Development of Lateral Control for Different Electric Vehicle Drive and Steering Systems

Abstract

It is necessary to implement a stable and robust longitudinal and lateral motion control for autonomous vehicles to follow the trajectory precisely, and especially crucial for emergency scenarios to avoid a crash or to minimize the crash severities. Automated driving vehicles have the possibility of different, mainly electric drive configuration as well as different steering systems. At the same time, lateral control is a requirement to achieve the desired vehicle's behavior. Within this scope, the objective of this paper is to develop a simplified and flexible lateral control that can be applied to the most current vehicle's drive and steering combination, reaching the vehicle's preferred behavior by adapting a reduced number of parameters. A complete vehicle dynamics simulation model in Matlab/Simulink is used to establish lateral control. A driver model is included to perform the necessary closed-loop test maneuvers and to calculate the necessary input parameters of the controller based on the environmental and vehicle constructive characteristics. After setting the desired path, the lateral control defines the optimum steering angle accordingly by using an improved path previewing approach. This approach makes use of look-ahead observation fields between the desired and the actual path, combined with a control strategy to correct the vehicle's center of gravity error. The performance of the lateral control is evaluated against different vehicle architectures. It includes a car with three possible powered electrical drive systems (all in-wheel, front-wheel, and rear-wheel drive) and three steering configurations (front-parallel, front-Ackermann and four-wheel steering). The results reveal that the parameters of the lateral control have a substantial influence on the gap between the actual and desired path. The path preview approach enables the quantity and property tuning of the look-ahead observation fields, which can result either in an aggressive or a smooth maneuver. Among the investigated architectures, the four-wheel steering with all in-wheel drive leads to the fastest response in an emergency scenario.