摘要
为获得高性能无人艇(USV)的航行路径,提出一种基于多策略改进的麻雀搜索算法(MISSA)。首先,设计了带转向角惩罚项的适应度函数;其次,利用黄金正弦法与参数自螺旋设定对位置更新策略进行改进,同时,在位置更新过程中加强了麻雀个体间的信息交流,以平衡全局探索与局部搜索过程;再次,引入混沌圆映射以提高初始麻雀种群的质量和多样性;最后,设计了局部搜索优化机制以解决原始麻雀算法(SSA)容易陷入局部最优的问题,得到适应度更佳的全局路径。仿真结果表明,相较改进A∗、结合遗传的改进蚁群及原始SSA等三种优秀算法,本文MISSA算法在路径距离、转向角度与次数等关键性能指标上均表现最佳,为无人艇自主安全运行提供了一种有效途径。
To obtain the navigation path of high⁃performance unmanned surface vehicle(USV),a multi⁃strategy improved sparrow search algorithm(MISSA)was proposed.Firstly,a fitness function with a steering angle penalty term was de⁃signed.Secondly,the position update strategy was improved by using the golden sine method and parameter self⁃spiral set⁃ting,at the same time,information exchange between sparrow individuals was strengthened during the position update process to balance global exploration and local search proces⁃ses,again,chaotic circular mapping was introduced to im⁃prove the quality and diversity of the initial sparrow popula⁃tion.Finally,a local search optimization mechanism was de⁃signed to solve the problem of the original sparrow algorithm(SSA)easily falling into local optima and obtain a global path with better fitness.Results show that compared with three ex⁃cellent algorithms,namely improved A∗,improved ant colony algorithm combined with genetics,and original SSA,the MIS⁃SA algorithm in this paper performs the best in key perform⁃ance indicators such as path distance,turning angle,and fre⁃quency,providing an effective path for autonomous and safe operation of USV.
作者
詹小飞
赵红
王宁
李汪洋
谢一哲
ZHAN Xiaofei;ZHAO Hong;WANG Ning;LI Wangyang;XIE Yizhe(Marine Electrical Engineering College,Dalian Maritime University,Dalian 116026,China;Marine Engineering College,Dalian Maritime University,Dalian 116026,China)
出处
《大连海事大学学报》
CAS
CSCD
北大核心
2024年第1期1-10,共10页
Journal of Dalian Maritime University
基金
国家高层次人才支持计划项目(SQ2022QB00329)
国家自然科学基金资助项目(U23A20680,52271306)
国防基础科研计划一般项目基础前沿寻宝项目(JCKY2022410C013)
中央引导地方科技发展专项资金项目(2023JH6/100100010)
辽宁省“兴辽英才计划”领军人才项目(XLYC2202005)
大连市科技创新基金重大基础研究项目(2023JJ11CG009)
中央高校基本科研业务费专项资金项目(3132023501)。
