Adaptive fuzzy self-learning controller for robotic manipulators

In this paper we propose a new control algorithm for trajectory tracking of a robotic manipulator. Our controller is designed by combining a fixed controller and an adaptive fuzzy controller for tracking a desired trajectory in the vicinity of uncertain environment and change in manipulator dynamics. The controlled system is characterized by feedforward and feedback components which can be computed separately and simultaneously. The feedforward component computes the gross torque by using computed torque method. The feedback component consisting of an adaptive fuzzy controller based on self learning computes the compensating torque which reduces errors caused by external disturbances and change in manipulator dynamics along the nominal trajectory. A computer simulation study is conducted to evaluate the performance of the proposed adaptive control. The manipulator payload is changed over a wide range and noise is introduced at the input to check the robustness of the designed control scheme. Under both these conditions, the response of the controlled system tracks the desired trajectory faithfully, thereby validating our proposed control scheme.