基于均值学习策略果蝇算法的工业机器人轨迹规划  

Trajectory planning of industrial robots based on fruit-fly optimization algorithm with mean learning strategy

作　　者：吴萍 田亮[2] WU Ping;TIAN Liang(School of Intelligent Control,Changzhou Institute of Industry and Technology,Changzhou 213164;College of Computer and Cyber Security,Hebei Normal University,Shijiazhuang 050024)

机构地区：[1]常州工业职业技术学院智能控制学院,江苏常州213164 [2]河北师范大学计算机与网络空间安全学院,河北石家庄050024

出　　处：《机械设计》2023年第2期126-133,共8页Journal of Machine Design

基　　金：2020年江苏省高等学校自然科学研究项目(20KJD510003)。

摘　　要：针对机器人空间轨迹的时间最优规划问题,文中在建立轨迹规划模型的基础上,对果蝇算法(FOA)进行改进,并由此提出基于均值学习策略果蝇算法(MLS-FOA)的机器人轨迹规划方法。相比于FOA,MLS-FOA增加了向种群均值学习的策略,平衡了算法的全局搜索能力和局部搜索能力。机器人轨迹规划实例表明,MLS-FOA能够在位置、速度、加速度满足约束条件的同时,使得运动曲线更为平滑,同时也缩短了机器人的运动时间,提高了运动效率。In this article,in view of time optimization of robot trajectory planning,based on the model of trajectory planning,the fruit-fly optimization algorithm(FOA)is improved,and the method of robot trajectory planning based on the fruit-fly optimization algorithm with the mean learning strategy(MLS-FOA)is proposed.Compared with FOA,MLS-FOA contains the mean learning strategy which can balances the algorithm’s global search ability and local search ability.The example of robot trajectory planning shows that MLS-FOA can make the robot’s movement curve smoother,shorten the motion time and improve the efficiency in motion,while the position,velocity and acceleration meet the constraint conditions.

关 键 词：果蝇优化算法 均值学习策略 工业机器人 轨迹规划 

分 类 号：U463.45[机械工程—车辆工程]