机械臂轨迹跟踪的小波神经网络模糊滑模算法  

Trajectory Tracking Control of Robot Manipulator Based on Wavelet Neural Network and Fuzzy Sliding Mode

作　　者：王飞[1] 晁智强[1] 李华莹[1] 张传清[1] WANG Fei;CHAO Zhi-qiang;LI Hua-ying;ZHANG Chuan-qing(Department of Mechanical Engineering, Academy of Armored Force Engineering, Beijing 100072, China)

机构地区：[1]装甲兵工程学院机械工程系,北京100072

出　　处：《计算机仿真》2017年第11期353-359,共7页Computer Simulation

基　　金：军内科研项目(2013BZ162)

摘　　要：为实现机械臂能够以理想的动态品质跟踪给定的轨迹,针对机械臂系统的非线性和不确定性,提出一种小波神经网络模糊滑模控制方法。上述方法通过小波神经网络来逼近滑模控制的等效部分,实现了无需模型信息的滑模控制。同时,通过设计模糊控制器,根据计算当前运动点到滑模面的距离来自适应地调节滑模控制的切换部分增益,有效地克服了抖振。上述方法在一定程度上提高了响应速度,减小了调整时间,并通过李雅普诺夫定理证明了系统的稳定性。仿真结果表明,针对三自由度机械臂系统,提出的方法是有效的。In order to make robot manipulator can track the desired trajectory under the better dynamic character, aimed at non-linearity and uncertainty of system of robot manipulator, a wavelet neural network fuzzy sliding mode controller is proposed. The wavelet neural network controller is introduced to approach the equivalent control part of sliding mode control,which realizes the sliding mode control without accurate model information. To reduce the chatte- ring of sliding mode control, a fuzzy controller is designed adaptively tune the switch part of sliding mode control by calculating the distance between current moving point and sliding mode surface. To a certain extent, the controller gets faster respone speed and shorter adjusting time. Lyapunov theorem is used to prove the stability of the control system. The simulation result shows that the proposed control scheme is effective aimed at 3-DOF robot manipulator system.

关 键 词：机械臂 轨迹跟踪 小波神经网络 模糊控制 滑模控制 仿真 

分 类 号：TP242.6[自动化与计算机技术—检测技术与自动化装置]