Design of an Optimal Fuzzy Controller of an Under-Actuated Manipulator Based on Teaching-Learning-Based Optimization

In this paper, an optimal fuzzy controller based on the Teaching-Learning-Based Optimization (TLBO) algorithm has been presented for the stabilization of a two-link planar horizontal under-actuated manipulator with two revolute (2R) joints. For the considered fuzzy control method, a singleton fuzzifier, a centre average defuzzifier and a product inference engine have been used. The TLBO algorithm has been implemented for searching the optimum parameters of the fuzzy controller with consideration of time integral of the absolute error of the state variables as the objective function. The proposed control method has been utilized for the 2R under-actuated manipulator with the second passive joint wherein the model moves in the horizontal plane and friction forces have been considered. Simulation results of the offered control method have been illustrated for the stabilization of the considered robot system. Moreover, for different initial conditions, the effectiveness and the robustness of the mentioned strategy have been challenged.