EKF estimation based PID Type-2 fuzzy control of electric cars

Abstract

In-wheel motor (IWM) configuration analysis in electric cars is essential for energy efficiency, design adaptability, and reduced vehicle mass. However, improving the vehicle design to accomplish an enhanced ride comfort and handling performance is still a challenging task. Hence, an active suspension system is developed through an Extended Kalman Filter (EKF)-based proportional-integral-derivative (PID)-fuzzy type-2 fuzzy neural network (T2FNN) controller. Jacobian of the modeled suspension system is trained and optimized through an EKF-NN framework. Additionally, one vibration absorber is employed to address the reduced force transmission to the motor bearing. Moreover, the optimized parameters of the suspension system for the objective design configuration are adopted during an optimization procedure. The effectiveness of the proposed control algorithm to withstand the external disturbances imposed on the system through the random road excitations is investigated in terms of the road holding and ride comfort criteria. The simulation results confirm the suitability of the designed control algorithm to address the major inherent drawbacks of the suspension system of IWM-based electric vehicles.