Embodied Lifelong Learning for Task and Motion Planning

Jorge Mendez-Mendez, Leslie Pack Kaelbling, Tomás Lozano-Pérez
Proceedings of The 7th Conference on Robot Learning, PMLR 229:2134-2150, 2023.

Abstract

A robot deployed in a home over long stretches of time faces a true lifelong learning problem. As it seeks to provide assistance to its users, the robot should leverage any accumulated experience to improve its own knowledge and proficiency. We formalize this setting with a novel formulation of lifelong learning for task and motion planning (TAMP), which endows our learner with the compositionality of TAMP systems. Exploiting the modularity of TAMP, we develop a mixture of generative models that produces candidate continuous parameters for a planner. Whereas most existing lifelong learning approaches determine a priori how data is shared across various models, our approach learns shared and non-shared models and determines which to use online during planning based on auxiliary tasks that serve as a proxy for each model’s understanding of a state. Our method exhibits substantial improvements (over time and compared to baselines) in planning success on 2D and BEHAVIOR domains.

Cite this Paper


BibTeX
@InProceedings{pmlr-v229-mendez-mendez23a, title = {Embodied Lifelong Learning for Task and Motion Planning}, author = {Mendez-Mendez, Jorge and Kaelbling, Leslie Pack and Lozano-P\'{e}rez, Tom\'{a}s}, booktitle = {Proceedings of The 7th Conference on Robot Learning}, pages = {2134--2150}, year = {2023}, editor = {Tan, Jie and Toussaint, Marc and Darvish, Kourosh}, volume = {229}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v229/mendez-mendez23a/mendez-mendez23a.pdf}, url = {https://proceedings.mlr.press/v229/mendez-mendez23a.html}, abstract = {A robot deployed in a home over long stretches of time faces a true lifelong learning problem. As it seeks to provide assistance to its users, the robot should leverage any accumulated experience to improve its own knowledge and proficiency. We formalize this setting with a novel formulation of lifelong learning for task and motion planning (TAMP), which endows our learner with the compositionality of TAMP systems. Exploiting the modularity of TAMP, we develop a mixture of generative models that produces candidate continuous parameters for a planner. Whereas most existing lifelong learning approaches determine a priori how data is shared across various models, our approach learns shared and non-shared models and determines which to use online during planning based on auxiliary tasks that serve as a proxy for each model’s understanding of a state. Our method exhibits substantial improvements (over time and compared to baselines) in planning success on 2D and BEHAVIOR domains.} }
Endnote
%0 Conference Paper %T Embodied Lifelong Learning for Task and Motion Planning %A Jorge Mendez-Mendez %A Leslie Pack Kaelbling %A Tomás Lozano-Pérez %B Proceedings of The 7th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2023 %E Jie Tan %E Marc Toussaint %E Kourosh Darvish %F pmlr-v229-mendez-mendez23a %I PMLR %P 2134--2150 %U https://proceedings.mlr.press/v229/mendez-mendez23a.html %V 229 %X A robot deployed in a home over long stretches of time faces a true lifelong learning problem. As it seeks to provide assistance to its users, the robot should leverage any accumulated experience to improve its own knowledge and proficiency. We formalize this setting with a novel formulation of lifelong learning for task and motion planning (TAMP), which endows our learner with the compositionality of TAMP systems. Exploiting the modularity of TAMP, we develop a mixture of generative models that produces candidate continuous parameters for a planner. Whereas most existing lifelong learning approaches determine a priori how data is shared across various models, our approach learns shared and non-shared models and determines which to use online during planning based on auxiliary tasks that serve as a proxy for each model’s understanding of a state. Our method exhibits substantial improvements (over time and compared to baselines) in planning success on 2D and BEHAVIOR domains.
APA
Mendez-Mendez, J., Kaelbling, L.P. & Lozano-Pérez, T.. (2023). Embodied Lifelong Learning for Task and Motion Planning. Proceedings of The 7th Conference on Robot Learning, in Proceedings of Machine Learning Research 229:2134-2150 Available from https://proceedings.mlr.press/v229/mendez-mendez23a.html.

Related Material