Fuzzy logic control of single-link flexible joint manipulator

Abstract

In this paper, a single-link flexible joint robot is designed, fabricated and controlled. Three different fuzzy logic controllers (FLCs) were used to remove link vibrations and to obtain accurate trajectory tracking of link end-point. The input variables of the first and the second FLCs are motor rotation angle error and its derivative, and end-effector deflection error and derivative of deflection error, respectively. The outputs of these controllers are inputs of the third FLC producing the control signal of the flexible joint system. All of the FLCs were embedded in ds1103 real time control board. In the step response experiments, the error of motor rotation angle was obtained less than 0.12 degree and there was no steady-state error in the end-effector deflection. In the different trajectory-tracking experiments with the same FLC structure, small errors and phase shift in the system variables were occurred. Also, parameters of flexible arm were changed to test robustness of the FLC. It is seen that FLC are very robust to internal and external disturbances. Considering the all results of the experiments, FLC shows efficient performance in flexible robot arm.