NeuROAM

NeuROAM is an effort initiated by the Institute for Experiential Robotics at Northeastern University to create a large-scale, multi-robot dataset for advancing the state-of-the-art in mapping and SLAM systems. With the collaborative efforts of multiple labs, PIs, postdocs, phd students, graduate and undergraduate students, we are collecting a diverse dataset using a fleet of 5 diverse robots - Unitree Go2, Unitree GO2W, Boston Dynamics Spot, Agile-X Hunter and Agile-X Scout.

Each robot is equipped with the same sensor payload which includes a lidar, RGB Stereo Camera, IMU, GPS and Mesh Radios. The dataset consists of synchronized data all robots operating in indoor and outdoor environments.

Key Features

• Multi-Robot Coordination: The dataset captures the dynamics of multiple robots working together in a shared environment, providing insights into coordination and communication strategies.
• Diverse Environments: Data is collected in a variety of settings, including indoor office spaces, outdoor terrains, and mixed environments, ensuring robustness and versatility.
• Rich Sensor Suite: Each robot is equipped with a comprehensive set of sensors, including LiDAR, RGB cameras, IMUs, and GPS, enabling multi-modal data analysis.
• High-Quality Annotations: The dataset includes detailed annotations for key features, obstacles, and landmarks, facilitating advanced research in perception and mapping.
• Open Access: The NeuROAM dataset is made publicly available to the research community, encouraging collaboration and innovation in the field of robotics.

Applications

The NeuROAM dataset is designed to support a wide range of research areas, including:

• Simultaneous Localization and Mapping (SLAM): Developing and testing SLAM algorithms in multi-robot scenarios.
• Multi-Agent Systems: Exploring strategies for coordination, communication, and task allocation among multiple robots.
• Perception and Sensor Fusion: Enhancing perception capabilities through the integration of data from diverse sensors.
• Autonomous Navigation: Improving navigation algorithms for robots operating in complex environments.
• Robotics Research and Development: Providing a valuable resource for researchers and developers working on cutting-edge robotics technologies.

Dataset Access

The NeuROAM dataset is available for download here. For more information on the dataset structure, sensor specifications, and usage guidelines, please refer to the NeuROAM Documentation.

Acknowledgements

We would like to thank all the members of the Institute for Experiential Robotics and our collaborators for their contributions to the NeuROAM project. This work was supported by [funding sources, if any].

For more details, visit the NeuROAM Project Page.