Keywords: Emerging control applications
Abstract: Distributed game theory and optimal control provide the foundation for the analysis and design of multi-agent systems. A central challenge in this area is game equilibrium seeking, which this talk presents from a systems and control perspective. First, variational analysis, operator theory, and systems theory areemployed to model and analyze equilibrium-seeking algorithms as dynamical systems, thereby leading to a unified framework for their convergence. Dynamic games for constrained systems are considered next. Leveraging optimal control theory, equilibrium control policies are devised in feedback form associated with lifted optimal value functions, thus enabling receding-horizon model-predictive control in dynamic games. The talk concludes with an outlook on data-driven game equilibrium seeking for systems with partially known objective functions, dynamics, and constraints.

Keywords: Markov processes, Network analysis and control, Nonlinear system theory
Abstract: Many complex systems exhibit interactions that depend on not only pairwise connections, but also group structures and memory effects. To capture such complex inter- actions, we develop a unified tensor framework for modeling higher-order Markov chains with memory. Our formulation introduces an even-order paired tensor that links folded and unfolded dynamics and characterizes their steady states and convergence. We further show that a Markov chain with memory can be approximated by a low-dimensional nonlinear tensor-based system and then provide a full system analysis. As an application, we define random walks on hypergraphs where memory naturally arises from the hyperedge structure, providing new tools for analyzing higher-order networks with time-dependent effects.

Keywords: Large-scale systems, Concensus control and estimation, Network analysis and control
Abstract: This paper studies the connectedness of graphons. It highlights that connectedness is related to some spectral property of the graphon-Laplacian operator, which is important for convergence of consensus and other diffusion-based dynamics on large-scale networks. Some equivalent characterizations of connectedness are given, and some subtleties in their definition are discussed through examples.

Keywords: Large-scale systems, Network analysis and control, Identification
Abstract: Fractional-order networks have emerged as powerful modeling frameworks across diverse domains, from neural systems to power grids, offering memory effects and multi-scale dynamics that classical linear time-invariant (LTI) systems cannot capture. While LTI systems admit arbitrary similarity transformations space realizations, whereas the structural constraints imposed by fractional-order dynamics remain unexplored.

In this paper we prove that non-commensurate discrete-time linear fractional-order networks (DTLFON), where fractional orders differ across states, admit only generalized permutation transformations that preserve their structure. Arbitrary similarity transformations destroy the memory property that characterizes fractional differentiation. Consequently, system parameters are uniquely identifiable up to generalized permutation, dramatically reducing the ambiguity compared to LTI systems. This structural rigidity further prevents classical Kalman decomposition for non-commensurate systems.

By contrast, commensurate DTLFON, where all fractional orders are equal, recover the full set of admissible similarity transformations of classical LTI systems, enabling Kalman decomposition but losing uniqueness. These results validate fractional exponents as intrinsic, physically meaningful system realization and their properties in applications of neurophysiology and network science.

Keywords: Concensus control and estimation, Agents networks, Agents and autonomous systems
Abstract: This paper investigates the approximation of the consensus value for a class of continuous-time opinion dynamics models. These models incorporate a state-dependent stubbornness factor that slows down each agent’s opinion update as gets closer to the two extreme viewpoints, 0 or 1. Due to the model’s nonlinear nature, determining the consensus value is a challenging task. The main contributions of this work are: (1) the development of a consensus value approximation approach, (2) the proposal of a relaxation methodology to compute practically useful lower and upper bounds for the consensus value. These contributions are illustrated in synthetic- and realistic-graph simulations.

Keywords: Concensus control and estimation, Quantized systems, Agents networks
Abstract: This paper studies the problem of distributed agreement over a directed communication network with quan- tized information exchange. We develop a quantized push- sum-based agreement algorithm embedded with quantization error correction to enable spatially distributed agents iteratively reach consensus on the average of their initial values. For standard average consensus problem, it has been shown that the introduction of quantized information exchange, even when an error-bounded quantizer is used can result in divergence of the consensus algorithm. Our proposed technique integrates a compression error correction mechanism to the push-sum consensus protocol to ensure accurate ratio tracking despite finite-bit communication. We present that the iterative procedure converges to a neighborhood of all agents’ initial average, with the steady-state characterized by the quantization level, the mixing rate, and spectral properties of the underlying communication graph. The analyses also present performance bounds that quantify the trade-off between the quantization level and solution accuracy. Our theoretical results are validated via numerical simulations illustrating the effectiveness of the proposed method under stringent communication constraints.

Keywords: Network analysis and control, Linear time-varying systems, Decentralized control
Abstract: This paper studies targeted opinion formation in multi-agent systems evolving over signed, time-varying directed graphs. The dynamics of each agent's state follow a Laplacian-based update rule driven by both cooperative and antagonistic interactions in the presence of exogenous factors. We formulate these exogenous factors as external control inputs and establish a suitable controller design methodology enabling collective opinion to converge to any desired steady-state configuration, superseding the natural emergent clustering or polarization behavior imposed by persistently structurally balanced influential root nodes. Our approach leverages upper Dini derivative analysis and Gr{"o}nwall-type inequalities to establish asymptotic convergence for opinion value towards the desired steady state configuration on networks with uniform quasi-strong delta-connectivity. Finally, the theoretical results are validated through extensive numerical simulations.

Keywords: Predictive control for nonlinear systems, Optimization algorithms, Stochastic control
Abstract: This paper presents a risk-aware model predictive control (MPC) framework that integrates Conditional Value-at- Risk (CVaR) into Temporal Deep Unrolling (TDU) for nonlinear vehicle navigation under uncertainty. The proposed formulation combines a standard stabilization cost with a CVaR-based penalty on restricted-region violations so that rare but severe safety-critical events are explicitly penalized. In contrast to our previous TDU-MPC formulation for risk-neutral trajectory tracking, the present study introduces a risk-sensitive objective for safe navigation in stochastic environments. Simulation results show that the proposed risk-aware TDU-MPC reduces the frequency of restricted-region violations compared with its risk neutral counterpart, while maintaining comparable stabilization performance. These results illustrate that TDU provides a differentiable optimization framework for incorporating CVaR-based risk sensitivity into nonlinear MPC.

Keywords: Predictive control for nonlinear systems, Safety critical systems, Robotics
Abstract: Ensuring inter-sample safety for continuous-time robotic systems under discrete Model Predictive Control (MPC) remains a key challenge. Standard discrete MPC risks safety violations between sampling instants, while existing conservative solutions often sacrifice performance. We introduce a novel Segment-Safe Control Barrier Function (SSCBF) integrated into a discrete-time MPC framework (MPC-SS). The SSCBF extends discrete-time CBF theory, providing a formal guarantee of safety along the line segment connecting consecutive predicted states. This linear approximation results in improved safety for the system, while avoiding overconservatism. The method is applied to obstacle avoidance problems, providing a practical choice of SSCBF constraints for both static and dynamic obstacles. Numerical validation is conducted on a 2D double integrator and a nonlinear quadrotor UAV, showing the effectiveness of the proposed approach also in cases where the system's true dynamics deviate significantly from the linear segment evolution. Safety and performance of the proposed method are compared with other CBF based approaches through theoretical analysis and simulations.

Keywords: Predictive control for nonlinear systems, Optimization
Abstract: This paper presents a novel method for computing terminal ingredients, i.e. terminal region and terminal cost, for nonlinear quasi-infinite horizon model predictive control (MPC). The method is optimization-based and aims at finding the volume-maximal terminal regions, which is of interest in MPC applications because it leads to an enlarged region of attraction. In contrast to state of the art methods our approach does not rely on manual tuning nor on approximations of the nonlinear dynamics. We discuss how to set up a suitable optimization problem and how to reformulate it for numerical optimization. The superiority of our method w.r.t. volume-maximal terminal regions is illustrated via a benchmark example.

Keywords: Predictive control for nonlinear systems, Robotics
Abstract: Model Predictive Control (MPC) has established itself as the primary methodology for constrained control, enabling autonomy across diverse applications. While model fidelity is crucial in MPC, solving the corresponding optimization problem in real time remains challenging when combining long horizons with high-fidelity models that capture both short-term dynamics and long-term behavior. Motivated by results on the Exponential Decay of Sensitivities (EDS), which imply that, under certain conditions, the influence of modeling inaccuracies decreases exponentially along the prediction horizon, this paper proposes a multi-timescale MPC scheme for fast-sampled control. Tailored to systems with both fast and slow dynamics, the proposed approach improves computational efficiency by i) switching to a reduced model that captures only the slow, dominant dynamics and ii) exponentially increasing integration step sizes to progressively reduce model detail along the horizon. We evaluate the method on three practically motivated robotic control problems in simulation and observe speed-ups of up to an order of magnitude.

Keywords: Adaptive systems, Predictive control for nonlinear systems, Output regulation
Abstract: This paper introduces a methodology that integrates a multi-model framework with model-free adaptive predictive control (MFAPC), a strategy that utilizes the full-form dynamic linearization (FFDL) data model. The standard FFDL model has no physical meaning because it excludes prior system knowledge and has limited capability for effective system behavior prediction. To overcome these limitations, we propose a novel FFDL model that incorporates a network of local linear models, thereby capturing partial knowledge of the nonlinear system at distinct operating points across the system’s operating surface. Owing to the structural similarity between FFDL and the autoregressive with exogenous input model (ARX), the novel Multi-ARX-MFAPC (M-ARX-MFAPC) can provide a more accurate and globally informed prediction of the system’s behavior, moving beyond a simple non-physical approximation. To verify these benefits, the controller’s effectiveness is demonstrated through an application to a nonlinear MIMO three-tank system. The results show that the proposed M-ARX-MFAPC reduces computational complexity and enhances robustness against measurement noise. By utilizing predefined multi-ARX coefficients, the method eliminates the need for real-time estimation of upcoming model parameter changes allowing a more efficient and reliable control strategy.

Keywords: Predictive control for nonlinear systems, Robust adaptive control, Output feedback
Abstract: The paper considers a method to safely adapt a discrete-time, linear, time-invariant model to a continuous-time nonlinear system with operating-point changes. We proposed a dual model-predictive-control strategy to properly excite the system to generate high-quality data for model adaptation. In the proposed strategy, event-based, model-free reactive feedback control safeguards the setpoint-tracking performance during data acquisition, that is, the system output provably evolves within a prescribed set about the reference signal. Furthermore, the event-based safeguard of the proposed method enhances the data quality by reducing the control action during data acquisition. A numerical simulation of a nonisothermal, continuous stirred tank reactor shows that the proposed strategy ensures that the temperature in the reactor evolves within a prescribed range. In addition, the proposed strategy provides a model with 32% less mean parameter error than a comparable method with a continuous safeguard.

Keywords: Constrained control, Cooperative control, Stability of nonlinear systems
Abstract: This paper investigates the control problem of steering a group of spherical mobile robots to cooperatively transport a spherical object. By controlling the movements of the robots to exert appropriate contact (pushing) forces, it is desired that the object follows a velocity command. To solve the problem, we first treat the robots’ positions as virtual control inputs of the object, and propose a velocity-tracking controller based on quadratic programming (QP), enabling the robots to cooperatively generate desired contact forces while minimizing the sum of the contact-force magnitudes. Then, we design position-tracking controllers for the robots. By appropriately designing the objective function and the constraints for the QP, it is guaranteed that the QP admits a unique solution and the QP-based velocity-tracking controller is Lipschitz continuous. Finally, we consider the closed-loop system as an interconnection of two subsystems, corresponding to the velocity-tracking error of the object and the position-tracking error of the robots, and employ nonlinear small-gain techniques for stability analysis. The effectiveness of the proposed design is demonstrated through numerical simulations.

Keywords: Cooperative control, Machine learning, Optimal control
Abstract: Achieving safe and coordinated behavior in dynamic, constraint-rich multi-agent systems remains a major challenge for learning-based control. Pure end-to-end learning often suffers from poor sample efficiency and limited reliability, while model-based methods depend on predefined references and struggle to generalize. We propose a hierarchical framework that combines tactical decision-making via reinforcement learning (RL) with low-level execution through Model Predictive Control (MPC). In the proposed multi-agent setting, each high-level policy selects a discrete tactical target and a continuous target point inside a structured region of interest (ROI), while MPC provides constraint-aware low-level execution. Tested on a predator–prey benchmark, our approach outperforms end-to-end and shielding-based RL baselines in terms of reward, safety, and consistency, underscoring the benefits of combining structured learning with model-based control.

Keywords: Distributed estimation over sensor nets, Optimization algorithms, Large-scale systems
Abstract: This paper presents a deterministic measurement scheduling method for state estimation in large multi-agent networks operating under communication and energy constraints. Each agent continuously exchanges low-cost relative information, while only a few occasionally take costly absolute measurements such as GPS. The central question is how to dynamically select which subset of agents should activate these high-cost measurements so that estimation accuracy remains comparable to full participation. By modeling the global state as a smooth graph signal, we combine relative and absolute data through a weighted least-squares estimator whose Fisher information quantifies accuracy. Building on ideas from spectral sparsification and online learning, we develop an algorithm that constructs sparse activation schedules whose information matrix remains within a guaranteed spectral bound of the fully active case. We show that the proposed schedule provides uniform accuracy guarantees under the classical D-, E-, and A-optimality criteria, with the required number of active agents scaling linearly with the signal’s bandlimited dimension. Simulations on random geometric networks confirm the spectral approximation guarantee and show that the method schedules only a small fraction of agents to take costly absolute measurements relative to full participation.

Keywords: Cooperative autonomous systems, Sliding mode control, Nonlinear system theory
Abstract: This paper presents a control strategy for reshaping a multi-agent formation to compress and safely pass through a rectangular window. The formation follows a two-level leader-follower structure within an affine formation framework. The level-I leader guides the agents through the centre of the window using a nonlinear bearings-only guidance law, while the level-II leaders control the reshaping of the formation. To track, both the paths of the leaders and the reshaped target formation, the followers utilize proposed control laws developed using sliding mode control design. The resulting formation can change over time, allowing all possible affine transformations, and needs no prior path planning, with potential applications for autonomous UAV operations in GPS-denied environments. The proposed strategy is validated using MATLAB simulations, demonstrating that the single integrator-based multi-agent system achieves the desired performance under bounded control inputs.

Keywords: Autonomous systems, Aerospace, Cooperative control
Abstract: This paper develops a displacement-based finite-time formation control law for agents arranged over a directed cycle. Building upon a finite-time extension of the classical cyclic pursuit algorithm with heterogeneous edge weights, we prove that, for any prescribed convergence time and desired formation shape at a specified Euclidean location, the proposed control law guarantees finite-time attainment of the formation. Notably, allowing heterogeneity with at most one negative edge weight expands the set of achievable formation reference points to the entire Euclidean space. Furthermore, we extend the approach to synthesize coordinated motion patterns in groups of unicycles. Theoretical results are corroborated through numerical simulations in both planar and three-dimensional settings.

Keywords: Optimal control, Cooperative autonomous systems, Agents and autonomous systems
Abstract: This paper presents a simulation-based framework for selecting rendezvous points for heterogeneous robot teams under polygonal geofence constraints. Building on earlier time-optimal rendezvous work, we combine a geofence-aware bidirectional astar{} formulation with distinct visibility graphs for aerial and ground robots, a simplified but explicit energy model for both classes, and a scalarized time--energy objective. These components are combined into a single pipeline that optimizes a lambda-weighted trade-off between maximal arrival time and total energy consumption over the evaluated candidate set.

By separating aerial and ground constraints and evaluating all candidate rendezvous nodes on the visibility graph, the method systematically explores the trade-off between temporal efficiency and energetic cost. A user-selected weighting factor~lambda controls the balance between temporal efficiency and energy consumption. Batch simulations reveal collision--free, near-synchronous rendezvous trajectories and show that the scalarization yields consistent time--energy trade-off responses across~lambda within the evaluated configurations. The approach provides a reproducible graph-based framework for energy-aware rendezvous planning and is amenable to integration into ROS~2 mission pipelines.

Keywords: Agents and autonomous systems, Game theoretical methods, Autonomous systems
Abstract: Classical fixed-point theorems guaranteeing the existence of equilibria in noncooperative games, such as those of Brouwer and Kakutani, rely on convexity assumptions. Nevertheless, many games are inherently nonconvex, rendering such results inapplicable. In this work, we establish the existence of equilibria in games by employing tools from algebraic topology. We introduce a local equilibrium notion, the first-order Nash equilibrium, that captures stationary strategy profiles. Using the Lefschetz fixed-point theorem, we prove that at least one local equilibrium exists under mild assumptions, even when cost functions are nonconvex. We further refine this result through the Nielsen extension of the Lefschetz theorem, which characterizes the persistence of equilibria under continuous deformations of the game.

Keywords: Game theoretical methods, Nonlinear system theory, LMI's/BMI's/SOS's
Abstract: In noncooperative systems, it is common for each agent to have multiple objectives, leading to a Nash equilibrium that forms a set rather than an isolated point. Analyzing stability of such a set under piecewise pseudo-gradient dynamics is often complicated. To address this, we propose a nonlinear, nonbijective transformation that simplifies the stability analysis. By showing that the origin (singleton) is stable in the transformed system, we guarantee stability of the Nash equilibrium set in the original system. A numerical example is provided to illustrate our results.

Keywords: Game theoretical methods, Autonomous systems, Optimal control
Abstract: We study a two-player dynamic Stackelberg game where the follower's intention is unknown to the leader. Classical formulations of the Stackelberg equilibrium (SE) assume that the follower's best response (BR) function is known to the leader. However, this is not always true in practice. We study a setting in which the leader receives updated beliefs about the follower BR before the end of the game, such that the update prompts the leader and subsequently the follower to re-optimize their strategies. We characterize the optimality guarantees of the SE solutions under this belief update for both open loop and feedback information structures. Interestingly, we prove that in general, assuming an incorrect follower's BR may lead to a lower leader cost over the entire game than knowing the true follower's BR. We support these results with numerical examples in a linear quadratic (LQ) Stackelberg game, and use Monte Carlo simulations to show that the instances of incorrect BR achieving lower leader costs are non-trivial in collision avoidance LQ Stackelberg games.

Keywords: Game theoretical methods, Machine learning, Transportation systems
Abstract: Understanding how agents coordinate or compete from limited behavioral data is central to modeling strategic interactions in traffic, robotics, and other multi-agent systems. In this work, we investigate the following complementary formulations of inverse game-theoretic learning: (i) a Closed-form Correlated Equilibrium Maximum-Likelihood estimator (CE-ML) specialized for 2x2 games; and (ii) a Logit Best Response Maximum-Likelihood estimator (LBR-ML) that captures long-run adaptation dynamics via stochastic response processes. Together, these approaches span the spectrum between static equilibrium consistency and dynamic behavioral realism. We evaluate them on synthetic "chicken-dare" games and traffic-interaction scenarios simulated in SUMO, comparing parameter recovery and distributional fit. Results reveal clear trade-offs between interpretability, computational tractability, and behavioral expressiveness across models.

Keywords: Game theoretical methods, Optimization algorithms, Randomized algorithms
Abstract: We study generalized Nash equilibrium (GNE) problems in games with quadratic costs and individual linear equality constraints. Departing from approaches that require strong monotonicity and/or shared constraints, we reformulate the KKT conditions of the (generally non-monotone) games into a tractable convex program whose objective satisfies the Polyak–Łojasiewicz (PL) condition. This PL geometry enables a distributed gradient method over a fixed communication graph with global geometric (linear) convergence to a GNE. When gradient information is unavailable or costly, we further develop a zero-order fully distributed scheme in which each player uses only local cost evaluations and their own constraint residuals. With an appropriate step size policy, the proposed zero-order method converges to a GNE, provided one exists, at rate O(1/t).

Keywords: Emerging control theory, Game theoretical methods, Optimal control
Abstract: In this paper, we present a unified framework for co-synthesizing control and communication strategies in discrete-time networked control systems (NCS). We propose a discrete-time zero-sum dynamic game formulation to jointly determine the control and communication policies. In this setting, the control policy functions as the minimizing player that seeks to reduce the overall cost, while the threshold policy that governs feedback transmission instances functions as the maximizing player. The Nash equilibrium solution is obtained by solving a discrete-time game Riccati-like equation (GRLE), which provides coupled optimal conditions for control and communication design. Using this solution, we formulate two feedback scheduling conditions within the proposed framework. The first condition determines transmission instants based on the control error, defined as the deviation between the periodically and aperiodically updated control inputs. The second condition is derived from the state error, leading to a state error-based feedback scheduling formulation that yields a suboptimal implementation. Leveraging Lyapunov stability theory, we derive sufficient conditions that ensure closed-loop stability of the system under the aperiodically updated Nash policies. Numerical studies on inverted pendulum demonstrate the effectiveness of the proposed co-synthesis method in achieving communication efficiency while preserving desired control performance.

Keywords: Predictive control for nonlinear systems, Autonomous robots, Modeling
Abstract: This paper presents a guidance strategy for assisting Blind and Visually Impaired (BVI) users in navigation. The system comprises an autonomous vehicle equipped with a Smart Tether System (STS). The dynamics between the robot and the user are formulated as a spring–mass model that captures the key human–robot interaction. The proposed interaction model is validated through experimental trials with multiple participants, enabling calibration of the model to match real-world user motion. A Model Predictive Planner optimizes the robot's trajectory, explicitly considering both the robot and the user's movements. In this way, it plans a safe and effective maneuver for both entities. The results indicate that the integrated system effectively balances path tracking accuracy and user safety, demonstrating promise for robust guidance in diverse real-world settings.

Keywords: Sensor and signal fusion, Stochastic filtering
Abstract: Accurate positioning, navigation, and timing (PNT) are fundamental to the operation of modern technologies and a key enabler of autonomous systems. A very important component of PNT is the Global Navigation Satellite System (GNSS) which ensures outdoor positioning. Modern research directions have pushed the performance of GNSS localization to new heights by fusing GNSS measurements with other sensory information, mainly measurements from Inertial Measurement Units (IMU). In this paper, we propose a loosely coupled architecture to integrate GNSS and IMU measurements using a Factor Graph Optimization (FGO) framework. Because the FGO method can be computationally challenging and often used as a post-processing method, our focus is on assessing its localization accuracy and service availability while operating in real-time in challenging environments (urban canyons). Experimental results on the UrbanNav-HK-MediumUrban-1 dataset show that the proposed approach achieves real-time operation and increased service availability compared to batch FGO methods. While this improvement comes at the cost of reduced positioning accuracy, the paper provides a detailed analysis of the trade-offs between accuracy, availability, and computational efficiency that characterize real-time FGO-based GNSS/IMU fusion.

Keywords: Feedback linearization, Neural networks, Maritime
Abstract: Addressing the control and maneuverability of surface vehicles with dynamically changing mass distributions is still an open problem. To solve the problem, we propose an intelligent controller for the path-following problem of a surface vehicle, which is controlled through mass distribution. This means that one of the control inputs is mass-centric. Specifically, we developed a Lyapunov-based nonlinear control scheme to enable an unmanned vessel to follow a smooth path according to a line-of-sight guidance law. The control inputs consist of the thrust force for forward motion and the position of a sliding mass that shifts the system’s overall mass distribution. Artificial neural networks are employed to estimate unmodeled dynamics and external disturbances. Simulation results demonstrate the effectiveness of the proposed controller in guiding the vessel along the desired path with minimal error.

Keywords: Identification for control, Maritime, Neural networks
Abstract: This paper investigates the application of Neural Network (NN)-based system identification for Nonlinear Model Predictive Control (NMPC) of an underactuated Unmanned Surface Vehicle (USV). We focus on practical NMPC implementation by employing neural networks to approximate system dynamics, specifically using a multi-step ahead prediction in a single inference pass. We compare a multi-step MLP predictor and a multi-step LSTM predictor against a conventional recursive single-step prediction approach. Simulations, using data from a Clearpath Robotics Heron USV, demonstrate that multi-step predictors achieve comparable trajectory tracking accuracy to the single-step approach. Critically, the multi-step predictors reduce the computational load of calculating the gradients of the optimal cost function with respect to the control input trajectory by up to an order of magnitude. This significant reduction in computational burden facilitates real-time NMPC implementation and enhances its deployment in dynamic and challenging environments.

Keywords: UAV's, Robotics
Abstract: This paper presents a cross-modal learning framework that exploits complementary information from depth and grayscale images for robust navigation. We introduce a Cross-Modal Wasserstein Autoencoder that learns shared latent representations by enforcing cross-modal consistency, enabling the system to infer depth-relevant features from grayscale observations when depth measurements are corrupted. The learned representations are integrated with a Reinforcement Learning-based policy for collision-free navigation in unstructured environments when depth sensors experience degradation due to adverse conditions such as poor lighting or reflective surfaces. Simulation and real-world experiments demonstrate that our approach maintains robust performance under significant depth degradation and successfully transfers to real environments.

Keywords: Aerospace, Optimal control, Linear systems
Abstract: Agile earth observation satellites employ multiple actuators to enable flexible and responsive imaging capabilities. While significant advancements in actuator technology have enhanced satellites' torque and momentum, relatively little attention has been given to control strategies specifically tailored to improve satellite agility. This paper provides a comparative analysis of different Model Predictive Control (MPC) formulations and introduces an augmented-MPC method that effectively balances agility requirements with hardware implementation constraints. The proposed method achieves the high-performance characteristics of nonlinear MPC while preserving the computational simplicity of linear MPC. Numerical simulations and physical experiments are conducted to validate the effectiveness and feasibility of the proposed approach.

Keywords: Stability of nonlinear systems, Quantized systems
Abstract: We address the output tracking problem for uncertain nonlinear systems under finite-level input quantization. A prescribed performance control (PPC) scheme is utilized to guarantee prescribed (user-defined) transient and steady-state response under quantized inputs. Stability and prescribed performance are guaranteed under a simple feasibility condition on the maximum and minimum quantization levels, while the control design remains entirely independent of the quantization characteristics. The proposed scheme preserves the low-complexity attribute of PPC providing low computational cost and easy implementation. Simulation results verify the approach and illustrate that proper controller gain selection can further improve hysteresis against chattering behavior.

Keywords: Stability of nonlinear systems, Optimization, Optimization algorithms
Abstract: This paper studies equality-constrained composite minimization problems. This class of problems, capturing regularization terms and inequality constraints, naturally arises in a wide range of engineering and machine learning applications. Inspired by recent results, we introduce the proportional--integral proximal gradient dynamics (PI--PGD): a closed-loop system where the Lagrange multipliers are control inputs and states are the problem decision variables. First, we establish the equivalence between the stationary points of the minimization problem and the equilibria of the PI--PGD. Then for the case of affine constraints, by leveraging tools from contraction theory we provide a comprehensive convergence analysis for the dynamics, showing linear--exponential convergence towards the equilibrium. In other words, we show that the distance between each solution and the equilibrium is upper bounded by a function that first decreases linearly and then exponentially. Our findings are illustrated numerically on a set of representative examples, including an exploratory application to nonlinear equality constraints.

Keywords: Stability of nonlinear systems, LMI's/BMI's/SOS's, Neural networks
Abstract: This paper analyzes the stability properties of Gated Recurrent Units (GRUs) under constant input signals. We propose Linear Matrix Inequality (LMI)-based conditions solely involving the GRU weights and biases. Paired with the knowledge of the region of validity of quadratic sector conditions associated to the GRU nonlinearities, the feasibility of these LMIs guarantees local exponential stability of a given equilibrium and provides an inner-approximation of its domain of attraction. Furthermore, by exploiting the inherent convergence properties of the GRU dynamics, we specialize the conditions to also ensure global asymptotic stability plus local exponential stability of the given equilibrium. Finally, we compare our conditions with results from the literature, which are shown to impose stronger constraints on the GRU parameters.

Keywords: Stability of nonlinear systems, Computer aided control design, Neural networks
Abstract: Estimating the Region of Attraction (RoA) for nonlinear dynamical systems is a fundamental problem in control theory, with direct implications for stability analysis and safe controller design. Traditional approaches rely on analytically derived Lyapunov functions, which are often conservative and challenging to construct for highly nonlinear systems. In this work, we propose a data-driven framework for learning and verifying RoA estimates for nonlinear systems using neural-network-based Lyapunov functions. Our method employs a composite Lyapunov function that combines a quadratic term with a neural-network-based component, providing both structure and flexibility. We introduce a novel homogeneous loss function for training, which removes the imbalance typically caused by the two non-homogeneous Lyapunov conditions. Together, these two aspects enable efficient training of the Lyapunov candidate. To guarantee the correctness of the learned Lyapunov function, we employ a Satisfiability Modulo Theories (SMT) solver to formally verify the stability results. Lastly, we perform a deeper analysis near the origin to overcome numerical artifacts, ensuring strict asymptotic stability. We demonstrate the effectiveness of our approach on benchmark nonlinear systems, showing that it significantly reduces conservatism compared to traditional Lyapunov methods while maintaining verifiability. This framework bridges the gap between function approximation and stability certification, paving the way for scalable safety analysis in learning-based control and safety-critical applications.

Keywords: Stability of nonlinear systems, Optimization algorithms, Complex systems
Abstract: Timescale separation is a powerful tool for analyzing interconnected dynamical systems. Meanwhile, operator theory provides a general framework for studying the convergence of iterative methods, including algorithms in optimization, learning, and control. In this paper, we bridge these two areas by establishing timescale separation results for a class of interconnected fixed-point iterations. As is customary in timescale separation, our results involve auxiliary systems that separately capture the dynamics induced by the slow and fast operators, arising from the original interconnection in the limit as the timescale parameter tends to zero. In particular, by assuming contractivity and paracontractivity of the respective auxiliary systems, we derive explicit bounds on the timescale parameter that guarantee linear convergence of the original, interconnected system. To illustrate the applicability of our result, we employ it to prove the convergence of a feedback optimization scheme.

Keywords: Stability of nonlinear systems
Abstract: The Koopman operator is a powerful approach to global stability analysis of nonlinear systems, which provides a systematic procedure for Lyapunov function design. In this framework, Lyapunov functions are obtained through the eigenfunctions of the Koopman operator associated with the eigenvalues of the Jacobian matrix at the equilibrium. In practice, the eigenfunctions are approximated via a finite-dimensional representation of the operator, and there is no guarantee that the approximated spectrum accurately matches the true one. In this paper, we develop a kernel-based method to compute Koopman eigenfunctions and preserve the spectrum of the Jacobian matrix. This approach is suitable for stability analysis of high-dimensional systems thanks to the kernel trick. Moreover, the Lyapunov function candidate is validated through a scenario-based optimization technique that provides a reliable estimation of the region of attraction of the system.

Keywords: Hybrid systems, Stability of hybrid systems, Observers for nonlinear systems
Abstract: This paper investigates the stability of a class of hybrid systems featuring rapidly occurring discrete transitions, analyzed through the lens of singular perturbation theory. The considered model consists of the interconnection of two hybrid subsystems, a timer governing the jump instants, and discrete variables determining the indices of the jump maps. The evolution of these variables during flows is described by singularly perturbed differential equations, where smaller values of the perturbation parameter correspond to increased jump frequency. In the limiting case of this parameter, the system is decomposed into a quasi steady-state subsystem, modeled by a continuous differential equation without jumps, and a boundary-layer subsystem governed by purely discrete dynamics. Building upon our previous work that established practical stability, this paper derives sufficient conditions for the asymptotic stability of a compact attractor by imposing suitable assumptions on both the quasi steady-state and boundary-layer subsystems. As an application, we address the design of observers for nonlinear systems with time-sampled measurements and show that detectability of the system ensures asymptotic stability of the estimation error under frequent sampling.

Keywords: Stability of hybrid systems, Switched systems, Stability of nonlinear systems
Abstract: We show how global input-to-state (ISS) can be concluded for a class of piecewise-affine (PWA) systems from a regional ISS property. Specifically, we combine regional ISS-Lyapunov functions to conclude global practical ISS (ISpS). Afterwards, we provide additional assumptions on the region containing the origin to arrive at global ISS.

Keywords: Switched systems, Lyapunov methods, Stability of hybrid systems
Abstract: This paper studies the stabilization of switched nonlinear systems with both unstable subsystems under time-dependent switching. Motivated by the observation that stability in such systems requires both frequent switching and proper activation ratios, we introduce a new stability condition based on average activation time (AAT), which can be viewed as a relaxation of strict periodic switching. Technically, we employ vector-valued Lyapunov functions (VLFs) combined with hybrid timers to construct a Lyapunov framework whose overall magnitude decreases despite the instability of each subsystem. We further derive explicit and easily verifiable stability conditions, which are validated through a numerical example.

Keywords: Switched systems, Stability of hybrid systems, Differential algebraic systems
Abstract: We are interested in the relationship between the stability properties of a singularly perturbed switched system and those of the corresponding limiting switched system. In some recent work, it has been shown that for a large class of singularly perturbed systems the worst case exponential growth rate for sufficiently small perturbation parameters is lower bounded by the growth rate of the limiting switched system. However, examples show that there could be a positive gap between these bounds. Here we want to investigate this gap further by first observing that the limiting switching system is in fact a switched differential-algebraic equation (switched DAE) for which numerous stability results are already available. Based on the underlying geometric structure of the switched DAE we introduce the concept of structurally aligned singular perturbations and show for the commuting case that indeed there is no gap. We also provide an example which has a commuting limiting switched DAE, but for which the singular perturbations are not structurally aligned and there is indeed a gap between the growth bounds.

Keywords: Switched systems, Lyapunov methods, Stability of hybrid systems
Abstract: This manuscript presents novel stabilizability-safety conditions for switched linear systems with arbitrary and unobservable underlying switching signals. The goal is to stabilize a discrete-time switched linear systems while avoiding certain (symmetric) cones of the state-space, supposed to be unsafe/undesirable for the system under analysis. The theoretical results are based on the use of directed and labeled graphs to design piecewise-quadratic control-barrier Lyapunov functions and piecewise linear state-feedback controllers. This results in novel matrix inequalities conditions in the form of BMIs, which are solved iteratively. Numerical case studies demonstrate how the suggested methods work and reveal the connections among graph size, system robustness, and closed-loop performance.

Keywords: Switched systems, Hybrid systems, Lyapunov methods
Abstract: In this paper, we present a novel data-driven model-reference control design approach for unknown switched linear systems with dwell-time switching. The proposed strategy aims to regulate the switched system such that the closed-loop behavior matches the desired dynamics defined by a reference model. By treating the switching signal as an external disturbance with sufficiently long dwell times between consecutive switching instants, we develop a model-reference design procedure that provides stability guarantees. Finally, simulation results demonstrate the effectiveness of the proposed approach.

Keywords: V&V of control algorithms, H2/H-infinity methods, Robust control
Abstract: The H∞ norm is a widely used metric to assess the performance of Linear Time-Invariant (LTI) systems. In the presence of uncertainties, guaranteed bounds on the worst-case H∞ norm can be efficiently computed by means of µ-based techniques. However, it is more difficult to obtain a guaranteed lower bound on all possible realizations of an uncertain linear system, particularly in the MIMO case. The precise characterization of such a bound is nevertheless very useful for quickly identifying the parametric regions where an H∞ criterion is not satisfied, and then for deducing reliable bounds on the failure probability of a control law. A new sufficient condition is proposed in this article for calculating this lower bound. Unlike previous results, our approach does not require any inversion of the system, which therefore does not need to be invertible or even square. The central result of the paper is based on an original use of Weyl's inequalities, which allow to bound more precisely the maximum singular value (rather than the set of singular values) of the uncertain transfer matrix. This new result is then integrated into a Branch-and-Bound scheme to improve the calculation of guaranteed bounds on the probability that a given H∞ performance requirement is verified or not. Since it quantifies the probability of rare events, such a probabilistic analysis offers a significant advantage over deterministic worst-case approaches by avoiding the invalidation of a controller on the basis of unlikely failures.

Keywords: V&V of control algorithms, Optimization, Signal processing
Abstract: During development of control engineering applications, the simulation-based or experimental testing of the closed-loop control system plays a crucial role, both for tuning and verification of the controller. Thereby, the choice of an appropriate test signal is of crucial importance for evaluating performance. For this purpose, a model free signal generation framework to generate worst-case reference trajectories for testing nonlinear closed-loop systems is presented in this paper. In our approach, no internal knowledge of the controller or plant is assumed, making it applicable even for black box systems. The generated input signals are physically plausible and represent real-world scenarios, ensuring practical relevance. Central to the method is a deterministic signal generator that enables systematic variation and finite dimensional specification of input trajectories. Two complementary strategies are employed to identify critical test cases: a probabilistic Monte Carlo approach provides statistical performance bounds and robustness margins, and an optimization-based method efficiently yields inputs likely to reveal control limitations, while minimizing simulation or measurement effort. The methodology is demonstrated on a pneumatic valve system, both in simulation and on a physical test bench.

Keywords: Control over communication, Control over networks, Communication networks
Abstract: Functional Mock-up Units in the Loop (FMUiL) is an open-source Python package that combines the Functional Mock-up Interface (FMI) and OPC UA standards. It allows co-simulation between software, hardware, and models created with different tools. This paper builds on FMUiL by introducing a set of bridge servers, that link the OPC UA communication layer with various industrial communication protocols. This improvement enables a wide range of external hardware to connect to the co-simulation environment, and allows for comparing communication protocol performance in control applications. In this study, bridge servers are configured for three popular industrial protocols: Modbus RTU, Modbus TCP, and MQTT. To demonstrate its effectiveness, the proposed architecture is validated with a case study that involves a simulated lube-oil cooling system connected to an external discrete-time proportional–integral (PI) controller using each of the supported protocols. The results demonstrate the flexibility of the extended FMUiL framework for hardware-in-the-loop testing, and its potential for broader applications in cyberphysical and industrial automation systems.

Keywords: Intelligent systems, Computer aided control design, Process control
Abstract: Simulation is central to model-based design and systems engineering, yet many workflows still rely on time-consuming manual configuration, parameter tuning, and experiment setup. Recent advances in generative AI have enabled agents that interpret natural-language instructions, plan tasks, invoke external tools, and access data from external resources. This creates opportunities to reduce engineering effort through automation, but it also raises a key question: can such agents reliably perform simulated experiments that require precise, deterministic computational steps rather than natural-language reasoning? These workflows depend on accurate tool execution and correct interpretation of intermediate numerical results, making it unclear whether current AI agents can manage them in a way that meaningfully improves engineering productivity.

This work investigates how effectively current LLM-based agents can plan and orchestrate multi-step tool chains. To support this study, we developed and released a prototype agent architecture, control-agent, along with a complementary toolset, agent-control-toolbox, designed for control-oriented workflows with agents. We evaluate the agent on benchmark tasks representative of common control-engineering tasks. The agent is evaluated on six benchmark tasks of increasing complexity, ranging from single-tool simulations to iterative controller tuning requiring multiple chained tool calls. Across six benchmark tasks, all runs satisfied the response evaluators, but redundant and non-optimal tool calls appeared as task complexity increased.

The results show that while the agent can reliably complete simpler workflows, its performance degrades as tool-chain complexity increases. Overall, the findings indicate that modern agent frameworks are mature enough to support tool-centric workflows. However, robust and efficient performance still requires new insights into how to design and implement agentic systems and how to efficiently and...

Keywords: V&V of control algorithms, Model validation
Abstract: We propose a human in the loop approach for black-box testing of Functional Mock-up Units (FMUs) using Large Language Models (LLMs). The goal is to reduce the manual effort in defining test scenarios for dynamic simulation models and to improve the interpretability of results. The approach takes the functional and interface specifications of an FMU as input, and prompts an LLM to generate structured scenario goals in Given-When-Then format that define the initial input conditions of the simulation, a possible change in those conditions, and the expected output behaviour of the system against those changes. The corresponding scenario plans specify input patterns and add assertion oracles that describe expected output patterns defined in test scenario goals. The approach generates a complete input time series for the scenario plans, runs the FMU simulation, and evaluates assertions on the recorded outputs. It produces human-readable logs and plots that show statistics for each scenario with overlays, aggregate pass rates, and per-goal outcomes. The generated tests and results are stored for evaluation and later re-execution. We evaluate the approach on a Lube Oil Cooling system and discuss design choices that make the approach practical for everyday use. Results suggest that LLM-assisted test generation can facilitate automatic test design and verification of dynamic simulation models.

Keywords: System reconfiguration, Modeling
Abstract: Sea trials are essential for verifying ship seaworthiness but require substantial resources. Virtual commissioning offers a promising alternative by enabling early testing and analysis of ship subsystems in a simulated environment. However, models are often developed independently, in different tools, and for different purposes, which creates challenges in interoperability, version control, and protection of proprietary information. This paper presents a model-based systems engineering (MBSE) framework that brings together established standards and tools, including the 150% model pattern from Product Line Engineering (PLE), the Functional Mockup Interface (FMI), and the System Structure and Parameterization (SSP) standard. The work focuses on integrating and demonstrating these existing concepts within a single environment that enables systematic configuration, management, and reuse of modular simulation models across different software toolchains. The framework is demonstrated on a ship engine case, where subsystem models from AVL CRUISEtextsuperscript{TM} M are exported as Functional Mockup Units (FMUs) and organized within the CATIA Magic MBSE environment. This setup enables variant configuration and SSP export while maintaining model confidentiality and consistency.

Keywords: Adaptive systems, Agents and autonomous systems, Game theoretical methods
Abstract: Incentive design constitutes a foundational paradigm for influencing the behavior of strategic agents, wherein a system planner (principal) publicly commits to an incentive mechanism designed to align individual objectives with collective social welfare. This paper introduces the Regret-Minimizing Adaptive Incentive Design (RAID) problem, which aims to synthesize incentive laws under information asymmetry and achieve asymptotically minimal regret compared to an oracle with full information. To this end, we develop the RAID algorithm, which employs a switching policy alternating between probing (exploration) and estimate-based incentivization (exploitation). The associated type estimator relies only on the weakest excitation condition required for strong consistency in least squares estimation, substantially relaxing the persistence-of-excitation assumptions previously used in adaptive incentive design. In addition, we establish the strong consistency of the proposed type estimator and prove that the incentive obtained asymptotically minimizes the planner’s average regret almost surely. Numerical experiments illustrate the convergence rate of the proposed methodology.

Keywords: Mechatronics, Flexible structures, Modeling
Abstract: In [1], we developed a higher-dimensional state-space model using Koopman operators to predict and monitor the nonlinear dynamics of a cable slab. In [2], we analyze the dynamics and bend geometry of a cable slab via trained neural networks. These studies focused on predicting slab motion but did not address the reaction forces that directly degrade positioning precision. In this paper, we present a method for visually estimating the cable slab reaction forces that affect positioning precision during both cyclic and rectilinear motion profiles. This paper introduces an analytical model of the reaction forces considering the cable slab dynamics that affect the kinematics of the attached markers. Utilizing the available marker positions, their kinematics are obtained and used to calculate the cable slab angular motion and its instantaneous center of rotation. Offline image processing is used to extract marker positions, from which we compute marker kinematics, the slab’s angular motion, and its instantaneous center of rotation. Using this geometric information, we express the acceleration of the cart–slab contact point with respect to the center of rotation and apply Newton’s second law with two tunable parameters: one representing the apparent mass seen by the cart, and another capturing pre-loading effects. These parameters can be fixed or adaptively updated. The resulting model provides direct estimates of cable-slab reaction forces and can be embedded in feedforward or feedback compensation schemes to improve motion-system precision. Experiments confirm the method’s accuracy, provided the camera maintains line-of-sight with the slab.

Keywords: Emerging control applications, Adaptive control, Identification for control
Abstract: High-performance computing (HPC) applications generate unpredictable and highly parallel input/output (I/O) operations. Therefore, storage systems in HPC stacks adopt complex, multi-layer architectures to process large volumes of data while meeting requirements for scalability, performance and parallelism. Yet overall application throughput is frequently limited by I/O bottlenecks across file systems and storage tiers. When many parallel jobs perform I/O concurrently, contention occurs and performance degrades. Static, model-based proportional–integral (PI) controllers can mitigate this only around a fixed operating point, but they fail when workload characteristics or device conditions change. This paper investigates event-triggered, adaptive PI tuning approaches that compute safe, near-optimal feedback policies from system load metrics and real-time measurements. These approaches control the unpredictable workload dynamics of the system and keep them within specific limits. Experiments on an HPC storage setup show that the adaptive approaches improve performance under changing workloads, compared to state-of-the-art PI controller with static gains.

Keywords: Robust adaptive control, Adaptive control, Lyapunov methods
Abstract: We extend the direct adaptive control approach by adding feedthrough of an error term to its update law. This reduces undesired oscillations of the calculated weights which can arise due to high adaptation rates. We show robustness (here: boundedness of all signals) of the closed loop against sufficiently fast, unmodeled dynamics and identify a class of harmless unmodeled dynamics for which adding feedthrough to the update law does not compromise the robustness margin if the damping injected into the adaptation process by the σ-modification is reduced as the gain of the feedthrough is increased. For harmless unmodeled dynamics, frequency-domain analysis of the purely linear case shows that such margin-preserving parameter adjustment simultaneously improves bandwidth, low-frequency disturbance rejection and attenuation of undesired oscillations.

Keywords: Constrained control, Adaptive control, Nonlinear system theory
Abstract: This paper investigates a predefined-time asymmetric prescribed performance adaptive control scheme for a class of uncertain nonlinear systems subject to unknown dead-zone nonlinearities, parametric uncertainties, and external disturbances. A novel tuning function and a tangent-based error transformation are proposed to achieve asymmetric prescribed performance with enhanced flexibility compared to existing methods. Dynamic surface control is employed to avoid the computation of derivatives of virtual control laws. In addition, unified adaptive laws are developed to simultaneously compensate for system uncertainties, disturbances, and dead-zone effects without requiring prior knowledge of their bounds. The proposed approach guarantees that the tracking error remains within the prescribed performance bounds and converges within a predefined settling time, independent of initial conditions. Lyapunov-based analysis shows that all closed-loop signals are semi-globally uniformly ultimately bounded. Simulation results verify the effectiveness of the proposed control scheme.

Keywords: Process control, Energy systems, Optimization
Abstract: This paper presents an adaptive steady-state control framework for improving the energy efficiency of industrial compressor trains. The approach combines a multidimensional piecewise linear (PWL) model with mixed-integer linear programming (MILP) to determine both compressor activation (on/off) and speed setpoints. Steady-state operating data are used to construct a PWL approximation of the nonlinear compressor model. The MILP formulation combines convex-combination encoding of the PWL model with binary activation constraints to minimize power while satisfying process constraints. To reduce plant–model mismatch from sparse sampling and drift, a data-driven online update scheme refines setpoints using gradient feedback and incorporates improved points to the PWL model. The method is evaluated in closed-loop simulations of a four-compressor train in an industrial process simulator. Results show energy savings of up to 42% at low load, consistently comparable or superior performance to the baseline strategy at other operating points, and better performance under gradual plant drift. Computation times remain on the order of seconds, confirming feasibility for practical deployment.

Keywords: Biological systems, Machine learning, Chemical process control
Abstract: In bioprocesses, as those based on microalgae, effective control plays a decisive role, as productivity ultimately depends on living cells that act as microscopic production units. Their intricate and self-regulated internal dynamics, however, introduce significant challenges to process control. Achieving and maintaining stable conditions, such as nutrient concentration, pH, and dissolved oxygen (DO), is essential for optimal cell growth and productivity, demanding the use of advanced control strategies. However, the integrated and coordinated regulation of pH and DO in microalgae photobioreactors remains largely unexplored in the current literature. This results in a highly nonlinear, multivariable control problem subject to multiple disturbances. To address this complexity, we propose an offline Multi-Agent Reinforcement Learning (MARL) approach that learns from historical data generated by expert systems, such as classical feedback controllers. The proposed strategy was validated in a real-world open, semi-industrial-scale photobioreactor at the University of Almería. Overall, this study demonstrates the strong potential of RL-based approaches for advanced bioprocess control and sets the foundation for their wider application to other complex, nonlinear, and disturbance-prone multivariable systems.

Keywords: Biomolecular systems, Nonlinear system theory, Signal processing
Abstract: We propose a biomolecular signal buffer (BSB) that stores and propagates molecular information in a modular manner. The BSB features a write-store-read architecture and is formulated as a simple chemical reaction network compatible with dynamic DNA nanotechnology. To clarify the operational principle of the BSB, we applied singular perturbation theory and developed a two-timescale model that rigorously explains how transient input signals are processed and retained within the system. We also demonstrated that the output asymptotically converges to the reference concentration and remains stable over an infinite time horizon under an appropriate parameter design.

The utility of the BSB is demonstrated through two applications. First, we integrated it into a molecular adder circuit, demonstrating that buffer-assisted computation ensures a stable and accurate output. Second, we designed a classical conditioning circuit that models associative learning by combining the BSB with a stimulus-processing module. Numerical simulations demonstrated that the system successfully acquired associative behavior, transitioning from YES logic (pre-learning) to OR logic (post-learning) in response to repeated co-stimulation. These results highlight the potential of the proposed BSB as a foundational buffering mechanism for constructing adaptive and intelligent biomolecular systems.

Keywords: Optimal control, Modeling, Biological systems
Abstract: Banana and plantain production is severely threatened by Radopholus similis, a parasitic nematode that lives and propagates within plant roots, causing substantial damage and making effective control particularly challenging. Building on a recently introduced and analyzed model, we upgrade the framework to account for the combined effect of two biological control levers against Radopholus similis: a biostimulant effort reducing infestation rate, and a biopesticide effort directly lowering nematode density. To investigate the most effective biocontrol strategies, we formulate an optimal control problem and derive first-order necessary conditions using Pontryagin’s Maximum Principle. The problem is solved numerically via direct methods using Bocop, yielding bang-bang-type solutions where the two controls are activated in a quasi-alternating pattern. We further compare this optimal strategy with a structured periodic control policy, showing that it provides close performance with simpler implementation

Keywords: Stability of nonlinear systems, Biological systems, Nonlinear system theory
Abstract: This paper is concerned with local stability properties of the so-called coexistence steady-state which arises when considering perturbed chemostat systems. Without perturbation, the competitive exclusion principle asserts that only one species is present asymptotically whereas in presence of perturbations, all species can be present asymptotically. The objective of this paper is twofold. First, we prove that under certain condition on the data defining a perturbed chemostat system, the coexistence steady-state is locally asymptotically stable (and not only for small perturbations of the chemostat system). Doing so, we give in a general setting sufficient conditions for a matrix B-de^top to be Hurwitz where Bin R^{ntimes n} is symmetric negative semi-definite and d,ein R^n. As an application of these conditions, we derive the desired local stability of the coexistence steady state. Next, to complement our analytical results, we numerically investigate properties of the coexistence steady-state for various perturbations of the chemostat system. Our computations reveal that when the system is subject to perturbations, the coexistence steady-state disperses around the index of the species that would win in the unperturbed case.

Keywords: Biological systems, Modeling, Stability of nonlinear systems
Abstract: Nowadays it is recognized that many ecosystems contain toxic stressors. Here it is shown how food chain dynamics can be affected by progressive contaminations and how predation mechanisms can distort and magnify pollutant effects in surprising ways. The aim is obtained by analyzing an expansion of a classical two preys (feeding on constant food) - one predator (subject to superpredator extra mortality) model, in which two new parameters have been introduced: the concentrations of toxicants for preys and predators, both affecting their growth rates (the prey intrinsic growth rate and the maximum predation rate, respectively). The analysis has been performed by determining the bifurcations of the model with respect to the following control parameters: toxic concentrations, prey’s enrichment (prey’s food) and of biological pressure (superpredator density) at the bottom and at the top of the food chain, respectively. The analysis shows that increasing values of toxicants imply a transition from cyclic to stationary coexistence up to herbivore extinction and that this transition is facilitate by impoverishing the ecosystem or increasing superpredator density. Moreover, multiplicity of dynamic behaviors, catastrophic transitions and hysteresis may arise. Finally, the average values of the populations have been evaluated. The results have shown a surprisingly increase of the biomass of the intoxicated species against an increase of the toxic concentration. The maximum is reached on the ''edge'' of the most complex behavior (transition from cyclic to stationary coexistence).

Keywords: Biological systems, Machine learning, Neural networks
Abstract: In this work, we focus on estimating water stress based on leaf turgor pressure in Arbequina olive orchards. Currently, this analysis is time-consuming and primarily conducted by experts, making it impractical for integration into commercial deficit irrigation applications. We propose an innovative automated system based on machine learning techniques to classify trees into three distinct water stress levels by analyzing their daily trajectory. Our study leverages a dataset comprising leaf turgor pressure measurements and meteorological variables, collected from 2014 to 2019 at an orchard in Spain, using off-the-shelf classification algorithms, with satisfactory results.

Keywords: Linear systems, LMI's/BMI's/SOS's, H2/H-infinity methods
Abstract: The notion of an L_{2+} induced norm, i.e., the L_2 induced norm for non-negative input signals, has been proposed. Furthermore, the L_{2+} induced norm has been shown to be larger than or equal to 1/sqrt{2} times the L_2 induced norm for any LTI system. Also, a numerically effective method for computing the lower bound of the L_{2+} induced norm of given systems is proposed. This paper proposes a numerical method to compute a less conservative lower bound of the L_{2+} induced norm for given systems, which consists of optimizing the frequency and signal waveform parts. The proposed method also provides information about the signal waveform that achieves the lower bound of L_{2+} induced norm. A numerical example demonstrates the effectiveness of the proposed method.

Keywords: Linear time-varying systems, Linear systems, Robust control
Abstract: This paper studies the (quasi) L_2/L_1 Hankel norm and (quasi) L_infty/L_2 Hankel norm of linear periodically time-varying (LPTV) systems, which are quantitative measures of the impact of the past input on the future output. Given an LPTV system and considering either one of the two norms for it, a clear perspective is laid on how the associated analysis is related through duality with the treatment in which its dual system is taken and the other of the two norms is considered instead. In addition, arguments on explicit expressions for the (quasi) L_infty/L_2 Hankel norm are provided through a somewhat different use of techniques relevant to duality, where we obtain a generalization of an existing result with a less restrictive assumption on LPTV systems. Furthermore, this result on the (quasi) L_infty/L_2 Hankel norm together with the properties on duality in the analysis of two types of (quasi) Hankel norms yields the corresponding expression for the (quasi) L_2/L_1 Hankel norm.

Keywords: Mechatronics, Machine learning, Sensor and signal fusion
Abstract: Accurate state and parameter estimation is essential for monitoring and control, yet high-fidelity models are often unavailable or costly to obtain. As a result, many applications rely on data-based virtual sensors that learn mappings from time-windowed measurements or derived features to the quantities of interest. While effective, such approaches are prone to changes in process parameters and operating conditions—such as wear, load variations, or disturbances—which alter the underlying data–state relationship and introduce concept shift. Ensuring robustness therefore requires both selecting features that are insensitive to such shifts and choosing a mapping that preserves discriminative information under varying conditions. We analyse this challenge in the concrete setting of electromagnetic solenoids, where accurate estimation of the armature position is crucial for control and diagnostics but typically requires costly position sensors. However, virtual-sensor performance often degrades under varying external loads and wear-induced changes, which disrupt the current–position relationship. To mitigate this effect, we propose a virtual sensor design that (i) employs physics-derived features capturing position-dependent electromagnetic effects and (ii) trains the mapping on a mixture of data from multiple operating contexts to reduce shift sensitivity. Experimental results on a solenoid test bench demonstrate that this combined strategy effectively improves robustness and maintains estimation accuracy under changing load conditions.

Keywords: Modeling, Robust control, Automotive
Abstract: We propose a robust nonlinear model predictive control (MPC) scheme for trajectory-tracking control of autonomous vehicles at the limits of handling on non-planar road surfaces. We derive the dynamics from first principles and selectively omit terms with negligible dynamic influence to maintain real-time capability. The resulting MPC with a three-dimensional (3D) dynamic single-track model integrates relevant dynamic effects directly into the prediction model and leverages them to improve prediction accuracy and therefore control performance. Even if the influence of terrain-induced vertical loads on the total acceleration potential is modeled, tire-road interactions are subject to uncertainty and disturbance. The uncertainty-aware constraint tightening scheme introduces a margin to constraint bounds to keep the vehicle controllable and stable in this environment. To validate our proposed approach, we perform high-fidelity dynamic double-track vehicle dynamics simulations on a model of a real circuit. We find that our algorithm can improve trajectory-tracking accuracy while maintaining low computation times.

Keywords: Stochastic control, Optimal control, Predictive control for linear systems
Abstract: The Linear Quadratic Gaussian (LQG) regulator is a cornerstone of optimal control theory, yet its performance can degrade significantly when the noise distributions deviate from the assumed Gaussian model. To address this limitation, this work proposes a distributionally robust generalization of the finite-horizon LQG control problem. Specifically, we assume that the noise distributions are unknown and belong to ambiguity sets defined in terms of a Sinkhorn discrepancy centered at a nominal Gaussian distribution. By deriving novel bounds on this entropy-regularized Wasserstein distance and proving structural and topological properties of the resulting ambiguity sets, we establish global optimality of linear policies for Sinkhorn distributionally robust LQG. Numerical experiments showcase improved distributional robustness of our control policy.

Keywords: Uncertain systems, Robust control, Optimization algorithms
Abstract: Uncertainty in model parameters presents a significant challenge in control and policy design. Domain Randomization (DR) is a generalization technique widely used in machine learning and control to enhance robustness by randomly varying simulator parameters during training. In this work, we address DR for Linear Quadratic Regulation (LQR) problems by introducing a semidefinite programming (SDP) framework that conveniently incorporates multiple constraints, including Lyapunov stability conditions. This formulation enables the design of a single controller that stabilizes all sampled system instances. Furthermore, we propose an algorithm that leverages iterative SDP steps and a decay factor to eliminate the need for an initial jointly stabilizing controller—often difficult to obtain. Empirical results on a classical inverted-pendulum example demonstrate the effectiveness of our approach.

Keywords: Maritime, Predictive control for nonlinear systems, Autonomous systems
Abstract: Energy efficiency is vital for Autonomous Surface Vehicles (ASV) performing missions in dynamic marine environments. Exploiting sea currents offers an effective pathway toward improved energy efficiency. To this direction, this work introduces a Gramian-Based, sea Current-Aware Model Predictive Control (MPC) framework that unifies short-horizon controllability Gramians, a corridor-based direction shaping strategy, and an Effort-Per-Progress (EPP) cost formulation. The controller dynamically blends a “Gramian corridor” - the most controllable motion direction - with the desired navigation corridor, guiding the vessel along directions that are both energy efficient and operationally effective. By incorporating real sea surface currents in the prediction horizon, the method adapts its trajectory to utilize them, while minimizing unnecessary thrust. The algorithm is extensively validated on a simulated autonomous marine platform performing energy-efficient cargo-transport under real environmental conditions. Results indicate remarkable enhancements in energy efficiency relative to standard nonlinear MPC and an Effort-aware MPC formulation. The method also produces smoother control inputs and improved dynamic stability by suppressing oscillatory behavior, highlighting the effectiveness of the proposed predictive controller for sustainable autonomous marine operations. The individual contribution of the proposed novel cost terms is also studied via simulated experiments.

Keywords: Maritime, Predictive control for nonlinear systems, Autonomous systems
Abstract: The maritime sector is undergoing a disruptive technological change driven by three main factors: autonomy, decarbonization, and digital transformation. Addressing these factors necessitates a reassessment of inland vessel operations. This paper presents the design and development of a decision support system for ferry operations based on a shrinking-horizon optimal control framework. The problem formulation incorporates a mathematical model of the ferry’s dynamics and environmental disturbances, specifically water currents and wind, which can significantly influence the dynamics. Real-world data and illustrative scenarios demonstrate the potential of the proposed system to effectively support ferry crews by providing real-time guidance. This enables enhanced operational efficiency while maintaining predefined maneuver durations. The findings suggest that optimal control applications hold substantial promise for advancing future ferry operations on inland waters. A video of the real-world ferry MS Insel Mainau operating on Lake Constance is available at: https://youtu.be/i1MjCdbEQyE

Keywords: Maritime, Sliding mode control, Lyapunov methods
Abstract: Precise motion control of underactuated surface vessels is a crucial task in various maritime applications. This work proposes a pursuit-guidance-based nonlinear path-following strategy for surface vessels. Unlike traditional methods, this work develops an integrated guidance and control approach capable of following a smooth path (not just a composition of straight-line and circular paths). Any sufficiently smooth path can be viewed as a continuum of virtual target points moving along it, which the vehicle can pursue. To achieve the path-following behavior, a two-fold objective is set. First, align the vehicles' velocity vector with the line-of-sight (the line joining the virtual target point moving on the path and the surface vessel). Second, nullify the range to the virtual target. Achieving these two objectives results in a tail-chasing scenario that leads to path-following behavior. Additionally, the control law is derived within a nonlinear framework using the sliding-mode control technique. The stability of the proposed control law is established using the Lyapunov method. The proposed strategy is evaluated against various specified paths via numerical simulations using the CyberShip II model.

Keywords: Maritime, Autonomous systems, Stability of nonlinear systems
Abstract: This paper presents a control framework for straight-line path following of a towed marine system consisting of an autonomous surface vehicle (ASV) pulling a passive payload via a cable. Unlike existing approaches that neglect payload dynamics, the proposed method explicitly accounts for the coupled motion of the towing vehicle and the payload. Moreover, the approach is payload-centric, directly controlling the position of the payload itself to follow the desired path. The design procedure for the controller is based on a simplified lumped-mass model of the towed system, and consists of two steps: First, the desired steady-state behavior of the system is obtained analytically. Second, a kinematic guidance law combined with a dynamic PI controller is designed to drive the system toward this behavior. Using Lyapunov analysis, we establish sufficient conditions under which the closed-loop system possesses an almost-everywhere globally asymptotically stable equilibrium point corresponding to the desired steady state. Simulations using both the lumped-mass model and a higher-fidelity scenario featuring an underactuated ASV towing a segmented cable validate the proposed approach.

Keywords: Maritime, Modeling, Nonlinear system identification
Abstract: Vessels navigating in confined waters are subject to banking effects, which are hydrodynamic forces and moments arising from pressure differentials between the vessel sides, significantly affecting manoeuvrability and safety. Existing numerical approaches such as computational fluid dynamics (CFD) can accurately capture these effects but are computationally expensive and unsuitable for real-time control or estimation. This paper presents a closed-form, first-principles model of banking effects. The model coefficients are identified using physics-informed regression on towing tank experiment data for a scaled container vessel. Validation through Shapley value analysis confirms the significance of the banking terms in reproducing the measured forces and moments. Lastly, the derived coefficients are shown to be non-dimensional, making the model applicable across different scales that preserve vessel geometry.

Keywords: Maritime
Abstract: Following recent adjustment of the azimuth sector configuration for the prototype ferry vessel milliAmpere1, we propose a control allocation pipeline supporting automatic control in all speed ranges. The findings are suitable for vessels with azimuth propellers. Nine different control allocation methods for low-speed maneuvering and six different allocation methods for path-following, which we term “virtual rudder”-allocation in respect to the resemblance of conventional rudder steering, are experimentally tested. Results indicate that singularity avoidance and nullspace methods perform well when considering reduced angular wear or power respectively, while algebraically simple allocation methods provide the best overall trade-off. For curve following, steering with the available thrusters within prescribed sectors achieved the smallest cross-track, while during straight-line-following, stern propellers alone is sufficient. Additional heading damping through feedback to the propeller revolutions achieved better robustness in accuracy to straight-line following at the cost of increased thrust wear, while this performed worse during curve following.

Keywords: Stochastic control, Stochastic systems, Statistical learning
Abstract: This part develops the central thesis that many modern generative AI frameworks can be interpreted as stochastic optimal control/transport problems over probability distributions. We begin with Schrödinger bridges and dynamic optimal transport, showing how diffusion models arise naturally from entropy-regularized control of stochastic dynamics. We then introduce the Path Integral Diffusion (PID) framework and demonstrate how its harmonic, adaptive, and guided variants provide analytically tractable and computationally efficient formulations for steering probability distributions. We further introduce the concept of Sampling Decisions, emphasizing control directly over distributions rather than trajectories, thereby unifying inference, generative modeling, and stochastic control. Finally, we discuss how transformer architectures can be interpreted as amortized control policies operating in probability space.

Keywords: Autonomous systems
Abstract: A future society in which autonomous systems contribute pervasively to sustainability, safety, and security is rapidly becoming a reality. From intelligent transportation to automated monitoring and assistive robotics, autonomy is transitioning from concept to deployment. In this context, control can play a central role by providing principled system-level design, robustness to uncertainty, and the ability to guarantee performance and safety in complex, real-world environments. At the same time, autonomous systems increasingly rely on the tight integration of control with perception, learning, and decision-making components, often driven by data and artificial intelligence. This evolution may require revisiting established design paradigms. Drawing on examples from industry research spanning transportation and logistics, automated monitoring, and space exploration and utilization —some of which are close to or already in deployment— I will discuss how the strengths of control can be leveraged to enable reliable and scalable autonomy, as well as emerging research opportunities to address the associated challenges.

Keywords: Energy systems, Machine learning
Abstract: This presentation describes advances in machine learning-based techniques for addressing systems problems that arise for lithium-ion batteries. The specific systems problems include the prediction and classification of battery cycle lifetime (aka remaining useful life), the determination of optimal charging protocols, and the identification of fundamental physicochemical expressions for electrochemical kinetics, thermodynamics, and mass transfer from real-time video imaging.

Keywords: Network analysis and control, Large-scale systems, Agents networks
Abstract: This paper proposes a mathematical model for the coevolution of actions and opinions for a population facing a social dilemma. In particular, we assume each person participates in a Public Goods Game (PGG), with their action being to cooperate or defect, and holds an opinion about which action they prefer. We propose a payoff function that combines the PGG with the Friedkin--Johnsen model from opinion dynamics to form a coevolutionary game. According to a discrete-time process, players asynchronously update their actions and opinions, aiming to maximise their individual payoff for the coevolutionary game using myopic best-response. We study the equilibria and provide conditions for the existence of the all-defection and all-cooperation consensus equilibria. We also establish conditions for global convergence to the all-defection equilibrium.

Keywords: Network analysis and control, Agents and autonomous systems, Complex systems
Abstract: We consider a verification problem for opinion dynamics based on binary observations. The opinion dynamics is governed by a Friedkin-Johnsen (FJ) model, where only a sequence of binary outputs is available instead of the agents' continuous opinions. At every time-step we observe a binarized output for each agent depending on whether the opinion exceeds a fixed threshold. The objective is to verify whether an FJ model with a given set of stubbornness parameters and initial opinions can generate the observed binary outputs up to a small error. The FJ model is formulated as a transition system, and an approximate simulation relation of two transition systems is defined in terms of the proximity of their opinion trajectories and output sequences. We then construct a finite set of abstract FJ models by simplifying the influence matrix and discretizing the stubbornness parameters and the initial opinions. It is shown that the abstraction approximately simulates any concrete FJ model with continuous parameters and initial opinions, and is itself approximately simulated by some concrete FJ model. These results ensure that consistency verification can be performed over the finite abstraction. Specifically, by checking whether an abstract model satisfies the observation constraints, we can conclude whether the corresponding family of concrete FJ models is consistent with the binary observations. Finally, numerical experiments are presented to illustrate the proposed verification framework.

Keywords: Network analysis and control, Robust control, Optimization
Abstract: This paper addresses the problem of transforming a structurally controllable system with symmetric dynamics matrix into an r-robust structurally controllable system through strategic modifications to the dynamics matrix (i.e., the problem of dynamics augmentation for robust structural controllability of bidirected networks, r-MARS-B). Unlike existing approaches that focus on input placement design, we investigate the minimal number of zero entries in the system's dynamics matrix that must be changed to non-zero values to ensure structural controllability is preserved when one input fails. We formulate r-MARS-B as an optimization problem and establish its NP-hardness. Additionally, we present a greedy algorithm for computing approximate solutions using a path-stitching approach. We provide numerical examples to illustrate our findings and analyze the performance of our algorithm.

Keywords: Network analysis and control, Agents networks
Abstract: The multipolar model of opinion dynamics describes a multi-topic opinion exchange between agents with an underlying prejudice, bias or preference for the topics at hand. In this work, we consider the case where agents are divided into two competing cohorts, each with their own topic preference. Analytic results on the location and stability of fixed points are provided in the case where discussions take place over a complete graph, as its size grows large. The opinion-dynamical model is shown to exhibit bifurcations, where the relative cohort size determines whether the discussion results in consensus or dissensus. Illustrative examples are provided, along with numerical analysis of the dynamics over sparse and dense graphs which demonstrate that our mechanisms persist beyond the analytically tractable setting.

Keywords: Network analysis and control, Adaptive control, Identification
Abstract: We propose an adaptive control protocol for identifying the topology of dynamical networks interconnected over undirected graphs with cooperative and antagonistic interactions. The signed network is modeled using a repelling Laplacian. Topology identification relies on an edge-based formulation of the network and adaptive control protocols through the design of a persistently excited auxiliary network. Our approach guarantees the simultaneous identification and synchronization of the unknown signed network and establishes uniform semiglobal practical asymptotic stability of the estimation errors. Numerical simulations validate our theoretical results.

Keywords: Reduced order modeling, Stability of nonlinear systems, Network analysis and control
Abstract: This paper studies model reduction for networks of nonlinear control-affine subsystems interconnected over a connected undirected graph. We first derive a sufficient condition that guarantees global exponential synchronization of the full network. Based on this condition, a differential-balancing framework is developed to reduce the order of the local nonlinear subsystem dynamics while preserving the network interconnection structure. For odd nonlinearities, we derive an explicit L_2 input-output error bound between the original subsystem and its reduced-order approximation. The proposed approach is illustrated on a network of mass-spring-damper systems with Coulomb friction.

Keywords: UAV's, Predictive control for nonlinear systems, Aerospace
Abstract: This paper presents a nonlinear model predictive control (NMPC) strategy for a quad-tiltrotor unmanned aerial vehicle (UAV). As a convertible aircraft, the quad-tiltrotor integrates fixed-wing efficiency with rotary-wing maneuverability, making it ideal for applications requiring extended endurance and operational flexibility through vertical takeoff and landing (VTOL), stable hovering, and efficient forward flight. However, controlling this type of aircraft is challenging due to its underactuated nature and complex multi-body dynamics. Therefore, our work addresses these issues by proposing an NMPC scheme that considers the UAV prediction model on an extended SE(3) manifold. The controller regulates position and attitude toward an equilibrium point, optimizing actions at each step without relying on predefined trajectories. Furthermore, this approach guarantees that state and control constraints are satisfied, ensuring both closed-loop stability and recursive feasibility, which are corroborated by numerical results

Keywords: Predictive control for nonlinear systems, Adaptive control, Robotics
Abstract: Variable Impedance Control (VIC) is crucial for robotic manipulators to achieve a safe and effective balance between precision and compliance in dynamic, uncertain environments. However, naive VIC schemes may violate passivity, which can lead to instability during physical interaction. We propose a novel Passive Model Predictive Variable Impedance Control (P-MVIC) f framework that jointly adapts impedance parameters and enforces passivity within a unified optimisation problem. The approach embeds an energy tank into a nonlinear Model Predictive Control (MPC) formulation, enabling the controller to regulate impedance parameters in an energy-aware manner while preventing non-passive behaviours. We employed a quaternion-based second-order impedance model and formulated VIC for full 6D pose regulation. The proposed method is validated on a Franka Emika 7-DOF manipulator during human–robot interaction.

Keywords: Predictive control for nonlinear systems, Constrained control, UAV's
Abstract: This paper proposes a finite-horizon optimal control strategy for set-point tracking using a nonlinear model predictive control framework with integrated avoidance capabilities. The formulation employs a smooth point-to-cloud distance metric that ensures continuously differentiable and numerically well-conditioned gradients, even in the presence of regions with complex and nonconvex geometries. This smoothness allows safety constraints to be formulated consistently and differentiably through control barrier functions, resulting in a reliable avoidance behavior for the closed-loop system. Additionally, stationary artificial variables are introduced in the optimal control problem to preserve feasibility under changing set-points. The proposed approach is validated through numerical experiments of an aerial robot, demonstrating accurate tracking and smooth obstacle avoidance in complex environments.

Keywords: Predictive control for nonlinear systems, Switched systems, Optimal control
Abstract: Real-time model predictive control of non-smooth switching systems remains challenging due to discontinuities and the presence of discrete modes, which complicate numerical integration and optimization. This paper presents a real-time-feasible non-smooth model predictive control scheme for a physical three-tank process, implemented without mixed-integer formulations. The approach combines Filippov system modeling with finite elements and switch detection for time discretization, leading to a finite-dimensional optimal control problem formulated as a mathematical program with complementarity constraints. The mathematical program is solved via a homotopy of smooth nonlinear programs. We introduce modeling adjustments that make the three-tank dynamics numerically tractable, including additional modes to avoid non-Lipschitz points and undefined function values. Hardware experiments demonstrate efficient handling of switching events, mode-consistent tracking across reference changes, correct boundary handling, and constraint satisfaction. Furthermore, we investigate the impact of model-mismatch and show that the tracking performance and computation times remain within real-time limits for the chosen sampling time. The complete controller is implemented using the non-smooth optimal control framework NOSNOC.

Keywords: Predictive control for nonlinear systems, Robotics, Optimization algorithms
Abstract: This paper presents a spatially lifted nonlinear model predictive control (NMPC) framework for an articulated loader operating under time-corridor constraints along a predefined path. The loader kinematics, including the articulation-rate contribution to yaw motion, are reformulated through a spatial reparameterization that yields a control-affine model suitable for successive convexification (SCvx). The resulting spatial NMPC problem incorporates convex actuator constraints and time-corridor constraints expressed as bounds on a lifted time state with convex slacks. We establish local well-posedness of the lifted dynamics and derive explicit bounds on the linearization-induced slacks, yielding recursive feasibility of the SCvx subproblems and a local practical stability result. Simulation studies on representative path-following scenarios demonstrate accurate tracking, time-corridor compliance, and consistency of the observed slack magnitudes with the conservative theoretical bounds. These results support the practical viability of the proposed lifted SCvx formulation for time-aware articulated path following.

Keywords: Predictive control for nonlinear systems, Optimization algorithms
Abstract: This paper presents a unified framework that connects sequential quadratic programming (SQP) and the iterative linear-parameter-varying model predictive control (LPV-MPC) technique. Using the differential formulation of the LPV-MPC, we demonstrate how SQP and LPV-MPC can be unified through a specific choice of scheduling variable and the 2nd Fundamental Theorem of Calculus (FTC) embedding technique and compare their convergence properties. This enables the unification of the zero-order approach of SQP with the LPV-MPC scheduling technique to enhance the computational efficiency of robust and stochastic MPC problems. To demonstrate our findings, we compare the two schemes in a simulation example. Finally, we present real-time feasibility and performance of the zero-order LPV-MPC approach by applying it to Gaussian process (GP)-based MPC for autonomous racing with real-world experiments.

Keywords: Concensus control and estimation, Communication networks, Network analysis and control
Abstract: This paper investigates output consensus in heterogeneous dynamical networks within a plug-and-play framework. The networks are interconnected through nonlinear diffusive couplings and operate in the presence of measurement and communication noise. Focusing on systems that are input feedforward passive (IFP), we propose a passivity-compensation approach that exploits the surplus passivity of coupling links to locally offset shortages of passivity at the nodes. This mechanism enables subnetworks to be interconnected without requiring global reanalysis, thereby preserving modularity. Specifically, we derive locally verifiable interface conditions, expressed in terms of passivity indices and coupling gains, to guarantee that consensus properties of individual subnetworks are preserved when forming larger networks.

Keywords: Hybrid systems, Nonlinear system theory
Abstract: Scaled relative graphs (SRGs) provide a promising framework for graphical analysis and design of nonlinear control systems. However, obtaining SRGs directly is a challenging task. This paper develops a systematic approach for over-approximating SRGs of piecewise linear systems through incremental dissipativity analysis. The key insight is to transform the geometric SRG computation problem into verifying multiple incremental dissipativity properties, and to translate each property to a region in the complex plane. The intersection of these regions provide tight over-approximations of the SRG. The computational framework utilizes piecewise quadratic storage functions and is particularly well-suited for piecewise linear systems where the verification of incremental dissipativity reduces to solving convex optimization problems.

Keywords: Model/Controller reduction, Linear systems
Abstract: This paper presents a structure-preserving model reduction framework for negative imaginary (NI) systems based on a state-space parametrisation. We first establish a constructive representation showing that any NI system without poles at the origin can be parametrised by a skew-symmetric, a positive semi-definite, an arbitrary, and a symmetric matrix, thereby revealing the internal state-space structure underlying the NI property. Leveraging this parametrisation, a projection-based model reduction method is developed that guarantees both the preservation of the NI property and exact interpolation of the transfer function at prescribed frequency points. Numerical examples on flexible structures are presented to demonstrate the validity of the proposed parametrisation, the interpolation accuracy and structure-preserving capability of the proposed method.

Keywords: Network analysis and control, Electrical power systems, Distributed control
Abstract: This paper presents a comprehensive dissipativity-based framework for co-designing distributed controllers and communication topologies in AC microgrids (MGs) to achieve robust guarantees for voltage regulation, frequency synchronization, and proportional power sharing among distributed generators (DGs). We formulate the closed-loop AC MG as a networked system where DGs, distribution lines, and loads are subsystems that are interconnected through cyber communication and physical transmission links. At each DG, we employ a steady-state controller for establishing the operating point, a local controller for voltage regulation, and a distributed droop-free controller for frequency synchronization and proportional power sharing through normalized power consensus. Next, we formulate the operating point design problem. Then, leveraging dissipativity theory, we establish necessary and sufficient conditions for subsystem dissipativity. Finally, the global co-design problem is formulated as a convex linear matrix inequality (LMI) problem to simultaneously design distributed controller gains and a sparse communication topology while handling highly coupled, complex nonlinear, dq-frame dynamics.

Keywords: Output regulation, Robust control, Nonlinear system theory
Abstract: Incremental passivity facilitates the development of output regulators via decoupled designs of a passivation controller and an internal model. While this approach is effective for data-driven output regulation with noiseless data, it fails to handle the noisy case, as noisy data leads to a data-based system representation with uncertainties. This work addresses this issue by robustifying the data-driven incremental passivation design. We present a robust characterization of incremental passivity for a class of uncertain nonlinear systems and design a data-driven feedback controller that renders the closed-loop system incrementally passive. The proposed robust data-driven incremental passivation controller is then applied to data-driven output regulation via noisy data. Finally, a numerical example validates the proposed data-driven regulator.

Keywords: Switched systems, Hybrid systems, Robust control
Abstract: We propose data-driven passivation strategies for switched linear systems with unknown system matrices. In particular, we consider two types of switching laws: controlled state-dependent switching and exogenous switching. Using offline open-loop input-state-output data, we derive linear state feedback controllers that render the switched system strictly passive via bilinear matrix inequalities (BMIs). Furthermore, passivation with noisy data, relaxation of the BMIs, and data-driven stabilization via passivity are discussed. Finally, numerical simulation results validate the performance of the proposed data-driven controllers.

Keywords: Game theoretical methods, Identification, Stochastic systems
Abstract: This work addresses cost identification in a finite-horizon linear quadratic Gaussian game. We characterize the set of cost parameters that generate a given Nash equilibrium policy and propose a backpropagation algorithm to compute these parameters. A probabilistic error bound is derived in case the cost parameters are identified from finite trajectories. Through numerical and driving simulations, our algorithm identifies the cost parameters that can reproduce the Nash equilibrium policy and trajectory observations.

Keywords: Optimization, Optimization algorithms, Game theoretical methods
Abstract: Inspired by the adaptive Golden Ratio Algorithm (aGRAAL), we propose two new methods for solving monotone variational inequalities. We show that by selecting the momentum parameter beyond the golden ratio in aGRAAL, the convergence speed can be improved, which motivates us to study the switching between small and large momentum parameters to accelerate convergence. We validate the performance of our proposed algorithms on several classes of variational inequality problems studied in the machine learning and control literature, including Nash equilibrium seeking, composite minimization, Markov decision processes, and zero-sum games, and compare them to that of existing methods.

Keywords: Transportation systems, Modeling, Game theoretical methods
Abstract: This paper studies competition between ride-hailing companies in a multi-regional market through a static non-cooperative game, in which each company allocates its fleet across regions to maximize profit. Under the assumptions of the proposed framework, we establish an equilibrium existence result on a restricted feasible set by analyzing the asymptotic quasi-concavity of the payoff functions. A numerical study of a two-company, two-region market shows that relative fleet size strongly affects equilibrium dispatching strategies: the smaller company tends to concentrate its fleet in one region and serves both regions only after its fleet share (in percentage) exceeds a threshold. In the numerical setting considered here and under the Logit demand specification, more balanced fleet sizes are associated with higher total industry profit and higher total served demand than strongly asymmetric fleet configurations.

Keywords: Transportation systems, Game theoretical methods
Abstract: Mobility-on-demand systems like ride-hailing have transformed urban transportation, but they have also exacerbated socio-economic inequalities in access to these services, also due to surge pricing strategies. Although several fairness-aware frameworks have been proposed in smart mobility, they often overlook the temporal and situational variability of user urgency that shapes real-world transportation demands. This paper introduces a non-monetary, Karma-based mechanism that models endogenous urgency, allowing user time-sensitivity to evolve in response to system conditions as well as external factors. We develop a theoretical framework maintaining the efficiency and fairness guarantees of classical Karma economies, while accommodating this realistic user behavior modeling. Applied to a simplified simulated mobility-on-demand scenario, we provide a proof-of-concept illustration of the proposed framework, showing that it exhibits promising behavior in terms of system efficiency and equitable resource allocation, while acknowledging that a full treatment of realistic MoD complexity remains an important direction for future work.

Keywords: Predictive control for nonlinear systems, Game theoretical methods, Optimal control
Abstract: We address the setpoint regulation problem in evolutionary game-theoretic dynamics within population games, aiming to design an algorithm that guides the population toward a desired collective behavior - particularly, an equilibrium point. In detail, we focus on a discrete-time replicator equation, which models the collective behavior of a population engaged in two-player, two-action matrix games with every other member of the population. To tackle the problem, we develop an optimal control strategy that manipulates the payoff matrix in a closed-loop manner by adding a nonnegative gain to one of its entries. This strategy is intended to direct the population behavior toward the desired equilibrium. The control problem is resolved using nonlinear model predictive control, ensuring both closed-loop stability and recursive feasibility through appropriate selection of the terminal ingredients.

Keywords: Agents networks, Network analysis and control, Markov processes
Abstract: In this paper, we study imitation-based evolutionary dynamics in potential population games, where the evolution of each agent's action is explicitly described. The global dynamics are modeled by a discrete-time Markov chain whose state is the vector of actions played by all agents. Although each agent's reward is determined by the underlying population game, we introduce an imitation mechanism that operates along the edges of a given network. Specifically, during each imitation step, an agent adopts the action of a neighboring agent with a probability that depends on the difference in their rewards. Moreover, to obtain the chain ergodicity, we impose also a spontaneous mutation mechanism in the chain dynamics in which each agent has the chance to randomly change her action. To ensure the ergodicity of the Markov chain, we also incorporate a spontaneous mutation mechanism, allowing each agent to randomly change their action with a small probability. We then analyze a double-limit regime, where the number of agents tends to infinity and the mutation intensity tends to zero. Under a suitable set of assumptions, we prove that the invariant distribution of the Markov chain concentrates on the set of Nash equilibria of the potential population game.

Keywords: Concensus control and estimation, UAV's, Optimal control of communication networks
Abstract: This paper presents a distributed estimation and control framework for UAV-assisted 5G networks, enabling synchronized drones to jointly localize ground user equipments (UEs) and adapt their spatial deployment. The proposed approach exploits the Physical Random Access Channel (PRACH) as a dual source of time-difference-of-arrival (TDOA) measurements and dynamic neighbor discovery, allowing the construction of UE-specific, time-varying communication graphs. Over these graphs, a consensus of an estimator with adaptive coupling achieves distributed TDOA-based localization. For density balance, a distributed K-means algorithm extracts from local estimates, supporting load-aware coverage control. Finally, a control layer based on combined Control Barrier Functions (CBF) and Control Lyapunov Functions (CLF) guarantees collision-free UAV motion and stability within feasible control limits. Simulation results demonstrate meter-level localization accuracy and robust convergence under realistic PRACH detection conditions, validating the proposed methods scalability and compliance with 3GPP TS 38.211/38.305 standards for 5G/6G positioning.

Keywords: Nonlinear system theory, Feedback linearization, UAV's
Abstract: We propose a quasi-static transverse feedback linearization (QSTFL) controller for a quadrotor to follow a prescribed geometric path, rather than a time-parameterized trajectory. In contrast to existing dynamic-feedback approaches, the controller does not introduce additional controller states. The thrust input is computed algebraically from the current state, eliminating the need for thrust-derivative measurements and numerical integration. The proposed design renders the path-following manifold invariant, ensuring that trajectories initialized on the path remain on it for all future time, while simultaneously regulating tangential velocity and yaw. We establish a diffeomorphic coordinate transformation and prove local exponential stability of the path-following manifold. In addition, closed-form expressions are derived for the thrust and torque inputs. Compared with dynamic-feedback constructions, the controller requires inversion of only a 3 × 3 decoupling matrix rather than a 4 × 4 one, leading to a simpler control law and reduced computational complexity. Numerical simulations demonstrate the effectiveness of the proposed method. Code and animations are publicly available at https://gitlab.com/a5akhtar/quasistatic-tfl-uav/.

Keywords: UAV's, Feedback linearization, Aerospace
Abstract: Considering the nonlinear dynamic model of a micro tandem helicopter with thrust vectoring, a feedback linearization control is derived that solves the state space exact linearization problem. The six degrees of freedom of the helicopter, position and attitude, can thus be controlled independently by manipulating each three thrust components of the two rotors appropriately. The basic control law is augmented next by a feedforward control for trajectory tracking and finally by integrators to mitigate the effect of disturbances and parameter mismatch. Simulation results for all three stages of the control law are presented.

Keywords: UAV's, Sliding mode control
Abstract: In this paper, a robust tracking control strategy with fixed-time performance is proposed for unmanned aerial vehicles (UAVs). The altitude and attitude control loops are designed relying on a fixed-time control, robust against matched perturbations. For under-actuated positions, a simple sliding mode control is employed to obtain a reference trajectory of tilting angle, whereas for actuated subsystems, global fixed-time stability is proved. The proposed technique is an easy-to-implement solution by virtue of the tuning of only two parameters. Its efficiency and robustness are illustrated in the paper through realistic simulations.

Keywords: UAV's, Fault estimation, Predictive control for nonlinear systems
Abstract: Despite their utility, quadcopters' underactuated dynamics make them highly vulnerable to failures. This research develops two fault-tolerant control strategies for quadcopters by combining nonlinear model predictive control (NMPC) with sparse identification of nonlinear dynamics with control (SINDYc). SINDYc identifies faulty dynamics as they occur mid-flight, adapting the internal model of the NMPC. A first controller shows the capability to pursue a trajectory, even after the loss of a single actuator. A second controller further interprets the identified dynamics from SINDYc to safely descend to the ground. The combination of NMPC with SINDYc was tested in simulations, showing the capability to adapt to various faults.

Keywords: UAV's, Robust control, Optimization
Abstract: The most conventional Multi-target tracking (MTT) techniques face challenging problems due to target deformation, occlusions, and clutters, thereby true tracks (target-track) becomes false track (clutter-track). In this paper, we have utilized a state-of-art sensor measurement method known as a motion capture system (mocap) sensor which is integrated with mocap Motive 2.2 sofware that is used to measure the position of targets such as unmanned aerial vehicles (UAVs). Although, mocap can track a UAV itself, however target occlusion, identification, and its trajectory behaviour remain challenging issues. Moreover, mocap measurement noise, UAV process noise, and unknown clutter measurements deteriorate the tracking performance. These technical issues can be mitigated by utilizing a Markov-Chain-Two (MC2) model for the UAV state dynamics, which effectively record the influence of the previous two state hypotheses, thereby the target behavior becomes more accurate. This work integrates linear multi-target tracking based on the integrated probabilistic data association (LMIPDA) by MC2 model and presents a MC2-based LMIPDA (LMIPDA-MC2) method to interface with the mocap system. The post tracking estimates are optimized using a novel bidding method which finds the closest state estimate to the desired target. The project prototype consists of a UAV, three robotics vehicles, mocap, computer, and a wireless network such as IEEE 802.11a WIFI protocol. The root-mean-squared position errors (RMSEs) of the proposed method is only 0.02 m, thereby achieving an improved estimation accuracy in occlusion and clutter as illustrated in the experimental results.

Keywords: Delay systems, Linear time-varying systems, Stability of linear systems
Abstract: In this paper, we study the problem of constructing the delay Lyapunov function for a scalar linear periodic time delay equation. When the delay is an integer multiple of the period of the coefficients, the delay Lyapunov function is known to be a periodic solution of the one-dimensional wave equation on a strip with specific boundary conditions. Our construction is based on the fundamental solution of the wave equation. First, we derive new representation formula for the delay Lyapunov function in terms of the fundamental solution and the initial function which is assumed given. Second, we construct a Fredholm integral equation of the second kind for the initial function. We demonstrate that our construction is consistent with the existing theory for a particular case of constant coefficients.

Keywords: LMI's/BMI's/SOS's, Linear time-varying systems, Stability of linear systems
Abstract: This paper introduces the concept of Generalized Finite-Time Stability (GFTS) for discrete-time Linear Time- Varying (LTV) systems, extending the classical Finite-Time Stability (FTS) framework. Unlike traditional FTS, which mandates that the initial domain is a subset of the trajectory domain (both centered at the origin), GFTS allows for decoupled, offorigin initial and target domains, featuring a two-stage temporal definition: a reach phase from an initial set X_0 at k_0 to a terminal set bar{X} at k_1, followed by a hold phase where the state remains in bar{X} until k_2. This decoupling is essential for applications where the terminal set bar{X} is a temporary goal region that is not necessarily positively invariant and may not contain the origin. By restricting the sets to ellipsoids, necessary and sufficient conditions for GFTS are derived in terms of a set of Linear Matrix Inequalities (LMIs) based on forward and backward reachability analysis. Furthermore, the analysis of Generalized Finite-Time Stabilizability via linear state-feedback is developed, providing a computationally tractable solution using convex relaxations.

Keywords: Algebraic/geometric methods, Lyapunov methods, Stability of nonlinear systems
Abstract: Recently, Kvalheim and Sontag provided a generalized global Hartman-Grobman theorem for equilibria under asymptotically stable continuous vector fields. By leveraging topological properties of Lyapunov functions, their theorem works without assuming hyperbolicity. We extend their theorem to a class of possibly discontinuous vector fields, in particular, to vector fields generating asymptotically stable semiflows.

Keywords: LMI's/BMI's/SOS's, Differential algebraic systems, Stability of nonlinear systems
Abstract: In the context of spacecraft attitude control, parametrizations such as direction vectors or quaternions are often used to avoid singularities in the attitude representation. This, however, complicates the stability analysis of the system since, given the additional unit constraints, the resulting dynamics evolve on non-contractible manifolds. In this paper, we present a framework to verify almost global asymptotic stability of such systems using LaSalle's invariance principle and sum-of-squares programming, simplifying the search for Lyapunov functions. The framework is then applied to two examples: two-axis attitude acquisition utilizing aerodynamics in very low Earth orbits, and three-axis attitude acquisition for a satellite subject to gravity gradient torques in a circular orbit.

Keywords: Stability of nonlinear systems
Abstract: This paper presents a unified theoretical framework that bridges Lyapunov stability analysis of fuzzy systems with the frequency-domain properties of Strictly Negative-Imaginary (SNI) systems. By formulating interval Type--2 fuzzy models under a frozen-parameter dissipativity perspective, it is shown that the footprint of uncertainty (FOU) embedded in Type--2 antecedents introduces a emph{dissipation consumption} term in the Lyapunov inequality via the usual Young bound on cross terms. A common quadratic storage function is employed to derive a generalized matrix inequality containing a positive-semidefinite penalty lambda_{mathrm{FOU}}P that must be dominated by the baseline Type--1 dissipation to preserve NI/SNI. The conventional Type--1 fuzzy stability condition emerges naturally as a limiting case when the FOU collapses to zero. Analytical results are validated using a symbolic two-rule example that confirms the theoretical predictions and yields computable bounds on allowable FOU widths that preserve the NI/SNI property. The proposed Lyapunov--SNI formulation thus provides a rigorous analytical bridge between fuzzy uncertainty modelling and dissipative systems theory.

Keywords: Robust control, Stability of nonlinear systems, Optimal control
Abstract: In this work, we present a rigorous stability analysis of nonlinear H-infinity controllers designed for a class of input-affine underactuated mechanical systems with input coupling. While existing studies assess stability exclusively in terms of the generalized input, the actual control implementation relies on the Moore–Penrose pseudo-inverse of the input coupling matrix. This mapping introduces a non-negligible residual term into the closed-loop dynamics, whose effect has not been addressed in prior analyzes. In this paper, we explicitly account for this residual term and establish conditions under which the closed-loop system achieves uniform asymptotic stability and uniform ultimate boundedness, even in the presence of exogenous disturbances and pseudo-inverse-induced errors.

Keywords: Discrete event systems, Petri nets, Neural networks
Abstract: This paper presents a model-free approach for tracking control of a Petri net (PN). The goal of the tracking control is to determine the optimal feasible firing sequence that drives the PN from a given initial marking to a destination marking. The proposed approach doesn’t need any pre-knowledge of the PN model of the plant. The basic idea is to integrate three artificial neural networks (ANNs) into the Monte-Carlo Tree Search (MCTS). Consequently, in the MCTS algorithm, the tree search expansion is guided by the ANNs predictions of action probability, state values and rewards, rather than by random rollouts. The action probabilities estimate the likelihood of each action, the state value evaluates the desirability of a state and the reward predicts the immediate outcome of a decision. The proposed approach is an unsupervised iterative training process where the ANNs are trained through self-play without requiring any PN model or prior knowledge of the system dynamics. This feature is particularly advantageous when explicit models are unavailable or difficult to obtain. An example of a manufacturing system is given to illustrate the proposed approach.

Keywords: Linear systems, Observers for linear systems, Discrete event systems
Abstract: This study addresses the problem of data informativity for the observability and detectability of discrete-time linear time-invariant state-space systems. In particular, we establish the algebraic characterizations on the given finite-length data for being informative for the observability/detectability of the system. Further, a necessary and sufficient condition has been established on the collected finite-length data from the system for the existence of data-driven observers. Two numerical examples illustrate the derived results.

Keywords: Discrete event systems, Automata
Abstract: This paper introduces a unified framework for the analysis and design of observers for timed Discrete Event Systems operating under deterministic and non-deterministic Dynamic Observation Mechanisms (DOM). Unlike the classical static observation setting, the proposed approach allows the observability of events to vary with time, enabling the modeling of adaptive sensing policies, intermittent sensor failures, and controlled information disclosure. The system behavior is modeled using a subclass of timed automata with a single clock, which are further transformed into Clock Interval Automata to discretize the observation of time through a discrete observable tick signal. Deterministic and non-deterministic dynamic labeling mapping are then defined to describe how event observability depends on both logical and temporal conditions. For deterministic DOM, a systematic determinization procedure is provided to compute the corresponding timed observer. For non-deterministic DOM, an extended construction is developed that captures all consistent evolutions under intermittent or uncertain observations. The proposed results contribute to enhance observability or enforce opacity while reducing sensor activation costs.

Keywords: Petri nets, Discrete event systems, Modeling
Abstract: The ubiquitous presence of uncertain parameters in dynamical systems hinders the development of appropriate models and methods for their analysis. Event nets are a modeling formalism for dynamical systems that can accommodate in a natural and compact way uncertainties related to the state changes caused by the occurrence of events. In contrast to related formalisms, such as Petri nets, event nets are associated with sets of linear inequalities that specify the potential state changes that the occurrence of an event can entail. Hence, the state trajectory produced by a sequence of events is allowed to be non-deterministic. This work compares the expressive power of event nets with that of continuous Petri nets and, after defining some key concepts of event nets, derives necessary and sufficient conditions for reachability.

Keywords: Optimization, Concensus control and estimation, Machine learning
Abstract: This paper presents a byzantine resilient approximate-Newton type distributed optimization algorithm combined with a soft-medoid robust aggregation scheme, which outperforms benchmark algorithms.

Keywords: Linear systems, Stability of linear systems, Uncertain systems
Abstract: In this work, we study data-driven stabilization of linear time-invariant systems using prior knowledge of system-theoretic properties, specifically stabilizability and controllability. To formalize this, we extend the concept of data informativity by requiring the existence of a controller that stabilizes all systems consistent with the data and the prior knowledge. We show that if the system is controllable, then incorporating this as prior knowledge does not relax the conditions required for data-driven stabilization. Remarkably, however, we show that if the system is stabilizable, then using this as prior knowledge leads to necessary and sufficient conditions that are weaker than those for data-driven stabilization without prior knowledge. In other words, data-driven stabilization is easier if one knows that the underlying system is stabilizable. We also provide new data-driven control design methods in terms of linear matrix inequalities that complement the conditions for informativity. A full-length version of this work is available as a preprint.

Keywords: Optimal control, Stochastic systems, Robotics
Abstract: Signal Temporal Logic (STL) has become a powerful formalism for specifying complex temporal-spatial behaviors in autonomous systems. Handling STL constraints within stochastic setting has received increasing research interest but still poses challenges. This paper proposes a general framework to efficiently solve continuous-time nonlinear stochastic optimal control problems under chance STL constraints. The STL formulae are implemented through extended dynamics, yielding a more classical chance constraint on the terminal state uniquely that we reliably relax via Conditional Value-at-Risk. The resulting new optimal control problem is then solved using established algorithms from risk-averse control. The efficiency and feasibility of the proposed approach are demonstrated through numerical simulations.

Keywords: Autonomous systems, Hybrid systems
Abstract: We study motion planning under Signal Temporal Logic (STL), a useful formalism for specifying spatial-temporal requirements. We pose STL synthesis as a trajectory optimization problem leveraging the STL robustness semantics. To obtain a differentiable problem without approximation error, we introduce an exact reformulation of the max and min operators. The resulting method is exact, smooth, and sound. We validate it in numerical simulations, demonstrating its practical performance.

Keywords: Agents and autonomous systems, Machine learning, V&V of control algorithms
Abstract: We propose a Reinforcement Learning (RL) based control design framework for handling complex tasks. The approach extends the concept of Reward Machines (RM) with Signal Temporal Logic (STL) formulas that can be used for event generation. The use of STL allows not only a more efficient representation of rewards for complex tasks but also guiding the training process to converge towards behaviors satisfying specified requirements. We also propose an implementation of the framework that leverages the STL online monitoring algorithms. We illustrate the framework with three case studies (minigrid, cart pole and high-way environments) with non-trivial tasks.

Keywords: Algebraic/geometric methods
Abstract: Set operations are well understood for convex sets but become considerably more challenging in the non-convex case due to the complexity of their representation, thus causing loss of structural properties. Constrained polynomial zonotopes (CPZs) offer an effective compromise, as they can capture complex, typically non-convex geometries while maintaining an algebraic structure suitable for further manipulation.

Building on this, we propose novel nonlinear encodings that provide sufficient conditions for testing inclusion between two CPZs and adapt them to ensure seamless integration within optimization frameworks.

Keywords: Autonomous systems, Optimization
Abstract: This paper introduces Roundabout Constrained Convex Generators (RCGs), a set representation framework for modeling multiply connected regions in control and verification applications. The RCG representation extends the constrained convex generators framework by incorporating an inner exclusion zone, creating sets with topological holes that naturally arise in collision avoidance and safety-critical control problems. We present two equivalent formulations: a set difference representation that provides geometric intuition and a unified parametric representation that facilitates computational implementation. The paper establishes closure properties under fundamental operations, including linear transformations, Minkowski sums, and intersections with convex generator sets. We derive special cases, including roundabout zonotopes and roundabout ellipsotopes, which offer computational advantages for specific norm selections. The framework maintains compatibility with existing optimization solvers while enabling the representation of non-convex feasible regions that were previously challenging to model efficiently. The present work focuses on the theoretical foundations of RCGs---namely, set representation and closure properties---while algorithmic reachability analysis and control synthesis are left for future work.

Keywords: Machine learning, Neural networks, Safety critical systems
Abstract: This abstract considers the problem of learning a neural network approximation of a discrete-time control barrier function (DTCBF) controller while preserving safety guarantees. We first synthesize and verify a robust control barrier function using a Lipschitz-bounded neural network parameterization of the DTCBF. The resulting optimization-based safety filter then serves as an expert policy for training a neural network controller. If the approximation error can be certified to remain within the input disturbance accounted for in the robust DTCBF formulation, the learned controller inherits the same safety guarantees.

Keywords: Biomedical systems, Lyapunov methods, Constrained control
Abstract: Accurate and safe regulation of tumor growth is a central challenge in cancer therapy, where chemotherapy and immunotherapy must be coordinated to suppress malignant cells while preserving immune function. Conventional control approaches often lack formal guarantees of stability or constraint satisfaction, which limits clinical reliability. This work presents an integrated framework for nonlinear cancer dynamics that unifies formal control abstraction and control Lyapunov function principles. The method ensures asymptotic stabilization to the benign equilibrium while enforcing a safety-critical constraint on immune preservation. The use of formal methods also significantly enlarges the region from which the tumor growth dynamics can be safely stabilized. The resulting control law is computationally efficient and suitable for real-time therapy scheduling. In-silico simulations on the Stepanova tumor–immune model demonstrate that the proposed controller achieves effective tumor suppression while strictly maintaining immune safety, and are compared against an unconstrained Pontryagin-based controller.

Keywords: Biomedical systems
Abstract: This paper assesses whether the performance improvement offered by a Proportional-Integral-Derivative-Acceleration (PIDA) controller, also known as Proportional-Integral-Double-Derivative (PIDD or PIDD2) controller, justifies its additional tuning effort and structural complexity compared to a standard PID for closed-loop control of neuromuscular blockade (NMB) with atracurium. To ensure a fair comparison, both controllers were tuned using the same optimization-based methodology. Their performance was evaluated in simulation on a dataset of fifty-three patient models. The PIDA controller demonstrated superior performance, achieving median improvements of 14% in integrated absolute error (IAE), 15% in control effort (TV), 27% in settling time, and 20% in overshoot, while also showing an increased robustness with respect to inter-patient variability. These results confirm that the performance gain provided by the PIDA may justify its additional complexity, making it a promising candidate to enhance quality and safety of NMB control in clinical practice.

Keywords: Biomedical systems, Constrained control, Intelligent systems
Abstract: Achieving safe and autonomous glycemic regulation for Type-1 Diabetes care is an urgent challenge. Although Reinforcement Learning (RL) emerged as a promising paradigm, practical deployment is hindered by the risk of uncontrolled hyperglycemia or hypoglycemia. This work adapts two safe deep RL approaches in the context of automated insulin delivery. The first consists of a Lagrangian constrained Markov decision process that solves a primal–dual scheme with adaptive multipliers, thereby delivering constraint satisfaction in expectation; the second adopts a Barrier–Lyapunov Actor–Critic framework that embeds discrete-time control-barrier conditions and Lyapunov decrease into the learning updates, ensuring stepwise feasibility and promoting stability by design. Simulations under randomized meal timing and size, benchmarked against a standard clinical practice protocol and an unconstrained DRL baseline, indicate improved time-in-range with reduced hypoglycemic events.

Keywords: Biomedical systems, Feedback linearization, Robust control
Abstract: This paper addresses the control design problem for critical care ventilators, focusing on those using two valves for flow regulation. While many valuable results exist in the literature for blower-based ventilators, very few are available for ventilators based on dual valves and, in particular, for those using pinch-valves. This work provides a mathematical model of the flow that circulates through each valve, enabling the compensation of their nonlinearity. Moreover, a virtual coupling between the valves is proposed to limit their saturation and to maximize the flow delivered by the ventilator. The presented approach simplifies the control design, facilitating robustness and performance optimization using frequency-domain techniques. Complex ventilation modes are translated into a cascaded controller, featuring a low-level flow regulator and a high-level pressure controller. The analysis considers flow measurement delays and actuator dynamics, offering insights applicable beyond medical ventilators to general pinch-valve-based flow control systems.

Keywords: Adaptive systems, Neural networks, Biomedical systems
Abstract: Blood glucose control is a critical aspect of type 1 diabetes patients' life. Glucose levels have to remain within a safe range to prevent critical events like hypo- and hyperglycemia. To ensure safety, predictive alarm systems are designed to alert patients in advance before a critical event occurs. In this work, an adaptive personalized alarm system for hypo- and hyperglycemia prevention is presented. It relies on a long short-term memory neural network that continuously adapts its parameters as new data become available. A transfer learning initialization allows the model to be immediately deployed on new subjects using a pre-trained reference network, which is then refined online to account for patient-specific metabolic dynamics and possible changes over time. The predictor estimates the glucose trajectory over a 1-hour horizon, and an alarm triggers early warnings when hypo- or hyperglycemia is predicted. Validation on 100 patients of the UVA/Padova simulator demonstrated high prediction accuracy (median FIT 88.6%, RMSE 3.5 mg/dL) and reliable alarm performance (median F1-scores of 90.9% and 88.2% for hypo- and hyperglycemic events, respectively). Overall, the combination of transfer learning and online adaptation provides a robust, self-personalizing predictive system capable of providing early and accurate alarms under realistic, time-varying conditions.

Keywords: Biomedical systems, Identification, Predictive control for nonlinear systems
Abstract: Tremor is an involuntary, rhythmic, oscillatory movement of a body part, commonly associated with neurological conditions such as Parkinson's disease (PD) and essential tremor (ET). Wearable exoskeletons have emerged as a non-invasive alternative to medication for mitigating the effects of tremor on a patient's activities of daily living. The Tremor Suppression Glove (TSG) is a jamming-based soft exoskeleton designed to attenuate wrist tremors. It consists of two layers of 3D-printed chainmail fabric, vacuum-sealed within a latex film that applies pneumatically controlled confining pressure on the chainmail links, thereby providing a tunable stiffness mechanism. In this work, an empirical model was developed for the confining pneumatic pressure dynamics, and it was used to implement three control strategies: a proportional (P) controller, a Lyapunov-based nonlinear (NL) controller, and model predictive controller (MPC). The evaluation results showed that the MPC outperformed the P and NL controllers, providing smoother control inputs, improved trajectory tracking, and greater energy efficiency. Furthermore, experiments using a tremor simulation device demonstrated that the TSG can effectively attenuate tremors within the frequency range for PD, achieving suppression between 40.21% and 52.35% depending on the applied confining pressure. Future work will focus on sensing tremor magnitude to enable closed-loop tremor suppression.

Keywords: Observers for nonlinear systems, Robotics, Optimization
Abstract: Accurate monitoring and control of contact forces are essential in many robot manufacturing processes. Conventional approaches rely on force–torque sensors mounted between the robot wrist and end-effector, as existing estimation methods often suffer from limited bandwidth and accuracy. Building upon a previously introduced moving horizon estimator (MHE) for external torque estimation in industrial robot joints equipped with secondary encoders, this paper focuses on the impact of different modeling concepts on estimation performance. Specifically, various nonlinear stiffness models of a cycloidal drive are investigated using a dedicated test bench representing an isolated robot joint. The models are compared in terms of estimation accuracy and computational effort. Furthermore, the embedded implementation of the MHE is evaluated across multiple optimization frameworks to assess its real-time feasibility for industrial applications.

Keywords: Nonlinear system identification, Robust adaptive control, Uncertain systems
Abstract: This paper proposes a method for constructing one-step prediction tubes for nonlinear systems using reproducing kernel Hilbert spaces (RKHSs). We approximate a bounded RKHS hypothesis set by a finite-dimensional subspace using bounds based on n-widths and a greedy algorithm for basis reduction. For kernels whose native spaces are norm-equivalent to Sobolev spaces, we derive how the required basis size scales with kernel smoothness and input dimension. This finite-dimensional representation enables the use of convex scenario optimization to obtain violation guarantees for the learned predictor without requiring an a priori bound on the true system’s RKHS norm or Lipschitz constant. The method is demonstrated on an obstacle-avoidance task. We also discuss the main limitations of the current analysis, including dimensional scaling and dependence on i.i.d. data.

Keywords: Predictive control for nonlinear systems, Machine learning, Observers for nonlinear systems
Abstract: Data-driven models are increasingly employed to control complex systems, avoiding first-principle modeling efforts. Model predictive control (MPC) provides a well-suited framework for integrating these data-driven models. In this work, Gaussian processes (GPs) are applied as data-driven models, due to their ability to represent nonlinear dynamics and quantify predictive uncertainty. As MPC performance depends heavily on model accuracy, offset-free strategies are required to mitigate model-plant mismatches and maintain satisfactory control performance under disturbances or faulty models. To ensure offset-free control, the MPC utilizes an augmented version of the GP-based state-space model (GP-SSM) that integrates additional disturbance states. All unmeasured states are estimated by applying moving horizon estimation (MHE) using the same augmented model as the MPC. The proposed offset-free MPC integrating MHE utilizes the same GP-SSM for both algorithms, which is learned from input-state data. This is validated on a simulated 1-tank system. The advantages of the offset-free method are investigated by varying the GP-SSM’s quality through different training data set sizes and by adding disturbances. Control performance is quantitatively evaluated using the MPC objective function.

Keywords: Signal processing, Sensor and signal fusion, Adaptive control
Abstract: Signal pre-processing in the mammalian cochlea is done by a sophisticated feedback loop, where the external stimuli is decomposed into its frequency components and is nonlinearly amplified. To mimic this pre-processing, frequency- selective oscillators, whose characteristic frequency can be adjusted by a controllable input, can be utilized. With this, sampling in the frequency domain is enabled. However, it is often difficult to determine the precise analytic relationship between the input and the characteristic frequency. To this end, the work proposes an adaptive frequency controller is proposed for an Andronov-Hopf oscillator to adjust its characteristic frequency without knowing the parameters of this relationship. For this, it is shown that an Andronov-Hopf oscillator has an unique and asymptotically stable response. By asserting a desired frequency in the harmonic excitation, an adaptive controller is designed that stabilizes the characteristic frequency in a neighborhood of the desired frequency.

Keywords: Manufacturing processes, Identification, Filtering
Abstract: Kalman filter based residual generation followed by statistical change detection is a standard approach for model-based cyberattack detection. The methods are often tested in idealized simulation studies or in case of tests in laboratory facilities typically a detailed analysis of the signal properties is lacking. In this contribution said approaches are tested in a model factory that uses smart industrial sensors, communication and control systems such that the signal properties are close to real word application.

Keywords: Sampled data control, Computational methods, H2/H-infinity methods
Abstract: This paper addresses the problem of designing an H_infty state-feedback controller for sampled-data systems. We propose a new numerical technique to handle infinite-dimensional Differential Linear Matrix Inequalities defined over a compact real interval. For this aim, we use an extended robust version of Finsler's Lemma to convert such infinite-dimensional conditions into a set of efficient, tractable Linear Matrix Inequalities. The proposed methodology provides a systematic way to eliminate the explicit dependence on the time variable, thereby avoiding the need for infinite pointwise evaluations.

Keywords: Coverage control, Autonomous robots, Constrained control
Abstract: This paper investigates coordinated image sampling strategies designed based on coverage control to reconstruct high-quality 3D structural models. Specifically, we experimentally demonstrate significance of appropriate camera rotational control in improving reconstruction accuracy. To this end, we implement two coverage controllers on a robotic testbed: one with fixed camera orientations and one with camera rotational control. The experimental results qualitatively show that rotational control of the cameras substantially enhances model fidelity, while also confirming the real-time feasibility and validity of the proposed controller in hardware experiments. Furthermore, comparative simulations are conducted on a simulator having the ground truth data of the environmental structure, contrasting the proposed method with a baseline approach employing fixed camera orientations. The results quantitatively validate the effectiveness of camera rotational control, as evidenced by improvements in the three metrics, Precision, Recall and F-score.

Keywords: V&V of control algorithms, Autonomous systems, Robust adaptive control
Abstract: Moving-base optoelectronic tracking systems are highly susceptible to carrier motion, which causes line-of-sight (LOS) pointing deviations. To address the degraded tracking accuracy in existing control strategies caused by neglecting the effects of base translational motion, this paper analyzes the coupling mechanism between base motion and LOS pointing. Subsequently, a translational disturbance error model incorporating base velocity, target distance, and azimuth angle is established, and a feedforward control method based on translational disturbance compensation is proposed. Experimental results demonstrate that compared with the traditional proportional-integral-derivative (PID) method, the proposed approach achieves a significant enhancement in the dynamic precision and robustness of the system.

Keywords: Autonomous systems, Cooperative autonomous systems, Safety critical systems
Abstract: Autonomous mining vehicles (AMVs) have demonstrated the potential to improve safety, efficiency, and productivity in open-pit operations through centralised fleet management and GNSS-based localisation. However, reliance on continuous satellite positioning and centralised coordination limits operational flexibility in underground and infrastructure-light environments, where GNSS signals are unavailable and communication is constrained. The high cost of full scale systems and operational complexity of practical scenarios pose challenges to experiment with decentralised cooperative path planning and collision avoidance algorithms for heavy, low-manoeuvrability vehicles under GNSS-denied and perception-constrained conditions, resulting in varying levels of localisation accuracy, sensor degradation, and intermittent vehicle-to-vehicle (V2V) communication. We developed and tested a scaled-down platform for investigating decentralised multi-agent navigation algorithms, enabling experimentation with perception, localisation, planning, control, and communication under conditions representative of real mining operations.

Keywords: Maritime, Aerospace, Computational methods
Abstract: To address the long-term error divergence of underwater inertial navigation system (INS) and the limitation of high-resolution marine gravity measurement in GNSS-denied environments, this paper proposes an inertial-based integrated autonomous navigation and gravity measurement system. The system takes a single-axis rotating ring laser gyro INS as the core, matched with a Doppler velocity log (DVL), depth gauge, and high-precision gravity sensing unit. It can output high-precision navigation information for autonomous underwater vehicles (AUVs) while completing real-time precise marine gravity field measurement. Key technologies including rotation modulation error suppression, underwater dynamic gravity compensation, and gravity-aided passive navigation are studied in this work. Experimental results show that the system achieves an underwater dynamic gravity measurement accuracy better than 1 mGal, and effectively suppresses the long-term error divergence of INS in GNSS-denied environments. This system provides core technical support for long-endurance autonomous navigation of deep-sea AUVs and high-resolution marine gravity field detection.

Keywords: Emerging control applications, Safety critical systems, Robotics
Abstract: Ensuring operational safety is critical for human-to-humanoid motion imitation. This paper presents a vision-based framework that enables a humanoid robot to imitate human movements while avoiding collisions. Human skeletal keypoints are captured by a single camera and converted into joint angles for motion retargeting. Safety is enforced through a Control Barrier Function (CBF) layer formulated as a Quadratic Program (QP), which filters imitation commands to prevent both self-collisions and human-robot collisions. Simulation results validate that the framework achieves real-time motion imitation with safety guarantees.

Keywords: Stability of nonlinear systems, Nonlinear system theory, Robust control
Abstract: We derive sufficient conditions for the stability of a negative feedback interconnection comprising a SISO, LTI system in the feedforward path and a memoryless, time-varying, sector-bounded nonlinearity in the feedback path. Using the notion of mixedness, we show that stability can be guaranteed via the circle criterion, extending prior Nyquist-based results for mixed systems to nonlinear settings.

Keywords: Robust adaptive control, Uncertain systems
Abstract: This paper addresses the problem of asymptotic tracking for high-order control-affine MIMO nonlinear systems with unknown dynamic terms subject to transient state constraints. We introduce Barrier Integral Control (BRIC), a novel algorithm designed to confine the system's state within a predefined funnel, ensuring adherence to the transient state constraints, and asymptotically drive it to a given reference trajectory from any initial condition. The algorithm leverages the innovative integration of a reciprocal barrier function and error-integral terms, featuring smooth feedback control. Notably, BRIC operates without relying on any information or approximation schemes for the (unknown) dynamic terms, which, unlike a large class of previous works, are not assumed to be bounded or to comply with globally Lipschitz/growth conditions. Additionally, the system's trajectory and asymptotic performance are decoupled from the uncertain model, control-gain selection, and initial conditions. Finally, comparative simulation studies validate the effectiveness of the proposed algorithm.

Keywords: Lyapunov methods, Computational methods, Switched systems
Abstract: Recently, continuous, piecewise linear Lyapunov functions were used to compute the minimum average-dwell time needed to assert global exponential stability for switched linear systems. Here, we use continuous, piecewise quadratic Lyapunov functions to do the same and compare the efficiency of the two methods. As in the case for piecewise linear Lyapunov functions, we use linear programming to parameterize our Lyapunov functions on a conical subdivision of the state-space.

Keywords: Sliding mode control, Robust control, Uncertain systems
Abstract: This paper presents a feasibility-limit analysis for prescribed-performance control in the presence of actuator saturation and matched disturbances. Conventional control methods enforce transient and steady-state error bounds through a time-varying performance envelope that converges to a prescribed steady-state value, typically treated as a free design parameter. However, in the presence of disturbances and actuation limits, not all performance specifications are achievable. For a first-order system controlled by a hybrid-gain finite-time sliding-mode controller with an error-function-based transformation, we derive an explicit closed-form lower bound on the achievable steady-state performance level. This bound represents the smallest envelope width that guarantees satisfaction of the prescribed performance constraints without violating actuator limits, thereby defining a hard feasibility boundary in the design space. Selecting a performance level below this bound renders the control problem structurally infeasible, irrespective of gain tuning. Simulation results validate the analytical bound and clearly demonstrate the transition between feasible and infeasible design regimes.

Keywords: Agents networks, Network analysis and control, Modeling
Abstract: This paper presents a theoretical convergence analysis for an opinion-action coevolution model that integrates the opinion updating rule of the Hegselmann-Krause model with a utility-based decision-making mechanism. The model is reformulated into an augmented state-space representation, where the state matrix induces a time-varying social interaction digraph. The convergence analysis is grounded on two existing theoretical findings that establish convergence for the Hegselmann-Krause type of models and containment control systems with multiple stationary leaders, respectively. Results indicate that, if the structure of the interaction digraph stabilizes within finite time, the model either converges to consensus, where all agents’ opinions and actions reach an identical state, or exhibits clustering, where some opinion nodes act as stationary leaders while the remaining nodes approach the convex hull formed by the leaders. Numerical simulations are then provided to validate the theoretical results.

Keywords: Agents networks, Network analysis and control, Optimization
Abstract: We study randomized asynchronous opinion dynamics in the Friedkin-Johnsen model, where agents update according to random activation. We establish a direct convergence proof with an explicit pathwise convergence rate and analyze its dependence on the activation probability vector. We then consider the problem of maximizing this rate. Although convexity of the objective remains open, the problem admits a reformulation with linear-bilinear constraints, which can be handled efficiently via sequential linear programming. In the single-agent update setting, the optimal activation distribution induces a novel centrality measure quantifying each agent's importance for rapid equilibration. Numerical experiments show that this measure differs from classical centralities, while remaining correlated with influence-based ones.

Keywords: Control over networks, Fault detection and identification, Safety critical systems
Abstract: Industrial control systems (ICSs) often consist of many legacy devices, which were designed without security requirements in mind. With the increase in cyberattacks targeting critical infrastructure, there is a growing urgency to develop legacy-compatible security solutions tailored to the specific needs and constraints of real-time control systems. We propose a least significant bits (LSBs) coding scheme providing message authentication and integrity, which is compatible with legacy devices and never compromises availability. The scheme comes with provable security guarantees, and we provide a simple yet effective method to deal with synchronization issues due to packet dropouts. Furthermore, we quantify the control performance loss for both a fixed-point and floating-point quantization architecture when using the proposed coding scheme. We demonstrate its effectiveness in detecting cyberattacks, as well as the impact on control performance, on a hydro power turbine control system.

Keywords: Identification, Agents networks, Optimization algorithms
Abstract: This paper addresses the problem of identifying the influence structure in Friedkin-Johnsen (FJ) opinion dynamics when the available data are incomplete or temporally aggregated. We consider scenarios in which agents’ opinions are only partially observed, either due to missing measurements or because observations represent averages over time intervals. Building upon the dynamic relation intrinsic to the FJ model, we formulate a unified optimization framework that jointly estimates the influence matrix and the latent opinion trajectories, while enforcing model consistency and structural constraints. The proposed alternating-optimization (AO) scheme alternates between convex subproblems, ensuring monotonicity of the overall objective. Numerical experiments on synthetic opinion dynamics demonstrate the robustness, accuracy, and monotonicity properties of the proposed framework compared to existing least-squares and covariance-based approaches.

Keywords: Modeling, Output regulation, Linear systems
Abstract: In this paper, the tracking of a given reference output trajectory is investigated for the class of Boolean control networks, by resorting to their algebraic representation. First, the case of a finite-length reference trajectory is addressed, and the analysis and algorithm first proposed in (Z. Zhang, T. Leifeld, and P. Zhang. IEEE Trans. Automatic Control, 2018) are extended to deal with arbitrary initial conditions and to identify all possible solutions. The approach developed for the finite-length case is then adjusted to cope with periodic reference output trajectories.

Keywords: Stability of nonlinear systems, Control over networks, Uncertain systems
Abstract: In this work, we address the output tracking problem with prescribed performance for the class of SISO pure-feedback systems subject to non-periodic communication in the controller-to-actuator channel. A dynamic threshold strategy is employed to regulate the information exchange. We propose two different approaches for updating the variable threshold of the event-triggered mechanism. It is proven that the proposed triggering scheme can further extend the inter-execution times compared to the conventional relative threshold strategy. Prescribed performance in terms of steady-state accuracy and convergence rate is guaranteed. Simulation results validate the theoretical findings and provide a comparative evaluation between the two proposed approaches and the relative threshold strategy.

Keywords: Optimization, Predictive control for linear systems, Power electronics
Abstract: This paper explores the combination of Model Predictive Control (MPC) and the Sphere Decoding Algorithm (SDA) to control a three-phase, variable-speed induction motor drive through a three-level inverter. It builds upon previous work in which a framework to simplify the integer optimization problem using Laguerre Polynomials (LPs) was introduced. The objective is to obtain a much faster convergence of the algorithm through a reduction in the dimensionality of the optimization problem. Specifically, this work explores different parametric selections to better understand their effects on the performance of the algorithm.

Keywords: Aerospace, Predictive control for linear systems, Stability of nonlinear systems
Abstract: This work introduces two lightweight model predictive control (MPC) approaches for attitude tracking with reaction wheels during spacecraft rendezvous synchronization. Both approaches are based on a novel attitude deviation formulation, which enables the use of inherently linear constraints on angular velocity. We develop a single-loop and a dual-loop MPC; the latter embeds a stabilizing feedback controller within the inner loop, yielding a linear time-invariant system. Both controllers are implemented with CasADi—including automatic code generation—evaluated across various solvers, and validated within the Basilisk astrodynamics simulation framework. The experimental results demonstrate improved tracking accuracy alongside reductions in computational effort and memory consumption. Finally, embedded delivery to an ARM Cortex-M7—representative of commercial off-the-shelf devices used in New Space platforms—confirms the real-time feasibility of these approaches and highlights their suitability for onboard attitude control in resource-constrained spacecraft rendezvous missions.

Keywords: Predictive control for linear systems, Optimal control, Linear systems
Abstract: Building mass dampers (BMDs) are crucial vibration control systems for supertall and high-aspect-ratio buildings. These systems mitigate seismic response by incorporating a relatively low-stiffness connecting layer and effectively utilizing the structure above this layer as a tuned mass damper (TMD) to tune the overall structural dynamics. However, practical applications of BMDs often face challenges such as limited displacement of the connecting layer, which can lead to deviations from optimal tuning. To address this, we previously proposed the model reference predictive control (MRPC) method as an active control technique designed to optimize BMD vibration control under such constraints. Although prior analytical studies and shaking table tests demonstrated MRPC's effectiveness, they also revealed some critical issues, including control force saturation and increased residual vibrations compared to uncontrolled scenarios in certain seismic events. In this study, we improved the MRPC control system to overcome these limitations. We evaluated the proposed enhanced control system through real-time hybrid tests; the system achieved zero occurrences of control-force constraint violations and reduced residual vibrations, demonstrating effective mitigation of the identified challenges.

Keywords: Predictive control for nonlinear systems, Optimal control, Automotive
Abstract: Thermal management is crucial system in electric vehicles that ensures passenger comfort, maintains battery and powertrain performance, and minimizes energy consumption for increased driving range. This paper presents a nonlinear model predictive control (NMPC) strategy for a novel integrated heat pump system that simultaneously conditions two cabin compartments, a battery pack, and powertrain components through shared coolant loop. The NMPC coordinates heat flow requests and coolant pumps operation to achieve temperature setpoints while minimizing total power consumption. To enable real-time feasibility, a move blocking scheme is implemented to reduce the size of the optimal control problem while maintaining long prediction horizon for improved performance. Controller performance is evaluated in simulation, demonstrating effective coordination of all actuators and achievement of desired temperature targets with minimal power usage for two scenarios. Results show that incorporating disturbance preview from superimposed controllers improves cabin inlet air temperature tracking performance by up to 50% without increasing power consumption. Additionally, NMPC is benchmarked against baseline coolant loop controllers in high-fidelity Dymola model-based simulation.

Keywords: Predictive control for nonlinear systems, Robotics
Abstract: This article presents a continuous nonlinear model predictive control formulation for the integrated control of irrigation canal networks with mobile robotic agents. Traditional approaches for incorporating robotic agents into control systems either decouple robot coordination from process control or resort to computationally taxing mixed-integer optimization problems. Our method addresses these limitations by employing smooth proximity functions that seamlessly integrate robot trajectory planning with canal water level control within a unified continuous optimization framework, enabling the controller to simultaneously generate process control inputs and robot movements while managing uncertainty propagation through stochastic model predictive control techniques. The effectiveness of the approach is demonstrated through simulations on both one-dimensional linear canal systems and two-dimensional branched networks.

Keywords: Automotive, Predictive control for nonlinear systems
Abstract: Autonomous racing controllers face the same dynamic uncertainties and external disturbances as professional racing drivers, while meeting strict real-time constraints. This work, replaces the driver with Model Predictive Contouring Control (MPCC), which offers a unified framework for local trajectory optimization and following. MPCC for autonomous racing is limited by bicycle vehicle models, which struggle to capture the dominant vehicle dynamics effects at the limit of handling, leading to suboptimal control decisions. A new formulation based on a two-track vehicle model, with extended tire dynamics is proposed and compared to the existing bicycle model-based controllers in a high-fidelity multi-body simulation environment. The controllers are benchmarked on a simulated FSAE racecar across multiple Formula Student track layouts, each one presenting distinct dynamic challenges. Assessment using consistent metrics demonstrates clear improvements in both on-track performance and real-time implementability for the proposed two-track controller.

Keywords: Adaptive systems, Identification, Robust adaptive control
Abstract: In this note, we propose a robust recursive least squares approach for the adaptive estimation of parameters in dynamic regressor systems. The approach provides linear dynamics of the covariance matrix. It is shown that, by choosing the tuning gains of the parameter update and the covariance updates, exponential convergence of the parameters is achieved while the covariance matrix vanishes to the origin at a slower exponential rate. The analysis also demonstrates that exponential convergence of the parameter estimates is preserved in the presence of bounded measurement error. A simulation study is presented to highlight the properties of the proposed technique.

Keywords: Distributed estimation over sensor nets, Observers for linear systems, Linear systems
Abstract: This paper addresses the distributed state estimation problem for a discrete-time, linear time-invariant system. Building on the framework proposed in (Gao & Yang, IEEE TAC 2025), we exploit the Jordan canonical form of the system matrix to develop a distributed estimation scheme that ensures the asymptotic convergence of the local state estimates to the true system state. In the proposed approach each node reconstructs the components of the system state that are detectable for it through a local Luenberger observer, while employing a consensus-based strategy to estimate the undetectable components. Necessary and sufficient conditions for the existence of a distributed observer that guarantees asymptotic estimation accuracy are derived. Compared with (Gao & Yang, IEEE TAC 2025), the proposed design offers greater flexibility in the selection of the coupling gains and leads to a less restrictive set of conditions for solvability.

Keywords: Distributed estimation over sensor nets, Observers for linear systems, Concensus control and estimation
Abstract: With the advancement of IoT technologies and the rapid expansion of cyber-physical systems, there is increasing interest in distributed state estimation, where multiple sensors collaboratively monitor large-scale dynamic systems. Compared with its continuous-time counterpart, a discrete- time distributed observer faces greater challenges, as it cannot exploit high-gain mechanisms or instantaneous communication. Existing approaches depend on three tightly coupled factors: (i) system observability, (ii) communication frequency and dimension of the exchanged information, and (iii) network connectivity. However, the interdependence among these factors remains underexplored. This paper identifies a fundamental trilemma among these factors and introduces a general design framework that balances them through an iterative semidefinite programming approach. As such, the proposed method mitigates the restrictive assumptions present in existing works. The effectiveness and generality of the proposed approach are demonstrated through a simulation example.

Keywords: Robust control, Concensus control and estimation, Optimal control
Abstract: In sensor networks, accurately estimating the desired states of a system is challenging, primarily due to the lack of an exact system model and the pervasive uncertainties. To overcome this challenge, in this paper, we address the state estimation problem for a time-varying system with a network of sensors with norm-bounded uncertainties. Our proposed method, a distributed robust Kalman filter, incorporates an upper bound for the error covariance matrix and a consensus mechanism to mitigate the effects of norm-bounded uncertainty. Our approach not only improves the state estimation accuracy but also guarantees agreement among the nodes in the sensor network. We demonstrate the effectiveness of our approach by calculating the Mean Squared Error (MSE) and comparing it with state-of-the-art (SOTA) methods [1], [2].

Keywords: Optimal control, Sensor and signal fusion, Control over networks
Abstract: We study remote state estimation in wireless networked control systems with energy-constrained sensors sharing an unreliable channel. Unlike traditional approaches assuming fixed sensor participation, we introduce a population-based framework for dynamic sensor admission control, modeling active, idle, and depleted sensors and accounting for collision-induced packet losses. We formulate a finite-horizon optimal control problem, solved analytically for a shared channel with constant reliability via a linear-quadratic admission policy, highlighting trade-offs between communication effort and estimation accuracy. When multiple sensors transmit simultaneously, collisions increase packet error probability. Motivated by this, we numerically derive an optimal admission policy to show how adaptive admission control improves estimation under energy constraints. Simulations demonstrate that the framework balances estimation accuracy, communication effort, and energy consumption, prolonging sensor lifetime.

Keywords: Robust adaptive control, Observers for linear systems, Electrical power systems
Abstract: A robust feedback–linearizing (FL) control for a DC–DC buck converter under parameter variations and dynamic uncertainties, utilizing only the output voltage and nominal parameters, is presented. The main contribution is a projected adaptive high-gain observer in which the observer gain is updated via a projection operator, ensuring that it remains within prescribed bounds. This approach reduces peaking and excessive noise amplification while preserving fast transients. The key feature is that the uncertainties are combined into a new state, and the observer estimates both the system states and the new lumped state in real time. These estimates are then fed into the FL law to achieve asymptotic voltage regulation. A concise result has been established, showing that the voltage loop is input-to-state stable (ISS) with respect to observer errors and bounded disturbances; the overall closed-loop system attains global practical robust stability against bounded uncertainties and measurement noise. Simulations with variations in input voltage and load resistance confirm the voltage regulation, reduced control effort, and robust performance.

Keywords: Optimization algorithms, Stability of nonlinear systems, Optimization
Abstract: Optimization with preference feedback is an active research area with many applications in engineering systems where humans play a central role, such as building control and autonomous vehicles. While most existing studies focus on optimizing a static user utility, few have investigated its closed-loop behavior that accounts for system transients. In this work, we propose an online feedback optimization controller that optimizes user utility using pairwise comparison feedback with both optimality and closed-loop stability guarantees. By adding a random exploration signal, the controller estimates the descent direction based on the binary comparison feedback between two consecutive time steps. We analyze its closed-loop behavior when interacting with a nonlinear plant and show that, under mild assumptions, the controller converges to the optimal point without inducing instability. Theoretical findings are further validated through numerical experiments.

Keywords: Optimization algorithms, Agents networks, Concensus control and estimation
Abstract: This article presents a second-order fully distributed optimization algorithm, HBNET-GIANT, driven by heavy-ball momentum, for L-smooth and mu-strongly convex objective functions. A rigorous convergence analysis is performed, and we demonstrate global linear convergence under certain sufficient conditions. Through extensive numerical experiments, we show that HBNET-GIANT with heavy-ball momentum achieves acceleration, and the corresponding rate of convergence is strictly faster than its non-accelerated version, NETWORK-GIANT. Moreover, we compare HBNET-GIANT with several state-of-the-art algorithms, both momentum-based and without momentum, and report significant performance improvement in convergence to the optimum. We believe that this work lays the groundwork for a broader class of second-order Newton-type algorithms with momentum and motivates further investigation into open problems, including an analytical proof of local acceleration in the fully distributed setting for convex optimization problems.

Keywords: Optimization algorithms, Constrained control, Computational methods
Abstract: Optimization problems with convex quadratic cost and polyhedral constraints are ubiquitous in signal processing, automatic control and decision-making. We consider here an enlarged problem class that allows to encode logical conditions and cardinality constraints, among others. In particular, we cover also situations where parts of the constraints are nonconvex and possibly complicated, but it is practical to compute projections onto this nonconvex set. Our approach combines the augmented Lagrangian framework with a solver-agnostic structure-exploiting subproblem reformulation. While convergence guarantees follow from the former, the proposed condensing technique leads to significant improvements in computational performance.

Keywords: Optimization algorithms, Concensus control and estimation, Optimal control
Abstract: Trajectory optimization methods provide an efficient and reliable means of computing feasible trajectories in nonconvex solution spaces. However, a well-known limitation of these algorithms is that they are inherently local in nature, and typically converge to a solution in the neighborhood of their initial guess. This paper presents a sequential operator-splitting framework, based on the alternating direction method of multipliers (ADMM), aimed at promoting exploration within the sequential convex programming (SCP) framework. In particular, diverse initial solutions are modeled as agents within the consensus ADMM framework. Driving these agents toward consensus facilitates exploration of the nonconvex optimization landscape. Numerical simulations demonstrate that the proposed method consistently yields equivalent or lower-cost solutions compared to the standard SCP approach, with the same number of or fewer agents.

Keywords: Optimization algorithms, Optimization, Robust control
Abstract: We consider exact tracking algorithms with integral action for solving online optimization problems char- acterized by quadratic cost functions with a time-varying optimal point whose motion is modelled by an (n − 1)th order polynomial. We particularly focus on an algorithm that achieves the fastest convergence factor ((sqrt(k)-1)/(sqrt(k)+1))^{1/n} where k is the condition number of the quadratic cost function. In this work, we provide a state-space realization for the algorithm that is derived from the transfer function interpretation of an associated Lur´e-type feedback system corresponding to the algorithm. The motivation for considering such a state- space realization is that as n becomes large, the dimension of the state-space realization becomes large, leading to issues of numerical ill conditioning associated with the standard controllable canonical form realization of the algorithm. With the proposed state space realization, it is found that as n increases, the condition number of the state-space realization converges to one. This perspective offers new insights into the structure and performance of emerging online optimization algorithms.

Keywords: Optimization algorithms, Optimization
Abstract: This paper investigates the use of a coordinate search algorithm in moving horizon estimation (MHE). In the standard MHE approach, the optimization problem is solved through full numerical optimization, which can be computationally expensive for real-time applications. To improve computational efficiency, iterative derivative-based approaches such as gradient or Newton-based MHE have been proposed. However, they still require the repeated evaluation of the gradient and Hessian of the cost function, which becomes particularly challenging for high-dimensional systems or long horizon lengths. In this work, we propose an iterative, derivative-free MHE formulation based on a coordinate search method. By solving a sequence of one-dimensional optimization problems along coordinate directions, the proposed approach significantly reduces the computational burden while maintaining accurate state estimates. The boundedness of the estimation error is theoretically proved under suitable assumptions. Simulation results on a benchmark nonlinear system confirm the effectiveness of the proposed method, particularly for long horizon lengths.