Intelligent Tracking Control of Redundant Robot Manipulators including Actuator Dynamics

In this paper, the trajectory tracking control problem of redundant robot manipulators at the actuator level is studied. In the proposed control scheme, RBF (Radial Basis Function) neural network and adaptive bound part is combined with the model based controller. All the existing uncertainties are learned with the RBF neural network without offline learning. The uncertain parameters, the bounded external disturbances and the neural network reconstruction error are approximated by the adaptive bound part. The hybrid controller is designed in such a way that both the trajectory tracking error and subtask task tracking error converges to zero as well as it controls the DC motors that are used to provide the required currents and torques. Finally, the Lyapunovs stability analysis is used to prove the overall closed loop system to be asymptotically stable.