摘要
为了求解机械臂关节空间的刚度最优轨迹,提出了基于多方式自适应学习蜂群的轨迹规划方法。以SR20A型串联机械臂为研究对象,基于刚度矩阵的最小特征值定义了机械臂刚度,并建立了机械臂关节轨迹的刚度最优化模型。针对蜜蜂差分搜索效率低、启发性不足等问题,根据个体差异设计了多方式自适应学习策略,并将多方式学习蜂群算法应用于刚度最优轨迹规划。使用5组轨迹规划实验进行验证,结果表明,多方式学习蜂群算法相比自适应蝙蝠算法和标准蜂群算法,在机械臂关节轨迹求解中具有显著的优越性。
In order to plan the stiffness-optimal trajectory of manipulator in joint space,a multi-mode adaptive learning bee colony based trajectory planning method is proposed.Taking SR20A series manipulator as the research object,the stiffness of the manipulator is defined based on the minimum eigenvalue of the stiffness matrix,and the stiffness optimization model of the manipulator joint trajectory is established.Aiming at the problems of low efficiency and lack of inspiration of bee differential search,a multi-mode adaptive learning strategy is designed according to individual differences,and the multi-mode learning bee colony algorithm is applied to the rigidity optimal trajectory planning.5 sets of trajectory planning experiments are used to verify the results.The results show that the order of solving the mean value of trajectory stiffness is improved by bee colony algorithm,adaptive bat algorithm,and standard bee colony algorithm.The superiority of the improved bee colony algorithm is verified in solving the joint trajectory of the manipulator by the experimental results.
作者
温承钦
黄维忠
赖道辉
陆润明
覃丽燕
温志力
Wen Chengqin;Huang Weizhong;Lai Daohui;Lu Runming;Qin Liyan;Wen Zhili(College of Logistics and Transportation,Guangxi Logistics Vocational and Technical College,Guangxi Guigang,537100,China;College of Vocational Education and Technology,Guangxi Science&Technology of Normal University,Guangxi Laibin,546100,China)
出处
《机械设计与制造工程》
2023年第6期91-96,共6页
Machine Design and Manufacturing Engineering
基金
广西高校中青年教师科研基础能力提升项目(2023KY2050,2023KY2048)
广西教育科学“十四五”规划2022年度专项课题(2022ZJY2115)。
关键词
机械臂
关节轨迹求解
刚度最优化
多方式自适应学习
蜂群算法
最小特征值
manipulator
joint trajectory solution
stiffness optimization
multi-mode adaptive learning
bee colony algorithm
minimum characteristic value
