Path Planning for Unmanned Aerial Vehicle Using Enhanced Dynamic Group Based Collaborative Optimization Algorithm

Abstract

The path planning optimization algorithm plays an important role in autonomous operation of the unmanned aerial vehicle (UAV). The swarm intelligence optimization algorithms have achieved good results in solving the shortest path NP problem. Recently, a Dynamic Group-based Collaborative Optimization (DGBCO) has shown good performance in the field of UAV path planning, but there are also several shortcomings, such as insufficient exploration space during the exploration phase and lacking information exchange between the exploitation and exploration groups. Inspired by these, we proposed an Enhanced Dynamic Group-based Collaborative Optimization (EDGBCO) to address those problems in UAV path planning. In EDGBCO, several strategies are designed to enhance the algorithm’s performance. First, the dynamic partitioning strategy is utilized to flexibly divide the exploration and exploitation groups. Second, an adaptive update strategy is employed to enhance the search capability during the exploration phase. Last, we design the local solution strategy to take deep information exchange between the exploration and exploitation groups. To demonstrate the effectiveness of the proposed algorithm for UAV path planning, we compare the proposed algorithm with four swarm intelligence optimization algorithms in various scale scenarios from different perspectives. The experimental results shows that the UAV trajectories produced by the proposed algorithm are far superior to other algorithms under the same conditions. Moreover, the proposed algorithm provides more robust performance in generating UAV paths than other algorithms.