RoboLoc: A Benchmark Dataset for Point Place Recognition and Localization in Indoor-Outdoor Integrated Environments

2025-11-30

AI Summary

Existing LiDAR datasets predominantly focus on outdoor scenes, lacking complex environments featuring seamless indoor–outdoor transitions, significant elevation changes, and inter-floor navigation—thus hindering robust GPS-denied localization and place recognition. To address this, we introduce RoboLoc, the first large-scale point cloud benchmark explicitly designed for indoor–outdoor integrated scenarios, incorporating real robot trajectories, multi-scale structural transitions, and continuous domain shifts. RoboLoc is the first to systematically formalize and evaluate cross-domain indoor–outdoor localization and floor-switching tasks, filling a critical gap in LiDAR-based localization benchmarks under complex multi-domain transitions. It supports multiple representations—including raw point clouds, voxelized inputs, and bird’s-eye view (BEV) projections—and validates cross-domain generalization across several state-of-the-art models. Experiments demonstrate that RoboLoc significantly enhances both the comparability and challenge level of multi-domain localization algorithms, establishing a unified, practical evaluation platform for cross-domain localization research.

Technology Category

• Intelligent Robots: Localization, Mapping, and Navigation
• Natural Language Processing: Safety and Robustness
• Application Domains: Mobility, Driving & Flight

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Location- and context-aware Web and WoT applications and services
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Security and Privacy: Large-scale security measurements

Abstract

Robust place recognition is essential for reliable localization in robotics, particularly in complex environments with fre- quent indoor-outdoor transitions. However, existing LiDAR-based datasets often focus on outdoor scenarios and lack seamless domain shifts. In this paper, we propose RoboLoc, a benchmark dataset designed for GPS-free place recognition in indoor-outdoor environments with floor transitions. RoboLoc features real-world robot trajectories, diverse elevation profiles, and transitions between structured indoor and unstructured outdoor domains. We benchmark a variety of state-of-the-art models, point-based, voxel-based, and BEV-based architectures, highlighting their generalizability domain shifts. RoboLoc provides a realistic testbed for developing multi-domain localization systems in robotics and autonomous navigation

Problem

• Benchmark dataset for indoor-outdoor place recognition
• Addresses lack of seamless domain shift datasets
• Evaluates model generalizability across environmental transitions

Innovation

• RoboLoc dataset for indoor-outdoor place recognition
• Benchmarking point, voxel, and BEV-based models
• Real-world trajectories with elevation and domain shifts