A Fast and Robust Algorithm with Reinforcement Learning for Large UAV Cluster Mission Planning

Large Unmanned Aerial Vehicle (UAV) clusters, containing hundreds of UAVs, have widely been used in the modern world. Therein, mission planning is core large UAV cluster collaborative systems. In this paper, we propose a method by introducing Simple Attention Model (SAM) into Dynamic Information Reinforcement Learning (DIRL), named DIRL-SAM. To reduce computational complexity original attention model, derive SAM with lightweight interactive model to rapidly extract high-dimensional features information. DIRL, dynamic training conditions are considered simulate different environments. Meanwhile, data expansion DIRL guarantees convergence these environments, which improves robustness algorithm. Finally, simulation experiment results show that proposed can adaptively provide feasible schemes second-level solution speed and it exhibits excellent generalization performance large-scale problems.