Kimera-Multi

Kimera-Multi is a multi-robot system that was first presented in (Chang et al., 2021). It is built to be robust to incorrect inter and intra-robot loop closures resulting from perceptual aliasing, fully distributed and only relies on local (peer-to-peer) communication to achieve distributed localization and mapping, and builds a globally consistent metric-semantic 3D mesh model of the environment in real-time. Kimera-Multi is field tested in many indoor and outdoor environments. We not only open-source the code, but also release a few large-scale hybrid indoor/outdoor datasets collected on the MIT campus.

Please check out our repo.

For a detailed description of the project, please refer to (Tian et al., 2022). (Tian et al., 2023) provide a deeper dive into the system and real world experiments.

  1. Tian, Y., Chang, Y., Quang, L., Schang, A., Nieto-Granda, C., How, J. P., & Carlone, L. (2023). Resilient and Distributed Multi-Robot Visual SLAM: Datasets, Experiments, and Lessons Learned. IEEE/RSJ Intl. Conf. on Intelligent Robots and Systems (IROS).
  2. Tian, Y., Chang, Y., Arias, F. H., Nieto-Granda, C., How, J. P., & Carlone, L. (2022). Kimera-Multi: Robust, Distributed, Dense Metric-Semantic SLAM for Multi-Robot Systems. IEEE Trans. Robotics.
  3. Chang, Y., Tian, Y., How, J. P., & Carlone, L. (2021). Kimera-Multi: a System for Distributed Multi-Robot Metric-Semantic Simultaneous Localization and Mapping. IEEE Intl. Conf. on Robotics and Automation (ICRA).