EUVIMEDEUVIMED entdecken
Crossref

PhyPlan: Learning to Plan Tasks with Generalizable and Rapid Physical Reasoning for Embodied Manipulation

Ankit Kanwar, Hartej Soin, Abhinav Barnawal, Mudit Chopra, Harshil Vagadia, Tamajit Banerjee, Shreshth Tuli, Rohan Paul, Souvik Chakraborty

Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (22) · 18360-18369

DOI: 10.1609/aaai.v40i22.38900 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Ankit Kanwar und 8 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Given the task of landing a ball in a goal region beyond direct reach, humans can often throw, slide, or rebound objects against the wall to attain the goal. Enabling robots to replicate such reasoning is non-trivial as it requires multi-step planning and involves a mixture of discrete and continuous action spaces, a sparse and sensitive reward structure, computationally expensive simulations, and an incomplete understanding of the environment's physics. We present PhyPlan, a physics-informed and adaptable planning framework for efficient multi-step physical reasoning. At its core, PhyPlan comprises of Generative Flow Networks (GFlowNets) and Monte Carlo Tree Search (MCTS) to explore and evaluate sequences of object interactions. GFlowNets sample discrete action sequences in proportion to their associated reward, enabling broad and reward-driven exploration of the discrete planning space. MCTS complements this by adaptively balancing the use of a fast but approximate pre-trained physics-informed dynamics predictor and costly but accurate environment rollouts, ensuring both speed and precision in planning. The known and actual physics discrepancy is captured using Gaussian Process Regression. Experiments on benchmark simulated tasks requiring composition of collisions, slides, and rebounds demonstrate that PhyPlan achieves a 45\% higher success rate and up to 3× efficiency gains over state-of-the-art model-based reinforcement learning approaches.

Zitieren

APA

Kanwar, A., Soin, H., Barnawal, A., Chopra, M., Vagadia, H., Banerjee, T., Tuli, S., Paul, R., & Chakraborty, S. (2026). PhyPlan: Learning to Plan Tasks with Generalizable and Rapid Physical Reasoning for Embodied Manipulation. Proceedings of the AAAI Conference on Artificial Intelligence, 40(22), 18360-18369. https://doi.org/10.1609/aaai.v40i22.38900

Vancouver

Kanwar A, Soin H, Barnawal A, Chopra M, Vagadia H, Banerjee T, et al. PhyPlan: Learning to Plan Tasks with Generalizable and Rapid Physical Reasoning for Embodied Manipulation. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(22):18360-18369. doi:10.1609/aaai.v40i22.38900

BibTeX
@article{Kanwar2026,
  author = {Kanwar, Ankit and Soin, Hartej and Barnawal, Abhinav and Chopra, Mudit and Vagadia, Harshil and Banerjee, Tamajit and Tuli, Shreshth and Paul, Rohan and Chakraborty, Souvik},
  title = {PhyPlan: Learning to Plan Tasks with Generalizable and Rapid Physical Reasoning for Embodied Manipulation},
  journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
  year = {2026},
  volume = {40},
  number = {22},
  pages = {18360--18369},
  doi = {10.1609/aaai.v40i22.38900}
}
Kostenlos starten →