Steering Control of Self-Driving Car Using Super-Twisting Sliding Mode Control

Abstract

Super-Twisting is an algorithm implemented on the typical Sliding Mode Control (SMC) that offers a different method of controlling complex systems, and commonly found in advanced robotic systems such as industrial robotic manipulators and mobile robotics. This paper presents a review of the steering control of a self-driving car using Super-Twisting algorithm. The algorithm offers optimal performance and speed when compared to other common controls such as the PID control due to its chattering resistant nature and speed. The algorithm is tested using a real-world servo motor applied on a real car body on the 2021 NXP Racetrack ©, using a set amount of time expected to complete the track at a certain speed. The Super-Twisting algorithm is applied on the motor with the required steering angles for the racetrack with respect to time as the reference input, and it is compared to an implemented PID Control. The results show the superior performance of the Super-Twisting SMC in this application compared to the PID control, offering the best option of control resulting in better line following and hence better overall time on the circuit.