Abstract
Synthetic aperture sonar (SAS) provides high-resolution acoustic imaging by processing coherently the backscattered signal recorded over consecutive pings as the bearing platform moves along a predefined path. Coherent processing requires accurate estimation and compensation of the platform's motion for high quality imaging. The motion of the platform carrying the SAS system can be estimated by cross-correlating redundant recordings at successive pings due to the spatiotemporal coherence of statistically homogeneous backscatter. This data-driven approach for estimating the motion of the SAS platform is essential when positioning information from navigational instruments is absent or inadequately accurate. Herein, the problem of platform motion estimation from coherence measurements of diffuse backscatter is formulated in a probabilistic framework. A variational autoencoder is designed to disentangle the ping-to-ping platform displacement from three-dimensional (3D) spatiotemporal coherence measurements. Unsupervised representation learning from unlabeled data offers robust 3D platform motion estimation. Including a small amount of labeled data during training improves further the platform motion estimation accuracy.
Highlights
Synthetic aperture sonar (SAS) utilizes the motion of the platform carrying the sonar system to synthesize an aperture that is much longer than the physical sonar antenna by coherently combining data from several pings
Synthetic aperture sonar (SAS) provides high-resolution acoustic imaging by processing coherently the backscattered signal recorded over consecutive pings as the bearing platform moves along a predefined path
The problem of platform motion estimation from coherence measurements of diffuse backscatter is formulated in a probabilistic framework
Summary
Synthetic aperture sonar (SAS) utilizes the motion of the platform carrying the sonar system to synthesize an aperture that is much longer than the physical sonar antenna by coherently combining data from several pings. This study proposes a machine learning approach for platform motion estimation in SAS from spatiotemporal coherence measurements based on probabilistic variational inference.. A factorized representation of independent latent features correlates better with the ground-truth generative factors of variation in the data and offers physically interpretable abstractions.22,26,27 To achieve such disentangled representations, b-VAE introduces an adjustable regularization parameter b in the objective function of the VAE optimization task, which controls the relative importance between the data likelihood and the resemblance of the approximate posterior to the prior latent distribution, which favours independent latent variables.. The results indicate that a b-VAE trained exclusively with unlabeled data learns to disentangle the latent features that cause the main variation in the coherence data, providing robust estimation of the complete motion of the platform.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
