A novel leader-follower-based hybrid particle swarm-grey wolf optimizer algorithm for the constrained UAV path planning

Purpose Unmanned aerial vehicles (UAVs) require the effective path planning method to navigate in the environments with obstacles and rugged terrain. This study aims to address the UAV path planning problem by formulating it as an optimization problem with multiple constraints and proposing an advanced algorithm to enhance the operational efficiency. Design/methodology/approach A novel leader–follower-based hybrid particle swarm-grey wolf optimizer algorithm (LFHPS-GWO) is proposed. In this approach, the particle swarm optimization (PSO) serves as the leader algorithm, facilitating the global exploration. And the reinforcement learning strategy is integrated to further optimize the search process. Meanwhile, the grey wolf optimizer (GWO) acts as the follower algorithm. It focuses on the local exploitation with the elite strategy and switch operation. A unique information exchange method between the leader and follower algorithms ensures the effective integration of two algorithms. Findings The extensive simulation experiments are conducted on different constrained flight environments, and the effectiveness and reliability of the proposed LFHPS-GWO algorithm are verified by the experimental results. Originality/value A novel LFHPS-GWO algorithm is proposed to deal with the constrained UAV three-dimensional path planning, where the leader–follower structure, special information exchange method and multistrategy reinforcement learning mechanism are introduced to improve the global and local search ability of the new algorithm.