Evolutionary State Estimation-Based Multi-Strategy Jellyfish Search Algorithm for Multi-UAV Cooperative Path Planning

Path planning is crucial for the successful mission execution of unmanned aerial vehicles (UAVs). However, planning feasible paths becomes challenging due to constraints imposed by complex environments and the inherent maneuverability of UAVs, particularly in large-scale scenarios involving multiple UAVs ( multi-UAV). This paper addresses the multi-UAV cooperative path planning problem, formulating it as a constrained optimization problem and proposing the evolutionary state estimation-based multi-strategy jellyfish search (ESE-MSJS) algorithm to search for high-quality paths. In the proposed algorithm, a switching framework based on evolutionary state estimation is constructed to prevent ineffective searches and enhance suitability for path planning. Within this framework, three distinct update modes are developed for each individual, enabling a more efficient and flexible selection of appropriate learning strategies. In addition, a neighborhood topology-based elite example learning strategy is employed to increase population diversity, and a best information guiding-driven adaptive scaling factor strategy exploits the surrounding space, strengthening local search capabilities. The Gaussian barebone mechanism is introduced to balance exploration and exploitation. To effectively cope with intricate constraints, a dynamic $\alpha$ -level comparison strategy is incorporated into the individual update stage of the ESE-MSJS. Experimental results demonstrate that ESE-MSJS outperforms state-of-the-art algorithms regarding accuracy, feasibility, and stability, proving to be an effective method for multi-UAV cooperative path planning in complex environments.