A Domain Knowledge Informed Approach for Anomaly Detection of Electric Vehicle Interior Sounds

2025-09-16

AI Summary

In-cabin sound anomaly detection for electric vehicles faces three key challenges: scarcity of fault annotations, difficulty in selecting appropriate unsupervised models, and unreliable evaluation metrics. Method: This paper proposes a domain-knowledge-guided unsupervised learning framework. Its core innovation is a surrogate anomaly generation mechanism based on structured perturbations of healthy-spectrum templates—replacing scarce real fault samples for robust model validation and selection. Leveraging acoustic feature extraction, spectrogram reconstruction, and advanced audio synthesis techniques, the framework synthesizes high-fidelity anomalous audio. Contribution/Results: Experiments across five representative fault types demonstrate substantial improvements over conventional unsupervised model selection strategies. Furthermore, the authors publicly release the first high-quality in-cabin audio dataset for electric vehicles, establishing a benchmark to advance research in this emerging domain.

Technology Category

• Machine Learning: Unsupervised & Self-Supervised Learning
• Natural Language Processing: Safety and Robustness
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs

Abstract

The detection of anomalies in automotive cabin sounds is critical for ensuring vehicle quality and maintaining passenger comfort. In many real-world settings, this task is more appropriately framed as an unsupervised learning problem rather than the supervised case due to the scarcity or complete absence of labeled faulty data. In such an unsupervised setting, the model is trained exclusively on healthy samples and detects anomalies as deviations from normal behavior. However, in the absence of labeled faulty samples for validation and the limited reliability of commonly used metrics, such as validation reconstruction error, effective model selection remains a significant challenge. To overcome these limitations, a domain-knowledge-informed approach for model selection is proposed, in which proxy-anomalies engineered through structured perturbations of healthy spectrograms are used in the validation set to support model selection. The proposed methodology is evaluated on a high-fidelity electric vehicle dataset comprising healthy and faulty cabin sounds across five representative fault types viz., Imbalance, Modulation, Whine, Wind, and Pulse Width Modulation. This dataset, generated using advanced sound synthesis techniques, and validated via expert jury assessments, has been made publicly available to facilitate further research. Experimental evaluations on the five fault cases demonstrate the selection of optimal models using proxy-anomalies, significantly outperform conventional model selection strategies.

Problem

• Detecting anomalies in electric vehicle cabin sounds
• Overcoming unsupervised model selection without labeled fault data
• Using proxy-anomalies for improved anomaly detection validation

Innovation

• Domain-knowledge-informed model selection method
• Proxy-anomalies via structured spectrogram perturbations
• Validation using engineered synthetic fault samples