GroundLoc: Efficient Large-Scale Outdoor LiDAR-Only Localization

2025-10-28

AI Summary

To address efficient large-scale outdoor localization for LiDAR-only mobile robots, this paper proposes a lightweight, pure-LiDAR localization system. Methodologically, it projects point clouds onto a bird’s-eye view (BEV), extracts and matches ground-region features—either R2D2 or SIFT—and performs fast registration against a sparse 2D grid-based prior map. The map is highly compressed to only 4 MB/km² and supports multi-LiDAR model compatibility. Its key innovation lies in being the first to integrate BEV-based ground-feature matching with hybrid learned/traditional feature descriptors for large-scale localization. Evaluated on SemanticKITTI and HeLiPR benchmarks, the system achieves superior performance: average trajectory error <50 cm on Ouster OS2-128 sequences, satisfying real-time online localization requirements.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Natural Language Processing: Language Grounding & Multi-modal NLP
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

In this letter, we introduce GroundLoc, a LiDAR-only localization pipeline designed to localize a mobile robot in large-scale outdoor environments using prior maps. GroundLoc employs a Bird's-Eye View (BEV) image projection focusing on the perceived ground area and utilizes the place recognition network R2D2, or alternatively, the non-learning approach Scale-Invariant Feature Transform (SIFT), to identify and select keypoints for BEV image map registration. Our results demonstrate that GroundLoc outperforms state-of-the-art methods on the SemanticKITTI and HeLiPR datasets across various sensors. In the multi-session localization evaluation, GroundLoc reaches an Average Trajectory Error (ATE) well below 50 cm on all Ouster OS2 128 sequences while meeting online runtime requirements. The system supports various sensor models, as evidenced by evaluations conducted with Velodyne HDL-64E, Ouster OS2 128, Aeva Aeries II, and Livox Avia sensors. The prior maps are stored as 2D raster image maps, which can be created from a single drive and require only 4 MB of storage per square kilometer. The source code is available at https://github.com/dcmlr/groundloc.

Problem

• Localizing mobile robots in large-scale outdoor environments using LiDAR data
• Achieving efficient map registration with minimal storage requirements per kilometer
• Supporting multiple sensor models while maintaining online runtime performance

Innovation

• LiDAR-only localization using prior maps
• BEV image projection with ground focus
• Keypoint selection via R2D2 or SIFT