EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Detecting Failures in Robot Manipulation via Distance‐Aware Temporal Modelling

Heayoun Choi, Eunwoo Kim

Electronics Letters · 2026-01 · 62 (1)

DOI: 10.1049/ell2.70629 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics Letters, online erschienen am 27.05.2026. Verfasst von Heayoun Choi und 1 weiteren Autor:innen. Die Arbeit nennt 19 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: Chung-Ang University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Failure detection is essential for ensuring the safety of robotic manipulation policies, yet modern imitation learning methods, such as diffusion‐based policies, remain vulnerable to subtle trajectory deviations that are not visually apparent. We propose a distance‐aware failure detection framework that models the temporal evolution of robot states and object‐relative distances using a lightweight residual‐based model and a stochastic consistency score, along with a rollout‐calibrated temporal prediction band to account for discrepancies between demonstrations and policy rollouts. Experiments on simulation manipulation benchmarks show that our method is effective when visual differences between successful and failed executions are subtle, demonstrating the benefit of distance‐aware failure detection.

Förderung

  • Chung-Ang University

Zitieren

APA

Choi, H., & Kim, E. (2026). Detecting Failures in Robot Manipulation via Distance‐Aware Temporal Modelling. Electronics Letters, 62(1). https://doi.org/10.1049/ell2.70629

Vancouver

Choi H, Kim E. Detecting Failures in Robot Manipulation via Distance‐Aware Temporal Modelling. Electronics Letters. 2026;62(1). doi:10.1049/ell2.70629

BibTeX
@article{Choi2026,
  author = {Choi, Heayoun and Kim, Eunwoo},
  title = {Detecting Failures in Robot Manipulation via Distance‐Aware Temporal Modelling},
  journal = {Electronics Letters},
  year = {2026},
  volume = {62},
  number = {1},
  doi = {10.1049/ell2.70629}
}
Kostenlos starten →