The paper introduces AWARE, a world‑model framework that detects anomalies in robotic systems using only external camera data for dynamics estimation and state prediction. Tested on 240 real‑world crane trajectories, AWARE outperforms the best baseline by 26% in anomaly classification under noisy, camera‑only conditions. It also decodes dynamics parameters to pinpoint fault sources, achieving 88% accuracy in localizing motor degradation.
arXiv:2607. 18142v1 Announce Type: cross Abstract: Industrial Video Anomaly Detection (IVAD) aims to identify anomalous objects and events in an industrial process, which is crucial for modern manufacturing and quality control systems.
The paper introduces MUSTARD, a multimodal spatiotemporal anomaly detection framework for industrial products that enhances large language models with fine‑grained motion awareness. It uses optical‑flow‑guided keyframe selection and augmented modalities to capture both local motion patterns and global dynamic evolution, enabling detection and reasoning of dynamics‑based defects with only a few normal reference samples. Experiments show that MUSTARD outperforms existing methods in identifying spatiotemporal anomalies and offers interpretable explanations.
arXiv:2509. 06419v2 Announce Type: replace Abstract: Time-series anomaly detection is crucial in AIOps for maintaining large-scale service reliability.
The paper introduces MUSTARD, a multimodal spatiotemporal anomaly detection framework that enhances large language models with fine‑grained motion awareness for industrial defect detection. It uses optical‑flow‑guided keyframe selection and augmented modalities to capture local motion patterns and global dynamic evolution, enabling detection of dynamics‑based anomalies without model updates. Experiments show that MUSTARD outperforms existing methods in identifying spatiotemporal defects and offers interpretable explanations.
The paper introduces a multimodal anomaly detection framework that uses cross‑modal reconstruction of heterogeneous time‑series sensor data to detect faults in industrial systems. By learning to reconstruct each modality from the others, the method leverages complementary information across sensing channels without requiring explicit temporal alignment or identical sampling rates. An adaptive test‑time thresholding mechanism further improves robustness to distribution shifts caused by changing operating conditions, as demonstrated by strong fault detection performance in three industrial case studies, especially under out‑of‑distribution regimes.
arXiv:2609.13940v1 Announce Type: new Abstract: Effective anomaly detection frameworks are a central pillar of the Industry 4.0 paradigm. In this paper, we introduce an Optimal Transport (OT)-based f...
arXiv:2606. 01691v1 Announce Type: cross Abstract: Industrial Internet systems face increasing threats from sophisticated industrial control system (ICS) attacks, resulting in critical safety incidents.
arXiv:2605.21957v2 Announce Type: replace Abstract: Video anomaly detection is critical for public safety and security, yet remains highly challenging despite extensive research due to large variatio...
arXiv:2608. 10587v1 Announce Type: new Abstract: Artificial Intelligence (AI)-based prospective anomaly detection methods are increasingly deployed in high-dimensional and nonlinear settings.
RoboVAD is a large-scale benchmark for video anomaly detection in robotic arm manipulation, featuring cross-domain evaluation where both tasks and anomaly types are unseen during training. The dataset challenges existing VAD methods, with state-of-the-art approaches, including a new method tailored for robotic arms, still achieving less than 70% micro-averaged frame-level AUC in the hardest setting. The authors provide the dataset and code publicly for further research.
The paper presents a Teacher-Student distillation framework for continuous online fault detection in mobile robots. An offline foundation model (TSPulse) generates pseudo‑labels from augmented time‑series data, while a lightweight MiniRocket Student, enhanced with a Recursive Least Squares estimator, performs real‑time inference with a 4.30 ms CPU latency. The Student adapts online to domain shifts, improving VUS‑PR scores from 0.26 to 0.75 and uses an uncertainty‑guided active learning strategy to request minimal operator interventions.
arXiv:2608.30804v1 Announce Type: new Abstract: Monitoring the health of heterogeneous industrial robot fleets is severely challenged by the multi-modal nature of their operational cycles and a persi...