A Modified Self-Adaptive Sparrow Search Algorithm for Robust Multi-UAV Path Planning  

一种面向鲁棒多无人机路径规划的自适应麻雀搜索 算法

作　　者：SUN Zhiyuan SHEN Bo PAN Anqi XUE Jiankai MA Yuhang 孙至远;沈波;潘安琪;薛建凯;马宇航(东华大学信息科学与技术学院,上海201620;东华大学数字化纺织服装技术教育部工程研究中心,上海201620)

机构地区：[1]College of Information Science and Technology,Donghua University,Shanghai 201620,China [2]Engineering Research Center of Digitized Textile&Fashion Technology,Ministry of Education,Donghua University,Shanghai 201620,China

出　　处：《Journal of Donghua University(English Edition)》2024年第6期630-643,共14页东华大学学报（英文版）

基　　金：Foundation items:National Natural Science Foundation of China(No.62303108);Fundamental Research Funds for the Central Universities,China(No.CUSF-DH-T-2023065)。

摘　　要：With the advancement of technology,the collaboration of multiple unmanned aerial vehicles(multi-UAVs)is a general trend,both in military and civilian domains.Path planning is a crucial step for multi-UAV mission execution,it is a nonlinear problem with constraints.Traditional optimization algorithms have difficulty in finding the optimal solution that minimizes the cost function under various constraints.At the same time,robustness should be taken into account to ensure the reliable and safe operation of the UAVs.In this paper,a self-adaptive sparrow search algorithm(SSA),denoted as DRSSA,is presented.During optimization,a dynamic population strategy is used to allocate the searching effort between exploration and exploitation;a t-distribution perturbation coefficient is proposed to adaptively adjust the exploration range;a random learning strategy is used to help the algorithm from falling into the vicinity of the origin and local optimums.The convergence of DRSSA is tested by 29 test functions from the Institute of Electrical and Electronics Engineers(IEEE)Congress on Evolutionary Computation(CEC)2017 benchmark suite.Furthermore,a stochastic optimization strategy is introduced to enhance safety in the path by accounting for potential perturbations.Two sets of simulation experiments on multi-UAV path planning in three-dimensional environments demonstrate that the algorithm exhibits strong optimization capabilities and robustness in dealing with uncertain situations.随着技术的进步,多无人机协作已成为军事和民用领域的普遍趋势。路径规划是多无人机执行任务的关键步骤,它是一个带约束的非线性问题。传统的优化算法很难找到在各种约束条件下成本函数最小化的最优解。同时,为确保无人机可靠、安全地运行,还需要考虑鲁棒性。该文提出了一种自适应麻雀搜索算法。在优化过程中,采用动态种群策略来分配搜索,在探索性和开发性之间取得平衡;提出了一种 t 分布扰动系数来自适应调整搜索范围;采用随机优化策略来帮助算法,以避免陷入原点和局部最优附近。自适应麻雀搜索算法的收敛性通过 CEC 基准测试集中的 29 个测试函数进行测试。在算法中进一步引入随机优化策略,通过考虑潜在的扰动来提高路径的安全性。两组关于三维环境中多无人机路径规划的仿真实验表明,该算法在处理不确定情况时表现出很强的优化能力和鲁棒性。

关 键 词：multiple unmanned aerial vehicle(multi-UAV) path planning sparrow search algorithm(SSA) stochastic optimization 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]