PPSwarm: A hybrid particle swarm optimization algorithm for multi-UAV path planning in complex scenarios is proposed to address challenges in path planning for multiple UAVs. The algorithm combines RRT* for initial path generation, prioritized planning to simplify collaboration, and path randomization to enhance particle diversity. Experimental results show that PPSwarm outperforms algorithms like DE, PSO, ABC, GWO, and SPSO in terms of path quality, convergence speed, and runtime for 40 UAVs across four scenarios. In larger-scale experiments with 500 UAVs, the algorithm demonstrates excellent processing capability and scalability. The algorithm uses a two-level path planning strategy, with high-level RRT* and prioritized planning for initial path generation and priority assignment, and low-level PSO for detailed path planning. The algorithm integrates the global search capabilities of RRT* with the local optimization of PSO, improving efficiency and avoiding local optima. The algorithm also incorporates a restart strategy to enhance diversity and adaptability. The results show that PPSwarm achieves better performance in terms of convergence accuracy, stability, and runtime compared to other algorithms. The algorithm is effective in complex environments with numerous obstacles and collaborative constraints.PPSwarm: A hybrid particle swarm optimization algorithm for multi-UAV path planning in complex scenarios is proposed to address challenges in path planning for multiple UAVs. The algorithm combines RRT* for initial path generation, prioritized planning to simplify collaboration, and path randomization to enhance particle diversity. Experimental results show that PPSwarm outperforms algorithms like DE, PSO, ABC, GWO, and SPSO in terms of path quality, convergence speed, and runtime for 40 UAVs across four scenarios. In larger-scale experiments with 500 UAVs, the algorithm demonstrates excellent processing capability and scalability. The algorithm uses a two-level path planning strategy, with high-level RRT* and prioritized planning for initial path generation and priority assignment, and low-level PSO for detailed path planning. The algorithm integrates the global search capabilities of RRT* with the local optimization of PSO, improving efficiency and avoiding local optima. The algorithm also incorporates a restart strategy to enhance diversity and adaptability. The results show that PPSwarm achieves better performance in terms of convergence accuracy, stability, and runtime compared to other algorithms. The algorithm is effective in complex environments with numerous obstacles and collaborative constraints.