Design Adaptive Fuzzy Inference Controller for Robot Arm

Design robust controller for uncertain nonlinear systems most of time can be a challenging work. One of the most active research areas in this field is control of the robot arm. The control strategies for robotics are classified in two main groups: classical and non-classical methods, where the classical methods use the conventional control theory and non-classical methods use the artificial intelligence theory such as fuzzy logic, neural networks and/or neuro-fuzzy. Control robot arm using classical controllers are often having lots of problems because robotic systems are always highly nonlinear. Accurate robot manipulator is difficult because some dynamic parameters such as compliance and friction are not well understood and some robot parameters such as inertia are difficult to measure accurately. Artificial control such as Fuzzy logic, neural network, genetic algorithm, and neurofuzzy control have been applied in many applications. Therefore, stable control of a nonlinear dynamic system such as a robot arm is challenging because of some mentioned issues. In this paper the intelligent control of robot arm using Adaptive Fuzzy Gain scheduling (AFGS) and comparison to fuzzy logic controller (FLC) and various performance indices like the RMS error, and Steady state error are used for test the controller performance.