# Adversarial Sensor Attack on LiDAR-based Perception in Autonomous Driving

## TL;DR

This paper investigates the security vulnerabilities of LiDAR-based perception in autonomous vehicles, demonstrating a strategic spoofing attack that can fool machine learning models and potentially compromise road safety.

## Contribution

It is the first to analyze LiDAR spoofing attacks in AVs, formulating an optimization-based attack method and achieving a 75% success rate, highlighting new security challenges.

## Key findings

- Blind spoofing is ineffective against ML-based detection.
- Strategic control of spoofing improves attack success rates.
- Potential safety risks demonstrated through attack scenarios.

## Abstract

In Autonomous Vehicles (AVs), one fundamental pillar is perception, which leverages sensors like cameras and LiDARs (Light Detection and Ranging) to understand the driving environment. Due to its direct impact on road safety, multiple prior efforts have been made to study its the security of perception systems. In contrast to prior work that concentrates on camera-based perception, in this work we perform the first security study of LiDAR-based perception in AV settings, which is highly important but unexplored. We consider LiDAR spoofing attacks as the threat model and set the attack goal as spoofing obstacles close to the front of a victim AV. We find that blindly applying LiDAR spoofing is insufficient to achieve this goal due to the machine learning-based object detection process. Thus, we then explore the possibility of strategically controlling the spoofed attack to fool the machine learning model. We formulate this task as an optimization problem and design modeling methods for the input perturbation function and the objective function. We also identify the inherent limitations of directly solving the problem using optimization and design an algorithm that combines optimization and global sampling, which improves the attack success rates to around 75%. As a case study to understand the attack impact at the AV driving decision level, we construct and evaluate two attack scenarios that may damage road safety and mobility. We also discuss defense directions at the AV system, sensor, and machine learning model levels.