DRIP: Discriminative Rotation-Invariant Pole Landmark Descriptor for 3D LiDAR Localization

TL;DR

The paper tackles 3D LiDAR self-localization by leveraging pole-like landmarks with discriminative ROI context. It introduces a discriminative ROI pole descriptor, a rotation-invariant CNN (RIConv++) trained in an unsupervised fashion with Chamfer Distance, and a learned pole dictionary that converts descriptors into compact pole words for fast Monte Carlo localization. These pole words enable a pole-word consistency check and integrate with traditional landmark matching to improve accuracy in dense pole environments. Evaluations on the NCLT dataset show improved localization performance over the prior baseline, with robustness across cross-domain and seasonal variations, highlighting practical scalability for large-scale urban sensing.

Abstract

In 3D LiDAR-based robot self-localization, pole-like landmarks are gaining popularity as lightweight and discriminative landmarks. This work introduces a novel approach called "discriminative rotation-invariant poles," which enhances the discriminability of pole-like landmarks while maintaining their lightweight nature. Unlike conventional methods that model a pole landmark as a 3D line segment perpendicular to the ground, we propose a simple yet powerful approach that includes not only the line segment's main body but also its surrounding local region of interest (ROI) as part of the pole landmark. Specifically, we describe the appearance, geometry, and semantic features within this ROI to improve the discriminability of the pole landmark. Since such pole landmarks are no longer rotation-invariant, we introduce a novel rotation-invariant convolutional neural network that automatically and efficiently extracts rotation-invariant features from input point clouds for recognition. Furthermore, we train a pole dictionary through unsupervised learning and use it to compress poles into compact pole words, thereby significantly reducing real-time costs while maintaining optimal self-localization performance. Monte Carlo localization experiments using publicly available NCLT dataset demonstrate that the proposed method improves a state-of-the-art pole-based localization framework.