Filippos Fotiadis
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Filippos Fotiadis has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Kyriakos G. Vamvoudakis
Daniel Guggenheim School of Aerospace Engineering, Georgia Institute of Technology, Atlanta, GA, USA
7 shared publications - George A. Rovithakis
Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki, Thessaloniki, Greece
2 shared publications - Ufuk Topcu
University of Texas at Austin
2 shared publications - Ali Bouland
Department of Mechanical Engineering, Virginia Tech, Blacksburg, VA, USA
1 shared publication - Andrea L'Afflitto
Grado Department of Industrial and Systems Engineering, Virginia Tech, Blacksburg, VA, USA
1 shared publication - Andrew J. Kurdila
Department of Mechanical Engineering, Virginia Tech, Blacksburg, VA, USA
1 shared publication
Latest Publications
- 15 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 10, 1237-1242
Linear–Quadratic Gaussian Games With Distributed Sparse Estimation
Linear-quadratic Gaussian games provide a framework for modeling strategic interactions in multi-agent systems, where agents must estimate system states from noisy observations while also making decisions to optimize a quadratic cost. However, these formulations usually require agents to utilize the full set of available observations when forming their state estimates, which can be unrealistic in large-scale or resource-constrained settings. In this paper, we study sparse distributed estimation for feedback implementations of linear-quadratic Gaussian (LQG) games with individual noisy observations. Agents implement feedback Nash strategies computed from the corresponding LQG game without measurement noise, while their state estimates are constructed using sparse subsets of available observations. To enforce sparsity in the estimation stage, we design a distributed estimator that balances estimation effectiveness with interagent measurement sparsity via a group lasso problem, while agents implement feedback Nash strategies based on their state estimates. We provide sufficient conditions under which the sparse estimator is guaranteed to trigger a corrective reset to the optimal estimation gain, ensuring that estimation quality does not degrade beyond a level determined by the regularization parameters. Simulations on a formation game show that the proposed approach yields a significant reduction in communication resources consumed while only minimally affecting the nominal equilibrium trajectories.
- 15 June 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 10, 1237-1242
Linear–Quadratic Gaussian Games With Distributed Sparse Estimation
Linear-quadratic Gaussian games provide a framework for modeling strategic interactions in multi-agent systems, where agents must estimate system states from noisy observations while also making decisions to optimize a quadratic cost. However, these formulations usually require agents to utilize the full set of available observations when forming their state estimates, which can be unrealistic in large-scale or resource-constrained settings. In this paper, we study sparse distributed estimation for feedback implementations of linear-quadratic Gaussian (LQG) games with individual noisy observations. Agents implement feedback Nash strategies computed from the corresponding LQG game without measurement noise, while their state estimates are constructed using sparse subsets of available observations. To enforce sparsity in the estimation stage, we design a distributed estimator that balances estimation effectiveness with interagent measurement sparsity via a group lasso problem, while agents implement feedback Nash strategies based on their state estimates. We provide sufficient conditions under which the sparse estimator is guaranteed to trigger a corrective reset to the optimal estimation gain, ensuring that estimation quality does not degrade beyond a level determined by the regularization parameters. Simulations on a formation game show that the proposed approach yields a significant reduction in communication resources consumed while only minimally affecting the nominal equilibrium trajectories.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- p. 1-16
Deception Against Data-Driven Linear-Quadratic Control
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- p. 1-16
Deception Against Data-Driven Linear-Quadratic Control
- Open Access
- 22 June 2025
- Published by Wiley in International Journal of Robust and Nonlinear Control
A Physics‐Informed Learning Framework to Solve the Infinite‐Horizon Optimal Control Problem
We propose a physics‐informed neural networks (PINNs) framework to solve the infinite‐horizon optimal control problem of nonlinear systems. In particular, since PINNs are generally able to solve a class of partial differential equations (PDEs), they can be employed to learn the value function of the infinite‐horizon optimal control problem via solving the associated steady‐state Hamilton‐Jacobi‐Bellman (HJB) equation. However, an issue here is that the steady‐state HJB equation generally yields multiple solutions; hence, if PINNs are directly employed on it, they may end up approximating a solution different from the optimal value function of the problem. We tackle this by instead applying PINNs to a finite‐horizon variant of the steady‐state HJB with a unique solution that uniformly approximates the optimal value function as the horizon increases. An algorithm to verify if the chosen horizon is large enough is also given, as well as a method to extend it—with reduced computations and robustness to approximation errors—in case it is not. Unlike many existing methods, the proposed technique works well with nonpolynomial basis functions, does not require prior knowledge of a stabilizing controller and does not perform iterative policy evaluations. Simulations are performed, which verify and clarify theoretical findings.
- 22 June 2025
- Published by Wiley in International Journal of Robust and Nonlinear Control
Open AccessA Physics‐Informed Learning Framework to Solve the Infinite‐Horizon Optimal Control Problem
We propose a physics‐informed neural networks (PINNs) framework to solve the infinite‐horizon optimal control problem of nonlinear systems. In particular, since PINNs are generally able to solve a class of partial differential equations (PDEs), they can be employed to learn the value function of the infinite‐horizon optimal control problem via solving the associated steady‐state Hamilton‐Jacobi‐Bellman (HJB) equation. However, an issue here is that the steady‐state HJB equation generally yields multiple solutions; hence, if PINNs are directly employed on it, they may end up approximating a solution different from the optimal value function of the problem. We tackle this by instead applying PINNs to a finite‐horizon variant of the steady‐state HJB with a unique solution that uniformly approximates the optimal value function as the horizon increases. An algorithm to verify if the chosen horizon is large enough is also given, as well as a method to extend it—with reduced computations and robustness to approximation errors—in case it is not. Unlike many existing methods, the proposed technique works well with nonpolynomial basis functions, does not require prior knowledge of a stabilizing controller and does not perform iterative policy evaluations. Simulations are performed, which verify and clarify theoretical findings.
- 20 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 8, 2145-2150
Convergence Rates of Online Critic Value Function Approximation in Native Spaces
This paper derives rates of convergence of online critic methods for the estimation of the value function for a class of nonlinear optimal control problems. Assuming that the underlying value function lies in reproducing kernel Hilbert space (RKHS), we derive explicit bounds on the performance of the critic in terms of the kernel functions, the number of basis functions, and the scattered location of centers used to define the RKHS. The performance of the critic is precisely measured in terms of the power function of the scattered bases, and it can be used either in an a priori evaluation of potential bases or in an a posteriori assessments of the value function error for basis enrichment or pruning. The most concise bounds in the paper describe explicitly how the critic performance depends on the placement of centers, as measured by their fill distance in a subset that contains the trajectory of the critic. To the authors’ knowledge, precise error bounds of this form are the first of their kind for online critic formulations used in optimal control problems. In addition to their general and immediate applicability to a wide range of applications, they have the potential to constitute the groundwork for more advanced “basis-adaptive” methods for nonlinear optimal control strategies, ones that address limitations due to the dimensionality of approximations.
- 20 June 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Control Systems Letters
- Vol. 8, 2145-2150
Convergence Rates of Online Critic Value Function Approximation in Native Spaces
This paper derives rates of convergence of online critic methods for the estimation of the value function for a class of nonlinear optimal control problems. Assuming that the underlying value function lies in reproducing kernel Hilbert space (RKHS), we derive explicit bounds on the performance of the critic in terms of the kernel functions, the number of basis functions, and the scattered location of centers used to define the RKHS. The performance of the critic is precisely measured in terms of the power function of the scattered bases, and it can be used either in an a priori evaluation of potential bases or in an a posteriori assessments of the value function error for basis enrichment or pruning. The most concise bounds in the paper describe explicitly how the critic performance depends on the placement of centers, as measured by their fill distance in a subset that contains the trajectory of the critic. To the authors’ knowledge, precise error bounds of this form are the first of their kind for online critic formulations used in optimal control problems. In addition to their general and immediate applicability to a wide range of applications, they have the potential to constitute the groundwork for more advanced “basis-adaptive” methods for nonlinear optimal control strategies, ones that address limitations due to the dimensionality of approximations.
- 13 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3, 6014-6019
A Physics-Informed Neural Networks Framework to Solve the Infinite-Horizon Optimal Control Problem
In this work, we leverage physics-informed neural networks (PINNs) to approximately solve the infinite-horizon optimal control problem for nonlinear systems. Specifically, since PINNs are generally able to solve a class of partial differential equations, they can be employed to approximate the value function in the infinite-horizon optimal control problem, via solving the associated steady-state Hamilton-Jacobi-Bellman (HJB) equation. However, the issue with such a direct approach is that the steady HJB equation generally yields more than one solution, hence directly employing PINNs to solve it can lead to divergence of the method. To tackle this problem, we instead apply PINNs to a finite-horizon variant of the steady-state HJB equation which has a unique solution, and which uniformly approximates the infinite-horizon optimal value function as the horizon increases. A method to verify whether the selected horizon is large enough is also provided, as well as an algorithm to increase it with reduced computations if it is not. Unlike conventional methods, the proposed approach does not require knowledge of a stabilizing controller, the execution of computationally expensive iterations, or polynomial basis functions for approximation.
- 13 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3, 6014-6019
A Physics-Informed Neural Networks Framework to Solve the Infinite-Horizon Optimal Control Problem
In this work, we leverage physics-informed neural networks (PINNs) to approximately solve the infinite-horizon optimal control problem for nonlinear systems. Specifically, since PINNs are generally able to solve a class of partial differential equations, they can be employed to approximate the value function in the infinite-horizon optimal control problem, via solving the associated steady-state Hamilton-Jacobi-Bellman (HJB) equation. However, the issue with such a direct approach is that the steady HJB equation generally yields more than one solution, hence directly employing PINNs to solve it can lead to divergence of the method. To tackle this problem, we instead apply PINNs to a finite-horizon variant of the steady-state HJB equation which has a unique solution, and which uniformly approximates the infinite-horizon optimal value function as the horizon increases. A method to verify whether the selected horizon is large enough is also provided, as well as an algorithm to increase it with reduced computations if it is not. Unlike conventional methods, the proposed approach does not require knowledge of a stabilizing controller, the execution of computationally expensive iterations, or polynomial basis functions for approximation.
- 31 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 69 (4) , 2249-2264
Data-Driven Actuator Allocation for Actuator Redundant Systems
In this paper, we consider the problem of optimally augmenting an actuator redundant system with additional actuators, so that the energy required to meet a given control objective is minimized. We study this actuator selection problem in two distinct cases; first, in the case where the control objective of the system is not known a priori, and second, in the case where the control objective is a linear state- feedback control law. In the latter scenario, knowledge of the system's state and input matrices is required to solve the corresponding actuator selection problem. However, we relax this requirement by exploiting trajectory data gathered from the system, and using them to iteratively approximate the anti-stabilizing solution of an associated algebraic Riccati equation (ARE). Notably, the proposed iterative procedure is proved to be small-disturbance input-to-state stable (ISS) even though the ARE associated with it entails no strictly positive-definite constant term; a result that significantly extends prior work. Finally, to further exploit the obtained trajectory data, we show that these can be used to perform online actuator fault detection without knowledge of the system's matrices, and with complexity lower than that of existing methods. Simulations showcase the theoretical findings.
- 31 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 69 (4) , 2249-2264
Data-Driven Actuator Allocation for Actuator Redundant Systems
In this paper, we consider the problem of optimally augmenting an actuator redundant system with additional actuators, so that the energy required to meet a given control objective is minimized. We study this actuator selection problem in two distinct cases; first, in the case where the control objective of the system is not known a priori, and second, in the case where the control objective is a linear state- feedback control law. In the latter scenario, knowledge of the system's state and input matrices is required to solve the corresponding actuator selection problem. However, we relax this requirement by exploiting trajectory data gathered from the system, and using them to iteratively approximate the anti-stabilizing solution of an associated algebraic Riccati equation (ARE). Notably, the proposed iterative procedure is proved to be small-disturbance input-to-state stable (ISS) even though the ARE associated with it entails no strictly positive-definite constant term; a result that significantly extends prior work. Finally, to further exploit the obtained trajectory data, we show that these can be used to perform online actuator fault detection without knowledge of the system's matrices, and with complexity lower than that of existing methods. Simulations showcase the theoretical findings.
- 6 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 69 (5) , 3301-3308
Input-Constrained Prescribed Performance Control for High-Order MIMO Uncertain Nonlinear Systems via Reference Modification
For a class of uncertain, multiple-input–multiple-output nonlinear systems, we consider the problem of prescribed performance tracking control under strict control input constraints. By prescribed performance, we mean that outputs of the system should track the given reference signals, with tracking errors confined within a user-specified performance envelope. However, as the controller imposing this desirable behavior is subject to inelastic control input constraints, the reference signals are temporarily modified whenever input saturation takes place, so that the tracking task becomes more feasible. It is proved that the proposed reference modification scheme leads to the exclusion of the internal instability phenomenon, from which conventional prescribed performance control suffers when subject to input saturation. In addition, it is proved that the proposed control architecture guarantees boundedness of all closed-loop signals given either an explicitly quantified feasibility condition on the saturation level, or provided that the system is input-to-state practically stable. Finally, the distortion of the reference signal owed to a possible saturation of the control input is quantified, and we additionally provide a level of saturation that can guarantee the reference signal will not be distorted, in the $\mathcal {L}_\infty$ norm, more than a prespecified value. Simulation results verify and clarify theoretical findings.
- 6 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 69 (5) , 3301-3308
Input-Constrained Prescribed Performance Control for High-Order MIMO Uncertain Nonlinear Systems via Reference Modification
For a class of uncertain, multiple-input–multiple-output nonlinear systems, we consider the problem of prescribed performance tracking control under strict control input constraints. By prescribed performance, we mean that outputs of the system should track the given reference signals, with tracking errors confined within a user-specified performance envelope. However, as the controller imposing this desirable behavior is subject to inelastic control input constraints, the reference signals are temporarily modified whenever input saturation takes place, so that the tracking task becomes more feasible. It is proved that the proposed reference modification scheme leads to the exclusion of the internal instability phenomenon, from which conventional prescribed performance control suffers when subject to input saturation. In addition, it is proved that the proposed control architecture guarantees boundedness of all closed-loop signals given either an explicitly quantified feasibility condition on the saturation level, or provided that the system is input-to-state practically stable. Finally, the distortion of the reference signal owed to a possible saturation of the control input is quantified, and we additionally provide a level of saturation that can guarantee the reference signal will not be distorted, in the $\mathcal {L}_\infty$ norm, more than a prespecified value. Simulation results verify and clarify theoretical findings.
- 2 August 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 68 (6) , 3712-3719
Concurrent Receding Horizon Control and Estimation Against Stealthy Attacks
In this work, we consider a game-theoretic framework for cyber-physical systems, where a defender develops a mitigation strategy against an intelligent attacker who exploits the system's uncertainty to remain undetected. The goal of the defender is to optimize a performance cost constructed specifically to account for robustness against stealthy attacks, so that the system is regulated. Conversely, the goal of the attacker is to disrupt the system's performance by leveraging its significant information advantage against the defender. Both players implement their policies in a moving horizon fashion, according to the principles of receding horizon control. However, because the defender has no access to the full state of the system, it concurrently employs receding horizon estimation to overcome this limitation. A rigorous theoretical analysis shows that such a concurrent policy can guarantee closed-loop boundedness, despite the stealthy attacks and the information disadvantage. Simulations verify and clarify these findings.
- 2 August 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 68 (6) , 3712-3719
Concurrent Receding Horizon Control and Estimation Against Stealthy Attacks
In this work, we consider a game-theoretic framework for cyber-physical systems, where a defender develops a mitigation strategy against an intelligent attacker who exploits the system's uncertainty to remain undetected. The goal of the defender is to optimize a performance cost constructed specifically to account for robustness against stealthy attacks, so that the system is regulated. Conversely, the goal of the attacker is to disrupt the system's performance by leveraging its significant information advantage against the defender. Both players implement their policies in a moving horizon fashion, according to the principles of receding horizon control. However, because the defender has no access to the full state of the system, it concurrently employs receding horizon estimation to overcome this limitation. A rigorous theoretical analysis shows that such a concurrent policy can guarantee closed-loop boundedness, despite the stealthy attacks and the information disadvantage. Simulations verify and clarify these findings.
- 28 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 34 (11) , 8467-8481
Recursive Reasoning With Reduced Complexity and Intermittency for Nonequilibrium Learning in Stochastic Games
In this article, we propose a computationally and communicationally efficient approach for decision-making in nonequilibrium stochastic games. In particular, due to the inherent complexity of computing Nash equilibria, as well as the innate tendency of agents to choose nonequilibrium strategies, we construct two models of bounded rationality based on recursive reasoning. In the first model, named level- $k$ thinking, each agent assumes that everyone else has a cognitive level immediately lower than theirs and—given such an assumption—chooses their policy to be a best response to them. In the second model, named cognitive hierarchy, each agent conjectures that the rest of the agents have a cognitive level that is lower than theirs, but follows a distribution instead of being deterministic. To explicitly compute the boundedly rational policies, a level-recursive algorithm and a level-paralleled algorithm are constructed, where the latter one can have an overall reduced computational complexity. To further reduce the complexity in the communication layer, modifications of the proposed nonequilibrium strategies are presented, which do not require the action of a boundedly rational agent to be updated at each step of the stochastic game. Simulations are performed that demonstrate our results.
- 28 February 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 34 (11) , 8467-8481
Recursive Reasoning With Reduced Complexity and Intermittency for Nonequilibrium Learning in Stochastic Games
In this article, we propose a computationally and communicationally efficient approach for decision-making in nonequilibrium stochastic games. In particular, due to the inherent complexity of computing Nash equilibria, as well as the innate tendency of agents to choose nonequilibrium strategies, we construct two models of bounded rationality based on recursive reasoning. In the first model, named level- $k$ thinking, each agent assumes that everyone else has a cognitive level immediately lower than theirs and—given such an assumption—chooses their policy to be a best response to them. In the second model, named cognitive hierarchy, each agent conjectures that the rest of the agents have a cognitive level that is lower than theirs, but follows a distribution instead of being deterministic. To explicitly compute the boundedly rational policies, a level-recursive algorithm and a level-paralleled algorithm are constructed, where the latter one can have an overall reduced computational complexity. To further reduce the complexity in the communication layer, modifications of the proposed nonequilibrium strategies are presented, which do not require the action of a boundedly rational agent to be updated at each step of the stochastic game. Simulations are performed that demonstrate our results.
- 21 December 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 66 (9) , 4409-4416
Prescribed Performance Control for Discontinuous Output Reference Tracking
This paper considers the problem of prescribed performance tracking control for a class of uncertain, multi-input multioutput (MIMO) nonlinear systems having a low-triangular structure, with possibly discontinuous reference signals. Prescribed performance means that the output tracking errors should converge to a prescribed maximum residual set in fixed time, exhibiting a predefined maximum overshoot. The controller we propose is of low-complexity and is designed according to the philosophy of the prescribed performance control (PPC) methodology. However, significant modifications on the standard PPC design are provided to successfully handle the presence of discontinuities in the reference. Key characteristic is the incorporation of impulsive dynamics in the control design, leading to an impulsive closed-loop system. Knowledge of neither the future values of the reference signal nor the time instants of the reference discontinuities is required. Simulation results clarify and verify theoretical findings.
- 21 December 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automatic Control
- Vol. 66 (9) , 4409-4416
Prescribed Performance Control for Discontinuous Output Reference Tracking
This paper considers the problem of prescribed performance tracking control for a class of uncertain, multi-input multioutput (MIMO) nonlinear systems having a low-triangular structure, with possibly discontinuous reference signals. Prescribed performance means that the output tracking errors should converge to a prescribed maximum residual set in fixed time, exhibiting a predefined maximum overshoot. The controller we propose is of low-complexity and is designed according to the philosophy of the prescribed performance control (PPC) methodology. However, significant modifications on the standard PPC design are provided to successfully handle the presence of discontinuities in the reference. Key characteristic is the incorporation of impulsive dynamics in the control design, leading to an impulsive closed-loop system. Knowledge of neither the future values of the reference signal nor the time instants of the reference discontinuities is required. Simulation results clarify and verify theoretical findings.