基于改进粒子群算法的无人机航迹规划  

Based on UAV Track Planning of Improved Particle Swarm Optimization Algorithm

作　　者：刘春玲[1] 冯锦龙 田玉琪 张琪珍 LIU Chun-ling;FENG Jin-long;TIAN Yu-qi;ZHANG Qi-zhen(College of Information Engineering,Dalian University,Dalian Liaoning 116622,China)

机构地区：[1]大连大学信息工程学院,辽宁大连116622

出　　处：《计算机仿真》2023年第10期38-43,共6页Computer Simulation

基　　金：辽宁省教育厅面上项目(LJKZ1184)。

摘　　要：标准粒子群算法在解决无人机航迹规划问题上容易陷入局部最优解。针对上述问题,提出了随机位置突变的自适应粒子群算法。算法选择基于凸函数的自适应惯性权重递减模型,能够同时兼顾全局搜索和局部精细搜索,引入早熟判定机制和位置突变机制,通过该机制使得群体符合位置突变条件时,各个粒子将会在保留记忆和速度的基础上,跳跃到一个新的位置进行搜索,增加了群体的多样性,具备了跳出局部最优的能力。通过MATLAB仿真表明,基于随机位置突变的自适应粒子群算法收敛效果有显著提升,并且改进算法的复杂度没有明显增加。所提算法在复杂环境下的适用性要远远优于标准粒子群算法,在航迹规划问题上具备可行性和优越性。The standard particle swarm optimization algorithm is easy to fall into the local optimal solution in solving the UAV trajectory planning problem.Aiming at this problem,an adaptive particle swarm algorithm with random position mutation is proposed.The algorithm selects an adaptive inertia weight-decreasing model based on convex functions,which can take into account both global development and local fine search.Then,a precocious decision mechanism and a position mutation mechanism were introduced.When the group meets the condition of position mutation,each particle will jump to a new position to search on the basis of retaining memory and speed,which increases the diversity of the group and has the ability to jump out of the local optimum.The MATLAB simulation experiments show that the convergence effect of the adaptive particle swarm based on random position mutation is significantly improved,and the complexity of the improved algorithm is not greatly improved.In addition,the applicability of the algorithm in complex environments is far better than the standard algorithm.It has feasibility and superiority in the problem of trajectory planning.

关 键 词：粒子群优化算法 随机位置突变 自适应惯性权重 三维航迹规划 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]