README.md

metadata

license: cc-by-nc-4.0
task_categories:
  - robotics

MCAP Robotics Dataset Collection

This repository contains a comprehensive collection of robotics datasets converted to the MCAP format, specifically designed for SLAM (Simultaneous Localization and Mapping) and multi-modal sensor fusion research.

Dataset Overview

Our collection includes three major robotics datasets:

🚀 FAST-LIVO Dataset

• Focus: LiDAR-Inertial-Visual Odometry systems
• Scenarios: University campus, LiDAR degenerate cases, visual challenges
• Applications: Algorithm robustness testing, multi-modal fusion

🌈 R3LIVE Dataset

• Focus: Real-time Radiance Reconstruction and RGB-colored mapping
• Scenarios: HKU campus sequences, indoor buildings, outdoor environments
• Applications: Photorealistic 3D reconstruction, real-time dense mapping

🛰️ MARS-LVIG Dataset

• Focus: Multi-sensor SLAM with LiDAR-Visual-Inertial-GNSS fusion
• Scenarios: Diverse outdoor environments with GNSS integration
• Applications: Robust state estimation, multi-sensor fusion, GNSS-aided navigation

Key Features

• Standardized Format: All datasets converted to MCAP for consistent processing
• Multi-modal Data: Synchronized LiDAR, IMU, and camera measurements
• Diverse Scenarios: Indoor, outdoor, and challenging environmental conditions
• Research Ready: Optimized for SLAM algorithm development and evaluation

File Format

• .mcap: Main data files containing synchronized sensor measurements (LiDAR, IMU, camera data)
• metadata.yaml: Configuration and metadata files describing recording parameters and sensor specifications

Usage

These MCAP files can be processed using standard robotics tools and libraries that support the MCAP format for multi-modal sensor data analysis and SLAM algorithm evaluation.

Recommended Tools

• ros2mcap - ROS 1 MCAP support
• MCAP CLI - Command-line tools for MCAP manipulation
• Foxglove Studio - Visualization and analysis platform
• ROS 2 - Robot Operating System with MCAP support

Getting Started

1. Browse datasets: Explore individual dataset folders for specific scenarios
2. Read metadata: Check metadata.yaml files for sensor specifications
3. Load MCAP files: Use your preferred robotics framework or MCAP tools
4. Analyze data: Process synchronized sensor streams for your research

Applications

This dataset collection is ideal for:

• SLAM algorithm development and benchmarking
• Multi-sensor fusion research
• Real-time mapping and localization studies
• Sensor calibration and synchronization validation
• Computer vision and robotics education

License

This dataset collection is licensed under CC-BY-NC-4.0 (Creative Commons Attribution-NonCommercial 4.0 International). This means you can use, share, and adapt the material for non-commercial purposes with proper attribution.

Contributing

We welcome contributions of additional datasets or improvements to existing ones. Please ensure all data follows our MCAP formatting standards and includes appropriate metadata.