Skip to content

Indoor Benchmark of 3D LiDAR SLAM at iilab - Industry and Innovation Laboratory

Abstract

This paper presents the IILABS 3D dataset, a novel and publicly available resource designed to address current limitations in indoor benchmarking of 3D LiDAR-based Simultaneous Localization and Mapping (SLAM) algorithms. Existing SLAM datasets often focus on outdoor environments, rely on a single type of LiDAR sensor, or lack ground-truth data suitable for evaluating diverse indoor conditions. IILABS 3D fills this gap by providing a sensor-rich, indoor-exclusive dataset recorded in a controlled laboratory environment using a wheeled mobile robot platform. It includes four heterogeneous 3D LiDAR sensors – Velodyne VLP-16, Ouster OS1-64, RoboSense RS-Helios-5515, and Livox Mid-360 – featuring both mechanical spinning and non-repetitive scanning patterns, as well as an IMU and wheel odometry. The dataset also features calibration sequences, challenging benchmark trajectories, and high-precision ground-truth poses captured with a Motion Capture (MoCap) system. By combining diverse sensor technologies, extensive calibration data, and carefully designed indoor scenarios, IILABS 3D enables more comprehensive and reproducible evaluation of LiDAR-based SLAM algorithms, fostering innovation in autonomous navigation within complex indoor environments. The dataset information and associated tools are available at project webpage jorgedfr.github.io/3d_lidar_slam_benchmark_at_iilab.

Keywords: dataset, ground mobile robot, indoor environment, Light Detection and Ranging (LiDAR), Simultaneous Localization and Mapping (SLAM).

This repository hosts the complete documentation and supporting resources for the project "Indoor Benchmark of 3D LiDAR SLAM at iilab - Industry and Innovation Laboratory". The project evaluates state-of-the-art 3D LiDAR SLAM algorithms using data captured in indoor environments. To this end, the study introduces the novel IILABS 3D dataset, which contains data from four diverse 3D LiDAR sensors (Velodyne VLP-16, Ouster OS1-64, RoboSense RS-Helios-5515, and Livox Mid-360), complemented by measurements from an IMU and wheel odometry.

In parallel with the dataset, the project presents a detailed benchmark analysis of nine leading SLAM algorithms. By comparing algorithm-generated odometry against high-accuracy ground-truth data using metrics such as ATE, RTE, and RRE, the study provides valuable insights into the performance, limitations, and integration of these algorithms in indoor environments.

The primary aim of this website and GitHub repository is to support researchers in developing, benchmarking, and applying 3D LiDAR-based SLAM solutions in indoor settings. As a result, the website includes the following information:

  • Usage: Step-by-step instructions for replicating the benchmark study using the provided scripts and dataset.
  • Dataset: A detailed overview of the IILABS 3D dataset, its structure, and key characteristics.
  • Sensors: Specifications and technical details of the sensors used during data collection.
  • Benchmark: A summary of the benchmark scripts, experimental setup, and key results from the analysis.

Lastly, this work is within the scope of the Mobile Robotics Development Team (MRDT) in the national project GreenAuto: Green innovation for the Automotive Industry. MRDT team is a Research & Development (R&D) team from the CRIIS - Centre for Robotics in Industry and Intelligent Systems at the iilab - Industry and Innovation Laboratory.

Videos

Contacts

If you have any questions or you want to know more about this work, please contact one of the following contributors:

Institutions

  • INESC TEC Logo
  • FEUP Logo

Funding

GreenAuto: Green innovation for the Automotive Industry

Citations

Article

TBC

Dataset

Plain Text

J.D. Ribeiro, R.B. Sousa, J.G. Martins, A.S. Aguiar, F.N. Santos and H.M. Sobreira, "IILABS 3D: iilab Indoor LiDAR-based SLAM Dataset", [Dataset], INESC TEC, 2025, doi: 10.25747/VHNJ-WM80.

BibTex

@MISC{ribeiro:2025:iilabs3d:dataset,
  author    = {J.D. Ribeiro and R.B. Sousa and J.G. Martins and A.S. Aguiar and F.N. Santos and H.M. Sobreira},
  title     = {{IILABS 3D}: iilab {I}ndoor {LiDAR}-based {SLAM} {D}ataset},
  year      = {2025},
  publisher = {INESC TEC},
  doi       = {10.25747/VHNJ-WM80},
  note      = {[Dataset]},}