Cloud-Based Digital Twins for Vehicle Dynamics Control with Application to Lateral Stability Enhancement

Abstract

Vehicle-to-Everything communication facilitates the creation of real-time, cloud-based traffic Digital Twins (DTs) that integrate diverse data streams to offer immediate and predictive insights into the driving environment. Although the application of traffic DTs for automated driving has been extensively studied, their potential to enhance vehicle dynamics control in manually driven vehicles remains underexplored. This paper addresses this gap by investigating the application of cloud-based traffic DTs for vehicle dynamics control. It presents key DT considerations such as model fidelity, data integration, and validation and verification, while exploring prospective DT services and cloud-based control’s inherent advantages and challenges. To illustrate these concepts, a use case for lateral vehicle stability control is presented and experimentally validated. The demonstration shows that using DT-derived information, Adaptive Cruise Control (ACC) and Torque Vectoring (TV) systems can proactively modify vehicle speed and torque distribution to maintain or improve stability and vehicle handling for human drivers. The paper concludes by evaluating the cloud execution of ACC and TV, highlighting the potential to reduce onboard computational requirements.