2023_programme: Self-supervised learning of platform motion in synthetic aperture sonar with coupled autoencoders

Back


  • Session: 16. Trends and Advances in Array Signal Processing
    Organiser(s): Angeliki Xenaki, Peter Gerstoft and Eliza Michalopoulou
  • Lecture: Self-supervised learning of platform motion in synthetic aperture sonar with coupled autoencoders
    Paper ID: 1937
    Author(s): Xenaki Angeliki, Pailhas Yan, Monti Alessandro
    Presenter: Xenaki Angeliki
    Abstract: 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 antenna by coherently combining data from several pings. Coherent processing requires platform motion estimation with sub-wavelength accuracy to achieve high-resolution SAS imaging. Micronavigation, i.e., platform motion estimation from spatio-temporal coherence measurements of diffuse backscatter on overlapping recordings between successive pings, is essential when positioning information from navigational instruments is absent or inadequately accurate. Representation learning with variational autoencoders offers an unsupervised data-driven micronavigation solution. Herein, we exploit the multistatic arrangement of a wideband SAS system and use coupled autoencoders for self-supervised learning of platform motion to improve the accuracy of the micronavigation solution.
  • Corresponding author: Dr Angeliki Xenaki
    Affiliation: Centre for Maritime Research and Experimentation STO-NATO
    Country: Italy
    e-mail:

Back