摘要
为改善基本烟花算法路径规划过程中有收敛速度慢、搜索空间小、易陷入局部极值等不足,提出一种融合竞争学习的导向烟花算法。在烟花爆炸的过程中利用优势烟花进行信息更新,设计了一种自适应策略,动态地调整优势烟花的爆炸范围;并且在其中加入导向因子对劣势烟花的迭代进化进行指引,从而提高算法的寻优效率;另外,为了提升烟花算法的局部及全局性,利用自由变异策略代替原有的变异方法;最后,在选择策略中提出了一种竞争学习机制,提高收敛效果。仿真结果表明,改进后的烟花算法与基本烟花算法相对比,路径各项指标均得以提升且平滑度更高,因此更加适应机器人作业环境。
In order to improve the path planning process of basic fireworks algorithm,such shortcomings as slow convergence speed,small search space and ready to get into local convergence,a new fireworks oriented algorithm based on competitive learning was proposed.An adaptive strategy is designed to dynamically modify the explosion range of the dominant fireworks by updating the information of the dominant fireworks in the process of fireworks explosion.In addition,guidance factors are added to guide the iterative evolution of inferior fireworks,so as to improve the optimization efficiency of the algorithm.In addition,in order to improve the local and global properties of fireworks algorithm,free variation strategy is used to replace the original variation method.Finally,a competitive learning mechanism is proposed in the selection strategy to promote the convergence effect.The simulation results show that,compared with the basic fireworks algorithm,the improved fireworks algorithm can improve all the indicators of the path and the smoothness is higher,so it is more suitable for the robot operating environment.
作者
莫海宁
钟友坤
MO Hai-ning;ZHONG You-kun(HTC VIVEDU School of Technology,Guangxi University of Science and Technology,Liuzhou 545006,China;College of Mathematics and Physics,Hechi University,Yizhou 546300,China)
出处
《组合机床与自动化加工技术》
北大核心
2022年第7期34-37,共4页
Modular Machine Tool & Automatic Manufacturing Technique
基金
广西省级重点研发计划(桂科AB17292053)
教育部产学合作协同育人项目(202002056051)。
关键词
竞争学习策略
导向因子
烟花算法
移动机器人
路径规划
competitive learning strategy
guidance factor
fireworks algorithm
mobile robot
path planning
