Optimization of Fuzzy Controller for an SMA-actuated Artificial Finger Robot

The purpose of this paper is to design and optimize an intelligent fuzzy-logic controller for a three-degree of freedom (3DOF) artificial finger with shape-memory alloy (SMA) wire actuators. The robotic finger is constructed using three SMA wires as tendons to bend each phalanx of the finger around its revolute joint and three torsion springs which return the phalanxes to their original positions. A PID controller is designed to control the rotation of each phalanx. The gains of the controller are defined and optimized using the genetic algorithm. Finally, a fuzzy PID controller is presented to improve the performance of the system. The performance of the designed controller to achieve the desired output is simulated and also tested. The rotation of each link of both prototype robot and simulated model is measured. The experimental results show that the fuzzy controller can reach the desired angle in less time and the output signal is uniform. Moreover, the simulation results indicate that the closedloop control system of the simulated robot is in good agreement with the prototype robot.