气动肌肉群驱动球关节机器人的无模型自适应控制  

Model-Free Adaptive Control for the Ball-Joint Robot Driven by PMA Group

作　　者：刘昱[1] 王涛[1] 范伟[1] 王渝[1] 

机构地区：[1]北京理工大学自动化学院,北京100081

出　　处：《机器人》2013年第2期129-134,共6页Robot

基　　金：国家自然科学基金资助项目(50475163)

摘　　要：设计了一种由气动肌肉群驱动的三轴球关节机器人,并对其进行逆运动学分析.根据该机器人的运动学逆解,提出了基于非线性反馈的气动肌肉群控制策略.气动肌肉系统是一个强非线性时变的自平衡系统,鉴于传统控制方法难以克服控制精度、响应速度和稳定性之间的矛盾,引入改进型无模型自适应控制器.实验证明,在保证稳定性的前提下,控制器提高了系统的调节速度和控制精度,使得系统的稳态误差小于0.3,取得了良好的控制效果.A triaxial ball-joint robot driven by PMA （pneumatic muscle actuator） group is designed. The inverse kinematics of the robot is analyzed. On the basis of its inverse kinematics analysis, the control strategy of the PMA group based on nonlinear feedback is proposed. The system of PMA is a strongly nonlinear and time-varying self-balanced system. Because it is difficult to overcome the conflict among the regulating speed, the stability and the control accuracy with traditional control methods, an improved model-free adaptive control is introduced. Experiments show that on the premise of ensuring the stability, the control accuracy and the adjusting speed are enhanced. The system＇s steady-error is less than 0.3° and a satisfactory control result is obtained.

关 键 词：气动肌肉 逆运动学分析 群控策略 无模型自适应控制 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]