Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Aiding autonomous underwater vehicles navigation using multistatic sonar data

 
: Daun, M.

5th Biennial Workshop on Digital Signal Processing for In-Vehicle Systems (DSP 2011) : Kiel, Germany, 4 - 7 September 2011
Red Hook, NY: Curran, 2014
ISBN: 978-1-62993-950-6
S.115-122
Biennial Workshop on Digital Signal Processing for In-Vehicle Systems (DSP) <5, 2011, Kiel>
Englisch
Konferenzbeitrag
Fraunhofer FKIE

Abstract
The use of Autonomous Underwater Vehicles (AUV) cooperating in a network for Anti Submarine Warfare (ASW) surveillance operations is of special interest. Research in this field is currently done e. g. at the NATO Undersea Research Centre NURC [1, 2, 14, 13, 12]. Each AUV has to localise targets (submarines) robustly and precisely by evaluating the target's sonar echoes. Avoiding detection, i. e. a covert operation, requires a minimisation of own emissions, like communication or navigation signals. For target localisation and racking, this is realised by a multistatic sonar configuration: The AUV is receiving the signals transmitted by acoustic sources located elsewhere. Our algorithm, called CoLT (Combined Localisation and Tracking), incoorperates multistatic sonar data into AUV navigation aiming at adding position measurements into the navigation filter. This increases robustness, because sufficient navigation is possible when the main navigation sensor, a Doppler velo city log (DVL), fails. It allows for covert operation since DVL sensors are based on emitting sonar signals. Furthermore, CoLT improves the target tracking accuracy by allowing to estimate time offsets between source and AUV as well as errors in the source positions. Research on CoLT started during a 2-month stay as visiting scientists at NURC in 2010. Therefore, we are able to evaluate CoLT with data from the sea trial GLINT'10 executed with NURC's AUVs and sensor arrays.

: http://publica.fraunhofer.de/dokumente/N-350238.html