Deep Learning-Based Autonomous Driving Systems: A Survey of Attacks and Defenses

TL;DR

This survey reviews various attacks on deep learning autonomous driving systems and discusses current defense strategies, highlighting challenges and future research directions for improving safety and security.

Contribution

It provides a comprehensive analysis of attack types and defense mechanisms in deep learning-based autonomous driving, covering physical, cyber, and adversarial threats.

Findings

Detailed categorization of attack types on ADSs

Overview of state-of-the-art defense mechanisms

Identification of promising research directions for safety enhancement

Abstract

The rapid development of artificial intelligence, especially deep learning technology, has advanced autonomous driving systems (ADSs) by providing precise control decisions to counterpart almost any driving event, spanning from anti-fatigue safe driving to intelligent route planning. However, ADSs are still plagued by increasing threats from different attacks, which could be categorized into physical attacks, cyberattacks and learning-based adversarial attacks. Inevitably, the safety and security of deep learning-based autonomous driving are severely challenged by these attacks, from which the countermeasures should be analyzed and studied comprehensively to mitigate all potential risks. This survey provides a thorough analysis of different attacks that may jeopardize ADSs, as well as the corresponding state-of-the-art defense mechanisms. The analysis is unrolled by taking an in-depth overview of each step in the ADS workflow, covering adversarial attacks for various deep learning models and attacks in both physical and cyber context. Furthermore, some promising research directions are suggested in order to improve deep learning-based autonomous driving safety, including model robustness training, model testing and verification, and anomaly detection based on cloud/edge servers.