Recent work by members of the DCIST alliance presents Kimera-Multi, a multi-robot system that (i) is robust and capable of identifying and rejecting incorrect inter and intra-robot loop closures resulting from perceptual aliasing, (ii) is fully distributed and only relies on local (peer-to-peer) communication to achieve distributed localization and mapping, and (iii) builds a globally consistent metric-semantic 3D mesh model of the environment in real-time, where faces of the mesh are annotated with semantic labels. Kimera-Multi is implemented by a team of robots equipped with visual-inertial sensors. Each robot builds a local trajectory estimate and a local mesh using Kimera, an open-source library developed under DCIST: https://github.com/MIT-SPARK/Kimera. When communication is available, robots initiate a distributed place recognition procedure to detect inter-robot loop closures. Subsequently, robots perform distributed trajectory estimation using distributed pose graph optimization (PGO), a technology recently developed under DCIST (https://github.com/mit-acl/dpgo). Distributed PGO, in combination with a newly developed distributed graduated non-convexity technique, allows the robots to accurately estimate their trajectories by leveraging inter-robot loop closures while being robust to outliers. Finally, each robot uses its improved trajectory estimate to correct the local mesh using mesh deformation techniques.
Kimera-Multi has been tested in photo-realistic simulations, SLAM benchmarking datasets, and challenging outdoor datasets collected using ground robots. Kimera-Multi has been demonstrated to (i) outperform the state of the art in terms of robustness and accuracy, (ii) achieve estimation errors comparable to a centralized SLAM system while being fully distributed, (iii) is parsimonious in terms of communication bandwidth, (iv) produces accurate metric-semantic 3D meshes, and (v) is modular and can be also used for standard 3D reconstruction (without semantic labels) or for trajectory estimation (without reconstructing a 3D mesh).
Points of Contact: Luca Carlone (PI), Jonathan How (PI), Yun Chang, Yulun Tian
Chang, Yun, et al. “Kimera-Multi: A System for Distributed Multi-Robot Metric-Semantic Simultaneous Localization and Mapping.” 2021 IEEE International Conference on Robotics and Automation (ICRA), IEEE, 2021, pp. 11210–18. DOI.org (Crossref), https://doi.org/10.1109/ICRA48506.2021.9561090.
Tian, Yulun, et al. “Kimera-Multi: Robust, Distributed, Dense Metric-Semantic SLAM for Multi-Robot Systems.” IEEE Transactions on Robotics (T-RO), 2021 (to appear). arXiv.org, http://arxiv.org/abs/2106.14386.