A robust sequential hypothesis testing method for brake squeal localisation.

Area of Science:

• Acoustics
• Automotive Engineering
• Signal Processing

Background:

• Brake squeal is a common issue in automobiles, originating from specific regions near the wheels.
• Accurate detection and localization of brake squeal are challenging due to complex and time-varying acoustic environments, sensor inaccuracies, and modeling difficulties.
• Existing methods often struggle with these uncertainties, relying on ad hoc formulations.

Purpose of the Study:

• To develop and validate a novel approach for in situ detection and localization of brake squeal in automobiles.
• To systematically address model uncertainties inherent in complex acoustic environments and sensor imperfections.
• To improve the consistency and reliability of brake squeal localization compared to prior state-of-the-art methods.

Main Methods:

• Utilizing distributed microphone arrays placed under the automobile to capture sound field directional properties.
• Formulating hypothesis tests based on spatial characteristics of the sound field.
• Developing a new approach with a stronger theoretical foundation to systematically account for model uncertainties, decoupling detection and localization.

Main Results:

• The proposed approach demonstrates greater consistency in real-world tests compared to previous methods.
• The study successfully decouples the tasks of brake squeal detection and localization for improved complexity management.
• Effective methods for brake squeal frequency detection and identification are presented.

Conclusions:

• The new method provides a more robust and theoretically sound framework for addressing uncertainties in brake squeal localization.
• The findings contribute to improved automotive diagnostics and noise reduction strategies.
• Practical considerations for microphone array design and placement are also discussed.