Prev-Talk | Next-Talk | All-Talks | Talks-Sorted

Talk-20: Model-based Adaptive Acoustic Sensing and Communication in the Deep Ocean with MOOS-IvP

Henrik Schmidt, Toby Schneider, MIT Laboratory for Autonomous Marine Sensing Systems (LAMSS), Center for Ocean Engineering

The use of autonomous underwater vehicles as platform for passive and active acoustic sensing provides the possibility of adaptively changing the heading, speed and depth for optimal sensing performance. For example, by measuring the local ambient noise directivity, the autonomy can choose a heading which minimizes the interference with the source of interest, or the vehicle can use its measured sound speed profile to choose an optimal depth for sensing or communication with other assets. The latter type of environmental adaptation requires that robust features of the acoustic environment are identified and modeled. In shallow water this is possible to a very limited degree, e.g. selecting a depth which is on the same side of the thermocline, whereas other features focusing the acoustic energy are in general unreliable due to fluctuations in the acoustic environment. In deep water, on the other hand, the deep pressure gradient of the sound speed is extremely stable and may be robustly exploited. Thus, a platform operating near the bottom in the deep ocean will have a direct acoustic path to and from a shallow receiver or source within the so-called RAP cone (Reliable Acoustic Path), extending to ranges of approximately 30 km. However, by moving up in the water column, the upward refractive sound speed profile will extend the direct path up to 60 km range. This so-called convergence zone path is extremely stable and predictable, and an optimal depth for acoustically connecting with a shallow source or receiver can be modeled very robustly. An IvP-Helm behavior has been developed that uses as an objective function the depth-dependent transmission loss or SNR, modeled onboard the vehicle. The behavior has not yet been field tested, but the performance will be demonstrated using the MIT-LAMSS simulation environment [Work supported by DARPA].


DOWNLOAD the brief given at MOOS-DAWG'11.

Categories:

  • Autonomy / Collaborative Autonomy
  • MOOS-IvP
  • Acoustic Communications
  • Unmanned Underwater Vehicles (UUVs) / Autonomous Underwater Vehicles (AUVs)
  • Anti-Submarine Warfare
  • IvP Helm Behavior Development