Dynamic coverage path planning method for UAV formations in multi-region aerial tasks

Abstract

Planning coverage paths for multiple discrete regions is a prerequisite for Unmanned Aerial Vehicle (UAV) formations to perform continuous coverage tasks. However, precisely solving for the globally optimal path is computationally challenging, and maintaining formation shape under the interference of static and dynamic obstacles is difficult. Traditional path planning methods often perform poorly in large-scale, multi-region coverage tasks due to getting trapped in local optima or low computational efficiency, making it challenging to maintain formation shape while achieving swarm obstacle avoidance. This study proposes a novel dynamic multi-region coverage path planning method to enhance the task efficiency and safety of UAV formations in complex environments. First, a heuristic optimization algorithm, PGS2, was developed, incorporating three optimized mechanisms to significantly enhance global search capabilities. In nine scenarios with varying numbers of access points and discrete regions, PGS2 reduced average path costs by 56.7% and 1.35% compared to six baseline algorithms, demonstrating superior optimization performance and stability. Second, the Orthogonal Artificial Potential Field (Orthogonal APF) path planning algorithm and a gradient-mapping-based swarm self-avoidance method were proposed, achieving dynamic path planning while maintaining formation shape through virtual target point design. Orthogonal APF achieved a 100% target arrival rate in nine multi-obstacle scenarios, with path deviation reduced by an average of 31.93% compared to four other algorithms, validating its effectiveness and unique path recovery capability. In a simulation environment with three regions and multiple obstacles, the UAV formation could avoid static obstacles in approximately 5 seconds and dynamic obstacles in about 2 seconds, while the virtual target point mechanism ensured formation recovery within approximately 3 seconds post-avoidance and supported formation reconfiguration for varying UAV counts. This study provides an innovative path planning method for efficient and safe UAV formation operations in multi-region, complex environments, with comparisons to traditional methods demonstrating its significant advantages in path optimization, obstacle avoidance, and task continuity.