Faïz Benamar
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Faïz Benamar has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
- Wenqian Du
School of Informatics The University of Edinburgh Edinburgh UK
5 shared publications - Mohamed Fnadi
University of the Littoral Opal Coast
4 shared publications - Oumar Sane
Sorbonne Université, Paris, France
2 shared publications - Stephane Doncieux
Sorbonne Université, Paris, France
2 shared publications - Abderraouf Benali
INSA Centre Val de Loire, Campus de Bourges, Univ. Orléans, PRISME, EA 4229, 88, Boulevard Lahitolle, 18020 Bourges, France
1 shared publication Alex Coninx
1 shared publication
Latest Publications
- Open Access
- 21 October 2025
- Published by IOS Press
- p. 3791-3798
Towards Language-Augmented Multi-Agent Deep Reinforcement Learning
Most prior works on communication in multi-agent reinforcement learning have focused on emergent communication, which often results in inefficient and non-interpretable systems. Inspired by the role of language in natural intelligence, we investigate how grounding agents in a human-defined language can improve the learning and coordination of embodied agents. We propose a framework in which agents are trained not only to act but also to produce and interpret natural language descriptions of their observations. This language-augmented learning serves a dual role: enabling efficient and interpretable communication between agents, and guiding representation learning. We demonstrate that language-augmented agents outperform emergent communication baselines across various tasks. Our analysis reveals that language grounding leads to more informative internal representations, better generalization to new partners, and improved capability for human-agent interaction. These findings demonstrate the effectiveness of integrating structured language into multi-agent learning and open avenues for more interpretable and capable multi-agent systems.1 1 Appendices and code are available at github.com/MToquebiau/LAMARL
- 21 October 2025
- Published by IOS Press
- p. 3791-3798
Open AccessTowards Language-Augmented Multi-Agent Deep Reinforcement Learning
Most prior works on communication in multi-agent reinforcement learning have focused on emergent communication, which often results in inefficient and non-interpretable systems. Inspired by the role of language in natural intelligence, we investigate how grounding agents in a human-defined language can improve the learning and coordination of embodied agents. We propose a framework in which agents are trained not only to act but also to produce and interpret natural language descriptions of their observations. This language-augmented learning serves a dual role: enabling efficient and interpretable communication between agents, and guiding representation learning. We demonstrate that language-augmented agents outperform emergent communication baselines across various tasks. Our analysis reveals that language grounding leads to more informative internal representations, better generalization to new partners, and improved capability for human-agent interaction. These findings demonstrate the effectiveness of integrating structured language into multi-agent learning and open avenues for more interpretable and capable multi-agent systems.1 1 Appendices and code are available at github.com/MToquebiau/LAMARL
- Open Access
- 13 February 2025
- Published by Springer Nature in Journal of Intelligent & Robotic Systems
- Vol. 111 (1) , 1-17
Large Workspace Frontal Human Following for Mobile Robots Utilizing 2D LiDAR
- 13 February 2025
- Published by Springer Nature in Journal of Intelligent & Robotic Systems
- Vol. 111 (1) , 1-17
Open AccessLarge Workspace Frontal Human Following for Mobile Robots Utilizing 2D LiDAR
- 28 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (10) , 8746-8753
Closed-Loop Shape Control of Deformable Linear Objects Based on Cosserat Model
The robotic shape control of deformable linear objects has garnered increasing interest within the robotics community. Despite recent progress, the majority of shape control approaches can be classified into two main groups: open-loop control, which relies on physically realistic models to represent the object, and closed-loop control, which employs less precise models alongside visual data to compute commands. In this letter, we present a novel 3D shape control approach that includes the physically realistic Cosserat model into a closed-loop control framework, using vision feedback to rectify errors in real-time. This approach capitalizes on the advantages of both groups: the realism and precision provided by physics-based models, and the rapid computation, therefore enabling real-time correction of model errors, and robustness to elastic parameter estimation inherent in vision-based approaches. This is achieved by computing a deformation Jacobian derived from both the Cosserat model and visual data. To demonstrate the effectiveness of the method, we conduct a series of shape control experiments where robots are tasked with deforming linear objects towards a desired shape.
- 28 August 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (10) , 8746-8753
Closed-Loop Shape Control of Deformable Linear Objects Based on Cosserat Model
The robotic shape control of deformable linear objects has garnered increasing interest within the robotics community. Despite recent progress, the majority of shape control approaches can be classified into two main groups: open-loop control, which relies on physically realistic models to represent the object, and closed-loop control, which employs less precise models alongside visual data to compute commands. In this letter, we present a novel 3D shape control approach that includes the physically realistic Cosserat model into a closed-loop control framework, using vision feedback to rectify errors in real-time. This approach capitalizes on the advantages of both groups: the realism and precision provided by physics-based models, and the rapid computation, therefore enabling real-time correction of model errors, and robustness to elastic parameter estimation inherent in vision-based approaches. This is achieved by computing a deformation Jacobian derived from both the Cosserat model and visual data. To demonstrate the effectiveness of the method, we conduct a series of shape control experiments where robots are tasked with deforming linear objects towards a desired shape.
- Open Access
E2R: a Hierarchical-Learning inspired Novelty-Search method to generate diverse repertoires of grasping trajectories
Robotics grasping refers to the task of making a robotic system pick an object by applying forces and torques on its surface. Despite the recent advances in data-driven approaches, grasping remains an unsolved problem. Most of the works on this task are relying on priors and heavy constraints to avoid the exploration problem. Novelty Search (NS) refers to evolutionary algorithms that replace selection of best performing individuals with selection of the most novel ones. Such methods have already shown promising results on hard exploration problems. In this work, we introduce a new NS-based method that can generate large datasets of grasping trajectories in a platform-agnostic manner. Inspired by the hierarchical learning paradigm, our method decouples approach and prehension to make the behavioral space smoother. Experiments conducted on 3 different robot-gripper setups and on several standard objects shows that our method outperforms state-of-the-art for generating diverse repertoire of grasping trajectories, getting a higher successful run ratio, as well as a better diversity for both approach and prehension. Some of the generated solutions have been successfully deployed on a real robot, showing the exploitability of the obtained repertoires.
Open AccessE2R: a Hierarchical-Learning inspired Novelty-Search method to generate diverse repertoires of grasping trajectories
Robotics grasping refers to the task of making a robotic system pick an object by applying forces and torques on its surface. Despite the recent advances in data-driven approaches, grasping remains an unsolved problem. Most of the works on this task are relying on priors and heavy constraints to avoid the exploration problem. Novelty Search (NS) refers to evolutionary algorithms that replace selection of best performing individuals with selection of the most novel ones. Such methods have already shown promising results on hard exploration problems. In this work, we introduce a new NS-based method that can generate large datasets of grasping trajectories in a platform-agnostic manner. Inspired by the hierarchical learning paradigm, our method decouples approach and prehension to make the behavioral space smoother. Experiments conducted on 3 different robot-gripper setups and on several standard objects shows that our method outperforms state-of-the-art for generating diverse repertoire of grasping trajectories, getting a higher successful run ratio, as well as a better diversity for both approach and prehension. Some of the generated solutions have been successfully deployed on a real robot, showing the exploitability of the obtained repertoires.
- Open Access
- 15 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 151-154
Diversity Search for the Generation of Diverse Grasping Trajectories
Robotic grasping refers to making a robotic system pick an object by applying forces and torques on its surface. Despite the recent advances in data-driven approaches, grasping still needs to be solved. In this work, we consider grasping as a Diversity Search problem, where we attempt to find as many solutions as possible that verify a sparse binary criterion. We propose a variant of a state-of-the-art QD method for grasping based on a divide-and-conquer paradigm to handle grasping discontinuities. Experiments conducted on 3 different robot-gripper setups and several standard objects show that this variant outperforms state-of-the-art for generating diverse repertoires of grasping trajectories.
- 15 July 2023
- Published by Association for Computing Machinery (ACM)
- p. 151-154
Open AccessDiversity Search for the Generation of Diverse Grasping Trajectories
Robotic grasping refers to making a robotic system pick an object by applying forces and torques on its surface. Despite the recent advances in data-driven approaches, grasping still needs to be solved. In this work, we consider grasping as a Diversity Search problem, where we attempt to find as many solutions as possible that verify a sparse binary criterion. We propose a variant of a state-of-the-art QD method for grasping based on a divide-and-conquer paradigm to handle grasping discontinuities. Experiments conducted on 3 different robot-gripper setups and several standard objects show that this variant outperforms state-of-the-art for generating diverse repertoires of grasping trajectories.
- 10 November 2022
- Published by ASME International in Journal of Mechanisms and Robotics
- Vol. 15 (4) , 1-12
A New Whole-Body Motion Generator and Adaptive Altitude Control for a Quadruped-On-Wheel Robot
This paper proposes a new whole-body motion generator for a quadruped-on-wheel robot that can cross various rough terrains, especially with a much-altitude difference. The motion generator combines the kinematics model, the wheel-center motion model, the robot centroidal momentum and dynamics models, and a proposed altitude-control model. Then, the whole-body motion references can be achieved by giving the robot centroidal-motion reference, including the centroidal translational motion in the forward and lateral directions in the inertial frame, and the centroidal-height motion with respect to the base frame. Especially, the relative motion depends only on the robot-self property and is independent of the terrain geometry. The robot is simulated to be driven on various tested rough terrains using our new motion generator and our compliant torque controller.
- 10 November 2022
- Published by ASME International in Journal of Mechanisms and Robotics
- Vol. 15 (4) , 1-12
A New Whole-Body Motion Generator and Adaptive Altitude Control for a Quadruped-On-Wheel Robot
This paper proposes a new whole-body motion generator for a quadruped-on-wheel robot that can cross various rough terrains, especially with a much-altitude difference. The motion generator combines the kinematics model, the wheel-center motion model, the robot centroidal momentum and dynamics models, and a proposed altitude-control model. Then, the whole-body motion references can be achieved by giving the robot centroidal-motion reference, including the centroidal translational motion in the forward and lateral directions in the inertial frame, and the centroidal-height motion with respect to the base frame. Especially, the relative motion depends only on the robot-self property and is independent of the terrain geometry. The robot is simulated to be driven on various tested rough terrains using our new motion generator and our compliant torque controller.
- 13 January 2022
- Published by Springer Nature in Multibody System Dynamics
- Vol. 54 (3) , 235-262
Integration of prioritized impedance controller in improved hierarchical operational-space torque control frameworks for legged locomotion robots
- 13 January 2022
- Published by Springer Nature in Multibody System Dynamics
- Vol. 54 (3) , 235-262
Integration of prioritized impedance controller in improved hierarchical operational-space torque control frameworks for legged locomotion robots
- 24 November 2021
- Published by Springer Nature in Journal of Intelligent & Robotic Systems
- Vol. 103 (4) , 1-34
Leg Centroidal Dynamics Based New Locomotion Principle of a Quadruped Robot with On-line Legged Motion Generation
- 24 November 2021
- Published by Springer Nature in Journal of Intelligent & Robotic Systems
- Vol. 103 (4) , 1-34
Leg Centroidal Dynamics Based New Locomotion Principle of a Quadruped Robot with On-line Legged Motion Generation
- 20 February 2020
- Published by Springer Nature in Multibody System Dynamics
- Vol. 49 (4) , 447-463
A compact form dynamics controller for a high-DOF tetrapod-on-wheel robot with one manipulator via null space based convex optimization and compatible impedance controllers
- 20 February 2020
- Published by Springer Nature in Multibody System Dynamics
- Vol. 49 (4) , 447-463
A compact form dynamics controller for a high-DOF tetrapod-on-wheel robot with one manipulator via null space based convex optimization and compatible impedance controllers
- Open Access
- 13 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 25 (5) , 2354-2363
Design and Control of a Compliant Wheel-on-Leg Rover Which Conforms to Uneven Terrain
This paper presents the design, control and implementation of a new compliant wheel-on-leg robot named Complios. This robot is able to spontaneously adapt its configuration while it is rolling on unknown uneven terrain. The constant contact of all wheels with the ground is ensured by a series elastic actuation at the top of each leg. It provides a force control that allows the robot to freely balance its posture while the load can be distributed as evenly as possible on each wheel for any relative height differences of the ground. The compliant structure of the Complios also gives access to the measurement of horizontal forces applied on the legs, which constitutes a valuable information when it comes to choose an appropriate response during obstacle crossing. Finally, the robot is tested on different ground shapes and proves its capability to adapt in real time to the ground profile.
- 13 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ASME Transactions on Mechatronics
- Vol. 25 (5) , 2354-2363
Open AccessDesign and Control of a Compliant Wheel-on-Leg Rover Which Conforms to Uneven Terrain
This paper presents the design, control and implementation of a new compliant wheel-on-leg robot named Complios. This robot is able to spontaneously adapt its configuration while it is rolling on unknown uneven terrain. The constant contact of all wheels with the ground is ensured by a series elastic actuation at the top of each leg. It provides a force control that allows the robot to freely balance its posture while the load can be distributed as evenly as possible on each wheel for any relative height differences of the ground. The compliant structure of the Complios also gives access to the measurement of horizontal forces applied on the legs, which constitutes a valuable information when it comes to choose an appropriate response during obstacle crossing. Finally, the robot is tested on different ground shapes and proves its capability to adapt in real time to the ground profile.