Keywords: Learning, Emerging control theory, Emerging control applications
Abstract: Distributed control of large-scale networked systems is a classical research topic, with practical applications in a variety of fields. While a rich theory is available, some assumptions such as availability of subsystem dynamics and topology and the subsystems following the prescribed controllers exactly have proven difficult to remove. An interesting direction in recent times to get away from these assumptions has been the utilization of learning for control. In this talk, we consider some problems in control design for distributed systems using learning. Our core message is that utilizing control-relevant properties in learning algorithms can not only guarantee concerns such as stability, performance, safety, and robustness that are important in control of physical systems, but also help with issues such as data sparsity and sample complexity that are concerns during the implementation of learning algorithms.

Keywords: Machine learning
Abstract: Software-defined vehicles (SDVs) are fundamentally reshaping the automotive industry through their data-driven capabilities, predominantly harnessing machine learning (ML) and deep learning. This presentation delves into the current landscape, hurdles, and solutions related to data annotation, ML model development, and cloud aspects within SDVs. Dealing with the deluge of data streams from diverse sensors demands precise and efficient annotation. We tackle real-time decision-making challenges, safety and reliability apprehensions, in SDV deployment. This underscores the paramount importance of continual research, development, testing, and adherence to industry standards for the secure and efficacious deployment of ML models. Perpetual research and collaboration between automakers, AI researchers, and regulatory bodies are imperative. To realize this, it is crucial to develop efficient annotation tools, optimize ML models for real-time decision-making, and efficient deployment of SDVs. Additionally, the presentation advocates for an open platform to validate ML models.

Keywords: Machine learning
Abstract: With the advent of commoditized sensing and scalable AI/ML-based techniques, the operations of many legacy physical systems and processes (e.g., building energy management; renewable energy generation; airline operations) are being digitally reimagined. The emphasis is on deriving business value from efficient operations through intelligent real-time decision making. In this talk, we identify some common pitfalls in real-time decision making for physical systems from an industrial practice perspective. We highlight our attempts to overcome these both using domain knowledge and novel solution techniques. We illustrate our approach with case-studies drawn from various industry verticals and conclude with some lessons learnt.

Keywords: Mechanical systems/robotics
Abstract: In a network of multiple agents, the behaviours of friendship and hostility arise naturally. This leads to signed networks in which the edge weights can take even negative values. Structural balance is an important property associated with such networks which stems from a certain bipartiteness of the networks. Under the Laplacian flow, signed networks exhibit richer dynamics as opposed to networks with strictly positive edge weights. The focus of the talk will be on the role of structural balance in achieving these behaviours of consensus or bipartite-consensus in signed networks.

Keywords: Cooperative control, Nonlinear systems, Robust control
Abstract: The article presents a method for decentralized attitude consensus of multi-agent rigid bodies in presence of both external disturbances as well as communication time-delays. To handle unknown but bounded disturbances an extended state observer (ESO) has been proposed on TSO(3), the estimate of which is feedback to the controller which guarantees that the attitude consensus is achieved with bounded error. It is also assumed that time-delays exist in transmission of states. To prove the stability of time-delayed feedback system, Lyapunov Razumikhin theorem is used. Simulation results have been given to demonstrate the effectiveness of the proposed approach in presence of time-varying disturbances and communication time-delays. Comparison result are also given with the scenario when ESO is not used.

Keywords: Cooperative control, Control of networks, Networked control systems
Abstract: In this work, we propose a linear continuous time opinion dynamics model that leads to the existence of diverse opinions called ‘clustering’ of opinions in a network at steady state. A clustering vector defines the number of clusters in the steady-state opinions pattern, the nodes belonging to each cluster and the ratio of opinion values of any two clusters. Desired clustering in the network refers to the convergence of opinions preserving the properties of the clustering vector. The clustering of opinions in the multiagent framework has applications ranging from task allocation in autonomous agents to social network analysis. Under the assumption that the network has at least one globally reachable node, We propose the design of a clustering matrix in the opinion model that leads to the desired clustering of opinions in a network with cooperative and antagonistic interactions. We also outline properties of the clustering matrix that are analogous to properties of the graph Laplacian matrix for undirected networks. All the results are verified with simulations.

Keywords: Cooperative control, Autonomous systems, Aerospace
Abstract: This paper addresses an active aircraft Defense problem wherein an aircraft is under attack by an external interceptor, and a defender is launched to neutralize the attacker. We develop a nonlinear guidance law that guarantees that the defender intercepts the attacker before it reaches the vicinity of the target. The guidance strategy is derived using the concept of line-of-sight guidance, where the defender is allowed to move toward the attacker while maintaining his position between the target and the attacker. We show that knowing the attacker's maximum acceleration capability is sufficient to guarantee the successful interception of the attacking interceptor. Nevertheless, the proposed guidance strategy is derived within a nonlinear engagement framework, which elucidates its applicability for broader operating conditions, unlike the methods where the linearized dynamics were considered. Finally, we demonstrate the merits and robust performance of the proposed guidance strategies with various engagement scenarios.

Keywords: Cooperative control, Multivehicle systems, Control applications
Abstract: Recent research on guidance strategies has seen an increased focus on cooperative guidance strategies for the three body engagement scenario, where the target aircraft and its defender interceptor cooperate to thwart the threat posed by an attacking interceptor (also called the pursuer). However, with an increase in the number of pursuers, such strategies are unlikely to be effective since even after the defender neutralizes one of the pursuers, the remaining pursuers can still intercept the target aircraft. This leads to a demand for the development of defensive guidance strategies for scenarios where non-cooperating pursuers pursue a target when the target does not have enough defenders to nullify the threat posed by each pursuer individually. This paper first proposes a variant of the pure proportional navigation (PPN) guidance law that allows a pursuer to successfully intercept its target while evading a defender released by the target. Subsequently, we show that when two such non-cooperating pursuers employ the aforesaid strategy against a target with a solitary defender, the pursuers’ strategy fail since they are lured into trajectories by the cooperating aircraft-defender team that leads to their mutual collision. Simulations are performed to validate the claims of the proposed guidance strategy.

Keywords: Agents-based systems, Cooperative control, Networked control systems
Abstract: In this paper, we study the effect of external biases on opinion formation in a group of n agents. To do so, we propose a biased Laplacian model. The proposed model shows that clustering and polarization of opinions are possible outcomes in the cooperative framework, even in the presence of an external bias. We also determine the conditions necessary for the stability of the system in the given framework. Numerical simulations are also presented to illustrate the theoretical results.

Keywords: Agents-based systems, Delay systems, Cooperative control
Abstract: In this paper, the problems introduced by homogeneous communication delay in a discrete average consensus rendezvous problem are investigated. First, we have shown that the amount of communication delay does not affect the system's stability conditions, as the conditions remain the same for delays ranging from zero to any arbitrary finite value. The deviation in the final convergence point due to communication delays from that of the ideal no-delay situation is studied. Based on this, we introduce a margin on the value of gain to keep this deviation in a predefined bound. In an attempt to maintain the deviation in a pre-defined bound, with an increase in communication delay, the system becomes sluggish as the gain needs to be reduced. Also for the same communication delay as the deviation bound reduces, the gain needs to be reduced due to which the system convergence also becomes sluggish. After finding out the gain corresponding to the fastest convergence, we use it to build an estimator with the help of which the agents can reach the average consensus without any deviation even in the presence of communication delay. We also estimate the amount of delay in the communication links in case of an unknown delay.

Keywords: Mechanical systems/robotics
Abstract: Multi-robot systems working in challenging outdoor environments should be robust and adaptable to external disturbances. These environmental disturbances cannot be easily modeled. Hence, designing an adaptive motion planning algorithm for teams working in such conditions is not trivial. In my research, I propose using reinforcement learning techniques to design motion planning algorithms for such multi-robot systems operating in challenging outdoor environments. One of the example applications of such systems is to sample and model physical phenomena at spatiotemporal scales. Spatial fields commonly occurring in nature consist of hotspots exhibiting extreme measurements and much higher spatial variability than the rest of the field, which is characterized by continuous, positively skewed, spatially correlated measurements. In this talk, I present algorithms and strategies to achieve efficient robotic sampling and reconstruction of non-uniform spatial fields. To collect data for modeling such fields, I apply informed path planning: a data collection strategy that computes paths to be traversed by a robot while considering resource constraints, such as power availability, and the uncertainty of the resulting model that should be minimized. I present informed non-myopic path planning techniques for robotic platforms to efficiently collect measurements from a spatio-temporal field and build a model of the underlying physical phenomenon with high accuracy.

Keywords: Mechanical systems/robotics, Autonomous systems, Nonlinear systems
Abstract: This paper presents a motion planning algorithm for quadruped locomotion based on density functions. We decompose the locomotion problem into a high-level density planner and a model predictive controller (MPC). Due to density functions having a physical interpretation through the notion of occupancy, it is intuitive to represent the environment with safety constraints. Hence, there is an ease of use to constructing the planning problem with density. The proposed method uses a simplified the model of the robot into an integrator system, where the high-level plan is in a feedback form formulated through an analytically constructed density function. We then use the MPC to optimize the reference trajectory, in which a low-level PID controller is used to obtain the torque level control. The overall framework is implemented in simulation, demonstrating our feedback density planner for legged locomotion.

Keywords: Mechanical systems/robotics, Modeling and simulation, Optimization algorithms
Abstract: In this study, we investigate the role of joint redundancy in robotic manipulators to enhance force production capabilities during task execution. Drawing inspiration from the adaptability found in nature and the benefits of redundancy in human arm manipulation, we propose that higher redundancy can lead to superior force production. In this context, the influence of prismatic and revolute joints from a redundancy perspective has not been thoroughly examined. We conduct a systematic analysis to assess the impact of joint type on the optimisation of force production or rejection. A novel secondary objective function is introduced to minimize required torques, which also serves as a performance metric for manipulators. We incorporate a prismatic joint in conjunction with traditional revolute joints, to examine the potential benefits of alternative redundancy configurations. Our simulations reveal a significant impact of joint type on manipulator performance, with revolute joints boosting force production and prismatic joints augmenting the isotropy of force handling. Notably, an alternating revolute-prismatic (RPRP) joint sequence significantly enhances force production, suggesting a promising direction for future research.

Keywords: Agents-based systems, Optimization, Simulation
Abstract: This paper addresses the music-playing robot problem, a benchmark problem for spatio-temporal multi-task assignment problem. In the music-playing robot problem, an algorithm must compute the trajectories for a dynamically sized team of robots that will play musical notes by traveling through the specific locations associated with them at their respective specific times. A two-step dynamic resource allocation based on spatio-temporal multi-task assignment problem has been implemented to assign robots to play a musical tune. The algorithm computes the number of robots required to play the music in the first step. In the second step, optimal assignments are computed for the updated team of robots, which minimizes the total distance traveled. Even for individual feasible trajectories, the multi-robot execution may fail if robots encounter a collision. As some time will be utilized for this conflict resolution, robots may not be able to reach the desired location on time. This paper analyses and proves that if robots are operating in a convex region, the solution of the DREAM approach provides collision-free trajectories. The working of the DREAM approach is illustrated using high-fidelity simulations in a Gazebo operated using ROS2. The result clearly shows that the DREAM approach computes the required number of robots and assigns multiple tasks to robots in at most two steps. A simulation of the robots playing the 'Happy Birthday' is available at https://youtu.be/XToicNm-CO8.

Keywords: Neural networks, Learning
Abstract: One of the exciting development in the previous decades has been the capacity to teach robots using Reinforcement Learning (RL) techniques to execute certain tasks. Deep Deterministic Policy Gradient (DDPG) is one of those RL techniques. This paper proposes an adaptive robust controller based on an improved DDPG algorithm for position and velocity control of an n-link robotic manipulator, which is nonlinear and uncertain. The designed controller takes care of model nonlinearities, uncertainties and also time-varying external disturbances. The controller is based on neural networks: multiple actor networks and a critic network. A reward function is proposed in order to guarantee a stable and effective learning of the proposed DDPG agent. Finally, numerical simulation are performed using two-link robot manipulator as an example. The outcomes demonstrate the robustness, adaptability and trajectory tracking accuracy of this control approach. Neural Lyapunov stability is also shown for the proposed controller.

Keywords: Modeling and simulation, Mechanical systems/robotics, Machine learning
Abstract: Most grasping solutions in the literature and industry rely on learning-based planners developed for grippers with rigid fingers, whose grasp geometries can be abstracted deterministically into simple shapes, typically in terms of a single grip width parameter. Soft grippers, on the other hand, have nonlinear relationships between the actuation input and final geometric shape of the grasping surface. Modeling this relationship is important for training accurate reinforcement learning-based grasp planners. In this paper, we present a prototype cable-driven soft robotic gripper, and describe a computer vision-based technique with LASSO regression to transfer the relationship between the length of the cable and the finger's shape parameters in terms of angular deflections to a PyBullet simulation environment. The average root mean-squared error for this regression model was 0.15 rad. This work forms the first step of a proposed real-to-sim-to-real pipeline for training physically accurate soft robotic grasp planners.

Keywords: Control applications, Autonomous systems, Simulation
Abstract: In this paper, we propose a new class of Control Barrier Functions (CBFs) for Unmanned Ground Vehicles (UGVs) that help avoid collisions with kinematic (non-zero velocity) obstacles. While the current forms of CBFs have been successful in guaranteeing safety/collision avoidance with static obstacles, extensions for the dynamic case have seen limited success. Moreover, with the UGV models like the unicycle or the bicycle, applications of existing CBFs have been conservative in terms of control, i.e., steering/thrust control has not been possible under certain scenarios. Drawing inspiration from the classical use of collision cones for obstacle avoidance in trajectory planning, we introduce its novel CBF formulation with theoretical guarantees on safety for both the unicycle and bicycle models. The main idea is to ensure that the velocity of the obstacle w.r.t. the vehicle is always pointing away from the vehicle. Accordingly, we construct a constraint that ensures that the velocity vector always avoids a cone of vectors pointing at the vehicle. The efficacy of this new control methodology is later verified by Pybullet simulations on TurtleBot3 and F1-Tenth.

Keywords: Machine learning
Abstract: In recent years, machine learning models have surpassed human performance in many applications e.g., machine translation, image recognition, etc. However, in critical applications where a small error can make consequential impacts as exemplified above, humans are preferred over ML models, as the latter can incur significant predictive errors. However, in most cases, the ML models are not aware of the presence of humans and therefore, they are trained for full automation. In this talk, we discuss how we can build machine learning models under triage--- a new machine learning setup--- where machines and humans together achieve better performance than what they can achieve at an individual level. More specifically, we report on progress towards making machine learning models aware of the presence of human decision-makers. Given a set of samples, we aim to find out a triage policy that decides which samples should be outsourced to a human expert and which ones should be assigned to the machines, so that machine and human together achieve superior performance than what they could have achieved individually. To that aim, we first introduce the convex learning problem under triage and show that it is NP-hard. Then, we derive an alternative representation of the corresponding objective function. Building on this representation, we further show that the objective is non-decreasing and satisfies approximate submodularity, a recently introduced notion of submodularity. These properties allow simple and efficient greedy algorithms to enjoy approximation guarantees at solving the problem.

Keywords: Machine learning, Neural networks
Abstract: Climate change and global energy crisis are two primary drivers behind the search of several forms of renewable energy sources e.g., wind energy. In studies of wind farm layout design and control, wind frequency map (WFM) plays a crucial role. WFM is a joint probability mapping between wind direction and speed considering their time series behavior. The limited past data suppresses the ability to represent long term variabilities in wind which leads to inaccurate estimation of WFMs resulting in unrealistic calculation of wind power for wind farms. Hence, in this paper, Generative Adversarial Networks (GANs) are explored for generation of wind frequency maps using limited wind time series data. GANs are data driven probabilistic techniques, which captures distribution hidden in data. The distribution of wind scenarios, once captured by GANs, can generate new scenarios. The success of GAN’s ability in capturing the probability distribution hidden in a wind dataset was demonstrated by calculating power production from an optimal layout using the available scenarios and scenarios generated by GANs. Trained GANs can generate many scenarios in future horizon which can be used for robust design of wind farms. Thus, this study can be helpful in efficient design and control of wind farms under wind state uncertainty.

Keywords: Process control, Identification
Abstract: Networked cyber-physical systems are central to several critical infrastructures such as process industries, energy systems, and transportation systems. However, as several recent incidents have shown, they are vulnerable to cyber attacks. Sensors or networks carrying sensor measurements can be compromised, and the resulting malfunctioning of the control system can cause misbehavior. Since the infrastructure systems are safety and economy critical, it is important to detect such attacks and take appropriate steps to make them resilient. This paper addresses the first step: How to detect such attacks? It studies this problem for the process industries by studying the cyber-security of the Tennessee Eastman Process (TEP), which is an open source benchmark that has been developed for the purpose of evaluating process control technology [1] used in industries such as chemical plants and oil refineries. We explore the use of Dynamic Watermarking [2], a pro-active method for detecting attacks. One can employ recursive system identification to develop ARMAX models of the system. The models show that the TEP is non-minimum phase, placing it outside the scope of previous methods. So motivated, we develop a new method that allows the use of Dynamic Watermarking for detecting attacks on non-minimum phase systems. We explore the use of this method to detect a range of attacks on the TEP, including replay attacks, noise injection attacks, and bias injection attacks.

Keywords: Stochastic systems, Control applications
Abstract: The Target Guarding Problem (TGP) or the Asset Defending Problem has acquired growing attention in recent years. This problem comprises of two players and a stationary target. One of the two players defends the target against the second player who tries to capture the target. In this paper, the TGP is analyzed in a game-theoretic framework and a Nash Equilibrium is derived. The Bayesian counterpart of the game, where the attacker’s information is ambiguous to the defender, is studied by extending the same framework and a Bayesian Nash Equilibrium (BNE) is derived for this case. Numerical experiments are provided to illustrate the effectiveness of the BNE strategy.

Keywords: Simulation, Linear systems, Control applications
Abstract: Appropriate control systems design for any nonlinear nuclear reactor is essential for efficient nuclear energy generation. This paper studies a design of emph{a single linear controller} for a nonlinear pressurized water nuclear reactor (PWNR). The considered model is point kinetics with one delayed neutron group, and the control objective is to track the reference reactor power of an emph{uncertain} nonlinear PWNR, while satisfying the desired specifications. The PWNR is linearized for the controller design, and a single linear PID controller is designed. The efficacy of the designed controller is tested on the nonlinear PWNR model for various test cases, including changes in reference reactor power levels and the controller performance in the presence of parametric uncertainty.

Keywords: Adaptive systems, Control applications, Simulation
Abstract: An adaptive Energy Management System (EMS) is proposed for Battery/Supercapacitor (SC) based Hybrid Energy Storage System (HESS) for use in Electric Vehicles (EVs). The objective of this study is to improve the performance and effectiveness of HESS by developing a filter based EMS using fuzzy control while taking actual topographical details into consideration. In this technique the demanded power is split into high frequency and low frequency parts using a low pass filter where the regulating frequency of the filter is adjusted by a Fuzzy Controller. The high and low frequency power components are delivered by the SC and battery respectively. The actual road slope data are computed using Contour Positioning System. The developed adaptive filter technique is examined for different road slope conditions such as city tour, downhill and uphill journeys. The proposed Fuzzy filter based EMS is assessed as per the energy consumption by the sources and its performance is compared with that of the Fuzzy based EMS. The performances of these two controllers are evaluated based on Mean battery power and Ampere hour (Ah) throughput of the battery. The Fuzzy filter based EMS showed a reduction of 28.25%, 44.08% and 23.85% in battery Ah throughput during uphill, city tour and downhill journeys respectively in comparison to only fuzzy controller. This shows that the proposed adaptive EMS technique is able to reduce the impact of power fluctuations on the battery and improve its performance.

Keywords: Nonlinear systems, Robust control
Abstract: In this paper, we present a super-twisting sliding mode controller that is robust in the presence of partial actuator failure for a coupled nonlinear 6-degree freedom torpedo-shaped autonomous underwater vehicle. We develop the appropriate sliding surfaces and control laws to effectively regulate the vehicle's speed, depth, and steering motion. The stability of the system is proven, and the appropriate gain is determined using the Lyapunov approach. Furthermore, range analysis is performed to determine an allowable range for the partial actuator failure cases corresponding to the thruster, rudder, and stern actuators. The proposed control laws are validated using simulation, and results are presented.

Keywords: Linear systems, Identification, Emerging control theory
Abstract: Models of the behaviour of processes form an important and pervasive role in the design, monitoring, control and optimization of processes. Models can either be derived from a physical understanding or from data. With the explosion in the quantity and quality of data, the development of data driven models (also known as system identification), is increasingly playing a crucial role. In classical approaches to modelling of engineered systems, the input variables are assumed to be exactly known without any errors or noise. This is an appropriate assumption when the data are obtained through controlled experiments. However, in many cases when the data are obtained from operating processes through sensors, all the variables are contaminated with errors, with possibly different variances. Deriving models from such data is known as errors-in-variables (EIV) modelling. Despite significant advances in solving this problem, it is not as popular perhaps due to a lack of a fully automated approach requiring little or no additional user inputs. In this talk, I will present our efforts over the last twenty years to develop such an automated approach based on Principal Components Analysis as a core technique. Our key contributions include a simultaneous approach for estimating a linear model along with error variances from data, a theoretical criteria for estimating the order of the system based on the eigenvalues of the covariance matrix of appropriately scaled data, and the relation between true rank of the lagged data matrix and order of a dynamical system. There exists a distinct possibility of extending this approach to solve the partial total least squares linear regression problem, and developing a unified approach for solving the classical or EIV system identification problem.

Keywords: Agents-based systems, Stochastic systems, Decentralized control
Abstract: We consider LQG teams with two agents receiving partial and asymmetric linear observations, where the agents are mistrustful of each other. An agent must choose actions that does not allow the other agent to infer its private information from the knowledge of this action. Our main finding is that privacy preservation is possible for an agent by choosing a strategy that constrains the environmental random vector to the subspace of the other agent's observations. Privacy constraints impose linear constraints on the agents' strategies, resulting in convex team problems. When both agents have privacy constraints, they must choose strategies that constrain the random vector to the same subspace, effectively reducing the problem to a centralized one. We provide examples to illustrate our findings.

Keywords: Agents-based systems, Robust control, Adaptive systems
Abstract: This work discusses the design of a robust control strategy for leader-follower consensus of multi-agent systems for agents modelled as perturbed double integrator dynamics. The prime objective is to achieve a finite time consensus in the presence of unknown bounded external perturbations. The adopted distributed consensus protocol is based on integral sliding mode (ISM) with the nominal control as a finite-time consensus protocol. The robust control is designed using an adaptive switching term, which rejects the external disturbance. The proposed combination of these two control laws guarantees robustness while preserving the finite time consensus, and additionally, the adaptive gain relaxes the knowledge of the upper bound of disturbance. The robust stability of the closed-loop system is established using Lyapunov analysis, ensuring the fixed time convergence of the MAS with uniform boundedness. A simulation example is included to verify the theoretical results.

Keywords: Quantized systems, Networked control systems, Switched systems
Abstract: This paper analysis the consensus of Multi-Agent systems in the presence of imperfect information exchange between the agents. In particular, the quantized values of the relative state measurements are exchanged. We allow for the communication occurring on each edge to have a possibly different quantization error, which has not been considered in the literature. The quantized consensus regions are derived for static graphs. A switching graph topology controller is proposed to obtain a smaller quantized consensus region while utilizing fewer communication edges and less accurate information transfers. Further, given a static graph that cannot be modified, a switching quantization scheme is proposed, which ensures the best consensus region is obtained. The results are supported through computer simulations.

Keywords: Agents-based systems, Automotive, Process control
Abstract: A hybrid electric vehicle (HEV) uses an electric motor (EM) driven by a battery alongside the conventional internal combustion engine (ICE) to provide the required energy to run the vehicle. Distribution of the required power demand between the ICE and the EM is one of the crucial tasks of a vehicle energy management system (EMS). Finding the optimal torque distribution between the two sources of energy while minimizing fuel consumption of the vehicle and maintaining the state of charge (SoC) of the battery is an optimal control problem. Offline control and real-time control of HEV are studied. For offline control, the optimal control problem is converted into a nonlinear programming (NLP) problem through control vector parameterisation (CVP) and solved using sequential quadratic programming (SQP) for a given predetermined drive cycle. For real-time control, a model-free reinforcement learning (RL) framework is proposed to handle external unknown disturbances such as acceleration / deceleration during a drive cycle. The results obtained with the RL-framework for urban and rural drive cycles show the efficacy of the RL-framework for real-time optimisation of fuel consumption in a HEV under uncertain conditions.

Keywords: Aerospace, Control applications
Abstract: This paper addresses cooperative guidance for multiple interceptors targeting fast-moving non-accelerating targets. It proposes a strategy using arbitrary time consensus to establish unanimous agreement on interceptors' time-to-go values early on. This parameter is crucial for successful simultaneous target interception. The paper introduces two salvo guidance strategies based on deviated pursuit and true proportional navigation (TPN), which are modified using a definitive consensus protocol to achieve consensus on time-to-go values. By integrating this consensus mechanism into the guidance laws, interceptors can maneuver to adjust trajectories to rendezvous with targets. The paper also emphasizes collaborative decision-making through neighbor-to-neighbor communication to enhance overall guidance performance. Extensive numerical simulations validate the proposed approach, highlighting its ability to achieve simultaneous target interception consistently and emphasizing the importance of cooperative time-constrained guidance for successful mission outcomes.

Keywords: Optimal control
Abstract: Lion and Man game is a classical Pursuit-Evasion game. Recently the game is considered with integral constraints. In this short article, we consider the game with the same integral constraints, but with an incomplete information about these constraints. We show that, under this incomplete information, the lion can always catch the man in a finite time.

Keywords: Nonholonomic systems, Aerospace, Optimal control
Abstract: This work constructs near-optimal continuous curvature paths for convergence to a straight line or a circular path in three-dimensional (3D) space to minimize tracking errors arising out of constraints on speed and turn radius of Unmanned Air vehicles (UAVs). It is developed in two stages: (i) determination of the optimal CSC path (Circular path of minimum turn radius (C)-Straight line path (S)-Circular path of minimum turn radius (C)) (3D Dubins path) from the initial pose to the target path, and (ii) construction of the continuous curvature path using Bezier curves with the aid of the Dubins reference path. This algorithm solves the problem of curvature discontinuity inherent with the CSC path for path convergence problems with minimal compromise in its optimal length by maintaining the maximum curvature limit. Simulation results demonstrate the efficacy of the proposed approach.

Keywords: Fault detection/accomodation, Optimization, Process control
Abstract: The current work proposes a game theory based approach for optimal Sensor Placement Design (SPD) for Fault Detection and Isolation (FDI) problems. In the proposed approach, both naturally occurring passive faults, as well as attacker induced active faults are considered. Similarly, apart from sensors (active sensors) to be specifically chosen from FDI perspective, existing (passive) sensors chosen for other requirements are also incorporated. The approach is thus suitable for SPD of modern day processes where the process is tightly coupled with automated and increasingly Industrial Internet of Things (IIOT) based control and monitoring systems. In such systems, apart from naturally occurring process faults, faults induced by malicious action of an attacker are also of concern. The SPD problem is formulated as a zero-sum game between attacker (a set of passive and active faults) and detector (a set of passive and active sensors). The game payoff considered in the proposed work incorporates stochastic events such as occurrence of faults and sensor failures. The detector’s Nash Equilibrium (NE) strategy is considered to be the optimal SPD for a given FDI problem. The utility of the proposed approach is demonstrated on an illustrative example and a CSTR case study.

Keywords: Optimal control, Aerospace
Abstract: In this paper, motion planning for a vehicle moving on a unit sphere with unit speed is considered, wherein the desired terminal location is fixed, but the terminal orientation is free. The motion of the vehicle is modeled to be constrained by a maximum geodesic curvature Umax, which controls the rate of change of heading of the vehicle such that the maximum heading change occurs when the vehicle travels on a tight circular arc of radius r = 1/sqrt(1 + Umax^2). Using Pontryagin's Minimum Principle, the main result of this paper shows that for r <= 1/2, the optimal path connecting a given initial configuration and a final location on the sphere belongs to a set of at most seven paths. The candidate paths are of type CG, CC, and degenerate paths of the same, where C = L, R denotes a tight left or right turn, respectively, and G denotes a great circular arc.

Keywords: Optimal control, Linear systems, Learning
Abstract: We consider a model free off-policy based reinforcement learning algorithm developed for the linear quadratic regulator control of a two wheeled self balancing robot (TWSBR) in discrete time setting. The usual methods to control TWSBR utilises its mathematical model that may not always be available or may not be accurate. Reinforcement learning (RL), which is an important branch of artificial intelligence, offers model free techniques that can be applied to systems whose complete model is unknown. The proposed approach uses a subtopic of reinforcement learning called as off-policy RL, employing separate policies for data generation and value function evaluation. The proposed model-free off-policy RL algorithm improves the control policy using generated data samples and offers the advantage of being immune to bias arising from probing noise added to the control input to satisfy the requirement of persistence of excitation. Numerical simulations are provided to demonstrate effectiveness of the approach.

Keywords: Observers for nonlinear systems, Stability of nonlinear systems, Grey-box modeling
Abstract: A novel approach to data-enabled control of discrete nonlinear control-affine systems with uncertain (“graybox”) dynamics is given here. The gray-box dynamics model accounts for known dynamics and lumps together the effects of disturbance inputs and poorly known dynamics into one time-varying unknown input, which is estimated in real time. This data-enabled approach leads to robust and stable real-time tracking control of desired output trajectories in the presence of model uncertainties. The lumped unknown input is estimated by a H¨older-continuous robustly stable learning scheme, using input-output data in discrete time. This leads to finite-time convergence of the estimation errors of the unknown dynamics to a bounded neighborhood of the zero vector, provided the system is Lipschitz-continuous with respect to states, inputs, outputs, and time. A Lyapunov analysis is carried out to show the nonlinear stability and robustness of the uncertainty observer and tracking control law. This results in simultaneous real-time uncertainty estimation and tracking control. Numerical simulation results for an inverted pendulum with imperfectly known dynamics are carried out, which agree with the theoretical results on stability and robustness.

Keywords: Computational methods, Nonlinear systems, Optimization
Abstract: In this work, we provide a novel data-driven technique for maximizing controllability and observability for a class of nonlinear systems, namely, control-affine nonlinear systems. In particular, we use the Koopman operator framework to derive a linear representation of control-affine systems in the space of functions (Koopman observables). This linear representation is then used to define two positive symmetric matrices, which are shown to capture the energy interpretation of gramians, thus leading to definitions of controllability and observability gramians for control-affine systems. These gramians are then used to derive optimization problems that can be solved by available solvers to obtain the optimal input and output matrices which maximize control- lability and observability for control-affine systems. Furthermore, since we use the Koopman operator framework, we show how the optimal solutions can be computed purely from time-series data of the states of the system. Finally, we demonstrate the efficacy of our approach by performing simulations on different systems, namely, two networked linear systems, one nonlinear network and finally on IEEE 68 bus system.

Keywords: Nonlinear systems, Systems biology
Abstract: Oscillatory networks are important in several biological contexts. However, establishing simple conditions for their existence is challenging, especially in higher dimensional models. Here, we apply an analytical technique to provide an elementary proof for the existence of oscillation in a benchmark cyclic three-dimensional oscillator. This analytical technique is based on the Brouwer's fixed point theorem. We establish the presence of a positively invariant region. A subset of this region is torus-like with the property that the flow maps a cross-section of the torus into itself. This leads to the existence of a limit cycle solution. These results should help in tracing the path of the limit cycle in the phase space.

Keywords: Stability of nonlinear systems, Robust control, Control applications
Abstract: Fixed-time convergence has recently gained much attention in the research community. Interest in Fixed-time control is because such techniques offer uniform boundedness on the settling time. In this paper, we propose a novel Fixed-time Sliding Mode Control (FxtSMC) scheme for a class of non-linear second-order systems with exogenous disturbances. The design provides a fast convergence due to the purposeful addition of specific non-linear terms in the switching manifold. The control law proposed is continuous and free from singularity. Lyapunov method augmented with homogeneous system theory is employed to prove closed-loop stability and establish fixed-time convergence of states. Simulations illustrate the theoretical claims and highlight the efficacy control allocation. Further, the applicability of the proposed FxtSMC in robotic systems is tested for trajectory tracking application of quadrotor and 2-DOF robot manipulator. Results show that the design indeed offers a robust, chatter-free, high-precision tracking performance.

Keywords: Automotive
Abstract: Heavy Commercial Road Vehicles (HCRVs) are more prone to rollover owing to their higher centre of gravity. A rollover prevention objective is formulated in this study as a Model Predictive Control (MPC) problem. The proposed MPC formulation combines the objectives of restricting load transfer ratio, a key indicator of a rollover, with the prevention of wheel lock. The controller was experimentally evaluated using a Hardware-in-Loop (HiL) setup for test manoeuvres based on National Highway Traffic Safety Administration (NHTSA) standards. The developed MPC controller effectively prevents rollover using differential braking while avoiding wheel lock

Keywords: Estimation, Mechatronics, Aerospace
Abstract: This paper presents an estimation of highly maneuverable targets traveling at high speeds. A highly maneuvering target can perform many types of maneuvers. It is not possible to make different types of estimators for different type of maneuvers. A Nonlinear Aerodynamic Model-based State Estimator is available in the literature, which works for ballistic and spiraling targets without any change. This paper has made an effort to evaluate the performance of a Nonlinear Aerodynamic Model-based State Estimator for other types of maneuvers so that a single estimator can be used for any maneuver. In this work, the dynamics of an incoming target are modeled with the nine states containing three relative positions, three relative velocities along with target's inverse ballistic coefficient and two maneuvering coefficients. The Extended Kalman filter-based state estimation approach is presented for tracking maneuvering targets, which provides satisfactory results for various target maneuvers without any change in filter formulation and tuning. The onboard sensor provides measurements, which are range, range rate, and two gimbal angles about azimuth and elevation.

Keywords: Control applications, Delay systems, PID control
Abstract: Time delay is an inherent characteristic of real-time systems, which can have detrimental effects on systems performance and may even result in instability. Therefore, in this paper, a two-degree-of-freedom (TDF) fractional order controller design for the fractional order time delay systems is proposed based on the internal model control (IMC) approach. It comprises set-point and feedback controllers, which can regulate the transient performance and robustness of the system from delay effects. The proposed approach comprises two tuning parameters, which are analytically derived based on the stability analysis. The proposed methodology’s performance is examined under servo, regulatory, and parametric analyses. Numerous performance parameters and indices are estimated for precise analysis. Its performance has been compared to recent state-of-the-art techniques and validated on the real-time four-tank system to analyze its effectiveness.

Keywords: Identification, Estimation, Modeling and simulation
Abstract: Modeling of a Li-ion battery plays a pivotal role in the battery management system (BMS) of electrical vehicles (EVs). In battery modeling, the fractional order model (FOM) becomes the better alternative compared to the integral order model (IOM) as it provides greater accuracy and deals with the fractional calculus in solving the differential equations of the battery. However, the precise estimation of the parameters is required for the proper use of the FOM in practical applications. In this paper, the recursive least square (RLS) algorithm is considered based on its recursive nature and its ability to estimate the parameter when the new data is already available. For improved parameter estimation, the forgetting factor (FF) has been taken into account in the RLS method. In order to add some weights to the estimates a variable forgetting factor (VFF) is introduced in the RLS algorithm. This paper analyzes the FOM and its parameters based on the RLS algorithm in which the VFF has been incorporated. A new approach to RLS has been proposed by considering the aging effect of the battery after that it went through a significant number of charging and discharging cycles. The comparative analysis for different cycles and variations in parameters for the aging effect and variable forgetting factor recursive least square (VFFRLS) method can be seen from the results.

Keywords: Nonlinear systems, Robust control
Abstract: In this paper, we present the control of a torpedo-shaped autonomous underwater vehicle (AUV) for payload delivery using an adaptive robust control method. Here we design a control algorithm that can adapt to changing system parameters and external disturbances for underwater vehicles. The approach estimates the unknown system parameters and also accounts for external disturbances. We use the Lyapunov stability theory to prove the control law's stability. The functionality of the designed controller is shown in the simulation by changing the mass of AUV, simulating the dropping of a payload. It is shown that the proposed controller with a particular choice of gains can handle payload changes up to 20 percent of its initial weight.