Howard M. Schwartz
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Howard M. Schwartz has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Sidney Nascimento Givigi
Queen’s University, Canada
8 shared publications - Mostafa D. Awheda
Department of Systems and Computer Engineering, Carleton University, Ottawa, Canada
4 shared publications - Amirhossein Asgharnia
Carleton Univ, Dept Syst & Comp Engn, 1125 Colonel By Dr, Ottawa, ON K1S 5B6, Canada
3 shared publications - Ioannis Lambadaris
Department of Systems and Computer Engineering, Carleton University, Ottawa, ON, Canada
3 shared publications - Michel Barbeau
Carleton University
2 shared publications - Mohammad Tayefe Ramezanlou
Carleton University,Department of Systems and Computer Engineering,Ottawa,Canada
2 shared publications
Latest Publications
- 18 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Control Systems Technology
- Vol. 34 (3) , 1625-1632
Trajectory Tracking of Autonomous Vehicles Using Data-Driven Modeling and Neural Predictive Controller (NPC)
The technology of autonomous vehicles is advancing rapidly, relying heavily on high-precision simulations and high-performance controllers. This study uses actual driving data to model coupled lateral and longitudinal vehicle dynamics with two neural network (NN) structures: a time-distributed multilayer perceptron (TD-MLP) and a long short-term memory (LSTM). Inputs include accelerator pedal position, brake pressure, steering wheel angle, gear number, and road slope, while outputs are acceleration and yaw rate. In addition, a bidirectional LSTM (Bi-LSTM) predicts vehicle control inputs based on expert driver data. A resolution of the path to be tracked is provided to the NN driver, which generates control commands that are subsequently fed into vehicle NN models. The vehicle’s output states are used to compute the quadratic cost function for trajectory tracking. The minimum cost is identified, and the corresponding control inputs are selected as the optimal inputs for each prediction horizon. A comparative analysis was performed between the neural predictive controller (NPC) algorithm and the traditional nonlinear model predictive controller (NMPC), utilizing the interior point optimizer (IPOPT). Incorporating high-fidelity models within NMPC leads to substantial computational overhead. The NPC method maintains both computational efficiency and is adaptive to road slope disturbances, without the need for a classical adaptive control design, making it suitable for real-time control applications.
- 18 February 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Control Systems Technology
- Vol. 34 (3) , 1625-1632
Trajectory Tracking of Autonomous Vehicles Using Data-Driven Modeling and Neural Predictive Controller (NPC)
The technology of autonomous vehicles is advancing rapidly, relying heavily on high-precision simulations and high-performance controllers. This study uses actual driving data to model coupled lateral and longitudinal vehicle dynamics with two neural network (NN) structures: a time-distributed multilayer perceptron (TD-MLP) and a long short-term memory (LSTM). Inputs include accelerator pedal position, brake pressure, steering wheel angle, gear number, and road slope, while outputs are acceleration and yaw rate. In addition, a bidirectional LSTM (Bi-LSTM) predicts vehicle control inputs based on expert driver data. A resolution of the path to be tracked is provided to the NN driver, which generates control commands that are subsequently fed into vehicle NN models. The vehicle’s output states are used to compute the quadratic cost function for trajectory tracking. The minimum cost is identified, and the corresponding control inputs are selected as the optimal inputs for each prediction horizon. A comparative analysis was performed between the neural predictive controller (NPC) algorithm and the traditional nonlinear model predictive controller (NMPC), utilizing the interior point optimizer (IPOPT). Incorporating high-fidelity models within NMPC leads to substantial computational overhead. The NPC method maintains both computational efficiency and is adaptive to road slope disturbances, without the need for a classical adaptive control design, making it suitable for real-time control applications.
- Open Access
- 3 June 2025
- Published by Elsevier in Fuzzy Sets and Systems
- Vol. 518
A fuzzy set-based methodology for autonomous navigation
- 3 June 2025
- Published by Elsevier in Fuzzy Sets and Systems
- Vol. 518
Open AccessA fuzzy set-based methodology for autonomous navigation
- Open Access
- 29 November 2024
- Published by Elsevier in Neurocomputing
- Vol. 617
Enhancing cooperative multi-agent reinforcement learning through the integration of R-STDP and federated learning
- 29 November 2024
- Published by Elsevier in Neurocomputing
- Vol. 617
Open AccessEnhancing cooperative multi-agent reinforcement learning through the integration of R-STDP and federated learning
- 14 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 12 (1) , 348-361
Altruism in Fuzzy Reinforcement Learning
We propose using a genetic algorithm to select hyperparameters in multiagent reinforcement learning (MARL) settings. In particular, we look at this in the context of cooperation and altruism. We show through the use of three continuous space games, that certain algorithmic hyperparameters are better suited to allow to agents learn altruistic behaviors. The agents learn using fuzzy actor critic learning algorithms in either a hierarchical structure or a single actor critic policy. The genetic algorithm selects the discount factors, the reward weights, and the standard deviation of noise applied to actor during learning. The genetic algorithm uses a fitness function based on the ratio of successful tests the group of agents can pass after training. This automated selection of these specific hyperparameters show that they are important for cooperation and also not trivial to select.
- 14 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 12 (1) , 348-361
Altruism in Fuzzy Reinforcement Learning
We propose using a genetic algorithm to select hyperparameters in multiagent reinforcement learning (MARL) settings. In particular, we look at this in the context of cooperation and altruism. We show through the use of three continuous space games, that certain algorithmic hyperparameters are better suited to allow to agents learn altruistic behaviors. The agents learn using fuzzy actor critic learning algorithms in either a hierarchical structure or a single actor critic policy. The genetic algorithm selects the discount factors, the reward weights, and the standard deviation of noise applied to actor during learning. The genetic algorithm uses a fitness function based on the ratio of successful tests the group of agents can pass after training. This automated selection of these specific hyperparameters show that they are important for cooperation and also not trivial to select.
- Open Access
- 5 August 2024
- Published by Elsevier in Neurocomputing
- Vol. 604
An adaptable fuzzy reinforcement learning method for non-stationary environments
- 5 August 2024
- Published by Elsevier in Neurocomputing
- Vol. 604
Open AccessAn adaptable fuzzy reinforcement learning method for non-stationary environments
- 5 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1421-1428
Hierarchical Reinforcement Learning for Non-Stationary Environments
What indications are there when the environment changes and the learned policy is no longer optimal? Is it possible to predict when a non-stationary environment changes in some way? In this paper we propose a method that helps agents know when to retrain their policies via reinforcement learning. The agents detect changes based on the temporal difference. A hierarchical learning model is used to aid in these non- stationary environments. The hierarchical learning model has two levels, the higher-level policy, which we call the learning switch, and the lower-level policy, which tells the agents their suitable action to play the game. The higher-level policy determines when reinforcement learning should be turned on or off based on the temporal differences calculated within a game or episode. Two multi-agent differential games are used as examples. The first two examples tackle the problem in cooperative games, while the last example addresses the competitive game. The results show that the agents can maintain suitable performance by switching on the learning process for a few iterations after environment changes occurs. In this paper we consider the change in environment to be long term occurrence within the dynamics of games; for example the mass of an agent may become heavier.
- 5 December 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1421-1428
Hierarchical Reinforcement Learning for Non-Stationary Environments
What indications are there when the environment changes and the learned policy is no longer optimal? Is it possible to predict when a non-stationary environment changes in some way? In this paper we propose a method that helps agents know when to retrain their policies via reinforcement learning. The agents detect changes based on the temporal difference. A hierarchical learning model is used to aid in these non- stationary environments. The hierarchical learning model has two levels, the higher-level policy, which we call the learning switch, and the lower-level policy, which tells the agents their suitable action to play the game. The higher-level policy determines when reinforcement learning should be turned on or off based on the temporal differences calculated within a game or episode. Two multi-agent differential games are used as examples. The first two examples tackle the problem in cooperative games, while the last example addresses the competitive game. The results show that the agents can maintain suitable performance by switching on the learning process for a few iterations after environment changes occurs. In this paper we consider the change in environment to be long term occurrence within the dynamics of games; for example the mass of an agent may become heavier.
- Open Access
- 7 March 2023
- Published by Wiley in Asian Journal of Control
- Vol. 25 (5) , 3456-3473
Fixed‐time sliding mode observer‐based controller for a class of uncertain nonlinear double integrator systems
This paper investigates a fixed‐time convergence issue using the sliding mode observer‐based controller for a class of uncertain nonlinear double integrator systems. This observer‐based controller is designed assuming that only the first state measurement is available and there is no information about external disturbances and modeling uncertainties. A new form of sliding mode observer in combination with a sliding mode controller is designed to estimate unmeasured state and unknown disturbances and uncertainties as well as provide the estimated data in the control law. A novel form of sliding surfaces for the robust observer‐based controller is proposed for which fixed‐time convergence is guaranteed to achieve trajectory tracking. In the proposed fixed‐time scheme, the bound on the settling time is user‐defined using design parameters regardless of the system's initial conditions. The control law and observer law are designed such that the chattering issue is alleviated in the control signal. The stability analysis of the closed‐loop system using the observer‐based controller is established via the Lyapunov theory. The validity of the controller design is tested by applying and simulating an example of a robot manipulator in Simulink/MATLAB. The superiority of the proposed method is demonstrated by comparing it with two other methods from the relevant literature.
- 7 March 2023
- Published by Wiley in Asian Journal of Control
- Vol. 25 (5) , 3456-3473
Open AccessFixed‐time sliding mode observer‐based controller for a class of uncertain nonlinear double integrator systems
This paper investigates a fixed‐time convergence issue using the sliding mode observer‐based controller for a class of uncertain nonlinear double integrator systems. This observer‐based controller is designed assuming that only the first state measurement is available and there is no information about external disturbances and modeling uncertainties. A new form of sliding mode observer in combination with a sliding mode controller is designed to estimate unmeasured state and unknown disturbances and uncertainties as well as provide the estimated data in the control law. A novel form of sliding surfaces for the robust observer‐based controller is proposed for which fixed‐time convergence is guaranteed to achieve trajectory tracking. In the proposed fixed‐time scheme, the bound on the settling time is user‐defined using design parameters regardless of the system's initial conditions. The control law and observer law are designed such that the chattering issue is alleviated in the control signal. The stability analysis of the closed‐loop system using the observer‐based controller is established via the Lyapunov theory. The validity of the controller design is tested by applying and simulating an example of a robot manipulator in Simulink/MATLAB. The superiority of the proposed method is demonstrated by comparing it with two other methods from the relevant literature.
- 19 July 2022
- Published by Springer Nature in International Journal of Fuzzy Systems
- Vol. 24 (7) , 3015-3038
Learning Deception Using Fuzzy Multi-Level Reinforcement Learning in a Multi-Defender One-Invader Differential Game
- 19 July 2022
- Published by Springer Nature in International Journal of Fuzzy Systems
- Vol. 24 (7) , 3015-3038
Learning Deception Using Fuzzy Multi-Level Reinforcement Learning in a Multi-Defender One-Invader Differential Game
- 28 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 500-507
Comparison of Cellular Network Controllers for Quadrotors Experiencing Time Delay
This paper presents a control framework including delay estimator, state estimator, and controller to compensate for random time delays in cellular networks. The effect of network delay on the control of a quadrotor is investigated. A comparative study is done between a linear-based PD and nonlinear Backstepping controller. A time delay estimator based on the Markov stochastic model is developed. The combination of the time delay estimator and the state estimator is used to compute the control signal. Results show that the performance of both controllers in low-variation delays is approximately equivalent. According to the results, the linear-based PD controller is a good choice since it satisfies the problem conditions with a more straightforward design process.
- 28 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 500-507
Comparison of Cellular Network Controllers for Quadrotors Experiencing Time Delay
This paper presents a control framework including delay estimator, state estimator, and controller to compensate for random time delays in cellular networks. The effect of network delay on the control of a quadrotor is investigated. A comparative study is done between a linear-based PD and nonlinear Backstepping controller. A time delay estimator based on the Markov stochastic model is developed. The combination of the time delay estimator and the state estimator is used to compute the control signal. Results show that the performance of both controllers in low-variation delays is approximately equivalent. According to the results, the linear-based PD controller is a good choice since it satisfies the problem conditions with a more straightforward design process.