Inversion of the sound speed profiles with an AUV carrying source using improved ensemble Kalman filter

Estimation of range‐dependent sound‐speed profile with dictionary learning method

“…In general, the SSP is closely related to the depth profiles of temperature and conductivity in the water body [5]. A conductivity‐temperature‐depth (CTD) device was usually deployed to measure the temperature and electric conductivity at different depths, from which to retrieve the SSP.…”

“…A conductivity‐temperature‐depth (CTD) device was usually deployed to measure the temperature and electric conductivity at different depths, from which to retrieve the SSP. However, it is impractical to monitor the SSP with such devices in a vast area over long time period [5, 6]. Alternative methods include an acoustic inversion approach to monitor rapidly changing SSP [5], travel‐time data collected with an ocean acoustic tomography (OAT) technique [7] or matched field inversion [8–10].…”

“…However, the requirement of orthogonality limits its efficacy of regularisation and resolution [11, 12]. Schmidt et al proposed an acoustically focused oceanographic sampling method to improve the resolution under rapidly changing environment [14, 5]. A conventional OAT method was applied first to get low‐resolution results, followed by dispatching a moving sensor like autonomous underwater vehicle (AUV) to high‐uncertainty areas to collect small‐scale high‐resolution data [5].…”

“…Schmidt et al proposed an acoustically focused oceanographic sampling method to improve the resolution under rapidly changing environment [14, 5]. A conventional OAT method was applied first to get low‐resolution results, followed by dispatching a moving sensor like autonomous underwater vehicle (AUV) to high‐uncertainty areas to collect small‐scale high‐resolution data [5]. However, significant Doppler shift from high‐frequency sources may compromise the data, even at slow moving speed of sensor [9].…”