Ready, Set, Plan! Planning to Goal Sets Using Generalized Bayesian Inference

Jana Pavlasek, Stanley Robert Lewis, Balakumar Sundaralingam, Fabio Ramos, Tucker Hermans
Proceedings of The 7th Conference on Robot Learning, PMLR 229:3672-3686, 2023.

Abstract

Many robotic tasks can have multiple and diverse solutions and, as such, are naturally expressed as goal sets. Examples include navigating to a room, finding a feasible placement location for an object, or opening a drawer enough to reach inside. Using a goal set as a planning objective requires that a model for the objective be explicitly given by the user. However, some goals are intractable to model, leading to uncertainty over the goal (e.g. stable grasping of an object). In this work, we propose a technique for planning directly to a set of sampled goal configurations. We formulate a planning as inference problem with a novel goal likelihood evaluated against the goal samples. To handle the intractable goal likelihood, we employ Generalized Bayesian Inference to approximate the trajectory distribution. The result is a fully differentiable cost which generalizes across a diverse range of goal set objectives for which samples can be obtained. We show that by considering all goal samples throughout the planning process, our method reliably finds plans on manipulation and navigation problems where heuristic approaches fail.

Cite this Paper

BibTeX
@InProceedings{pmlr-v229-pavlasek23a, title = {Ready, Set, Plan! Planning to Goal Sets Using Generalized Bayesian Inference}, author = {Pavlasek, Jana and Lewis, Stanley Robert and Sundaralingam, Balakumar and Ramos, Fabio and Hermans, Tucker}, booktitle = {Proceedings of The 7th Conference on Robot Learning}, pages = {3672--3686}, 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/pavlasek23a/pavlasek23a.pdf}, url = {https://proceedings.mlr.press/v229/pavlasek23a.html}, abstract = {Many robotic tasks can have multiple and diverse solutions and, as such, are naturally expressed as goal sets. Examples include navigating to a room, finding a feasible placement location for an object, or opening a drawer enough to reach inside. Using a goal set as a planning objective requires that a model for the objective be explicitly given by the user. However, some goals are intractable to model, leading to uncertainty over the goal (e.g. stable grasping of an object). In this work, we propose a technique for planning directly to a set of sampled goal configurations. We formulate a planning as inference problem with a novel goal likelihood evaluated against the goal samples. To handle the intractable goal likelihood, we employ Generalized Bayesian Inference to approximate the trajectory distribution. The result is a fully differentiable cost which generalizes across a diverse range of goal set objectives for which samples can be obtained. We show that by considering all goal samples throughout the planning process, our method reliably finds plans on manipulation and navigation problems where heuristic approaches fail.} }
Endnote
%0 Conference Paper %T Ready, Set, Plan! Planning to Goal Sets Using Generalized Bayesian Inference %A Jana Pavlasek %A Stanley Robert Lewis %A Balakumar Sundaralingam %A Fabio Ramos %A Tucker Hermans %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-pavlasek23a %I PMLR %P 3672--3686 %U https://proceedings.mlr.press/v229/pavlasek23a.html %V 229 %X Many robotic tasks can have multiple and diverse solutions and, as such, are naturally expressed as goal sets. Examples include navigating to a room, finding a feasible placement location for an object, or opening a drawer enough to reach inside. Using a goal set as a planning objective requires that a model for the objective be explicitly given by the user. However, some goals are intractable to model, leading to uncertainty over the goal (e.g. stable grasping of an object). In this work, we propose a technique for planning directly to a set of sampled goal configurations. We formulate a planning as inference problem with a novel goal likelihood evaluated against the goal samples. To handle the intractable goal likelihood, we employ Generalized Bayesian Inference to approximate the trajectory distribution. The result is a fully differentiable cost which generalizes across a diverse range of goal set objectives for which samples can be obtained. We show that by considering all goal samples throughout the planning process, our method reliably finds plans on manipulation and navigation problems where heuristic approaches fail.
APA
Pavlasek, J., Lewis, S.R., Sundaralingam, B., Ramos, F. & Hermans, T.. (2023). Ready, Set, Plan! Planning to Goal Sets Using Generalized Bayesian Inference. Proceedings of The 7th Conference on Robot Learning, in Proceedings of Machine Learning Research 229:3672-3686 Available from https://proceedings.mlr.press/v229/pavlasek23a.html.

Related Material