中国机械工程 ›› 2012, Vol. 23 ›› Issue (23): 2792-2796.

• 机械基础工程 • 上一篇    下一篇

基于模糊RBF神经网络动态摩擦分块补偿的机器人数字鲁棒滑模控制算法

李敏;王家序;肖科;黄超;徐超   

  1. 重庆大学机械传动国家重点实验室,重庆,400030
  • 出版日期:2012-12-10 发布日期:2012-12-21
  • 基金资助:
    国家自然科学基金资助重点项目(50735008);国家自然科学基金资助项目(50905189, 50905191) 
    Key Program of National Natural Science Foundation of China(No. 50735008)
    National Natural Science Foundation of China(No. 50905189, 50905191)

Digital Robust Sliding Mode Control of Robot Manipulator with Dynamic Friction Block Compensation Using Fuzzy RBF Neural Network

Li Min;Wang Jiaxu;Xiao Ke;Huang Chao;Xu Chao    

  1. State Key Laboratory of Mechanical Transmission,Chongqing University,Chongqing,400030
  • Online:2012-12-10 Published:2012-12-21
  • Supported by:
     
    Key Program of National Natural Science Foundation of China(No. 50735008)
    National Natural Science Foundation of China(No. 50905189, 50905191)

摘要:

结合非线性、强耦合的机器人动力学模型,提出了采用3个模糊RBF神经网络对机器人中的不确定项——LuGre动态摩擦进行分块补偿的机器人数字鲁棒滑模控制算法,在线自适应训练非线性动态摩擦项的参数,并分析了该算法的Lyapunov稳定性。通过在二自由度机器人上的仿真,证明了该算法具有高精度、高可靠性、高品质、稳定、强鲁棒性等特点。同时发现了该机器人的摩擦模型中存在类菱形吸引子等非线性动力学现象。

关键词: 模糊RBF神经网络, 摩擦补偿, LuGre摩擦模型, 不确定性, 机器人数字控制

Abstract:

Combining with nonlinear,strong coupling dynamics model of  a robot manipulator,this paper presented a digital robust sliding mode robot control
algorithm,which compensated for the uncertainties of robot manipulator-LuGre
dynamic friction with three fuzzy RBF neural network,and trained parameters of nonlinear dynamic friction on-line and adaptively.Then this paper analyzed the Lyapunov stability
of the algorithm.The simulation of a two degrees of freedom robot manipulator proves
that the algorithm is of high accuracy,high reliability,high quality,
stable and strong robustness.Meanwhile,nonlinear kinetic phenomena,such as rhombus
attractor,lie in the kinetic properties of the friction model of the robot manipulator. 

Key words: fuzzy radial base function(RBF) neural network, friction compensation, LuGre friction model, uncertainty, robot digital control

中图分类号: