Machine Learning based intrusion detection systems for connected autonomous vehicles: A survey

Connected and Autonomous Vehicles (CAVs) expect to dramatically improve road safety and efficiency of the transportation system. However, CAVs can be vulnerable to attacks at different levels, e.g., attacks on intra-vehicle networks and inter-vehicle networks. Those malicious attacks not only result in loss of confidentiality and user privacy but also lead to more serious consequences such as bodily injury and loss of life. An intrusion detection system (IDS) is one of the most effective ways to monitor the operations of vehicles and networks, detect different types of attacks, and provide essential information to mitigate and remedy the effects of attacks. To ensure the safety of CAVs, it is extremely important to detect various attacks accurately in a timely fashion. The purpose of this survey is to provide a comprehensive review of available machine learning (ML) based IDS for intra-vehicle and inter-vehicle networks. Additionally, this paper discusses publicly available datasets for CAV and offers a summary of the many current testbeds and future research trends for connected vehicle environments.