Robust back-stepping control for flexible-joint robot manipulators

This paper proposes a robust back-stepping control method for flexible joint robot manipulators to overcome parameter uncertainty. In the first step, a virtual control is designed for the rigid link dynamic by the nonlinear H-infinity method. In second and third steps, the other virtual control and real control are designed using the saturation-type nonlinear control based on the Lyapunov´s second method. In each step, the designed robust inputs satisfy the L₂-gain, which is equal to or less than gamma in the input-output sense of the closed loop system. In contrast with the previous researches, the proposed control method does not need the information of the acceleration and desire jerk of the links for the feedback control law. The performance robustness of the proposed control is verified through a series of computer simulations of a 2-DOF robot manipulator with joint flexibility.