Estimating mechanical impedance from hydrophone measurements.

Area of Science:

• Acoustics
• Mechanical Engineering
• Signal Processing

Background:

• Integrating machine learning into autonomous underwater vehicles faces challenges like accuracy trade-offs and limited training data.
• Accurate identification of underwater objects is crucial for various applications, including defense and environmental monitoring.

Purpose of the Study:

• To introduce a reduced-order identification approach for thin spherical shells using sparse hydrophone measurements.
• To explore the potential of mechanical impedances as unique identifiers for underwater scatterers.
• To develop a method enabling efficient, real-time object identification for autonomous underwater vehicles.

Main Methods:

• Analytically reconstruct the scattered acoustic field from sparse hydrophone measurements.
• Obtain surface pressures and velocities for each spherical harmonic.
• Compute the in vacuo mechanical impedance to identify the object's properties.

Main Results:

• The Modal Mechanical Impedance Estimation (MMIE) method accurately approximates the first two modal mechanical impedances with less than 10% error in the low-frequency range (ka ≤ 2.1).
• The method requires only 10 hydrophones for effective identification.
• Performance is limited by directivity at higher frequencies.

Conclusions:

• MMIE provides an efficient, training-free approach for underwater object identification.
• The method is suitable for real-time and low-energy applications in autonomous underwater vehicles.
• This technique offers a promising alternative to traditional machine learning methods for underwater acoustics.