Zobrazeno 1 - 10
of 137
pro vyhledávání: '"Leslie Pack Kaelbling"'
Publikováno v:
The International Journal of Robotics Research. 41:210-231
We present a framework for learning to guide geometric task-and-motion planning (G-TAMP). G-TAMP is a subclass of task-and-motion planning in which the goal is to move multiple objects to target regions among movable obstacles. A standard graph searc
Publikováno v:
The International Journal of Robotics Research. 40:866-894
The objective of this work is to augment the basic abilities of a robot by learning to use sensorimotor primitives to solve complex long-horizon manipulation problems. This requires flexible generative planning that can combine primitive abilities in
Motion planning is a ubiquitous problem that is often a bottleneck in robotic applications. We demonstrate that motion planning problems such as minimum constraint removal, belief-space planning, and visibility-aware motion planning (VAMP) benefit fr
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::849535af3fda2fd6619bda096438de22
http://arxiv.org/abs/2206.02305
http://arxiv.org/abs/2206.02305
Publikováno v:
AAAI
Other repository
Scopus-Elsevier
Web of Science
Other repository
Scopus-Elsevier
Web of Science
Many important applications, including robotics, data-center management, and process control, require planning action sequences in domains with continuous state and action spaces and discontinuous objective functions. Monte Carlo tree search (MCTS) i
Autor:
Mirko Klukas, Tomás Lozano-Pérez, Sugandha Sharma, Yilun Du, Leslie Pack Kaelbling, Ila Fiete
When animals explore spatial environments, their representations often fragment into multiple maps. What determines these map fragmentations, and can we predict where they will occur with simple principles? We pose the problem of fragmentation of an
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::0dd777345ae42eb1f67b5d9f35d65f31
https://doi.org/10.1101/2021.10.29.466499
https://doi.org/10.1101/2021.10.29.466499
Publikováno v:
2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
Anytime motion planners are widely used in robotics. However, the relationship between their solution quality and computation time is not well understood, and thus, determining when to quit planning and start execution is unclear. In this paper, we a
Publikováno v:
2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS).
Robotic planning problems in hybrid state and action spaces can be solved by integrated task and motion planners (TAMP) that handle the complex interaction between motion-level decisions and task-level plan feasibility. TAMP approaches rely on domain
Autor:
Jiayuan Mao, Tomer Ullman, Zhezheng Luo, Chuang Gan, Jiajun Wu, Joshua B. Tenenbaum, Leslie Pack Kaelbling
Publikováno v:
IJCAI
We present Temporal and Object Quantification Networks (TOQ-Nets), a new class of neuro-symbolic networks with a structural bias that enables them to learn to recognize complex relational-temporal events. This is done by including reasoning layers th
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7edf847926366618785401ff3254dd68
http://arxiv.org/abs/2106.05891
http://arxiv.org/abs/2106.05891
Publikováno v:
ICRA
We propose a new, data-efficient approach for skill transfer to novel objects, accounting for known categorical shape variation. A low-dimensional shape representation embedding is learned from a set of deformations, sampled between known objects wit
In robotic domains, learning and planning are complicated by continuous state spaces, continuous action spaces, and long task horizons. In this work, we address these challenges with Neuro-Symbolic Relational Transition Models (NSRTs), a novel class
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4e9f3923e843db79707554684fbb0181
http://arxiv.org/abs/2105.14074
http://arxiv.org/abs/2105.14074