Scanning the Ocean with Motion Tomography
Abstract
The first goal for this project is to develop an algorithm called motion tomography(MT), a novel way to construct GEMs autonomously and systematically. The method fusesthe data collected by multiple mobile platforms along their paths to formulate an inverse problemthat has been the core problem underlying medical CT imaging. By solving this inverse problem, ahigh-resolution spatial map of ocean flow in the volume traversed by the mobile platforms can bereconstructed. Modeling and predicting ocean currents are great challenges for physical oceanography due to the lack of direct measurements. Mobile sensor networks have been proven to be an effective tool to answer this challenge, providing estimated flow information along the Lagrangian trajectories. To incorporate these flow estimates into ocean models, existing approaches based on data assimilation usually require significant amount of computing power. Generic environmental models (GEMs) developed by the PI combine computational ocean models with real-time data streams collected by mobile sensing platforms to provide high-resolution predictions of ocean current in a small spatial area around the mobile platforms. The GEM provides data assimilation algorithms an Eulerian view of the flow, which is much easier to assimilate than flow along Lagrangian trajectories. GEM can also provide real-time support for navigational control algorithms to plan the paths of the mobile platforms. But the actual trajectory of the mobile platforms will differ from the planned path with an offset. This offset, named the controlled Lagrangian prediction error (CLPE), reflects the differences between the GEM and the actual flow. GEM has shown great potential that needs to be further advanced towards an agile maritime autonomy. The first goal for this project is to develop an algorithm called motion tomography (MT), a novel way to construct GEMs autonomously and systematically. The method fuses the data collected by multiple mobile platforms along their paths to formulate an inverse problem that has been the core problem underlying medical CT imaging. By solving this inverse problem, a high-resolution spatial map of ocean flow in the volume traversed by the mobile platforms can be reconstructed. While a similar inverse problem has been formulated and solved in ocean acoustic tomography to reconstruct spatial maps of sound speed, motion tomography provides a directly measured Eulerian map of ocean current, which has never been achieved through other means before. MT may significantly increase the spatial resolution of GEMs. The second goal of this project is to extend CLPE to evaluate the accuracy of GEM and its capability to support feedback control laws used for navigation of marine robots. Under feedback control, CLPE can be bounded, especially near the Lagrangian coherent structures of the flow field. The CLPE can be computed and evaluated using the GEM for different feedback control laws. The bounded CLPE can serve as a performance metric for the accuracy of the GEM. This performance metric can then be used to decide the optimal structure of a GEM so that the CLPE can be reduced. More accurate GEMs also feed higher quality data to data assimilation algorithms, hence eventually improve operational ocean models used by the Navy. PI has obtained promising experimental results validating MT and the effectiveness of GEM using underwater gliders in ocean sensing experiments. The project will analyze a significant amount of data collected by underwater gliders over a period of four years to evaluate the performance of MT and the accuracy of GEM in realistic ocean experiments.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Nov 23, 2016
- Source ID
- N000141612667
Entities
People
- Fumin Zhang
Organizations
- Georgia Tech Research Corporation
- Office of Naval Research
- United States Navy