2D LiDAR Indoor Localization Dataset for Line-Relation-Based Mapping and Localization

Description

This dataset contains 2D LiDAR data for indoor localization and lightweight map construction experiments based on line-segment features and line-relation descriptors. The data were used to evaluate the RoLiL algorithm, a line-relation-based localization method for indoor mobile robots. The dataset includes LiDAR scans collected in feature-rich indoor environments and geometrically degenerate environments. This dataset can be used for line-segment extraction, relation-based feature matching, lightweight line-feature map construction, localization accuracy evaluation, occlusion robustness analysis, and fusion experiments with Cartographer. It provides both raw and preprocessed LiDAR measurements, as well as experimental scene data. The dataset is intended to support reproducible research on 2D LiDAR-based indoor localization, feature-based map construction, robust line-segment fitting, and relation-based localization for mobile robots.

Steps to reproduce

The real point cloud data were obtained using a Pepperl+Fuchs OMD30M-R2000 LiDAR scanner, and the simulated data were generated in MATLAB.

Institutions

• Chongqing University

  Chongqing, Chongqing

Categories

Licence