See it, Think it, Sorted: Large Multimodal Models are Few-shot Time Series Anomaly Analyzers

Time series anomaly detection (TSAD) is becoming increasingly vital due to the rapid growth of time series data across various sectors. Anomalies in web service data, for example, can signal critical incidents such as system failures or server malfunctions, necessitating timely detection and response. However, most existing TSAD methodologies rely heavily on manual feature engineering or require extensive labeled training data, while also offering limited interpretability. To address these challenges, we introduce a pioneering framework called the Time Series Anomaly Multimodal Analyzer (TAMA), which leverages the power of Large Multimodal Models (LMMs) to enhance both the detection and interpretation of anomalies in time series data. By converting time series into visual formats that LMMs can efficiently process, TAMA leverages few-shot in-context learning capabilities to reduce dependence on extensive labeled datasets. Our methodology is validated through rigorous experimentation on multiple real-world datasets, where TAMA consistently outperforms state-of-the-art methods in TSAD tasks. Additionally, TAMA provides rich, natural language-based semantic analysis, offering deeper insights into the nature of detected anomalies. Furthermore, we contribute one of the first open-source datasets that includes anomaly detection labels, anomaly type labels, and contextual description, facilitating broader exploration and advancement within this critical field. Ultimately, TAMA not only excels in anomaly detection but also provides a comprehensive approach for understanding the underlying causes of anomalies, pushing TSAD forward through innovative methodologies and insights.

MIMO ILC for Precision SEA robots using Input-weighted Complex-Kernel Regression

This work improves the positioning precision of lightweight robots with series elastic actuators (SEAs). Lightweight SEA robots, along with low-impedance control, can maneuver without causing damage in uncertain, confined spaces such as inside an aircraft wing during aircraft assembly. Nevertheless, substantial modeling uncertainties in SEA robots reduce the precision achieved by model-based approaches such as inversion-based feedforward. Therefore, this article improves the precision of SEA robots around specified operating points, through a multi-input multi-output (MIMO), iterative learning control (ILC) approach. The main contributions of this article are to (i) introduce an input-weighted complex kernel to estimate local MIMO models using complex Gaussian process regression (c-GPR) (ii) develop Ger\v{s}gorin-theorem-based conditions on the iteration gains for ensuring ILC convergence to precision within noise-related limits, even with errors in the estimated model; and (iii) demonstrate precision positioning with an experimental SEA robot. Comparative experimental results, with and without ILC, show around 90% improvement in the positioning precision (close to the repeatability limit of the robot) and a 10-times increase in the SEA robot's operating speed with the use of the MIMO ILC.