Self-localization of a mobile swarm using noise correlations with local sources of opportunity

Abstract

Groups of coordinated underwater vehicles or sensors are powerful tools for monitoring the ocean. A requirement of many coordinated surveys is to determine a spatial reference between each node in a swarm. This work considers the self-localization of a swarm of independently moving vehicles using acoustic noise from a dominating incoherent source recorded by a single hydrophone onboard each vehicle. This method provides an inexpensive and infrastructure-free spatial reference between vehicles. Movement between the vehicles changes the swarm geometry and a self-localization estimate must be generated from data collected on short time scales. This challenges past self-localization approaches for acoustic arrays. To overcome this challenge, the proposed self-localization algorithm jointly estimates the vehicle geometry and the directionality of the ambient noise field, without prior knowledge of either estimate. To demonstrate this method, experimental results are provided when a boat is the main dominating source. The results demonstrate the ability to both estimate the direction of arrival of the boat and the relative positions of the vehicles in the swarm. The approach in this paper is not limited to moving vessels. Simulations are provided to examine three different factors that affect the proposed solution: inter-vehicle motion, vehicle geometry, and the azimuthal variance of the noise field.

Geometry estimation for multiple autonomous underwater vehicles using active sound source

“…On the one hand, we compared the performance of different methods in time delay estimation, including GCC, FS‐GCC, the method used by Naughton.et al. [1] and the proposed FS‐GCC + autoencoder under the same SNR (−13 dB); On the other hand, to evaluate the dependent conditions of far‐field hypothesis, we analysed the influences of different factors to the final geometry estimation accuracy, including rotation angles of sound source, the distance between sound source and the geometry, and SNR. Anechoic pool experiment. Based on the experiment results in simulation data, we use FS‐GCC + autoencoder to estimate the time delay from the dataset collected from the anechoic pool, and further estimate the geometry using SVD. Lake trial.…”

“…In this experiment, we used different methods to estimate the time delay, including GCC [10], FS‐GCC, FS‐GCC + U‐Net [38], the method adopted in ref. [1], and the proposed FS‐GCC + autoencoder method. All the estimated time delay matrix are sent into SVD step for the final geometry estimation.…”

“…Left to right (a and f, b and g, c and h, d and i, e and j) corresponds to generalise cross correlation (GCC), the method used in ref. [1], frequency‐sliding generalised cross‐correlation (FS‐GCC), FS‐GCC + U‐Net in ref. [38] and FS‐GCC + Autoencoder respectively.…”

“…2) Compared with the method used in ref. [1], the RMSE of the proposed method drops by 9% and 54% for two different geometries respectively. 3) Although FS‐GCC is helpful in alleviating the influence of noise through sub‐band low‐pass filtering, the introduction of autoencoder network can further improve the accuracy and robustness of delay estimation.…”

“…Under the far‐field assumption, Naughton et al. [1] took the radiated noise of a moving ship as the dominant noise, calculated the time delay between multiple AUVs with cross correlation and estimate the geometry by decomposing the time delay matrix with SVD.…”