Learning over Subgoals for Efficient Navigation of Structured, Unknown Environments

Gregory J. Stein, Christopher Bradley, Nicholas Roy
; Proceedings of The 2nd Conference on Robot Learning, PMLR 87:213-222, 2018.

Abstract

We propose a novel technique for efficiently navigating unknown environments over long horizons by learning to predict properties of unknown space. We generate a dynamic action set defined by the current map, factor the Bellman Equation in terms of these actions, and estimate terms, such as the probability that navigating beyond a particular subgoal will lead to a dead-end, that are otherwise difficult to compute. Simulated agents navigating with our Learned Subgoal Planner in real-world floor plans demonstrate a 21% expected decrease in cost-to-go compared to standard optimistic planning techniques that rely on Dijkstra’s algorithm, and real-world agents show promising navigation performance as well.

Cite this Paper


BibTeX
@InProceedings{pmlr-v87-stein18a, title = {Learning over Subgoals for Efficient Navigation of Structured, Unknown Environments}, author = {Stein, Gregory J. and Bradley, Christopher and Roy, Nicholas}, booktitle = {Proceedings of The 2nd Conference on Robot Learning}, pages = {213--222}, year = {2018}, editor = {Aude Billard and Anca Dragan and Jan Peters and Jun Morimoto}, volume = {87}, series = {Proceedings of Machine Learning Research}, address = {}, month = {29--31 Oct}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v87/stein18a/stein18a.pdf}, url = {http://proceedings.mlr.press/v87/stein18a.html}, abstract = { We propose a novel technique for efficiently navigating unknown environments over long horizons by learning to predict properties of unknown space. We generate a dynamic action set defined by the current map, factor the Bellman Equation in terms of these actions, and estimate terms, such as the probability that navigating beyond a particular subgoal will lead to a dead-end, that are otherwise difficult to compute. Simulated agents navigating with our Learned Subgoal Planner in real-world floor plans demonstrate a 21% expected decrease in cost-to-go compared to standard optimistic planning techniques that rely on Dijkstra’s algorithm, and real-world agents show promising navigation performance as well. } }
Endnote
%0 Conference Paper %T Learning over Subgoals for Efficient Navigation of Structured, Unknown Environments %A Gregory J. Stein %A Christopher Bradley %A Nicholas Roy %B Proceedings of The 2nd Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2018 %E Aude Billard %E Anca Dragan %E Jan Peters %E Jun Morimoto %F pmlr-v87-stein18a %I PMLR %J Proceedings of Machine Learning Research %P 213--222 %U http://proceedings.mlr.press %V 87 %W PMLR %X We propose a novel technique for efficiently navigating unknown environments over long horizons by learning to predict properties of unknown space. We generate a dynamic action set defined by the current map, factor the Bellman Equation in terms of these actions, and estimate terms, such as the probability that navigating beyond a particular subgoal will lead to a dead-end, that are otherwise difficult to compute. Simulated agents navigating with our Learned Subgoal Planner in real-world floor plans demonstrate a 21% expected decrease in cost-to-go compared to standard optimistic planning techniques that rely on Dijkstra’s algorithm, and real-world agents show promising navigation performance as well.
APA
Stein, G.J., Bradley, C. & Roy, N.. (2018). Learning over Subgoals for Efficient Navigation of Structured, Unknown Environments. Proceedings of The 2nd Conference on Robot Learning, in PMLR 87:213-222

Related Material