimization framework to improve urban cycling infrastructure by upgrading bike lanes' degree of separation from motorized traffic. The six-level degree of separation classification reflects increasing safety and comfort as cyclists' exposure to annual average daily traffic decreases. The lower level estimates cyclist flows using an all-or-nothing assignment method that accounts for factors like slope, angular changes, and traffic, ensuring realistic flow distribution. The upper level minimizes traffic exposure by upgrading segments to higher DoS levels within a fixed budget. A case study on Grenoble's cycling network applies this framework using a modified genetic algorithm to optimize safety, comfort, and budget efficiency. Results are compared to a specialized algorithm designed to improve overall perceived safety and comfort. Findings highlight that targeted infrastructure upgrades enhance both local and network-wide safety, better balancing cyclist flows. This framework provides urban planners with a data-driven tool for prioritizing cycling infrastructure investments.

Keywords: Traffic control, Iterative learning control, Modeling
Abstract: Highway congestion leads to significant delays and pollution. Regulating the outflow from the Service Station can help alleviate this congestion. Notably, traffic flows follow recurring patterns over days and weeks, allowing for the application of Iterative Learning Control (ILC). Building on these insights, we propose an ILC approach based on the Cell Transmission Model with service stations (CTM-s). It is shown that ILC can effectively compensate for potential inaccuracies in model parameter estimates by leveraging historical data.

Keywords: Transportation systems, Traffic control
Abstract: Bus bunching, where buses arrive at stops in quick succession rather than at evenly spaced intervals, disrupts public transit systems by increasing passenger waiting times and decreasing system reliability. To address this issue, we propose a novel strategy featuring automated modular bus units that can either operate individually or be combined en route to form a larger bus. Our strategy systematically combines this new modular technology with bus bunching mitigation techniques, including stop skipping and bus substitution, to minimize passenger waiting times and operational costs. Our strategy is formulated as a Markov Decision Process and solved by an Advantage Actor-Critic algorithm. Our approach is validated through numerical experiments using real-world data, demonstrating significant improvements in system reliability and passenger satisfaction in various traffic conditions.

Keywords: Automotive, Optimal control
Abstract: This paper presents an energy and thermal management system for electric race cars, where we tune a lift-off-throttle signal for the driver in real-time to respect energy budgets and thermal constraints. First, we compute the globally optimal state trajectories in a real-time capable solving time, optimizing a 47-kilometer horizon in 2.5 seconds. Next, for safe operation with a human driver, we simplify it to a maximum-power-or-coast operation in full-throttle regions (straights). Thereby, both the positions from which the vehicle should start coasting and the optimal throttle map are subject to tuning. To this end, we define the coasting sections with a threshold on the costate trajectory of the kinetic energy from the optimal solution. We devise an online implementable bisection algorithm to tune this threshold and adapt it using PI feedback. Finally, we validate the proposed approach for an electric endurance race car and compare three variants with varying implementation challenges: one re-optimizing and updating the kinetic costate trajectory online, one applying only the bisection algorithm online, and one relying exclusively on feedback control. Our results show that, under typical racing disturbances, our energy management can achieve stint times ranging from less than 0.056% to 0.22% slower compared to offline optimization with a priori knowledge of disturbances, paving the way for on-board implementations and testing.

Keywords: Traffic control, Machine learning, Transportation systems
Abstract: Traffic signal control is essential for managing urban traffic, reducing congestion, and minimizing environmental impact by optimizing both vehicular and pedestrian flow. This paper investigates the application of Reinforcement Learning (RL) in traffic signal control within mixed traffic environments, emphasizing the development of a synergistic RL approach, named Advantage Actor-Critic with Maximum Pressure (A2CMP). A2CMP leverages actor-critic techniques in combination with real-time pressure metrics to dynamically adjust traffic signals based on prevailing traffic conditions. Additionally, the paper introduces a pedestrian-friendly phase-skipping mechanism for further enhancing the efficiency of the proposed algorithm in real-world traffic management. Simulation results across diverse traffic scenarios show significant reductions in CO2 emissions and waiting time. Particularly, A2CMP can reduce waiting time by 12% compared to other RL-based algorithms.

Keywords: Automotive, Traffic control, Autonomous systems
Abstract: This paper analyzes the performance of decentralized motion planning algorithms for autonomous vehicles, when complementing them with future movement predictions of other vehicles. When multiple predictions are available (e.g., future states as predicted by behavior models (BM) used as prediction model within the controller or external predictions (EP) from prediction algorithms), the challenge is to take advantage of all predictions and find an optimal trajectory and control strategy. To this end, this paper presents a model predictive control (MPC)-framework that combines EP with BM used as prediction model within the MPC and analyzes the effect of combining multiple predictions. The framework is applied to and analyzed for a two-vehicle, multilane lane merging (LM) scenario. The proposed framework is compared with motion planners that use only EP or BM. It is shown that the framework inherits reactive performance properties from the EP and can better anticipate future motions of other road users, which is lacking in the MPC with only BM (e.g., the target vehicle cannot be predicted to do lane changes, but EP can). The integrated framework is also able to use the BM target vehicle proactive behavior, which is lacking when only using EP. Therefore, the integrated framework can handle a wider variety of scenarios and improve traffic behavior.

Keywords: V&V of control algorithms, Automotive
Abstract: Latest automated driving vehicle functionalities present major challenges, where a large number of scenarios needs to be validated for assuring passenger safety and comfort. There is a significant and ongoing effort in legislative, regulatory and industrial standards to cover such necessities, such as UNECE R157 regulation for Automated Lane Keeping Systems (ALKS). At the same time, simulating and testing of such behavior becomes increasingly complex, as vehicles need to cover a large number of driving scenarios. In this paper we present how such testing and processes can be modeled and automated utilizing a tool that assists on quick validation of ALKS functions along with exploration and verification of the space in which the ALKS function is specifically designed to operate within, i.e. its Operational Design Domain (ODD).

Keywords: Autonomous systems, Agents and autonomous systems
Abstract: In the semiconductor manufacturing industry, the path planning of Automated Guided Vehicles (AGVs) is directly related to production efficiency and logistics costs. However, the logistics systems inside semiconductor factories typically present a complex multi-level three-dimensional environment, constrained by factors such as dense equipment distribution, dynamic task generation, and path conflicts. To address this, this paper proposes an AGV path planning method based on a three-dimensional spatiotemporal A* algorithm. This method extends the traditional A* algorithm by incorporating the three-dimensional spatial characteristics and dynamic scheduling requirements of semiconductor manufacturing environments, introducing the time dimension into the path search process. By constructing a spatiotemporal state graph, it dynamically evaluates the reachability of each node in both the three-dimensional space and time dimensions, and implements efficient path optimization with conflict-avoidance strategies. Experimental results demonstrate that the proposed algorithm can effectively solve the path planning problem in semiconductor manufacturing environments, improving the overall operational efficiency of the logistics system. This research provides an efficient and practical path planning solution for intelligent logistics systems in complex manufacturing environments.

Keywords: Biological systems, Predictive control for nonlinear systems
Abstract: This extended abstract presents a scenario-based model predictive control (MPC) scheme designed to control an evolving pandemic via non-pharmaceutical intervention (NPIs). The proposed approach combines predictions and possible pandemic evolutions to decide on a level of NPIs severity to be implemented during multiple weeks to maintain hospital pressure below a prescribed threshold, while minimizing their impact on society. Specifically, we first introduce a compartmental model which divides the population into Susceptible, Infected, Detected, Threatened, Healed, and Expired (SIDTHE) sub-populations. This model is expressive enough to explicitly capture the fraction of hospitalized individuals while preserving parameter identifiability w.r.t. publicly available datasets. Second, we devise a scenario-based MPC scheme with recourse actions that captures potential uncertainty of the model parameters. e.g., due to population behavior or seasonality. Our results show that the scenario-based nature of the proposed controller manages to adequately respond to all scenarios, keeping the hospital pressure at bay also in very challenging situations, whereby conventional MPC methods fail.

Keywords: Energy systems, Emerging control applications, Machine learning
Abstract: Implementing data-driven model predictive control in dynamical system control is challenging due to the need for precise data management, real-time computation, and effective user interaction. This paper introduces a digital control platform that integrates data-driven model predictive control with data management, model identification, and real-time control. Simulations demonstrate its efficiency in regulating swimming pool temperatures, outperforming classical methods.

Keywords: Neural networks, Machine learning
Abstract: This study focuses on the machine learning-based predictive modeling of conversion in an industrial visbreaker unit. Visbreaking is a moderate thermal cracking technique used to lower the viscosity of Vacuum Distillation Residue (VDR), thereby enhancing distillate yield by producing fuel oil and lighter products. The process transforms heavy components into gas, gasoline, and distillates while ensuring that the resulting fuel oil meets required specifications. Principal Component Analysis (PCA) was applied to identify the most influential variables affecting conversion. Subsequently, a predictive model was developed using machine learning algorithms such as Support Vector Regression (SVR) and Random Forest (RF). The performance of these models was evaluated using real-time series data.

Keywords: Transportation systems, Energy systems, Machine learning
Abstract: We introduce an energy management system for an energy hub in an electric vehicle charging station with photovoltaic generation and a battery energy storage system. First, we design a scenario-based stochastic model predictive control that uses probabilistic day-ahead forecasts of charging load and solar generation to achieve cost-optimal operation of the energy hub. Second, the probabilistic forecasts leverage conformal prediction providing calibrated distribution-free confidence intervals starting from machine learning models that generate no uncertainty quantification. In preliminary results, we run a 10-day evaluation in a closed-loop simulated environment to compare the observed cost of the scenario-based stochastic control (100.21%) with two deterministic alternatives: a version with point forecast (100.32%) and a version with perfect forecast (100%).