3-RRR平面并联机器人神经网络滑模控制研究  

Research on Neural Network Sliding Mode Control for 3-RRR Planar Parallel Robot

作　　者：任鹏飞 耿世勇[2] REN Pengfei;GENG Shiyong(College of Electrical Information and Engineering,Henan Institute of Engineering,Zhengzhou Henan 451191,China;School of Electric and Information Engineering,Zhongyuan University of Technology,Zhengzhou Henan 450007,China)

机构地区：[1]河南工程学院电气信息工程学院,河南郑州451191 [2]中原工学院电子信息学院,河南郑州450007

出　　处：《机床与液压》2018年第15期16-19,共4页Machine Tool & Hydraulics

基　　金：国家自然科学基金资助项目(51407035);广东省自然科学基金资助项目(S2013040013776);河南省高等学校重点科研项目(17A470007)

摘　　要：针对具有强耦合和高非线性并联机器人的轨迹跟踪控制研究,设计了一种基于神经网络滑模控制器的控制系统。在传统滑模控制的基础上,利用神经网络算法实时修正系统非线性项和不确定参数的功能,有效抑制了SMC系统的抖振现象。建立了3-RRR平面并联机器人的结构简图和Matlab模型,并采用闭环矢量法得到了机器人的运动学反解,为控制系统提供了参考输入。基于机器人的简化动力学方程,设计了一种RBF神经网络滑模控制器,并构造Lyapunov函数证明控制器的稳定性。分别采用传统滑模和神经网络滑模控制方式对机器人的轨迹跟踪进行仿真分析。仿真结果表明:神经网络滑模控制器具有更好的轨迹跟踪精度和较小的稳态误差,验证了神经网络SMC控制器的有效性。A control system based on neural network sliding mode controller is designed for the trajectory tracking of parallel robot with strong coupling and high nonlinearity. On the basis of traditional sliding mode control,the neural network algorithm was used to correct the function of nonlinear and uncertain parameters of system,which could effectively suppress the chattering phenomenon of Sliding Mode Control（ SMC） system. Firstly,the structure diagram and Matlab model of 3-RRR planar parallel robot were established,and the inverse kinematics solution of the robot was obtained by the closed loop vector method,which was the reference input for control system. Then,based on the simplified dynamic equation of robot,a Radial Basis Function（ RBF） neural network SMC was designed,and a Lyapunov function was constructed to prove the stability of the controller. Finally,the traditional SMC and neural network SMC were used to simulate the trajectory of robot. The simulation results show that the neural network sliding mode controller has better trajectory tracking accuracy and small steady-state error,and verifies the effectiveness of neural network SMC controller.

关 键 词：并联机器人 运动学分析 神经网络 滑模控制 仿真分析 

分 类 号：TH112[机械工程—机械设计及理论]