基于遗传算法的平面欠驱动机器人模糊控制  

Fuzzy Control of Planar Underactuated Robots Based on Genetic Algorithm

作　　者：刘庆波[1] 余跃庆[1] 苏丽颖[1] 刘善增[1] 

机构地区：[1]北京工业大学机电学院,北京100124

出　　处：《机械设计与研究》2008年第5期42-46,共5页Machine Design And Research

基　　金：国家自然基金项目(50575002;60705036);北京市自然基金项目(3062004);北京市教委人才强教计划拔尖人才项目(PHR(IHLB));北京市教委科技发展计划项目(KM200610005003)

摘　　要：以平面欠驱动机器人为研究对象,对其运动控制问题进行了研究,提出了一种基于遗传算法的模糊控制新方法。控制目标为实现欠驱动机器人的任意位置控制。控制器直接以关节角度误差变量作为模糊控制器的输入,引入遗传算法对模糊控制规则和隶属度函数同时进行优化,进而得到最优模糊控制系统,最后进行数值仿真,仿真结果表明所设计的控制器能快速精确的实现2自由度及3自由度欠驱动机器人的任意位置控制,且具有很好的稳定性。Motion control of planar underactuated robots is investigated in this paper. A new fuzzy control method which based on genetic algorithm is proposed. The control objective is to realize the arbitrary position control of underaetuated robots. The joint angular errors are chosen to be the input of fuzzy logic controller directly. The best control rules and optimal membership functions are obtained by the global optimization of genetic algorithm automatically. Numerical simulations show that the proposed method has good stability and can realize the arbitrary position control of 2 - dof and 3-dof underactuated manipulators quickly and accurately.

关 键 词：欠驱动 运动控制 模糊控制 遗传算法 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]