analiza aiqbench
This paper applies learned multimodal world models (vision, action, and force) to predict anomalies in robotic manipulation before they become visible, rather than reacting after the fact. The reported gains are moderate: bACC of 0.765 and 0.710 on seen and novel tasks, and task success rising from 52.6% to 61.5% without retraining the policy. The new CRIA dataset and the claim that the approach transfers to unseen tasks are the main points to verify against the full paper.
MAIA: Anticipating Interaction Anomalies in Contact-Rich Manipulation via Multimodal World Models
arXiv:2610.21686v1 Announce Type: new Abstract: Reliable contact-rich manipulation requires detecting execution deviations early enough for corrective action. Yet most monitoring methods are reactive, identifying anomalies only after they become observable. Anticipating such anomalies is challenging because visual observations only partially reveal physical interaction, while planned actions describe intended motion rather than their actual physical consequences. We address this problem with MAIA, a multimodal framework for anomaly anticipation. Its vision-action-force world model learns nominal interaction dynamics and predicts the visual and force responses of planned actions, while a lightweight anomaly head predicts impending interaction anomalies. To enable systematic evaluation, we introduce CRIA, a real-world multimodal dataset of 2,852 trajectories spanning 18 contact-rich manipulation tasks, with frame-level annotations of anomaly onset and evolution. Experiments on both seen and novel tasks show that MAIA achieves more accurate anomaly discrimination than representative baselines while providing earlier warnings, with bACC scores of 0.765 and 0.710 and median anticipation lead times of 28.0 and 25.1 steps, respectively. Moreover, using these predictions for anomaly-aware action selection improves end-to-end task success from 52.6% to 61.5% without retraining the underlying manipulation policy. These results highlight the value of modeling future physical consequences for proactive anomaly anticipation.
Abstract:Reliable contact-rich manipulation requires detecting execution deviations early enough for corrective action. Yet most monitoring methods are reactive, identifying anomalies only after they become observable. Anticipating such anomalies is challenging because visual observations only partially reveal physical interaction, while planned actions describe intended motion rather than their actual physical consequences. We address this problem with MAIA, a multimodal framework for anomaly anticipation. Its vision-action-force world model learns nominal interaction dynamics and predicts the visual and force responses of planned actions, while a lightweight anomaly head predicts impending interaction anomalies. To enable systematic evaluation, we introduce CRIA, a real-world multimodal dataset of 2,852 trajectories spanning 18 contact-rich manipulation tasks, with frame-level annotations of anomaly onset and evolution. Experiments on both seen and novel tasks show that MAIA achieves more accurate anomaly discrimination than representative baselines while providing earlier warnings, with bACC scores of 0.765 and 0.710 and median anticipation lead times of 28.0 and 25.1 steps, respectively. Moreover, using these predictions for anomaly-aware action selection improves end-to-end task success from 52.6% to 61.5% without retraining the underlying manipulation policy. These results highlight the value of modeling future physical consequences for proactive anomaly anticipation.
Robotics (cs.RO)
arXiv:2610.21686 [cs.RO]
(or arXiv:2610.21686v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2610.21686 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 285 cuvinte
Subiecte conexe
ForeTac-VLA: A Forecasting-Based Tactile-Vision-Language-Action Model for Contact-Rich Robotic Manipulation
arXiv · 21 sept. 2026
Task aware Dynamic Movement Primitives for failure detection and recovery in contact rich manipulation
arXiv · 22 sept. 2026
ContactWorld: What Matters in Vision-Tactile World Models for Contact-Rich Manipulation
arXiv · 16 iun. 2026
PSR: Predictive Sensorimotor Representation Learning for Contact-Rich Manipulation
arXiv · 21 sept. 2026