Robust intelligent neural network controller for uncertain steer-by-wire electrical power steering with driver's feedback

This study develops a novel robust and robust neural network (NN) controller for the Steer-by-Wire (SbW) system under uncertain conditions with steering motor pulsating disturbances and other unknown external uncertainties. Under the proposed NN-based controller for the SbW system, the front wheel (FW) angle is followed by the reference driver Steering-Wheel (SW) angle under the effect of uncertainties like self-aligning torque, external disturbances in the rack and steering motor torque pulsation disturbances. Unlike conventional electrical power steering (EPS) systems, the main challenge in the SbW system is generating the feedback torque for the driver perceived steering feel due to the absence of any mechanical connection between the steering wheel and the front wheel. In order to address this problem, a novel feedback torque generation scheme is investigated based on steering motor angle information. This approach incorporates a feeling gain that makes it more modular in the context of vehicle dynamics. Apart from its modularity, the main advantage of the proposed control scheme is that it relaxes the condition of having prior knowledge of the system's parameters, leveraging function approximate properties of neural architecture. To prove the stability and robustness of the proposed control method, an extensive Lyapunov functional study is presented. Finally, a simulation analysis of the practical system is performed to verify the effectiveness of the proposed method with different driving manoeuvres and operating points.