Towards Artificial-Intelligence-Based Cybersecurity for Robustifying Automated Driving Systems Against Camera Sensor Attacks

CARAMEL is a European project that aims amongst others to improve and extend cyberthreat detection and mitigation techniques for automotive driving systems. This paper highlights the important role that advanced artificial intelligence and machine learning techniques can have in proactively addressing modern autonomous vehicle cybersecurity challenges and on mitigating associated safety risks when dealing with targetted attacks on a vehicle's camera sensors. The cybersecurity solutions developed by CARAMEL are based on powerful AI tools and algorithms to combat security risks in automated driving systems and will be hosted on embedded processors and platforms. As such, it will be possible to have a specialized anti-hacking device that addresses newly introduced technological dimensions for increased robustness and cybersecurity in addition to industry needs for high speed, low latency, functional safety, light weight, low power consumption.© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works. https://www.ieee.org/publications_standards/publications/rights/rights_policies.html C. Kyrkou et al., "Towards Artificial-Intelligence-Based Cybersecurity for Robustifying Automated Driving Systems Against Camera Sensor Attacks," 2020 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), Limassol, Cyprus, 2020, pp. 476-481, doi: 10.1109/ISVLSI49217.2020.00-11. https://ieeexplore.ieee.org/document/915490