SyncAI.news, a Varaisys broadcasting
FailureGen: Scalable Robot Failure Data Generation with World Models
AS

Akul Santhosh, Chaitra Sathe, Aastha Jhunjhunwala

· 1 min read

ResearcharXiv cs.AI

FailureGen: Scalable Robot Failure Data Generation with World Models

arXiv:2611.22367v1 Announce Type: cross Abstract: Robot failures are difficult to collect at scale, limiting the range of failure scenarios represented in robot learning datasets. We present FailureGen, a method for generating synthetic videos with temporal event annotations by reusing action trajectories from recorded failures. The method comprises four stages: (1) fine-tuning an action-conditioned Cosmos world model on successful and failed real executions; (2) annotating source recordings to identify failure-associated donor segments and compatible successful hosts; (3) transferring donor motion into host contexts, with optional spatial offsets, to generate continuous video rollouts; and (4) annotating and filtering the generated samples according to observed events and visual quality. We study two implementations with different tasks and robot configurations, with visual failure detection serving as one downstream application. In a preliminary comparison, a detector trained on real and generated data achieves an average precision of 0.51, compared with 0.41 using real data alone, and detects 33 of 36 failure intervals, compared with 28 of 36. By reusing observed failures in additional contexts, FailureGen expands the data available for studying robot failures and developing downstream learning methods.

Original source

This story was published by arXiv cs.AI and written by Akul Santhosh, Chaitra Sathe, Aastha Jhunjhunwala. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News