We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.RO

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Robotics

Title: Proactive Robot Assistance via Spatio-Temporal Object Modeling

Abstract: Proactive robot assistance enables a robot to anticipate and provide for a user's needs without being explicitly asked. We formulate proactive assistance as the problem of the robot anticipating temporal patterns of object movements associated with everyday user routines, and proactively assisting the user by placing objects to adapt the environment to their needs. We introduce a generative graph neural network to learn a unified spatio-temporal predictive model of object dynamics from temporal sequences of object arrangements. We additionally contribute the Household Object Movements from Everyday Routines (HOMER) dataset, which tracks household objects associated with human activities of daily living across 50+ days for five simulated households. Our model outperforms the leading baseline in predicting object movement, correctly predicting locations for 11.1% more objects and wrongly predicting locations for 11.5% fewer objects used by the human user.
Subjects: Robotics (cs.RO); Artificial Intelligence (cs.AI)
Cite as: arXiv:2211.15501 [cs.RO]
  (or arXiv:2211.15501v1 [cs.RO] for this version)

Submission history

From: Maithili Patel [view email]
[v1] Mon, 28 Nov 2022 16:20:50 GMT (33269kb,D)

Link back to: arXiv, form interface, contact.