Personalized Dynamics Models for Adaptive Assistive Navigation Systems

Eshed OhnBar, Kris Kitani, Chieko Asakawa
Proceedings of The 2nd Conference on Robot Learning, PMLR 87:16-39, 2018.

Abstract

Consider an assistive system that guides visually impaired users through speech and haptic feedback to their destination. Existing robotic and ubiquitous navigation technologies (e.g., portable, ground, or wearable systems) often operate in a generic, user-agnostic manner. However, to minimize confusion and navigation errors, our real-world analysis reveals a crucial need to adapt theinstructional guidance across different end-users with diverse mobility skills. To address this practical issue in scalable system design, we propose a novel model based reinforcement learning framework for personalizing the system-user interaction experience. When incrementally adapting the system to new users, we propose to use a weighted experts model for addressing data-efficiency limitations in transfer learning with deep models. A real-world dataset of navigation by blind users is used to show that the proposed approach allows for (1) more accurate long-term human behavior prediction (up to 20 seconds into the future) through improved reasoning over personal mobility characteristics, interaction with surrounding obstacles, and the current navigation goal, and (2) quick adaptation at the onset of learning, when data is limited.

Cite this Paper

BibTeX
@InProceedings{pmlr-v87-ohnbar18a, title = {Personalized Dynamics Models for Adaptive Assistive Navigation Systems}, author = {OhnBar, Eshed and Kitani, Kris and Asakawa, Chieko}, booktitle = {Proceedings of The 2nd Conference on Robot Learning}, pages = {16--39}, year = {2018}, editor = {Billard, Aude and Dragan, Anca and Peters, Jan and Morimoto, Jun}, volume = {87}, series = {Proceedings of Machine Learning Research}, month = {29--31 Oct}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v87/ohnbar18a/ohnbar18a.pdf}, url = {https://proceedings.mlr.press/v87/ohnbar18a.html}, abstract = {Consider an assistive system that guides visually impaired users through speech and haptic feedback to their destination. Existing robotic and ubiquitous navigation technologies (e.g., portable, ground, or wearable systems) often operate in a generic, user-agnostic manner. However, to minimize confusion and navigation errors, our real-world analysis reveals a crucial need to adapt theinstructional guidance across different end-users with diverse mobility skills. To address this practical issue in scalable system design, we propose a novel model based reinforcement learning framework for personalizing the system-user interaction experience. When incrementally adapting the system to new users, we propose to use a weighted experts model for addressing data-efficiency limitations in transfer learning with deep models. A real-world dataset of navigation by blind users is used to show that the proposed approach allows for (1) more accurate long-term human behavior prediction (up to 20 seconds into the future) through improved reasoning over personal mobility characteristics, interaction with surrounding obstacles, and the current navigation goal, and (2) quick adaptation at the onset of learning, when data is limited.} }
Endnote
%0 Conference Paper %T Personalized Dynamics Models for Adaptive Assistive Navigation Systems %A Eshed OhnBar %A Kris Kitani %A Chieko Asakawa %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-ohnbar18a %I PMLR %P 16--39 %U https://proceedings.mlr.press/v87/ohnbar18a.html %V 87 %X Consider an assistive system that guides visually impaired users through speech and haptic feedback to their destination. Existing robotic and ubiquitous navigation technologies (e.g., portable, ground, or wearable systems) often operate in a generic, user-agnostic manner. However, to minimize confusion and navigation errors, our real-world analysis reveals a crucial need to adapt theinstructional guidance across different end-users with diverse mobility skills. To address this practical issue in scalable system design, we propose a novel model based reinforcement learning framework for personalizing the system-user interaction experience. When incrementally adapting the system to new users, we propose to use a weighted experts model for addressing data-efficiency limitations in transfer learning with deep models. A real-world dataset of navigation by blind users is used to show that the proposed approach allows for (1) more accurate long-term human behavior prediction (up to 20 seconds into the future) through improved reasoning over personal mobility characteristics, interaction with surrounding obstacles, and the current navigation goal, and (2) quick adaptation at the onset of learning, when data is limited.
APA
OhnBar, E., Kitani, K. & Asakawa, C.. (2018). Personalized Dynamics Models for Adaptive Assistive Navigation Systems. Proceedings of The 2nd Conference on Robot Learning, in Proceedings of Machine Learning Research 87:16-39 Available from https://proceedings.mlr.press/v87/ohnbar18a.html.

Related Material