Physical Adversarial Attacks on LiDAR-Based Perception Systems in Autonomous Vehicles: A Taxonomy, Analysis, and Defense Survey

Author(s)	Ms. Prajna Anand
Country	India
Abstract	Somewhere underneath all the excitement around self-driving cars is an assumption most people never think to question: that the sensors feeding these vehicles actually see the world as it is. For LiDAR, that assumption has held up well enough in ordinary driving conditions. But a growing pile of research — going back roughly five years and picking up pace recently — has started to poke serious holes in it. Researchers have inserted phantom objects into point clouds, erased genuine obstacles from them, and done both things using nothing more elaborate than a mirror. This paper looks at what that body of work actually shows. It organises the main attack types — spoofing, relay attacks, jamming, and adversarial surfaces — examines the evidence behind each, and then asks honestly what the proposed defences are actually worth. The short answer, unfortunately, is that no individual defence is anywhere near sufficient on its own, and that the most realistic path to robustness involves stacking multiple independent protective layers rather than betting everything on any single mechanism.
Keywords	Autonomous vehicles, LiDAR, adversarial attacks, point cloud, sensor spoofing, sensor fusion, deep learning, physical security, perception systems.
Field	Engineering
Published In	Volume 8, Issue 3, May-June 2026
Published On	2026-05-11

About IJFMR Fees & Payment Current Issue Publication Archive	Submit Research Paper Track Submission Status Publication Guidelines Publication Ethics Peer Review & Plagiarism	Join as a Reviewer Editors & Reviewers Reviewer Referral Program Get Reviewer Membership Certi.	Website/Journal Policies Usage Policy Content Policies Privacy Policy

Contact Us		+91-9687-828-838	editor@ijfmr.com

International Journal For Multidisciplinary Research