Adaptive Fuzzy PID Control of Two‐DOF Pneumatic Artificial Muscle Manipulators for Leg Rehabilitation

ABSTRACT Rehabilitation robots are machines that allow patients to perform practice movements and ought to be sufficiently compliant for safe human interaction. Most robots today use hydraulic systems and continually rotating servo motors, making them less suitable for rehabilitation. The Pneumatic Muscle Manipulators (PAMs), however, offer an intriguing substitute for electric actuators due to their inherent compliance, but their nonlinear properties among force, displacement, and pressure make precise position control impossible in the classical controllers that neglect the significant phenomena of nonlinearities and handle the unmodeled dynamics of a pneumatic artificial muscle (PAM) manipulator. Hence, this paper presents adaptive fuzzy PID control for two‐degree‐of‐freedom (2‐DOF) PAM manipulators for leg rehabilitation to alleviate this problem. A 2‐DOF robotic manipulator is designed to freely rehabilitate the upper and lower leg arms. MATLAB/Simulink software was used to perform the simulation. The results demonstrate that the fuzzy PID controller exhibits excellent transient performance, effectively handles nonlinearities, and efficiently rejects disturbances and parameter variations.