# A Dual-Stream Physics-Augmented Unsupervised Architecture for Runtime Embedded Vehicle Health Monitoring

arXiv: https://arxiv.org/abs/2602.10432 · HTML: https://mlsanity.com/paper/2602.10432/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 28 references that are in the map.
- First posted on arXiv 11 Feb 2026.

## Builds on (most cited first, 7 of 7)

- [CARLA: An Open Urban Driving Simulator](https://mlsanity.com/paper/1711.03938/) (2017)
- [Deep Learning for Time Series Anomaly Detection: A Survey](https://mlsanity.com/paper/2211.05244/) (2022)
- [Deep Metric Learning for Few-Shot Image Classification: A Review of Recent Developments](https://mlsanity.com/paper/2105.08149/) (2021)
- [Time Series Embedding Methods for Classification Tasks: A Review](https://mlsanity.com/paper/2501.13392/) (2025)
- [Self-Supervised Learning for Data Scarcity in a Fatigue Damage Prognostic Problem](https://mlsanity.com/paper/2301.08441/) (2023)
- [Systemic Decarbonization of Road Freight Transport: A Comprehensive Total Cost of Ownership Model](https://mlsanity.com/paper/2410.21026/) (2024)
- [Sensor-Driven Predictive Vehicle Maintenance and Routing Problem with Time Windows](https://mlsanity.com/paper/2412.04350/) (2024)

## Abstract (arXiv)

Runtime quantification of vehicle operational intensity is essential for predictive maintenance and condition monitoring in commercial and heavy-duty fleets. Traditional metrics like mileage fail to capture mechanical burden, while unsupervised deep learning models detect statistical anomalies, typically transient surface shocks, but often conflate statistical stability with mechanical rest. We identify this as a critical blind spot: high-load steady states, such as hill climbing with heavy payloads, appear statistically normal yet impose significant drivetrain fatigue. To resolve this, we propose a Dual-Stream Architecture that fuses unsupervised learning for surface anomaly detection with macroscopic physics proxies for cumulative load estimation. This approach leverages low-frequency sensor data to generate a multi-dimensional health vector, distinguishing between dynamic hazards and sustained mechanical effort. Validated on a RISC-V embedded platform, the architecture demonstrates low computational overhead, enabling comprehensive, edge-based health monitoring on resource-constrained ECUs without the latency or bandwidth costs of cloud-based monitoring.