Collaborative Research: A Simulation Environment to Enable Development of Adaptive Sampling Algorithms for Autonomous Underwater Vehicles

Virginia Polytechnic Institute And State University, Blacksburg VA

Investigators

Abstract

The PIs have requested a Small Grant for Exploratory Research (SGER) as the first stage in the development of an adaptive sampling algorithms tool which would enable an autonomous underwater vehicle (AUV) to determine the best locations, routes, and times to collect scientific data. The PIs believe that adaptive sampling algorithms have the potential to enable rapid, high fidelity mapping and tracking of chemical and biological plumes, fronts, small scale circulation phenomena, and other oceanographic features of scientific interest. The key feature of these algorithms is that sampling strategies are determined by autonomous vehicles based on data collected in real-time and without human intervention. In contrast to the current state of the art, in which AUVs run pre-programmed routes through areas of potential interest, adaptive sampling has the potential to dramatically increase the quality of collected data and to significantly decrease the time and expense required to collected it.