OpenTechnical
This is a record of a document published elsewhere, first published by the United States Patent and Trademark Office, as US-2026097508-A1, on 9 April 2026. OpenTechnical did not first disclose it. The identifier, hashes and timestamps below record what this library served and when, not when the material became public. View the original.

OT-2026-000000990

Uncertainty-aware failure detection for imitation learning robot policies

Published
Permanent URL
Classification
Robotics & Automation › Robot Architectures
Inventors
AMBRUS RARES A, ITKINA MIKHAL, LEE ROBERT, NISHIMURA HARUKI, SHAH PAARTH, XU CHEN
Assignee
TOYOTA JIDOSHA KK, TOYOTA RES INSTITUTE INC
Original source
US-2026097508-A1, the United States Patent and Trademark Office, 9 April 2026
Document SHA-256
b4ac641f4a18b46c48482f344cba5bfeb2da51d853d894c6086e20ad9a30ab9d

Abstract

Systems and methods described herein relate to generating a discrimination scores set based on observation data and action data obtained from a robot trained to perform a task, determining a threshold value based on the discrimination score set, and comparing a discrimination score obtained while the task is being performed with the threshold value to determine if a failure condition is present. This may be performed by utilizing random network distillation or other out-of-distribution detectors and conformal band prediction computed on a set of successful rollouts by the robot.

Publication information

Plain textMarkdownManifestIntegrity record

Citation

OpenTechnical Publication OT-2026-000000990, “Uncertainty-aware failure detection for imitation learning robot policies,” published September 15, 2026 at 19:03:42 UTC, https://opentechnical.org/p/OT-2026-000000990.

To cite specific content, add the paragraph or figure numbers.

Terms

This record reproduces a document published elsewhere and is provided for reference and study. Rights in the original remain with its owner. See the publication policy.