Hybrid Fuzzy-Expert System Control for Robotic Manipulator Applications

This research examines a hybrid fuzzy-expert system for the control of robotic manipulators, integrating the flexibility of fuzzy logic with the analytical decision-making capabilities of expert systems. The hybrid system switches dynamically between triangle membership functions, which facilitate smooth transitions, and trapezoidal membership functions, which efficiently manage sudden step changes. This adaptive technique mitigates the shortcomings of independent fuzzy logic controllers, particularly their inconsistency across varied setpoints. Simulation outcomes demonstrate substantial decreases in overshoot and settling time, as well as enhanced adaptability and flexibility in dynamic settings. A comparison test shows that the hybrid system is better than separate triangular and trapezoidal fuzzy controllers because it chooses the best control strategy based on the setpoint attributes in real time. Although there are occasional compromises in accuracy (IAE and RMSE), the hybrid controller provides balanced performance appropriate for various robotic applications. The results confirm its viability as a dependable option for industrial and medical robots, particularly in applications necessitating precision control and adaptability.