A Survey on Adversarial Robustness of LiDAR-based Machine Learning Perception in Autonomous Vehicles

TL;DR

This survey reviews the vulnerabilities of LiDAR-based perception systems in autonomous vehicles to adversarial attacks, highlighting threats, defenses, and the importance of robust security measures for safe autonomous driving.

Contribution

It provides a comprehensive overview of adversarial threats and defense strategies specific to LiDAR-based systems in autonomous vehicles, summarizing recent advances and challenges.

Findings

LiDAR systems are vulnerable to various adversarial attacks

Current defenses include sensor data validation and adversarial training

Securing LiDAR perception is critical for autonomous vehicle safety

Abstract

In autonomous driving, the combination of AI and vehicular technology offers great potential. However, this amalgamation comes with vulnerabilities to adversarial attacks. This survey focuses on the intersection of Adversarial Machine Learning (AML) and autonomous systems, with a specific focus on LiDAR-based systems. We comprehensively explore the threat landscape, encompassing cyber-attacks on sensors and adversarial perturbations. Additionally, we investigate defensive strategies employed in countering these threats. This paper endeavors to present a concise overview of the challenges and advances in securing autonomous driving systems against adversarial threats, emphasizing the need for robust defenses to ensure safety and security.