Detection of Water on Road Surface with Acoustic Vector Sensor

Area of Science:

• Acoustics
• Environmental Sensing
• Smart City Technology

Background:

• Road surface condition monitoring is crucial for traffic safety.
• Existing methods for detecting road wetness have limitations.
• Acoustic sensing offers a potential alternative for real-time road state assessment.

Purpose of the Study:

• To develop and evaluate a new acoustic-based method for detecting water on road surfaces.
• To utilize sound intensity analysis in the frequency domain for road wetness detection.
• To assess the performance of the proposed method against traditional approaches and reference devices.

Main Methods:

• Employing an acoustic vector sensor to capture sound signals from the road.
• Analyzing sound intensity in the frequency domain to identify acoustic events from vehicles.
• Calculating the direction of incoming sound to filter out irrelevant noise.
• Estimating road surface state based on the sound intensity spectrum and comparing it to a threshold for wetness detection.

Main Results:

• The proposed algorithm achieved 89% precision, recall, and F1 score in real-world tests.
• The method successfully detected the presence of water on the road surface.
• The acoustic sound intensity approach outperformed traditional sound pressure analysis methods.

Conclusions:

• The developed acoustic method is effective for detecting water on road surfaces.
• This technology can be integrated into smart city monitoring systems using acoustic sensors.
• The findings support the use of sound intensity analysis for environmental sensing applications.