UAVs team flight training based on a virtual leader: application to a fleet of quadrirotors
Résumé
This paper discusses the control strategies of a fleet of robots, especially the control by the virtual leader. The main contribution consists is to achieve the control of a group of vehicles while following a predefined mission carried out thanks to a virtual leader, and simultaneously avoiding the collisions between the different agents of the group. The approach proposed in this paper is based on optimization methods, inspired by the metaheuristics. The goal is to find at each time step the best trajectory for each robot to reach a desired flight training without collision. The proposed method is independent from the model or the control of a particular robot. The method is tested in simulation on a UAV fleet of a quadrirotor type. Test results and comparisons between different simulation conditions are presented.