BEGIN:VCALENDAR
PRODID:-//University of Hawaii//UH Events Calendar//EN
VERSION:2.0
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Pacific/Honolulu
BEGIN:STANDARD
TZOFFSETFROM:-1000
TZOFFSETTO:-1000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CLASS:PUBLIC
CREATED:20260420T125819Z
DESCRIPTION:​This MS work aims to contribute to underwater acoustic sampling techniques through machine​ learning. It has two objectives: (1) Optimizing the sampling process for underwater sound fields​ and (2) Optimizing data assimilation for ocean acoustic tomography.​ \n\nTo address the first objective, we developed an approach that leverages autonomous underwater​ vehicles (AUVs) to sample unknown sound fields. Unlike fixed sensor networks with spatial​ constraints, AUVs can make real-time decisions and adaptively survey a region. The proposed​ algorithm adaptively samples a survey region based on the sound field characteristics. It uses an​ active learning strategy based on Gaussian Process (GP) regression to characterize a static sound​ field in a survey region. With each location​ sampled, the algorithm employs a GP to estimate the​ field and quantify the uncertainty in the predicted sound field. The uncertainty metric is used to​ choose the next sampling location. This dynamic approach maximizes the information gained by​ the AUV at the locations that it samples. It also ensures efficient convergence toward the true​ distribution of underwater static sources in the sample region. Our algorithms were developed​ via simulation and were validated with a controlled experiment in a swimming pool [work​ funded by the NSF AI Institute in Dynamic Systems].\n\nFor the second objective, we aimed to optimize the process of integrating acoustic data into​ ocean models via ocean acoustic tomography. Ocean acoustic tomography derives water column​ properties from acoustic observations. It traditionally uses ray tracing and requires making\nfrozen ray approximations that can be limiting in some cases. Another approach involves​ iteratively updating sound speed profiles and re-running sound propagation models until the​ modeled travel times agree with measured travel times. However, this approach is​ computationally expensive. To optimize the iterative approach, we developed a machine-learning​ pipeline to map perturbations in sound speed profiles to corresponding changes in acoustic ray​ paths. The 2010–11 North Pacific Acoustic Laboratory (NPAL) Philippine Sea experiment​ dataset was used to develop and validate a neural network. To constrain the model’s learning to​ small perturbations in ocean states and observed acoustic travel times, sound speed profiles were​ decomposed using empirical orthogonal functions, with principal components used for training,​ while ray paths were represented as Fourier functions. The proposed neural network focuses on the variabilities in sound speed profiles and ray paths so that a predicted decomposed ray can be​ obtained for small changes in the ocean state [work funded by the Office of Naval Research].\n\nZoom meeting ID: 832 8033 9161, Passcode: 2025
DTEND;TZID=Pacific/Honolulu:20250423T210000Z
DTSTAMP:20260420T125819Z
DTSTART;TZID=Pacific/Honolulu:20250423T190000Z
LAST-MODIFIED:20260420T125819Z
LOCATION:POST 723 & Zoom (Please see description for Meeting ID & Passcode)
PRIORITY:5
SEQUENCE:0
SUMMARY;LANGUAGE=en-us:Ocean and Resources Engineering Lecture
TRANSP:OPAQUE
UID:177672589943880web-support-l@lists.hawaii.edu
END:VEVENT
END:VCALENDAR