Learning where to trust unreliable models in an unstructured world for deformable object manipulation
Abstract
We plan for rope manipulation with an unreliable model by learning where to trust the model and how to recover when stuck.
Document Details
- Document Type
- Pub Defense Publication
- Publication Date
- May 12, 2021
- Source ID
- 10.1126/scirobotics.abd8170
Entities
People
- Dale McConachie
- Dmitry Berenson
- Peter Paul Mitrano
Organizations
- National Science Foundation
- Office of Naval Research
- Toyota Research Institute
- University of Michigan