UAV Path Planning Based on an Improved Chimp Optimization Algorithm

Abstract

Path planning is one of the key issues in the research of unmanned aerial vehicle technology. Its purpose is to find the best path between the starting point and the destination. Although there are many research recommendations on UAV path planning in the literature, there is a lack of path optimization methods that consider both the complex flight environment and the performance constraints of the UAV itself. We propose an enhanced version of the Chimp Optimization Algorithm (TRS-ChOA) to solve the UAV path planning problem in a 3D environment. Firstly, we combine the differential mutation operator to enhance the search capability of the algorithm and prevent premature convergence. Secondly, we use improved reverse learning to expand the search range of the algorithm, effectively preventing the algorithm from missing high-quality solutions. Finally, we propose a similarity preference weight to prevent individuals from over-assimilation and enhance the algorithm’s ability to escape local optima. Through testing on 13 benchmark functions and 29 CEC2017 complex functions, TRS-ChOA demonstrates superior optimization capability and robustness compared to other algorithms. We apply TRS-ChOA along with five well-known algorithms to solve path planning problems in three 3D environments. The experimental results reveal that TRS-ChOA reduces the average path length/fitness value by 23.4%/65.0%, 8.6%/81.0%, and 16.3%/41.7% compared to other algorithms in the three environments, respectively. This indicates that the flight paths planned by TRS-ChOA are more cost-effective, smoother, and safer.