To address the collision-free flight and cooperative control problems of large-scale Unmanned Aerial Vehicle (UAV) swarms in complex dynamic environments, a hierarchical framework integrating trajectory planning and distributed control is proposed. At the global planning layer, an improved A* algorithm incorporating trajectory smoothing strategies generates reference trajectories satisfying kinematic constraints, resolving the poor dynamic adaptability of discrete paths. At the low-level control layer, a distributed nonlinear control law based on the Composite Barrier Lyapunov Function is designed. By transforming hard obstacle avoidance constraints into barrier functions, the forward invariance and asymptotic stability of the closed-loop system are mathematically proven. The hierarchical architecture effectively bypasses the local minima trap typical of pure artificial potential field methods. Simulation results indicate that, under strong crosswind disturbances, the proposed method reduces trajectory tracking error by 80.5% and energy consumption by 74.2% compared to classic PID control. In a high-density cross-flow scenario with 100 UAVs, zero-conflict interwoven operation is achieved, verifying the convergence and robust feasibility of the method in non-convex constrained environments.
Gao et al. (2026) studied this question.