An improved moth-flame optimization algorithm with periodic mutation and Gaussian mutation for safety-enhanced UAV path planning
Xiaodong Zhao, Zhiqiang Hu, Xiaoqian Li, Xianliang Zhang · 2024
To handle the path planning problem of unmanned aerial vehicles (UAV) meeting numerous obstacles in complicated environments, an improved moth-flame optimization(MFO) algorithm with periodic and Gaussian mutations(PGMFO) is proposed. For PGMFO, firstly, moths are able to jump out of local optima and better their position by means of periodic mutation. Secondly, the Gaussian mutation is used to select the optimal flame in each iteration can improve the development capability of MFO. Finally, PGMFO is contrasted with other 9 optimization algorithms using 23 test problems. According to the findings, PGMFO performs faster and more accurately during convergence in the majority of optimization issues. In addition, in the generation of benchmark scenes in real digital elevation model maps, PGMFO is used to solve the optimization problem of feasible and safe operation requirements and constraints for UAVs. The results show that the validity of the PGMFO.