# SJTU-ViSYS/M2DGR

M2DGR： a Multi-modal and Multi-scenario Dataset for Ground Robots(RA-L2021 & ICRA2022)

Repository: https://github.com/SJTU-ViSYS/M2DGR
Canonical: https://ross.abutalabs.com/products/m2dgr
License: MIT
License Family: permissive
Topics: slam, dataset, robotics
Last push: 2026-06-15T14:06:06+00:00

## Health v2 (maintenance only)
Score: 71/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 87, release rhythm 35, longevity 100
- inputs: {"age_days": 1829, "days_push": 79, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1194, forks 130 (observed 2026-08-28T04:03:56.733316+00:00)

## What it is
M2DGR is a multi-modal, multi-scenario SLAM dataset collected with ground robots, providing synchronized vision, lidar, IMU, GNSS, event, and thermal-infrared data across indoor and outdoor environments. It serves as a benchmark for evaluating state-of-the-art SLAM and localization algorithms.

## Use cases
- benchmark a new SLAM algorithm against LIO-SAM and ORB-SLAM3
- download multi-sensor robot datasets for lidar-inertial-visual odometry research
- evaluate GNSS-degraded indoor localization with lidar and IMU data
- test event camera or thermal-infrared SLAM methods
- find challenging sequences with lifts, streets, and halls for robot navigation
- compare ground robot localization systems on a public benchmark

## When to choose
- you need real-world multi-modal sensor data for ground robot SLAM research
- you want a recognized benchmark to validate a newly proposed SLAM system
- you need sequences with degenerate motion or GNSS-denied scenarios

## When to avoid
- you need aerial or underwater robot data
- you want a ready-to-run SLAM implementation rather than raw sensor data
- you need synthetic or simulated environments

## Facets
- artifact type: dataset
- maturity: stable
- function: simulation, benchmarking, computer-vision
- domain: robotics, autonomous-vehicles, computer-vision
- platform: cross-platform
- tags: slam, multi-modal, lidar, imu, gnss, ground-robots, benchmark, localization, mapping, algorithms, linux, ros

## Member repositories
- SJTU-ViSYS/M2DGR (main) score 71

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:56.733316+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:22:08.857290+00:00, confidence not recorded.
  - readme: https://github.com/SJTU-ViSYS/M2DGR (fetched 2026-08-28T04:03:56.733316+00:00, sha a1ec9a5e0813)
- Data as of 2026-08-30T08:39:29.467469+00:00.