受限机械臂的自适应小波滑模位置/力混合控制  

Hybrid position/force control of a constrained manipulator using adaptive wavelet sliding mode

作　　者：周芳[1] 朱齐丹[1] 姜迈 汪瞳[1] 

机构地区：[1]哈尔滨工程大学自动化学院,黑龙江哈尔滨150001

出　　处：《华中科技大学学报（自然科学版）》2009年第11期9-12,共4页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：人事部出国留学人员基金资助项目(20001024A2)

摘　　要：针对终端运动受约束的机械臂位置/力混合控制问题,提出了一种自适应小波滑模控制算法.该控制方案将滑模控制的鲁棒性及自适应调整能力与小波神经网络相结合,根据坐标变换得到降阶位置/力模型,针对降阶模型采用小波神经网络在线学习系统未知动力学模型中的非线性部分,同时引入滑模控制自动调整小波网络权值参数,从而对神经网络的固有逼近误差进行有效补偿,达到期望的跟踪性能.二自由度机械臂的仿真结果表明该控制器能保证系统快速有效跟踪指定参考信号.The hybrid position/force control of the robotic manipulator whose motion was constrained in the end was discussed. An adaptive wavelet sling mode control strategy was proposed, in which integrating the robustness and self-adjustment advantages of sliding mode control were integrated into the wavelet network controller. A reduced position/force model was obtained according to the nonlinear translation, a wavelet network was used to learn the unknown nonlinear dynamic of system. Simultaneous, the sliding mode technique was designed such that the wavelet weighted parameters could be automatically tuned thereby compensating the approximate error of the wavelet network in order to obtain the desired tracking performances. Simulation results of two-degree of freedom manipulator show the designed controller ensures the system track the desired reference signals quickly and efficiency.

关 键 词：受限机械臂 小波神经网络 滑模控制 自适应控制 位置/力混合控制 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]