基于改进混合灰狼优化算法的无人机三维路径规划OA

To address the problems of slow convergence and falling into local optimum easily when solving UAV 3D path planning problems by the traditional Grey Wolf Optimization(GWO)algorithm,an improved hybrid GWO-CLGWO algorithm is proposed.Firstly,the gray wolf population is initialized based on Cat chaotic mapping and backward learning strategy,which lays the foundation for enriching the population diversity in the global search process of the algorithm.An improvement strategy of a new nonlinear convergence factor is proposed to improve the global search capability of the algorithm.Secondly,the introduction of perturbation factors and dynamic weights of the Lion Swarm Optimization(LSO)algorithm is proposed in the gray wolf position update to make the gray wolf have active search ability and avoid the gray wolf losing population diversity and falling into local optimum.Finally,to verify the effectiveness of the improved algorithm,eight internationally used standard test function convergence comparison experiments and UAV 3D path planning simulation experiments are conducted.The experimental results show that the CLGWO algorithm has better convergence and higher optimization-seeking accuracy on single-peak and multi-peak functions;the average path length,average number of iterations,and average running time of the CLGWO algorithm are optimized by 33％,31％,and 52％,respectively compared with the GWO algorithm under the 3D path simulation environment,and there are fewer path transitions,which can better obtain the global optimum.The simulation results verify the effectiveness of CLGWO algorithm.