Exposing Adaptive Cruise Control in Advanced Driving Assistance Systems

Abstract

Advanced Driving Assistance Systems (ADAS) are expected to improve the urban driving experience by reducing accidents and enabling safe driving. Such systems provide different features to automate driving and ease the driver's tasks. ADAS features highly depend on the data received from in-vehicle networks, such as Controller Area Network (CAN) bus. CAN play an important role in in-vehicle communication by enabling efficient data exchange among electronic control units (ECUs). However, the CAN bus is vulnerable to different attack strategies, which can manipulate CAN data and cause fatal consequences on ADAS features. This paper shows how CAN bus data can be exploited to expose an industry-grade advanced driving assistance system's Adaptive Cruise Control (ACC) feature. The attack technique demonstrates the requirements of timely human intervention or automatic recovery techniques. We also provide a detection mechanism based on the high-definition vector map and Global Positioning System device installed in the vehicle.