AUV-based seabed target detection and tracking

Abstract

The GOATS (Generic Oceanographic Array Technology Systems) Joint Research Program explores the development of environmentally adaptive autonomous underwater vehicles (AUV) specifically directed toward Rapid Environment Assessment and Mine Counter Measurement (MCM) in coastal environments. As part of the effort, MIT is developing the GOATS multi-static sonar concept which uses a low-frequency source on one AUV to sub-critically insonify the seabed over a wide area, while a formation of multiple AUVs are used for mapping the associated 3D scattered acoustic field in the water column. Based on the significantly different characteristics of the scattered field of various buried targets, an online algorithm has been developed for concurrent detection and tracking the buried seabed targets. The use of subcritical angle isonification essentially results in the measurement of the reception of weak signals from the buried targets. Tracking these targets in such a scenario encounters the difficulties of detecting the targets in the presence of extremely low signal-to-noise ratio. Furthermore, combining with the uncertainties of navigation of AUVs presents another challenge for detection and tracking the targets. The method proposed in this work applies the techniques of Track-Before-Detection (TBD) to solve this problem. This technique tracks the targets first using the slowly changing environment information, and then the weak signal detection is declared after confidence of the track estimation is established. However, enormous possible trajectories of AUVs needed to be searched while the TBD algorithm is applied, which makes direct online implementation of this technique impossible. A dynamic programming (DP) algorithm is introduced to solve the highly interconnected stochastic network which TBD creates in a much more efficient way. Therefore, together with the DP algorithm, the TBD algorithm is feasible to implement online. This new algorithm has been applied on the GOATS 2000 bistatic data. The result shows successful detection of three buried targets where two of those are relatively weak targets. The result also shows the satisfactory performance of the algorithm in simultaneously detecting multiple, mixed strong and weak targets online.