Qi-Dun Li

FALCON-MASAC: Formation-Aware Attention-Enhanced Leader-Guided Control-Barrier Optimization for Safe Multi-UAV Formation Navigation in Dynamic 3-D Environments

FALCON-MASAC is presented, a safety-integrated multi-agent reinforcement learning framework that decomposes this task into four complementary layers: a hierarchical leader-follower paradigm that pairs a pre-trained virtual leader with followers learning a distributed cooperative policy, and a bypass-side commitment coordination layer that suppresses trajectory chattering and mitigates crossing conflicts among neighboring UAVs.