Optimal Control Strategies for UAV Formation Recovery Based on APF and PSO

作者：Mai Chang, Shihao Cai, Mingqian Wang, Yanyan Chen, Zixuan Xu, Jianshan Zhou, Guixian Qu, Daxin Tian · 年份：2024 · DOI：10.1109/icus61736.2024.10840043 · 被引用次数：14 · 研究领域：Robotic Path Planning Algorithms、Aerospace Engineering and Control Systems、Distributed Control Multi-Agent Systems

In coordinated UAV swarm operations, maintaining formation while effectively avoiding obstacles presents significant challenges that can disrupt the swarm's original configuration. This research explores a control strategy designed to minimize the time required for formation recovery following obstacle avoidance. We utilized the Artificial Potential Field (APF) method to compute virtual forces that guide UAVs through complex environments, facilitating obstacle avoidance. To address the high-dimensional nature of the optimization problem, characterized by limited feasible solutions, we employed Particle Swarm Optimization (PSO). PSO's capability to explore extensive search spaces and avoid local optima allowed us to optimize controller parameters effectively. We established a relationship between recovery time and controller gain and optimized the gain to improve formation recovery. MATLAB simulations demonstrated that the proposed method achieved formation recovery in 21.1 seconds, delivering stable control and enhanced performance. This study underscores the effectiveness of integrating APF with PSO for improving UAV swarm navigation and formation control, with significant implications for practical applications in dynamic environments.