A Comparative Study to Evaluate the Performance of ANFIS and PID Controllers in Electric Bicycle Pedal Assist Systems

Electric bikes (E-bikes) equipped with Pedal Assist Systems (PAS) represent a forward-thinking and ecofriendly mode of transportation, merging the traditional cycling experience with the efficiency of electric motor technology. The purpose of this research is to explore the potential of an Adaptive Neuro-Fuzzy Inference System (ANFIS) controller to enhance the performance and adaptability of a PAS E-bike by optimizing motor output and responsiveness under nonlinear conditions. In this paper, a detailed E-bike model is developed and evaluated utilizing MATLAB-Simulink to examine the performance of the proposed ANFIS controller. Also, a comparative study has also been conducted to highlight the advantage of using a ANFIS controller over conventional Proportional-Integral-Derivative (PID) controller at different load conditions in terms of response time to rider input and adaptability to diverse events. The obtained results demonstrate that the proposed ANFIS provides a more adaptable response and relative assistance to the input better than the PID controller.