This organisation contains resources around CBGL, the Cumulative-Absolute-Error-per-Ray-based Global Localisation method. CBGL may localise a 2D LIDAR sensor in a map of its environment given (i) a single LIDAR measurement and (ii) the map. CBGL is fast, and robust to sensor FOV, sensor noise, environment arbitrariness and repetition, and map-to-environment discrepancies.

• cbgl contains the roscpp source code and docker-related files for building and deploying it
• IROS-2024-paper contains the source tex code of the paper titled "CBGL: Fast Monte Carlo Passive Global Localisation of 2D LIDAR Sensor", presented at the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems, held at Abu Dhabi, UAE
• IROS-2024-presentations-resources contains material related to the presentation of CBGL, namely the poster and pitch required by IROS 2024, CBGL in graphical abstract form, its full presentation, and video illustration of its methodology and key properties
• ros-experiments-src contains the source C++ code of CBGL as well as that of state-of-the-art methods that was used to test the methods' performance in global localisation with ROS, featured in section V "Experimental Evaluation" of the paper
• sim-experiments-src contains the source C++ code of CBGL that was used to test its performance in global localisation with respect to varying scan--to--scan-matching methods and environment area, featured in sections V "Experimental Evaluation" and VI "Characterisation and Limitations" of the paper
• experiments-logs contains raw data used to illustrate and support CBGL's contributions in sections V "Experimental Evaluation" and VI "Characterisation and Limitations" of the paper