STREAM-VAE: Dual-Path Routing for Slow and Fast Dynamics in Vehicle Telemetry Anomaly Detection

2025-11-19

AI Summary

Automotive telemetry data exhibit concurrent slow drifts and fast spikes, rendering conventional single-latent-variable reconstruction methods (e.g., sequential VAEs) inadequate for capturing multi-scale temporal dynamics—often oversmoothing spikes or amplifying variance, thereby degrading anomaly separability. To address this, we propose the Dual-Path Variational Autoencoder (DP-VAE), which employs decoupled slow and fast encoding pathways to separately model long-term drifts and transient anomalies, coupled with independent decoding branches that explicitly disentangle normal patterns from transient deviations. By integrating latent-variable decomposition with multi-scale reconstruction error analysis, DP-VAE achieves fine-grained dynamic feature disentanglement. Evaluated on an automotive telemetry dataset and the SMD benchmark, DP-VAE significantly outperforms prediction-based, attention-based, graph neural network, and various VAE-based baselines. It yields robust and stable anomaly scores while supporting both edge-efficient deployment and cloud-based high-precision analysis.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Planning, Routing, and Scheduling: Replanning and Plan Repair
• Data Mining & Knowledge Management: Anomaly/Outlier Detection

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Automotive telemetry data exhibits slow drifts and fast spikes, often within the same sequence, making reliable anomaly detection challenging. Standard reconstruction-based methods, including sequence variational autoencoders (VAEs), use a single latent process and therefore mix heterogeneous time scales, which can smooth out spikes or inflate variances and weaken anomaly separation. In this paper, we present STREAM-VAE, a variational autoencoder for anomaly detection in automotive telemetry time-series data. Our model uses a dual-path encoder to separate slow drift and fast spike signal dynamics, and a decoder that represents transient deviations separately from the normal operating pattern. STREAM-VAE is designed for deployment, producing stable anomaly scores across operating modes for both in-vehicle monitors and backend fleet analytics. Experiments on an automotive telemetry dataset and the public SMD benchmark show that explicitly separating drift and spike dynamics improves robustness compared to strong forecasting, attention, graph, and VAE baselines.

Problem

• Separates slow drift and fast spike dynamics in telemetry
• Prevents anomaly score instability across operating modes
• Improves robustness over forecasting and VAE baselines

Innovation

• Dual-path encoder separates slow and fast dynamics
• Decoder represents transient deviations from normal patterns
• Produces stable anomaly scores across operating modes