摘要
针对水下无人航行器在三维环境中的全局路径规划问题,从优化初始种群和提高收敛精确度寻得最优路径的角度改进遗传算法,并对遗传算法和差分进化算法的融合进行了研究。采用精英反向学习的方式筛选初始种群,寻得较优初始种群并融合差分进化算法思想改进遗传算法,提升了算法的全局搜索能力。结果表明,改进算法的初期收敛速度较快,规划的曲线能够降低UUV能耗,在一定程度上改善了陷入局部最优解的情况。
Aiming at the global path planning problem of underwater unmanned vehicles in a three-dimensional environment,the genetic algorithm was improved from the perspective of optimizing the initial population and improving convergence accuracy to find the optimal path.The fusion of genetic algorithm and differential evolution algorithm was studied.Using elite reverse learning to screen the initial population,finding the optimal initial population and integrating the idea of differential evolution algorithm to improve the genetic algorithm,enhancing the algorithm′s global search ability.The results show that the improved algorithm has a faster initial convergence speed,and the planned curve can reduce UUV energy consumption,which to some extent improves the situation of getting stuck in local optimal solutions.
作者
赵鹏
丁雪
程婷婷
莫小艳
ZHAO Peng;DING Xue;CHENG Tingting;MO Xiaoyan(School of Intelligent Science and Engineering,Harbin Engineering University,Harbin,Heilongjiang 150001,China)
出处
《自动化应用》
2024年第5期25-28,共4页
Automation Application
关键词
遗传算法
差分进化算法
全局路径规划
genetic algorithm
dfferential evolution algorithm
global path planning
