Modeling and analyzing cyberattack effects on connected automated vehicular platoons

Abstract

Vehicular platooning requires an open-access environment of vehicle communications to enable cooperating connected automated vehicles (CAVs) to share their travel information. However, the open-access environment makes the cooperating platoon CAVs vulnerable to cyberattacks, causing concerns on traffic safety and mobility. So far, due to the lack of analytical models integrating malicious effects of manipulated information into CAV dynamics, we have limited knowledge on CAV platoon performance under cyberattack. It necessities a modeling framework to support a comprehensive analysis of the CAV platoon performance in a capricious travel environment. This study seeks to bridge this critical gap from the following three aspects. First, we review potential safety-related cyberattacks faced by CAVs, related to manipulation of vehicle position, speed, and acceleration. Based on the review, these cyberattacks are categorized into three types: bogus messages, replay/delay, and collusion attack. Second, we develop a generalized vehicle dynamics model that accounts for the cyberattack effects on CAV dynamics. The proposed model, labeled cooperative intelligent driver model (CIDM), integrates the dynamic communication topological structure to enable the analysis of effects of manipulated information on CAV dynamics. The dynamic communication structure is a time-varying function which is determined by the communication range and the distance between two vehicles. Upon the proposed CIDM, the third aspect focuses on simulation analysis of CAV platoon safety and efficiency under cyberattacks, which demonstrate the cyberattack effects at the platoon level. This study contributes to the fundamental understanding of CAV platoon dynamics under cyberattacks and lays a foundation to enhance the cybersecurity of CAV systems.