柔性关节机器人高精度自适应反步法控制  

High-precision Adaptive Backstepping Control of Flexible Joint Robots

作　　者：王雪竹[1,2] 李洪谊[1] 王越超[1] 崔龙[1] 

机构地区：[1]中国科学院沈阳自动化研究所机器人学国家重点实验室,辽宁沈阳110016 [2]中国科学院大学,北京100049

出　　处：《信息与控制》2016年第1期1-7,共7页Information and Control

基　　金：国际热核聚变实验堆(ITER)计划专项资助项目(2012GB102005)

摘　　要：为实现多连杆柔性关节机器人的高精度运动控制,首先对其建立完整的动力学模型,包含了连杆和关节动力学的耦合项、Lu Gre动态摩擦模型和关节回差等因素.然后针对该模型设计带观测器的自适应反步法控制器,对不可测项进行在线估计和补偿.理论分析证明了观测器的收敛性和闭环系统的稳定性.该方法在一个3DOF(degree of freedom)柔性关节机器人上进行仿真,仿真结果验证了观测器的有效性,并表明该控制器能够降低连杆跟踪误差,实现良好的轨迹跟踪效果.In order to achieve high-precision motion control performance in multi-link flexible joint robots,we present a complete dynamics model,which includes the coupling between the links and the actuators,a Lu Gre dynamic friction model,and backlash. Based on this dynamics model,we propose an adaptive backstepping control method with observers,in which non-measurable terms are estimated and compensated online. Our theoretical analysis proves the convergence of the observer and the stability of the closed-loop system. The simulation results of this method on a 3-DOF flexible joint robot verify the effectivity of the observer,and demonstrate that the proposed controller can reduce link errors and achieve good trajectory tracking performance.

关 键 词：柔性关节机器人 反步法 自适应控制 观测器 LuGre摩擦 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]