Incremental Belief Space Planning

Decision-making is a fundamental problem in robotics and Artificial Intelligence. Under belief space planning (BSP), in a partially observable setting, it involves calculating the expected cumulative belief-dependent re- ward (cost) concerning all future measurements. Since solving this general problem quickly becomes intractable, state-of-the-art approaches turn to approximations while still calculating planning sessions from scratch. In this research, we investigate a shift from the theoretical formulation, incremental eXpectation BSP (iX-BSP), based on our key insight that calculations across planning sessions are similar and thus can be appropriately re-used. We demonstrate how iX-BSP could benefit existing approximations of the general problem. Introducing iX-BSP and iML-BSP, which re-use calculations across planning sessions for an open-loop sampling-based BSP estimator and the common Maximum-Likelihood assumption respectively.

Related Publications:

Journal Articles

  1. E. Farhi and V. Indelman, “Bayesian Incremental Inference Update by Re-using Calculations from Belief Space Planning: A New Paradigm,” Autonomous Robots, Aug. 2022.
    Farhi22arj.pdf DOI: 10.1007/s10514-022-10045-w

Technical Reports

  1. M. Novitsky, M. Barenboim, and V. Indelman, “Previous Knowledge Utilization In Online Anytime Belief Space Planning,” 2024.
    arXiv: https://arxiv.org/abs/2412.13128
  2. E. Farhi and V. Indelman, “iX-BSP: Incremental Belief Space Planning,” 2021.
    arXiv: https://arxiv.org/pdf/2102.09539

Theses

  1. E. Farhi, “Joint Incremental Inference and Belief Space Planning for Online Operations of Autonomous systems,” PhD thesis, Technion - Israel Institute of Technology, 2021.
    Farhi21thesis.pdf Farhi21thesis.slides Farhi21thesis.video

Conference Articles

  1. E. Farhi and V. Indelman, “iX-BSP: Belief Space Planning through Incremental Expectation,” in IEEE International Conference on Robotics and Automation (ICRA), May 2019.
    Farhi19icra.pdf Farhi19icra.supplementary Farhi19icra.poster
  2. K. Elimelech and V. Indelman, “PIVOT: Predictive Incremental Variable Ordering Tactic for Efficient Belief Space Planning,” in Workshop on Toward Online Optimal Control of Dynamic Robots, in conjunction with the IEEE International Conference on Robotics and Automation (ICRA), May 2019.
    Elimelech19icra_ws.poster
  3. E. Farhi and V. Indelman, “Tear Down that Wall: Calculation Reuse Across Inference and Belief Space Planning,” in Workshop on Toward Online Optimal Control of Dynamic Robots, in conjunction with the IEEE International Conference on Robotics and Automation (ICRA), May 2019.
    Farhi19icra_ws.pdf Farhi19icra_ws.poster
  4. E. Farhi and V. Indelman, “Towards Efficient Inference Update through Planning via JIP - Joint Inference and Belief Space Planning,” in IEEE International Conference on Robotics and Automation (ICRA), May 2017.
    Farhi17icra.pdf Farhi17icra.slides