SLACK: Attacking LiDAR-based SLAM with Adversarial Point Injections
This addresses a critical security problem for autonomous vehicles by introducing a novel attack method on LiDAR-based SLAM, which is incremental as it builds on adversarial techniques but targets an unexplored area.
The paper tackles the vulnerability of LiDAR-based SLAM in autonomous vehicles to adversarial point injections, proposing SLACK, a deep generative adversarial model that achieves superior performance in degrading navigation and map quality while maintaining accurate scan quality on KITTI and CARLA-64 datasets.
The widespread adoption of learning-based methods for the LiDAR makes autonomous vehicles vulnerable to adversarial attacks through adversarial \textit{point injections (PiJ)}. It poses serious security challenges for navigation and map generation. Despite its critical nature, no major work exists that studies learning-based attacks on LiDAR-based SLAM. Our work proposes SLACK, an end-to-end deep generative adversarial model to attack LiDAR scans with several point injections without deteriorating LiDAR quality. To facilitate SLACK, we design a novel yet simple autoencoder that augments contrastive learning with segmentation-based attention for precise reconstructions. SLACK demonstrates superior performance on the task of \textit{point injections (PiJ)} compared to the best baselines on KITTI and CARLA-64 dataset while maintaining accurate scan quality. We qualitatively and quantitatively demonstrate PiJ attacks using a fraction of LiDAR points. It severely degrades navigation and map quality without deteriorating the LiDAR scan quality.