摘要
针对标准灰狼算法(GWO)在解决移动机器人路径规划问题时存在初始参数依赖性强、缺乏多样性及易陷入局部极值的缺陷,提出一种基于多策略融合灰狼算法(LTGWO)。首先运用精英化思想将Logistic-Tent复合混沌映射与反向学习结合,优化灰狼种群分布序列;然后引入sigmoid函数修改收敛因子a,平衡算法全局探索与局部开发能力,并改进控制参数C以更好地拟合灰狼实际捕猎过程;最后加入随适应度值变化的比例权重,提高灰狼个体搜索能力,同时采用种群淘汰策略,淘汰适应度值差的个体,促进种群进化。选用3组不同的栅格地图进行实验,实验结果表明:由LTGWO算法生成的平均路径长度、路径长度标准差都优于对比算法。
Aimed at the problems that robot path planning problems being solved by the standard gray wolf algorithm,initial parameters are strong in dependence,lack of variety,and liable to sink into local extreme value,a Logistic-Tent based Grey Wolf Optimizer(LTGWO)is proposed.Firstly,by an elite method,the Logistic-Tent composite chaotic mapping is in combination with inverse learning to improve the population distribution.And then,by introducing the sigmoid function,the factor of concentration and balances between global exploration and local exploitation is adjusted,while the improved control parameters are fitted even more in line with the actual hunting behavior.Lastly,Proportional weights being in company with the changes of adaptable value are added to enhance the search ability of individual gray wolves.A population culling strategy is adopted to eliminate the individuals with poor fitness,promoting evolution.Three different groups of raster maps are selected for the experiments,and the experimental results show that the average path length and the standard deviation of the path length generated by the LTGWO algorithm are better than the comparison algorithm.
作者
黄琦
陈海洋
刘妍
都威
HUANG Qi;CHEN Haiyang;LIU Yan;DU Wei(School of Electronics and Information,Xi’an Polytechnic University,Xi’an 710048,China)
出处
《空军工程大学学报》
CSCD
北大核心
2024年第3期112-120,共9页
Journal of Air Force Engineering University
基金
国家自然科学基金(51905405)。
