MultiRRT-PSO-Based Cooperative Path Planning for Unmanned Aerial Vehicles

With the increasing demand for multiple unmanned aerial vehicles (UAVs) to perform complex missions, the importance of cooperative path planning is becoming more and more prominent. This paper proposes a novel multi-UAV cooperative path planning algorithm named MultiRRT-PSO, which integrates the multi-objective rapidly exploring random tree (MultiRRT) with safety and dynamic constraints and the particle swarm optimization (PSO) algorithm. The proposed algorithm leverages the global optimization capability of PSO and the search efficiency of MultiRRT to achieve efficient and optimized cooperative path planning for multiple UAVs.Firstly, a particle initialization strategy based on the MultiRRT tree is designed to accelerate the convergence of PSO by utilizing the existing search information from MultiRRT. Secondly, a hybrid sampling mechanism is constructed to balance the global search capability and convergence speed of the algorithm. Thirdly, a dynamic fitness adjustment mechanism is proposed to intelligently optimize the multi-objective weight distribution according to the planning progress. Finally, a post-optimization path mechanism is introduced, which combines redundant node pruning and Bézier curve interpolation techniques to further enhance the smoothness and safety of the planned paths.Experimental results demonstrate that the proposed algorithm significantly outperforms Theta-RRT, FN-RRT, RRT, and RRT-Smart in terms of path length, smoothness, and computational efficiency. The optimal path length achieved by the proposed algorithm is 488.27 m, with a smoothness of 0.09 and a typical computation time of 0.37 s. This study provides an efficient and safe solution for multi-UAV cooperative path planning in complex environments, which holds significant theoretical and practical value.