VisionGuard: Secure and Robust Visual Perception of Autonomous Vehicles in Practice
Open Access
- 2 December 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1864-1878
Abstract
Modern Autonomous Vehicles (AVs) implement the Visual Perception Module (VPM) to perceive their surroundings. This VPM adopts various Deep Neural Network (DNN) models to process the data collected from cameras and LiDAR. Prior studies have shown that these models are vulnerable to physical adversarial examples (PAEs), which pose a critical safety risk to the autonomous driving task. While a few defense methods have been proposed to safeguard AVs, most of them only target a limited set of attack types and specific scenarios, making them impractical for real-world protection. In this paper, we introduce VisionGuard, a novel and practical methodology to comprehensively detect and mitigate various types of PAEs to the VPM. The key of VisionGuard is to leverage the spatiotemporal inconsistency property of PAEs to detect anomalies. It predicts the motion states from historical ones and compares them with the current driving states to identify any motion inconsistency caused by physical attacks. We evaluate 9 state-of-the-art PAEs against both camera and camera-LiDAR fusion-based object classification & detection models. Experimental results in both simulation and physical world validate the effectiveness and robustness of VisionGuard. Codes, demo videos and appendix can be found on our anonymous website: https://sites.google.com/view/visionguard.Keywords
Funding Information
- National Research Foundation, Singapore and DSO National Laboratories under its AI Singapore Programme (AISG2-GC-2023-008)
- Nanyang Technological University (NTU)-DESAY SV Research Program under Grant 2018-0980
This publication has 44 references indexed in Scilit:
- Research on GNSS INS & GNSS/INS Integrated Navigation Method for Autonomous Vehicles: A SurveyIEEE Access, 2023
- TransWeather: Transformer-based Restoration of Images Degraded by Adverse Weather Conditions2022
- Shadows can be Dangerous: Stealthy and Effective Physical-world Adversarial Attack by Natural Phenomenon2022
- Learning Multiple Adverse Weather Removal via Two-stage Knowledge Learning and Multi-contrastive Regularization: Toward a Unified Model2022
- PointGuard: Provably Robust 3D Point Cloud Classification2021
- Adversarial Laser Beam: Effective Physical-World Attack to DNNs in a Blink2021
- The Translucent Patch: A Physical and Universal Attack on Object Detectors2021
- Towards Corner Case Detection by Modeling the Uncertainty of Instance Segmentation Networks2021
- A Model-Driven Deep Neural Network for Single Image Rain Removal2020
- Robust Physical-World Attacks on Deep Learning Visual Classification2018