Orateurs
Description
Pursuit–evasion games provide a stimulating framework for studying decision-making, control, and interaction among autonomous agents. They involve pursuers and evaders evolving in complex environments where speed, safety, and adaptability are critical. These problems have direct applications in mobile robotics, autonomous navigation, security, and multi-robot cooperation. The objective of this project is to design and evaluate control and learning strategies for autonomous agents engaged in pursuit–evasion scenarios.
Three main directions will be explored: (1) dynamic modeling of the agents, (2) implementation of predictive control strategies (MPC) and safety-critical Control Barrier Functions (CBFs), and (3) integration of reinforcement learning (RL) approaches to enable adaptation in uncertain environments.
The methods will first be developed and benchmarked in simulation (Python,MATLAB/Simulink), then validated in a ROS2/Gazebo environment, with possible experiments on TurtleBot platforms.
| Master | Automatique et Robotique |
|---|---|
| Laboratoire d'accueil | LAGEPP |
| Composante ou Département Composante | MECA |
