Aerospace Controls Laboratory

Resource-aware collaborative SLAM

Yulun Tian, Kasra Khosoussi, Matthew Giamou


Inter-robot loop closure detection, e.g., for collaborative simultaneous localization and mapping (CSLAM), is a fundamental capability for many multirobot applications in GPS-denied regimes. In real-world scenarios, this process is resource-intensive as it involves exchanging many observations and verifying a large number of potential matches. This poses severe challenges especially for small-size and low-cost robots with various operational and resource constraints that limit, e.g., energy consumption, communication bandwidth, and computation capacity.

This project aims to develop resource-aware algorithms for distributed inter-robot loop closure detection and measurement selection. In particular, we seek to maximize a monotone submodular performance metric without exceeding computation and communication budgets. This problem is in general NP-hard. We have developed efficient approximation algorithms with provable performance guarantees that handle such resource constraints.


Related Publications

  • Tian, Y., Khosoussi, K., and How, J. P., “Resource-Aware Algorithms for Distributed Loop Closure Detection with Provable Performance Guarantees,” Proceedings of the Workshop on the Algorithmic Foundations of Robotics (WAFR), 2018.
  • Tian, Y., Khosoussi, K., Giamou, M., How, J. P., and Kelly, J., “Near-Optimal Budgeted Data Exchange for Distributed Loop Closure Detection,” Robotics: Science and Systems, 2018.
  • Giamou, M., Khosoussi, K., and How, J. P., “Talk Resource-Efficiently to Me: Optimal Communication Planning for Distributed Loop Closure Detection,” IEEE International Conference on Robotics and Automation (ICRA), 2018.
  • Khosoussi, K., Giamou, M., Sukhatme, G., Huang, S., Dissanayake, G., and How, J. P., “Reliable Graphs for SLAM,” International Journal of Robotics Research (IJRR), 2019.