Abstract: A fuzzy PID (proportional-integral-differential) control algorithm based on Prandtl-Ishlinskii (P-I) inverse model is proposed to improve the control accuracy of position tracking for a 3-DOF (degree of freedom) parallel platform actuated by PAM (pneumatic artificial muscle). Firstly, the hysteresis characteristics of a single pneumatic artificial muscle are analyzed through dynamic tests to establish P-I inverse model. Then, the PAM actuated 3-DOF parallel test platform is built, and the fuzzy PID controller is designed. Tests result show that the maximum error of the model is 0.3904 mm, and the average error is 0.0793 mm, which verifies that the proposed control algorithm can greatly reduce the influence of PAM hysteresis characteristics on the dynamic control accuracy. And the error of angle γ of the 3-DOF platform can be kept within 0.13°, and the overall error fluctuation of the fuzzy PID controller based on P-I inverse model is small, which shows that the effect of dynamic trajectory tracking is good.