              |
MARINE ENVIRONMENT
Projects:
 Feature Based Navigation for AUV's in Very Shallow Water There are two main foci of this research: The first is the operation of vehicles within feature maps. Such maps are critical navigational tools especially in operations where traditional sensors may not be applicable, for example in cluttered environments. The on-line creation of maps, the generation of paths within maps, and the control of the vehicles from these maps are major themes. The second element is moving-baseline navigation, wherein several key vehicles maintain the absolute reference for a larger fleet; this paradigm allows drastic reduction in the system cost. A major theme is the combined navigation and control problem of such a distributed agent system, especially in the case of reconfiguration.
 Autonomous Underwater and Sea Surface Vehicles  Autonomous Navigation
Contact PI: John Leonard, Franz Hover
MEMS Pressure Arrays for Near-Field Flow Patterns This project will deploy inexpensive, low-power sensors, passively detecting dynamic and static pressure fields with sufficient resolution to detect objects and bodies generating the disturbance. The novel sensors consist of arrays of tens to hundreds of MEMS-based pressure micro-transducers. These sensors and processing software emulate and extend the capabilities of the lateral line in fish. The arrays will be capable of detecting near-field flow patterns and near- and far-body obstacles and vehicles, as well as mapping near-body objects. This will provide a unique capability for navigation in shallow-water and/or cluttered environments, for use with multiple AUVs, and for flow control in conventional and biomimetic vehicles.
 Lateral Line Sensing of Fish  Design of MEMS Pressure Array  Use of Passive MEMS Sensor Arrays
Contact PI: Michael Triantafyllou
Guided Wave Optics for Flow-Sensing MEMS Arrays This project will design low-cost high-performance integrated sub-wavelength optics ("nano-photonics") as a complement to electronic sensing in MEMS-based pressure micro-transducer emulations of the lateral line in fish. Optical sensing offers significant potential benefits over electronic sensing because the amount of required wiring is reduced and the large bandwidth of optical signals can provide better measurement accuracy. Recent developments in the technology of integrated sources and detectors also promise that within 2-3 years these elements will be widely available at low cost for integration in autonomous micro-systems.
Contact PI: George Barbastathis
Free-Space Optics for AUV Navigation and Map Generation This project will contribute high-performance imaging equipment, based on digital holography, to complement acoustic and flow sensing for vehicle navigation and map building. Advanced algorithms will be used to "clean up" the images and increase the optical imaging range under conditions of high absorbance and scattering.
 MIT-WHOI Aquatic Imaging Camera (Barbastathis, Millgram & Davis)  VHI: Real Time Optical Slicing & Imaging Spectroscopy
Contact PI: George Barbastathis
Algorithms for Adaptive Sampling in Coastal Zone Environment This project will develop and test algorithms for planning cooperative adaptive sampling of the coastal zone and ocean environments via multiple autonomous underwater vehicles and autonomous surface craft. This would include measurements in all modalities available from other parts of the project via sensors mounted on these vehicles as well as on buoys and on bottom mounted sensors, and would be in association with the data assimilation, modeling and forecasting activities of the project.
 Time-Progressive Path Planning of AUVs for Adaptive Sampling (2-Vehicle Case)
Contact PI: Nicholas M. Patrikalakis
Algorithms for Creation of Solid Models from AUV Sensing Systems In the context of ship hull and near shore or harbor maritime structure inspection, we plan to develop algorithms for the creation of geometric models of underwater solid features using different sensing modalities (acoustical, vision, laser as appropriate) based on sensors mounted on autonomous surface and underwater vehicles. Handling uncertainty is a major component of this work.
Contact PI: Nicholas M. Patrikalakis
High Resolution Modeling of Biosphere-Atmosphere-Ocean The modeling effort will use the Finite Volume Coastal Ocean Model (FVCOM) adapted to the configuration of the South China Sea (90E-140E; 20S-30N) with very fine space resolution in the basin surrounding Singapore. The simulations of the ocean circulation and property distributions (temperature, salinity, density) will be carried out under surface forcings of wind stress, heat and moisture fluxes and will be validated with altimetric, tidal and Sea Surface Temperature (SST) data. Three assimilation packages based on sequential data assimilation (Kalman filtering) have already been adapted to FVCOM. They will be used for the assimilation of different observations (altimetry, tide gauges, in situ data). Specific emphasis will be given to the design of fixed and adaptive observational arrays in the Singapore regional environment. FVCOM will also be coupled to the bottom boundary layer and sediment transport model developed by Prof. Madsen. The final goal and outcome of the project will be to achieve operational and real time assimilation and forecasting capabilities for the Singapore region and surrounding seas.
Contact PI: Paola Rizzoli
Related References:
Cyberinfrastructure for CENSAM During the initial year of the project, we intend to identify and work with a subset of CENSAM researchers to develop a prototype data archive that will tag and record real time geo-referenced data from sensors. The archive will optimize the availability of the data for present and future researchers through metadata tagging of the archived data and an organization that will optimize query processing. The initial design phase of the archive will require discussions with groups generating sensor data (Whittle, Patrikalakis), modelers fusing and processing the resulting data sets (Rizzoli, Entekhabi), and the computer graphics team at CAMTech (Mueller-Wittig). While this project will not generate new algorithms for data assimilation, the proposed architecture should be able to accommodate a variety of data assimilation and data fusion techniques.
Contact PI: Judson Harward
Related References:
|
 |
|
© 2008 Massachusetts Institute of Technology