Dynamic Cooperative Perception and Path Planning for Collision Avoidance

This paper deals with cooperative path and cooperative perception of Unmanned Aerial Vehicles (UAVs) in dynamic environment. Differential geometry techniques based on Pythagorean Hodographs (PH) are adapted to plan cooperatively path of UAVs in order to avoid obstacles. These obstacles are proposed in this paper to be both static and moving obstacles. In case of moving obstacles, we propose to develop cooperative perception algorithms based on multiple views geometry optimization/estimation scheme to accurately predict obstacle trajectories. These estimated trajectories in conjunction with PH curves to amend current UAV paths. Simulation results based on swarm of UAVs scenario show the efficiency of the proposed techniques.