Abstract: Robotic systems suffer from different types of uncertainties which may affect the system control accuracy and even cause the system unstable. In this paper, for a class of uncertain robotic systems with external disturbance, internal dynamic parametric uncertainty and unknown dead-zone, an adaptive controller based on disturbance observer is proposed. First, the nonlinear mathematical model of robotic systems subject to external disturbance is established and the internal dynamic parametric uncertainty and unknown dead-zone in the system are analyzed. Then nonlinear disturbance observer is employed to estimate and compensate for the external disturbance. Based on the disturbance observer, an adaptive controller is then designed to deal with the uncertain dynamic parameters and unknown dead-zone. Finally, the stability and the convergence of the position tracking error of the system are proved via Lyapunov function analysis. Digital simulations demonstrate the validity of the proposed approach.