Coordinating Lunar Construction Swarms: Bridging Task Allocation through Dual-Layer Bilateral Matching

This study proposes a method for rapidly generating flyable trajectories for fixed-wing UAV swarms in a 3D environment with static mountain terrain and dynamic obstacles. We introduce a multilayered structure that integrates global and local planning. At the global level, a differential evolutionary (DE) algorithm generates a feasible trajectory that satisfies UAV physical constraints. Locally, a distributed model predictive control (DMPC) approach enables UAVs to follow the rough trajectory while avoiding unexpected moving obstacles. Additionally, trajectory tracking refines MPC inputs in real time, ensuring flyability. Simulations validate the applicability and efficiency of the proposed approach.