COMPARATIVE STUDY OF PD, FUZZY AND NEURAL NETWORK BASED. CONTROLLER FOR -A ROBOT ARM WITH TIME VARYING LOAD

Abstract

The precise control of robot manipulator to track the desired trajectory is a very tedious job and almost unachievable with the help of conventional controllers. This task is achievable to a certain extent with the help of adaptive controllers but they. also have their own .limitations that the system parameters being controlled, change relatively slowly. In this thesis an algorithm is proposed to minimize thewdeficiencies. It is called computed torque method. This is further improved to fuzzy logic based controller. It gives the guaranteed trajectory tracking with uncertainties from environment. Finally these results are compared with neural network based controller.