Small Business Innovation Research/Small Business Tech Transfer
Massively Parallel Processing for Dynamic Airspace Configuration
Project Description
Through extensive research conducted by Mosaic ATM in the area of Dynamic Airspace Configuration (DAC), we have identified the significant benefit of the use of Dynamic Density (DD) as the DAC objective function. The use of DD as the objective function allows the DAC algorithm to directly address critical aspects of sector design beyond simple balancing of the flight counts. These sector design considerations include the alignment of sector boundaries with flow direction, proximity of conflict points to sector boundaries, and boundary alignment with respect to vertical traffic movement. By using DD as the objective function, we generate a multi-objective optimization approach that considers both efficiency and complex controller workload issues. The SectorFlow DAC algorithm has performed well in NASA's DAC algorithm comparison experiments. However, due to the additional computational complexity caused by the use of DD as the objective function, only limited application of DD as the objective function was conducted. In this proposed SBIR effort, Mosaic ATM will apply a massively parallel computing architecture to the DAC algorithm using DD as an objective function to demonstrate and evaluate both the computational advantages of massively parallel processing, and the benefits of using DD as the objective function in DAC.
More »
Anticipated Benefits
The FAA is considering Dynamic Airspace as a potential component of NextGen. However, it is clear that decision support tools will be required to facilitate evaluation of such dynamic airspace designs. As we have found in our DAC research, the required computational performance to algorithmically design and evaluate airspace sectors is significant. The results of this research could be applied by the FAA in a decision support tool for DAC. Modeling and simulation of Air Traffic Management operations is computationally intensive. However, the availability of real-time ATM modeling could be applied to numerous decision-making situations by both the FAA and by Flight Operators. Currently, the level of detailed modeling required to achieve beneficial and useful recommendations from such models in prohibitive. However, the results of this research may provide the necessary computational speed to overcome this obstacle and create the opportunity for commercial real-time ATM modeling and simulation tools.
The most likely Phase III activities involve further development of the SectorFlow software and algorithms to support NASA's continued aeronautics mission. The SectorFlow DAC algorithms have been analyzed by NASA in comparison to other DAC concepts, and SectorFlow's performance was found to be high. NASA would benefit from the ability to continue to conduct DAC research using a variety of DAC approaches, including the use of Dynamic Density as an objective function as has been implemented in SectorFlow. The application of massively parallel computing can also be applied by NASA to numerous aeronautics and Air Traffic Management algorithms and analysis efforts. This increase in computational power may allow the necessary increase in modeling samples required to generate robust decision support tool recommendations using a stochastic optimization approach. More »
The most likely Phase III activities involve further development of the SectorFlow software and algorithms to support NASA's continued aeronautics mission. The SectorFlow DAC algorithms have been analyzed by NASA in comparison to other DAC concepts, and SectorFlow's performance was found to be high. NASA would benefit from the ability to continue to conduct DAC research using a variety of DAC approaches, including the use of Dynamic Density as an objective function as has been implemented in SectorFlow. The application of massively parallel computing can also be applied by NASA to numerous aeronautics and Air Traffic Management algorithms and analysis efforts. This increase in computational power may allow the necessary increase in modeling samples required to generate robust decision support tool recommendations using a stochastic optimization approach. More »
Primary U.S. Work Locations and Key Partners
| Organizations Performing Work | Role | Type | Location |
|---|---|---|---|
| Mosaic ATM, Inc. | Lead Organization | Industry | Leesburg, Virginia |
Ames Research Center
(ARC)
|
Supporting Organization | NASA Center | Moffett Field, California |
Primary U.S. Work Locations
-
California
-
Virginia
Suggest an Edit
Recommend changes and additions to this project record.
This is a historic project that was completed before the creation of TechPort on October 1, 2012. Available data has been included. This record may contain less data than currently active projects.