人工智能研究院学术报告 第2021-12-28期

Feiyang Cai is currently a visiting scholar at the Institute for Artificial Intelligence affiliated with Harbin Institute of Technology. He will obtain his Ph.D. degree in Electrical Engineering from Vanderbilt University, USA, in Jan 2022, where he works under Professor Xenofon Koutsoukos with a research affiliation at the Institute for Software Integrated Systems.

He received his B.S. and M.S. degrees in Electrical Engineering from Harbin Institute of Technology in 2014 and 2016, respectively. His research focuses on safety assurance in learning-enabled cyber-physical systems.

Real-time Detection of Sensor Replay and Controller Integrity Attacks in Cyber-Physical Systems

Cyber-Physical Systems (CPS) involve the integration of sensors, actuators, and computation into physical systems using networking, and they are increasingly deployed in a wide range of domains. Networking interconnections between components expose attack surfaces to adversaries who can launch various cyber-attacks. Recent work demonstrates that successful attacks can lead the system to abnormal behavior which may significantly undermine the safety of the system and even cause loss of human life. Therefore, detection of attacks is paramount to the safe and reliable operation of CPS. In this talk, we narrow down the scope to sensor replay and controller integrity attacks in the CPS domain of autonomous driving. I will present an approach for efficiently and robustly detecting such attacks in real-time.

报告时间

2021年12月28日 19:00-19:40

腾讯会议

会议号：824 455 316