多连杆机械臂GA-RBF神经网络轨迹跟踪控制

肖凡; 李光; 周鑫林

doi:10.13433/j.cnki.1003-8728.2018.0503

多连杆机械臂GA-RBF神经网络轨迹跟踪控制

doi: 10.13433/j.cnki.1003-8728.2018.0503

1.
湖南工业大学机械工程学院, 湖南株洲 412007

基金项目:

湖南省自然科学基金项目（2018JJ4079）资助

详细信息

作者简介:
肖凡(1991-),硕士研究生,研究方向为机器人智能控制,15116429175@163.com

通讯作者:
李光,教授,硕士生导师,liguanguw@126.com

计量
- 文章访问数: 297
- HTML全文浏览量: 36
- PDF下载量: 14
- 被引次数: 0
出版历程
- 收稿日期: 2017-06-28
- 刊出日期: 2018-05-05

GA-RBF Neural Network Control for Trajectory Tracking of Multilink Robot Arm

1.
School of Mechanical Engineering, Hunan University of Technology, Hu'nan Zhuzhou 412007, China

摘要

摘要: 针对多连杆机械臂模型系统信息不完整、存在外界干扰等问题，设计了一种新型的GA-RBF神经网络闭环自适应控制系统。该系统利用径向基函数（RBF）神经网络来逼近并补偿系统的模型误差和外界扰动，在基于计算力矩法的基础上实现对机械臂的轨迹跟踪控制，并采用遗传算法（GA）对RBF神经网络权值进行在线优化，确保机械臂控制系统能在更短时间内获得稳定，实现了高精度的轨迹跟踪，提高了轨迹跟踪的性能。MATLAB数值仿真的结果验证了该方法的有效性。
- 计算力矩法 /
- RBF神经网络 /
- 遗传算法 /
- 机械臂 /
- 轨迹跟踪 /
- MATLAB
Abstract: A new closed loop adaptive control system of GA-RBF neural network is designed to solve the problem of incomplete information and external disturbance of multilink robot arm model system. The system uses radial basis function (RBF) neural network to approximate and compensate the system model errors and external disturbance. Based on the computed torque method of manipulator, it realizes trajectory tracking control; based on genetic algorithm (GA) and the online optimization of RBF network weights, it ensures that the manipulator control system can get stable in a shorter period of time, to achieve high precision tracking trajectory, and improves the performance of trajectory tracking. The effectiveness of the proposed method is verified by the results of MATLAB simulation.
- computational torque method /
- RBF neural network /
- genetic algorithm /
- mechanical arm /
- trajectory tracking /
- MATLAB

HTML全文

参考文献(15)

[1]	张宏敏,刘延斌.基于计算力矩法的3-RRRT并联机器人控制研究[J].机械设计与制造,2009,(1):174-176 Zhang H M, Liu Y B. Study on control of a 3-RRRT parallel robot based on computed-torque[J]. Machinery Design & Manufacture, 2009,(1):174-176(in Chinese)
[2]	龚捷,鲍金锋,衣冠超,等.基于计算力矩法的装载机工作装置轨迹控制[J].机械工程学报,2010,46(13):141-146 Gong J, Bao J F, Yi G C, et al. Trajectory-following control for manipulator of wheel loaders based on computed Torque[J]. Journal of Mechanical Engineering, 2010,46(13):141-146(in Chinese)
[3]	陈鹏,李洪谊.FS-SEA柔性臂改进的反馈计算力矩控制方法[J].载人航天,2016,22(2):233-240 Chen P, Li H Y. An improved feedback computed torque control method for FS-SEA flexible manipulators[J]. Manned Spaceflight, 2016,22(2):233-240(in Chinese)
[4]	唐晓腾,陈旻辉,唐诚焜.基于RBF神经网络的空间机械臂关节空间轨迹跟踪补偿控制[J].闽江学院学报,2011,32(2):34-37 Tang X T, Chen M H, Tang C K. Trajectory tracking compensating control for the joint space of space manipulator based on RBF neural network[J]. Journal of Minjiang University, 2011,32(2):34-37(in Chinese)
[5]	冯治国.基于计算力矩的助行腿机器人神经网络补偿控制[J].中国机械工程,2013,24(16):2173-2179 Feng Z G. Neural-network compensation control for exoskeleton robot based on computed torque control[J]. China Mechanical Engineering, 2013,24(16):2173-2179(in Chinese)
[6]	刘金琨.RBF神经网络自适应控制MATLAB仿真[M].北京:清华大学出版社,2014:116-124 Liu J K. RBF neural network control for mechanical systems design[M]. Beijing:Tsinghua University Press, 2014:116-124(in Chinese)
[7]	钟斌.不确定关节机器人模型的神经网络补偿自适应控制[J].机械科学与技术,2017,36(3):372-377 Zhong B. Adaptively controlling neural network compensation with uncertain joint robot model[J]. Mechanical Science and Technology for Aerospace Engineering, 2017,36(3):372-377(in Chinese)
[8]	Chen S, Chng E S, Alkadhimi K. Regularized orthogonal least squares algorithm for constructing radial basis function networks[J]. International Journal of Control, 1996,64(5):829-837
[9]	Tobar F A, Kung S Y, Mandic D P. Multikernel least mean square algorithm[J]. IEEE Transactions on Neural Networks and Learning Systems, 2014,25(2):265-277
[10]	Paul A, Mukhopadhyay S. An improved ant system using least mean square algorithm[C]//Proceedings of 2012 Annual IEEE India Conference (INDICON). Kochi, India:IEEE, 2012:897-902
[11]	Ding S F, Xu L, Su C Y, et al. An optimizing method of RBF neural network based on genetic algorithm[J]. Neural Computing and Applications, 2012,21(2):333-336
[12]	Jia W K, Zhao D A, Shen T, et al. A new optimized GA-RBF neural network algorithm[J]. Computational Intelligence and Neuroscience, 2014,2014:982045
[13]	汪定伟,王俊伟.智能优化方法[M].北京:高等教育出版社,2007:20-29 Wang D W, Wang J W. Intelligent optimization methods[M]. Beijing:Higher Education Press, 2007:20-29(in Chinese)
[14]	杨瑞.基于遗传算法优化RBF神经网络控制器[D].哈尔滨:哈尔滨理工大学,2011 Yang R. Optimize RBF neural network controller based on the genetic algorithm[D]. Harbin:Harbin University of Science and Technology, 2011(in Chinese)
[15]	龙亿,杜志江,王伟东.GA优化的RBF神经网络外骨骼灵敏度放大控制[J].哈尔滨工业大学学报,2015,47(7):26-30 Long Y, Du Z J, Wang W D. RBF neural network with genetic algorithm optimization based sensitivity amplification control for exoskeleton[J]. Journal of Harbin Institute of Technology, 2015,47(7):26-30(in Chinese)