차량용 LiDAR 센서 물리적 신호교란 공격 중심의 실험적 분석과 대응방안 제안 = Experimental Analysis of Physical Signal Jamming Attacks on Automotive LiDAR Sensors and Proposal of Countermeasures

다국어 초록 (Multilingual Abstract)

LiDAR(Light Detection And Ranging) sensors, which play a pivotal role among cameras, RADAR(RAdio Detection And Ranging), and ultrasonic sensors for the safe operation of autonomous vehicles, can recognize and detect objects in 360 degrees. However, since LiDAR sensors use lasers to measure distance, they are vulnerable to attackers and face various security threats. In this paper, we examine several security threats against LiDAR sensors: relay, spoofing, and replay attacks, analyze the possibility and impact of physical jamming attacks, and analyze the risk these attacks pose to the reliability of autonomous driving systems. Through experiments, we show that jamming attacks can cause errors in the ranging ability of LiDAR sensors. With vehicle-to-vehicle (V2V) communication, multi-sensor fusion under development and LiDAR anomaly data detection, this work aims to provide a basic direction for countermeasures against these threats enhancing the security of autonomous vehicles, and verify the practical applicability and effectiveness of the proposed countermeasures in future research.

국문 초록 (Abstract)

자율주행 자동차의 안전한 운행을 위해 카메라, RADAR(RAdio Detection And Ranging), 초음파 센서 중 중추적인 역할을 하는 LiDAR(Light Detection And Ranging) 센서는 360도에서 사물을 인식하고 탐지할 수 있다. 하지만 이러한 LiDAR 센서는 레이저를 통해서 거리를 측정하기 때문에 공격자에 노출되기 쉬우며 다양한 보안 위협에 직면해 있다. 따라서 본 논문에서는 LiDAR 센서를 대상으로 한 여러 가지 보안 위협인 Relay, Spoofing, Replay 공격을 살펴보고 물리적 신호교란(Jamming) 공격의 가능성과 그 영향을 분석하며, 이러한 공격이 자율주행 시스템의 안정성에 미치는 위험을 분석한다. 실험을 통해, 물리적 신호교란 공격이 LiDAR 센서의 거리 측정 능력에 오류를 유발할 수 있음을 보여준다. 개발이 진행 중인 차량 간 통신(Vehicle-to-Vehicle, V2V), 다중 센서 융합과 LiDAR 비정상 데이터 탐지를 통해 이러한 위협에 대한 대응방안과 자율주행 차량의 보안 강화를 위한 기초적인 방향을 제시하고 향후 연구에서 제안된 대응방안의 실제 적용 가능성과 효과를 검증하는 것을 목표로 한다.

참고문헌 (Reference)

1 HYUNDAI, "ioniq5 robotaxi"

2 Cao, Yulong, "You Can’t See Me : Physical Removal Attacks on LiDAR-based Autonomous Vehicles Driving Frameworks" 2993-3010, 2023

3 ParaVeiw, "VeloView"

4 Velodyne, "VLP-16 user manual"

5 LUMISOL, "Types of Lidar Sensors Based on Operation"

6 Jiseong Jeong, "Trends in the Lidar Sensor Technology for Autonomous Vehicles" 45 (45): 22-26, 2023

7 Sun, Jiachen, "Towards robust {LiDAR-based} perception in autonomous driving: General black-box adversarial sensor attack and countermeasures" 877-894, 2020

8 Ren, Kui, "The Security of Autonomous Driving:Threats, Defenses, and Future Directions" 108 (108): 357-372, 2020

9 SAE, "Taxonomy and Definitions for Terms Related to Driving Automation Systems for On-Road Motor Vehicles"

10 Hallyburton, R Spencer, "Securing Autonomous Vehicles Under Partial-Information Cyber Attacks on LiDAR Data"

11 Yang, Kaichen, "Robust roadside physical adversarial attack against deep learning in lidar perception modules" 349-362, 2021

12 Petit. J, "Remote Attacks on Automated Vehicles Sensors: Experiments on Camera and LiDAR" 11 (11): 995-, 2015

13 Y. Li, "Lidar for Autonomous Driving : The Principles, Challenges, and Trends for Automotive Lidar and Perception Systems" 37 (37): 50-61, 2020

14 Pierrick Boulay, "LiDAR industry report:Hesai wins top rankings in the global market" Yole Group 2022

15 Chris Clonts, "LIdar vs. Everybody in the Onboard Sensor Race" 10-15, 2023

16 Shin, Hocheol Kim, "Illusion and Dazzle: Adversarial Optical Channel Exploits against Lidars for Automotive Applications" 445-467, 2017

17 RON AMADEO, "Google’s Waymo invests in LiDAR technology, cuts costs by 90 percent" ars TECHNICA 2017

18 G.D Choi, "Development Trends and Expectations of Three-Dimensional Imager based on LIDAR Technology for Autonomous Smart Car Navigation" 31 (31): 0-0, 2016

19 Eunbin Na, "Autonomous vehicles and LiDAR" 50 (50): 44-49, 2023

20 ResearchInChina, "Automotive LiDAR Industry Report, 2023" 2023

21 Khader, Motaz, "An introduction to automotive lidar"

22 Cao, Yulong, "Adversarial sensor attack on lidar-based perception in autonomous driving" 2019