Control, Planning, and Learning for Multi-UAV Cooperative Hunting

Download or read book Control, Planning, and Learning for Multi-UAV Cooperative Hunting written by Jacky Juan Ze Liao and published by . This book was released on 2020 with total page 0 pages. Available in PDF, EPUB and Kindle. Book excerpt: This thesis presents a model predictive control (MPC) algorithm for solving the cooperative hunting problem with multiple unmanned aerial vehicles (UAVs) in obstacle rich and moving obstacle environments and builds upon the MPC algorithm using a reinforcement learning method called Guided Policy Search (GPS). The cooperative hunting problem deals with surrounding a target while navigating around obstacles and avoiding collisions, and the MPC algorithm combines the cooperative hunting objective with planning and obstacle avoidance. Simulation results show that the MPC algorithm successfully surrounds the target in both scenarios. Local minima are avoided for obstacle rich environments and the obstacle interaction is minimized in moving obstacle environments. Furthermore, the MPC algorithm was validated with physical quadrotors. Finally, the GPS algorithm improves upon the MPC solutions by reducing the trajectory length and capture time, and then uses the improved trajectories to train a global neural network policy.