Model-free adaptive control for the ball-joint robot driven by PMA group

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.