Predicting Routine Object Usage for Proactive Robot Assistance

Maithili Patel, Aswin Gururaj Prakash, Sonia Chernova
Proceedings of The 7th Conference on Robot Learning, PMLR 229:1068-1083, 2023.

Abstract

Proactivity in robot assistance refers to the robot’s ability to anticipate user needs and perform assistive actions without explicit requests. This requires understanding user routines, predicting consistent activities, and actively seeking information to predict inconsistent behaviors. We propose SLaTe-PRO (Sequential Latent Temporal model for Predicting Routine Object usage), which improves upon prior state-of-the-art by combining object and user action information, and conditioning object usage predictions on past history. Additionally, we find some human behavior to be inherently stochastic and lacking in contextual cues that the robot can use for proactive assistance. To address such cases, we introduce an interactive query mechanism that can be used to ask queries about the user’s intended activities and object use to improve prediction. We evaluate our approach on longitudinal data from three households, spanning 24 activity classes. SLaTe-PRO performance raises the F1 score metric to 0.57 without queries, and 0.60 with user queries, over a score of 0.43 from prior work. We additionally present a case study with a fully autonomous household robot.

Cite this Paper

BibTeX
@InProceedings{pmlr-v229-patel23a, title = {Predicting Routine Object Usage for Proactive Robot Assistance}, author = {Patel, Maithili and Prakash, Aswin Gururaj and Chernova, Sonia}, booktitle = {Proceedings of The 7th Conference on Robot Learning}, pages = {1068--1083}, 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/patel23a/patel23a.pdf}, url = {https://proceedings.mlr.press/v229/patel23a.html}, abstract = {Proactivity in robot assistance refers to the robot’s ability to anticipate user needs and perform assistive actions without explicit requests. This requires understanding user routines, predicting consistent activities, and actively seeking information to predict inconsistent behaviors. We propose SLaTe-PRO (Sequential Latent Temporal model for Predicting Routine Object usage), which improves upon prior state-of-the-art by combining object and user action information, and conditioning object usage predictions on past history. Additionally, we find some human behavior to be inherently stochastic and lacking in contextual cues that the robot can use for proactive assistance. To address such cases, we introduce an interactive query mechanism that can be used to ask queries about the user’s intended activities and object use to improve prediction. We evaluate our approach on longitudinal data from three households, spanning 24 activity classes. SLaTe-PRO performance raises the F1 score metric to 0.57 without queries, and 0.60 with user queries, over a score of 0.43 from prior work. We additionally present a case study with a fully autonomous household robot.} }
Endnote
%0 Conference Paper %T Predicting Routine Object Usage for Proactive Robot Assistance %A Maithili Patel %A Aswin Gururaj Prakash %A Sonia Chernova %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-patel23a %I PMLR %P 1068--1083 %U https://proceedings.mlr.press/v229/patel23a.html %V 229 %X Proactivity in robot assistance refers to the robot’s ability to anticipate user needs and perform assistive actions without explicit requests. This requires understanding user routines, predicting consistent activities, and actively seeking information to predict inconsistent behaviors. We propose SLaTe-PRO (Sequential Latent Temporal model for Predicting Routine Object usage), which improves upon prior state-of-the-art by combining object and user action information, and conditioning object usage predictions on past history. Additionally, we find some human behavior to be inherently stochastic and lacking in contextual cues that the robot can use for proactive assistance. To address such cases, we introduce an interactive query mechanism that can be used to ask queries about the user’s intended activities and object use to improve prediction. We evaluate our approach on longitudinal data from three households, spanning 24 activity classes. SLaTe-PRO performance raises the F1 score metric to 0.57 without queries, and 0.60 with user queries, over a score of 0.43 from prior work. We additionally present a case study with a fully autonomous household robot.
APA
Patel, M., Prakash, A.G. & Chernova, S.. (2023). Predicting Routine Object Usage for Proactive Robot Assistance. Proceedings of The 7th Conference on Robot Learning, in Proceedings of Machine Learning Research 229:1068-1083 Available from https://proceedings.mlr.press/v229/patel23a.html.

Related Material