Hybrid Neural Predictive-Fuzzy Controller for Motorized Robot Arm

In this study, a design methodology is introduced that blends the neural predictive and fuzzy logic controllers in an intelligent way developing a new intelligent hybrid controller has been achieved. In this design methodology, the fuzzy logic controller works in parallel with neural predictive controller and adjusts the output of the predictive controller in order to enhance system predicted input. The performance of our proposal controller is demonstrated on the motorized robot arm with disturbances. The simulation shows that the new hybrid neural predictive-fuzzy controller provides better system response in terms of transient and steady-state performance when compared to neural predictive or fuzzy logic controller applications. The simulation is performed on MATLAB/Simulink toolbox to illustrate the efficiency of the proposed method.