Hybrid Belief POMDP Planning

Data Association Aware Belief Space Planning

Real-world problems often require reasoning about hybrid beliefs, over both discrete and continuous random variables. As an example, one might consider matching images from two different but similar in appearance places (possibly observed by different robots), or attempting to recognize an object that is similar in appearance, from the current viewpoint, to another object. Both cases are examples of ambiguous situations, where naive and straightforward approaches are likely to yield incorrect results, i.e. mistakenly considering the two places to be the same place, and incorrectly associating the observed object. These and numerous other applications necessitate reasoning about hybird beliefs, where the discrete variables correspond to hypotheses (e.g. data association and/or classification hypotheses). Yet, such a setting has hardly been investigated in the context of planning. Moreover, existing online Partially Observable Markov Decision Processes (POMDPs) solvers do not support hybrid beliefs directly. In particular, these solvers do not address the added computational burden due to an increasing number of hypotheses with the planning horizon, which can grow exponentially. In this research project we develop online hybrid belief POMDP planning approaches that address these challenges.

Related Publications:

Journal Articles

  1. M. Barenboim, M. Shienman, and V. Indelman, “Monte Carlo Planning in Hybrid Belief POMDPs,” IEEE Robotics and Automation Letters (RA-L), no. 8, Aug. 2023.
    Barenboim23ral.pdf Barenboim23ral.slides DOI: 10.1109/LRA.2023.3282773 Barenboim23ral.supplementary
  2. M. Barenboim, I. Lev-Yehudi, and V. Indelman, “Data Association Aware POMDP Planning with Hypothesis Pruning Performance Guarantees,” IEEE Robotics and Automation Letters (RA-L), no. 10, Oct. 2023.
    Barenboim23ral2.pdf Barenboim23ral2.slides DOI: 10.1109/LRA.2023.3311205 Barenboim23ral2.supplementary
  3. O. Shelly and V. Indelman, “Hypotheses Disambiguation in Retrospective,” IEEE Robotics and Automation Letters (RA-L), no. 2, Apr. 2022.
    Shelly22ral.pdf DOI: 10.1109/LRA.2022.3143298 Shelly22ral.supplementary Shelly22ral.poster
  4. V. Tchuiev and V. Indelman, “Distributed Consistent Multi-Robot Semantic Localization and Mapping,” IEEE Robotics and Automation Letters (RA-L), no. 3, Jul. 2020.
    Tchuiev20ral.pdf DOI: 10.1109/LRA.2020.3003275 Tchuiev20ral.supplementary Tchuiev20ral.video
  5. S. Pathak, A. Thomas, and V. Indelman, “A Unified Framework for Data Association Aware Belief Space Planning and Perception,” International Journal of Robotics Research (IJRR), no. 2-3, 2018.
    Pathak18ijrr.pdf DOI: 10.1177/0278364918759606

Theses

  1. M. Barenboim, “Simplified POMDP Algorithms with Performance Guarantees,” PhD thesis, Technion - Israel Institute of Technology, 2024.
    Barenboim24thesis.pdf Barenboim24thesis.slides Barenboim24thesis.video
  2. O. Shelly, “Hypotheses disambiguation in retrospective for robust perception in ambiguous environments,” Master's thesis, Technion - Israel Institute of Technology, 2022.
    Shelly22thesis.pdf Shelly22thesis.slides Shelly22thesis.video
  3. A. Thomas, “Incorporating Data Association Within Belief Space Planning For Robust Autonomous Navigation,” Master's thesis, Technion - Israel Institute of Technology, 2017.
    Thomas17thesis.pdf Thomas17thesis.slides

Book Chapters

  1. M. Shienman and V. Indelman, “Nonmyopic Distilled Data Association Belief Space Planning Under Budget Constraints,” in Robotics Research, Springer, 2023.
    Shienman23chapter.pdf DOI: 10.1007/978-3-031-25555-7_8

Conference Articles

  1. M. Shienman and V. Indelman, “D2A-BSP: Distilled Data Association Belief Space Planning with Performance Guarantees Under Budget Constraints,” in IEEE International Conference on Robotics and Automation (ICRA), *Outstanding Paper Award Finalist*, May 2022.
    Shienman22icra.pdf Shienman22icra.supplementary Shienman22icra.poster
  2. M. Shienman and V. Indelman, “Nonmyopic Distilled Data Association Belief Space Planning Under Budget Constraints,” in International Symposium on Robotics Research (ISRR), Sep. 2022.
    Shienman22isrr.pdf Shienman22isrr.supplementary
  3. V. Tchuiev, Y. Feldman, and V. Indelman, “Data Association Aware Semantic Mapping and Localization via a Viewpoint Dependent Classifier Model,” in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Nov. 2019.
    Tchuiev19iros.pdf Tchuiev19iros.slides
  4. S. Pathak, A. Thomas, and V. Indelman, “Nonmyopic Data Association Aware Belief Space Planning for Robust Active Perception,” in IEEE International Conference on Robotics and Automation (ICRA), May 2017.
    Pathak17icra.pdf Pathak17icra.slides
  5. S. Pathak, A. Thomas, A. Feniger, and V. Indelman, “Towards Data Association Aware Belief Space Planning for Robust Active Perception,” in AI for Long-term Autonomy, workshop in conjunction with IEEE International Conference on Robotics and Automation (ICRA), May 2016.
    Pathak16icra_ws.pdf Pathak16icra_ws.poster
  6. S. Pathak, A. Thomas, A. Feniger, and V. Indelman, “DA-BSP: Towards Data Association Aware Belief Space Planning for Robust Active Perception,” in European Conference on Artificial Intelligence (ECAI), accepted for short paper presentation, Sep. 2016.
    Pathak16ecai.pdf Pathak16ecai.poster
  7. S. Pathak, S. Soudjani, V. Indelman, and A. Abate, “Formal and Data-association aware Belief Space Planning,” in Eighth European Starting AI Researcher Symposium (STAIRS), co-located with European Conference on Artificial Intelligence (ECAI), Sep. 2016.
    Pathak16stairs.pdf Pathak16stairs.slides

Technical Reports

  1. S. Pathak, A. Thomas, A. Feniger, and V. Indelman, “Robust Active Perception via Data-association aware Belief Space Planning,” 2016.
    arXiv: http://arxiv.org/abs/1606.05124