多策略鼠群优化算法的无人机三维航迹规划  

Multi-Strategy Rat Swarm Optimizer for Unmanned Aerial Vehicle 3D Flight Path Planning

作　　者：解瑞云 海本斋[3] XIE Rui-yun;HAI Ben-zhai(He’nan Institute of Technology,College of Cable Engineering,He’nan Xinxiang 453000,China;He’nan Key Laboratory of Wire and Cable Structures and Materials,He’nan Xinxiang 453000,China;He’nan Normal University,Faculty of Education,He’nan Xinxiang 453000,China)

机构地区：[1]河南工学院电缆工程学院,河南新乡453000 [2]河南省线缆结构与材料重点实验室,河南新乡453000 [3]河南师范大学教育学部,河南新乡453000

出　　处：《机械设计与制造》2024年第9期112-119,共8页Machinery Design & Manufacture

基　　金：河南省科技攻关项目(242102210119,242102230175,242102230167);河南省高等学校重点科研项目(23A520020);河南省高等教育教学改革研究与实践项目(2024SJGLX0559);2023年度河南省普通本科高等学校智慧教学研究专项项目—基于智慧教学平台的“新工科”混合教学模式构建与应用研究;横向课题:高密度人群课堂环境下的学生学习行为识别技术开发;2023年校级科研反哺教学专项课题(XJ2023002402,XJ2023001702);河南工学院校级教改项目(2024JG-ZD009)。

摘　　要：针对在复杂电力检测环境中无人机三维航迹规划问题,提出一种基于跳跃式自适应小孔成像反向学习鼠群优化(JAPRSO)算法的无人机三维航迹规划方法。JAPRSO算法引入了Sobol序列初始化种群以增强种群多样性;引入了非线性自适应因子实现动态权衡局部开发和全局搜索能力;嵌入了跳跃式围捕猎物机制以避免算法陷入局部最优;同时,引入跳跃式自适应小孔成像反向学习追赶猎物机制以提高算法的全局寻优能力。仿真结果表明:所提出的路径规划方法寻优性能优于RSO算法、灰狼优化(GWO)算法,金枪鱼群优化(TSO)算法和海鸥优化(SOA)算法,能够有效地躲避威胁区,快速获得航迹代价最小的安全可行航迹,可适用于求解电力检测方面的无人机三维航迹规划问题。Aiming at the problem of UAV three-dimensional trajectory planning in complex power detection environment,a novel unmanned aerial vehicle three-dimensional flight path planning method based on jumping and adaptive pin-hole imaging opposition-based learning rat swarm optimizer(JAPRSO)was proposed.JAPRSO algorithm introduced Sobol sequence initial⁃ization population to enhance population diversity.A nonlinear adaptive factor was introduced to dynamically balance local devel⁃opment and global search capabilities.In order to avoid the algorithm falling into local optimum,the jumping attacked prey mechanism is embedded.In order to improve the global optimization capability of the algorithm,a jumpy and adaptive pin-hole imaging opposition-based learning mechanism is introduced into hunting prey behavior.The simulation results show that the per⁃formance of the proposed method is better than RSO algorithm,the Gray Wolf Optimization(GWO)algorithm,Tuna Swarm Op⁃timization(TSO)algorithm and Seagull Optimization Algorithm(SOA),meanwhile,the proposed method can effectively avoid threat area,safe and feasible to get the track is least costly fast track,and it can be applied to solve complex unmanned aerial ve⁃hicle three-dimensional path planning problem for power detection.

关 键 词：鼠群优化算法 无人机三维航迹规划 非线性自适应因子 跳跃式围捕机制 自适应小孔成像反向学习 

分 类 号：TH16[机械工程—机械制造及自动化] TP242.6[自动化与计算机技术—检测技术与自动化装置] TP18[自动化与计算机技术—控制科学与工程]