In this work, the sliding mode control (SMC) theory, genetic algorithm
(GA) and input estimation method (IE) are adopted to design a control
system, which is capable of implementing the disturbance torque estimation
and good control performance for robotic systems. The IE algorithm is an
on-line inverse estimation method involving the Kalman filter (KF) and
recursive least square estimator (RLSE), which can estimate the
disturbance torque without additional torque sensor. The SMC has the characteristics of low sensitivity with variable system parameters. Furthermore, the GA is proposed to search the optimal parameters for SMC controller. The design and analysis processes of the proposed controller are also discussed in this paper. The joint control of a flexible-joint robot is utilized to test and verify the control performance of the proposed controller. The simulation results demonstrate that the IE is a good observer for estimating the disturbance torque, and the proposed controller can effectively cope with the situation that the disturbance exists.