An Adaptive Disturbance Observer for A Two-link Robot Manipulator

In this paper, an adaptive disturbance observer based control algorithm is proposed for a two-link robot manipulator. The dynamics of the multi-body manipulator is coupled each link and highly nonlinear. So many sorts of robot manipulators have to be robust against internal and external variations in the work place. The disturbance observer has to solve those problems by estimating unknown states and rejecting decoupled terms between links. It is difficult to control worked robot manipulators because of changing many factors which are effect on the system especially such as the inertia moment of links in the robot manipulator. Conventional disturbance observer method ignoring some robot dynamics' factors was suggested to linearize the multi-body system. But this paper suggests that it makes the most use of some important factors of the dynamics that includes variable terms in the inertia moment of links. A suggested adaptive disturbance observer makes a more robust system against the excessive disturbance, namely payloads and external forces which are much bigger than inertia moment of each motor of links in the system, by updating the nominal inertia moment in the disturbance observer simultaneously. The effectiveness of the designed disturbance observer is investigated through the simulation results on the two-link robot manipulator.