基于CEA-GA的多无人机三维协同曲线航迹规划方法  

Multi-UAVs 3D cooperative curve path planning method based on CEA-GA

作　　者：文超 董文瀚[2] 解武杰[2] 蔡鸣 WEN Chao;DONG Wenhan;XIE Wujie;CAI Ming(School of Graduate,Air Force Engineering University,Xi’an 710038,China;College of Aeronautics Engineering,Air Force Engineering University,Xi’an 710038,China)

机构地区：[1]空军工程大学研究生院,西安710038 [2]空军工程大学航空工程学院,西安710038

出　　处：《北京航空航天大学学报》2023年第11期3086-3099,共14页Journal of Beijing University of Aeronautics and Astronautics

摘　　要：针对多无人机协同航迹规划求解计算复杂度高,收敛效率差等问题,提出一种基于混沌精英适应遗传算法(CEA-GA)的多无人机三维协同曲线航迹规划方法。利用层级规划思想,建立基于单机规划层-航迹平滑层-多机协同规划层的多无人机三维协同曲线航迹层级规划模型,将复杂约束规划问题分解为子函数优化求解问题,减小计算量;考虑到遗传算法(GA)求解高维复杂约束优化问题存在的性能局限,采用Tent混沌映射均匀初始化种群,以扩大个体搜索空间,丰富种群多样性,在此基础上,通过引入自适应遗传算子平衡算法的全局搜索与局部开发能力,帮助个体跳出局部最优,并采用适应度动态更新策略进一步提高算法的局部探索能力和收敛速度。将精英保留策略引入GA以更好地保证改进算法的全局收敛性。将CEA-GA应用于模型求解,仿真实验结果表明:CEA-GA具有较强的鲁棒性、较好的寻优性能和收敛效率,且能够为集群规划满足约束条件的协同曲线航迹,从而验证了所提方法的有效性和CEA-GA的优越性。To address the problems of high computational complexity and poor convergence efficiency of multi-UAVs cooperative path planning,a multi-UAVs 3D cooperative curve path planning method based on chaos elite adaptive genetic algorithm(CEA-GA)is proposed.A multi-UAVs 3D cooperative curve path hierarchical planning model based on single UAV planning layer—path smoothing layer—multiple UAVs cooperative planning layer is established with the idea of hierarchical planning to transform the complex constrained planning problems into the sub-functional optimization solution problems to reduce the computational effort.Considering the performance limitations of genetic algorithm(GA)in solving high-dimensional complex constrained optimization problems,Tent chaotic mapping is used to uniformly initialize the population in order to expand the individual search space and enrich the population diversity.On this basis,the adaptive genetic operators are introduced to balance the global search and local exploitation capability of the algorithm,so as to help individuals jump out of the local optimum.Then,the fitness dynamic update strategy is adopted to further improve the local exploration ability and convergence speed of the algorithm.The elite retention strategy is introduced into the GA to better ensure the global convergence of the improved algorithm.CEA-GA is used to solve the proposed model,and the simulation results show that CEA-GA has strong robustness,good search performance and convergence efficiency,and can plan the cooperative curve path to satisfy the constraints for the swarms,thus verifying the effectiveness of the proposed method and the superiority of CEA-GA.

关 键 词：协同航迹规划 多无人机 混沌映射 遗传算法 精英保留 

分 类 号：V221.3[航空宇航科学与技术—飞行器设计] TB553[理学—物理]