Multi-Strategy Deployment-Time Learning and Adaptation for Navigation under Uncertainty

Abhishek Paudel, Xuesu Xiao, Gregory J. Stein
Proceedings of The 8th Conference on Robot Learning, PMLR 270:3908-3923, 2025.

Abstract

We present an approach for performant point-goal navigation in unfamiliar partially-mapped environments. When deployed, our robot runs multiple strategies for deployment-time learning and visual domain adaptation in parallel and quickly selects the best-performing among them. Choosing between policies as they are learned or adapted between navigation trials requires continually updating estimates of their performance as they evolve. Leveraging recent work in model-based learning-informed planning under uncertainty, we determine lower bounds on the would-be performance of newly-updated policies on old trials without needing to re-deploy them. This information constrains and accelerates bandit-like policy selection, affording quick selection of the best-performing strategy shortly after it would start to yield good performance. We validate the effectiveness of our approach in simulated maze-like environments, showing improved navigation cost and cumulative regret versus existing baselines.

Cite this Paper

BibTeX
@InProceedings{pmlr-v270-paudel25a, title = {Multi-Strategy Deployment-Time Learning and Adaptation for Navigation under Uncertainty}, author = {Paudel, Abhishek and Xiao, Xuesu and Stein, Gregory J.}, booktitle = {Proceedings of The 8th Conference on Robot Learning}, pages = {3908--3923}, year = {2025}, editor = {Agrawal, Pulkit and Kroemer, Oliver and Burgard, Wolfram}, volume = {270}, series = {Proceedings of Machine Learning Research}, month = {06--09 Nov}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v270/main/assets/paudel25a/paudel25a.pdf}, url = {https://proceedings.mlr.press/v270/paudel25a.html}, abstract = {We present an approach for performant point-goal navigation in unfamiliar partially-mapped environments. When deployed, our robot runs multiple strategies for deployment-time learning and visual domain adaptation in parallel and quickly selects the best-performing among them. Choosing between policies as they are learned or adapted between navigation trials requires continually updating estimates of their performance as they evolve. Leveraging recent work in model-based learning-informed planning under uncertainty, we determine lower bounds on the would-be performance of newly-updated policies on old trials without needing to re-deploy them. This information constrains and accelerates bandit-like policy selection, affording quick selection of the best-performing strategy shortly after it would start to yield good performance. We validate the effectiveness of our approach in simulated maze-like environments, showing improved navigation cost and cumulative regret versus existing baselines.} }
Endnote
%0 Conference Paper %T Multi-Strategy Deployment-Time Learning and Adaptation for Navigation under Uncertainty %A Abhishek Paudel %A Xuesu Xiao %A Gregory J. Stein %B Proceedings of The 8th Conference on Robot Learning %C Proceedings of Machine Learning Research %D 2025 %E Pulkit Agrawal %E Oliver Kroemer %E Wolfram Burgard %F pmlr-v270-paudel25a %I PMLR %P 3908--3923 %U https://proceedings.mlr.press/v270/paudel25a.html %V 270 %X We present an approach for performant point-goal navigation in unfamiliar partially-mapped environments. When deployed, our robot runs multiple strategies for deployment-time learning and visual domain adaptation in parallel and quickly selects the best-performing among them. Choosing between policies as they are learned or adapted between navigation trials requires continually updating estimates of their performance as they evolve. Leveraging recent work in model-based learning-informed planning under uncertainty, we determine lower bounds on the would-be performance of newly-updated policies on old trials without needing to re-deploy them. This information constrains and accelerates bandit-like policy selection, affording quick selection of the best-performing strategy shortly after it would start to yield good performance. We validate the effectiveness of our approach in simulated maze-like environments, showing improved navigation cost and cumulative regret versus existing baselines.
APA
Paudel, A., Xiao, X. & Stein, G.J.. (2025). Multi-Strategy Deployment-Time Learning and Adaptation for Navigation under Uncertainty. Proceedings of The 8th Conference on Robot Learning, in Proceedings of Machine Learning Research 270:3908-3923 Available from https://proceedings.mlr.press/v270/paudel25a.html.

Related Material