|
|
| |
Research: Greater Capacity |
|
Project Summaries:
The following are the NEXTOR research projects in alignment with FAA Flight Plan Goal 2: Greater Capacity.
Strategic Traffic Flow Models based on Data-Mining and System-Identification
Techniques
NEXTOR Team: A. Bayen, UCB
The objectives of this project are: 1) develop strategic traffic flow models for the
National Airspace System (NAS) based on historical data using a data-driven/data mining
approach, 2) use these relationships to improve the demand forecasting models and 3)
adjust these models in real-time via observation of NAS data. Subsequently, the models
can be used for designing optimal flow control strategies to achieve the desired
demand-capacity balance in the NAS.
View Project Website
Air Traffic Flow Management under Uncertainty and Dynamics Conditions
Industry Partner: Volpe Center
NEXTOR Team: Amedo Odoni, MIT
Develop an objective methodology to measure information complexity associated
with Air Traffic Management displays. One of the major shortcomings
of the decision support tools that the FAA currently uses for Air
Traffic Flow Management (TF<) is that they assume that the
predictions made by the Enhanced Traffic Management System (ETMS)
are always accurate. That is, no explicit account is take of the
uncertainty in the predictions and the fact that these predictions
are updated over time. Traffic managers must therefore allow for
uncertainty by making intuitive adjustments based on their
experience and expertise. Because this is not a satisfactory way to
deal with the ubiquitous problem of “dynamic” uncertainty, one of
the recent research directions in TFM has been to determine how to
satisfactorily incorporate this complexity into TFM decision support
tools.
The proposed work will focus specifically on uncertainty in airport acceptance rates (AARs)
under dynamic conditions. This uncertainty is most commonly caused
by uncertainty over weather, but it could have other causes. It is
expected that the ideas that arise from this work would have more
general application to uncertainty over airport demand and also
uncertainty in en route problems.
Collaborative Decision Making
NEXTOR Team: M. Ball, UMD; M. Hansen, UCB; T. Trani, VPI; J-P. Clarke, MIT
Develop models to evaluate the impact of various sources of traffic flow uncertainty:
arrival flow uncertainty, weather, and congestion in order to
minimize delays and maximize airspace.
JPDO Policy
NEXTOR Team: A. Weigel, MIT
Assess the policy and architecture interdependencies of the National Air
Transportation System to assist JPDO in the formulation and
implementation of a successful technology policy strategy.
Using Real Options and System Dynamics to Evaluate Bond Issues for Aviation
Infrastructure Improvements
NEXTOR Team: J-P. Clarke, MIT
Despite the potential benefits to all stakeholders, the adoption of new
technology is often hindered by a lack of consensus within industry,
between industry and government, and within government regarding the
infrastructure improvements that need to be made and the timetable
for making the requisite improvements. The FAA faces a challenge in
modernizing the National Airspace System (NAS).
NEXTOR plans to study whether the commitment by government to uninterruptible
long-term financing for the requisite infrastructure improvements
can be provided through bonds. Specifically, we propose to study
the feasibility of issuing bonds for the development and deployment
of ADS-B infrastructure, and then restructuring the NAS so that
ADS-B users who have paid a subscription can reap the benefits of
early adoption—thereby incentive for others to adopt ADS-B
technology.
Enhancements to SIMMOD
Sponsoring Agency: ASD-430
NEXTOR Team: T. Trani, VPI; G. Gosling and M. Hansen, UCB; E. Feron, MIT; P. Schonfeld, UMD
NEXTOR Research Report RR-97-8 “Enhancements to SIMMOD: A Neural Network
Post-processor to Estimate Aircraft Fuel Consumption” by A.A. Trani
and F.C. Wing-Ho, was completed and delivered to the FAA in December
1997. It details the findings of a study conducted at Virginia Tech
to improve the accuracy and flexibility of SIMMOD’s fuel burn
postprocessor. A neural network model was developed to estimate fuel
consumption of sample aircraft. Results were compared to the actual
performance provided in the aircraft performance manual and found to
be accurate within 2%. The model developed can be implemented in
SIMMOD and other fast-time simulation programs.
NEXTOR Research Report RR-97-9, “Development of an On-Site Ground Operations Model
for Logan International Airport,” by E. Feron and B. Declare was
completed and delivered to the FAA in December 1997. It details the
findings of a study conducted at MIT to model Boston Logan Airport’s
ground operations. Using SIMMOD was particularly challenging in
modeling Logan Airport’s operations due to the complex layout of
runways and taxiways.
The Feasibility of Using Low Earth Orbit
(LEO) Satellite Systems for Air Traffic Control
Communications
Sponsoring Agency: AUA-570
NEXTOR Team: M. Ball, UMD
LEO satellite systems such as Iridium are now becoming operational. These systems
offer significant advantages over traditional (GEO) systems that
employ satellites in geosynchronous orbits. Of particular interest
to aeronautical communications is the drastic reduction in
communications delay. Since LEO systems would seem to have natural
advantages for air-ground communications, various efforts have been
initiated to develop standards and requirements for their use in
this setting. The NEXTOR project is focusing on the evaluation of
various hybrid communications architectures in which LEO satellite
systems are used to augment ground-based systems. Some of the roles
being investigated for LEO systems are as follows:
- providing coverage where ground-based systems are not feasible, e.g., over
water and in remote areas such as parts of Alaska,
- augmenting the capacity of ground-based systems, and
- reducing redundancy requirements for ground based system.
The NEXTOR effort will concentrate on comparing alternatives by evaluating system-wide
metrics of various architectures.
Louisville Area Navigation Development
NEXTOR Team: J-P. Clarke, MIT
Develop a continuous-descent approach design to minimize noise interruptions
in United Parcel Service landings at Louisville Airport.
Complexity Models and Metrics for the Support of Air Traffic
Management Tools &
Operations—Dynamic Density and Resectorization
Sponsoring Agency: FAA
NEXTOR Team: J. Hansman, MIT
This project recognizes that airspace and traffic
complexity as being key limitations in the current operation
of the NAS. For reasons of safety, it is important that the level
of traffic complexity in any sector does not exceed the capabilities
of the controllers to safely and reliably manage traffic. Because
of a lack of understanding of the real basis for cognitive
complexity in air traffic control, only the crudest metrics for
complexity (e.g., number of aircraft in a sector) are used to manage
complexity.
The goal of this research is to first
identify and, second, potentially adapt methods from other
disciplines as complexity metrics. These modeling approaches would
then be evaluated for their applicability to current airspace and
airway structures. These same evaluation tools would be appropriate
for reporting changes in airspace or operations, both in the short
term (e.g., Dynamic Resectorization) and in the long term (e.g.,
Airspace or Airway Redesign). One unique aspect of the proposed
research is that it will be conducted in parallel with CENA in
France.
Development of Massport Planning and
Tactical Response Capabilities for Irregular Operations
Sponsoring Agency: Massachusetts Port Authority
NEXTOR Team: John-Paul Clarke, MIT
Members of the NEXTOR project, Development of
Massport Planning and Tactical Response Capabilities for Irregular
Operations, seek to determine how Logan Airport interacts with other
agencies in the National Airspace System (NAS). Team members are
interested in exploring how changes or disruptions in the activities
of other agents are likely to affect Logan Airport. Assessing the
quality and quantity of information available for irregular
operations planning, the sources of that information, and the
prediction tools available, is vital to the examination of the
relationship between Logan Airport and the other NAS agents.
Researchers are interested in determining the infrastructure
required for real-time adaptive operations. The ultimate objective
of the project is to demonstrate the potential benefits of
monitoring the activities of other NAS agents through implementation
of a prototype prediction methodology in controlled case studies.
CPDLC Benefits Assessment and Extensibility Analysis
Sponsoring Agency: Massachusetts Port Authority
NEXTOR Team: M. Hansen, J. Rakas, UCB; T. Trani, D. Teodorovi, VPI; J. Hansman, MIT
The Virginia Polytechnic Institute and the University of California, Berkeley
will serve as co-leads in this research project. The Massachusetts
Institute of Technology will serve in a consulting role with the
University of Maryland contributing with the participation of a
post-doctoral researcher at UCB.
Two areas of research will be explored.
1. Area 1 will set the baseline on how operations in the nonintegrated FFPI
environment in 2005 will be conducted. Deficiencies and problems
caused by the non-integrated nature of the toolsets will be
described. Initially, there will be an analysis of benefits of the
independent implementations of the decisions support tools URET and
TMA and the implementation of the data link. Research focus will be
in the extrapolation of how the benefits would occur in an
integrated data link—decision support tool environment.
2. Research area 2 is to assess the benefits of the implementation of a data link
service integrated with the passive FAST decision support tool,
which will be in place in the 2005 time frame as part of FFP1. This
implementation of data link services will represent the first
application of air traffic control data link in a terminal
environment. The challenge will be synthesizing, postulating and
defending the benefits from utilization of both tools, the FAST
decision support tool and the data link service in a terminal
environment.
Impact of AATT Technologies on Air Traffic Management Concept
Definition
Sponsoring Agency: NASA Ames
NEXTOR Team: J. Hansman, A. Odoni, MIT; M. Hansen, A. Kanafani, UCB
Industry Partners: Boeing; Seagull Technologies
This project involved two tasks: the first
task was to define and document the probable evolution of the NAS
through the year 2015, based on current documents and on-going work
by the FAA, NASA, and industry. The work involved interaction with
major ATM stakeholders and documented the definition of the ATM
system evolution as of the end of the reporting period. The second
task studied the impact of technologies being developed under the
NASA Advanced Air Transportation Technology (AATT) program on the
overall definition of the future ATM system concept. This task
defined and documented the systems concepts being developed under
the AATT program and documented how these concepts would be
implemented within the context of current ATM modernization plans.
More information about the themes in this
project can be found in NEXTOR Research Reports RR-97-3, “Air
Traffic Management Concept Baseline Definition” and RR-97-4,
“National Airspace System Stakeholder Needs.” These reports are
available through the NEXTOR program office at UC Berkeley.
Investigation on the Integration of Airfield and Airspace Simulation
Models through an Open
Systems Architecture
Sponsoring Agency: ATAC
NEXTOR Team: M. Hansen, G. Gosling, UCB; A. Trani, J. Kobza, H. Sherali, VPI
This NEXTOR project, funded by industry
partner ATAC Corporation, seeks to review the current state of the
art of incorporating open architecture principles into existing
airport and airspace simulation models, and develop recommendations
for the most appropriate way to provide these capabilities in future
versions of the FAA Airport and Airspace Simulation Model (SIMMOD).
The issues associated with providing users with the capability to
intervene in the logic of such models during model execution and
access intermediate data flows are being explored through a case
study approach addressing the impact of new air traffic control
technology (the Center-TRACON Automation System) and improved ground
movement logic using the current version of SIMMOD.
The research is divided into three phases.
The first phase reviewed the current state of the art of open
architecture principles in existing airspace and airfield simulation
models and implementation issues of how to integrate inputs from
various sources. This phase also included a review of a proposed
open system architecture termed the SIMBUS concept, as well as
development of a detailed work plan for the case study analysis in
Phase 2. The second phase includes analysis of existing SIMMOD
model structure, the case study analysis, and development of
preliminary recommendations for the best way to provide an open
architecture within the SIMMOD code. Reports documenting the
findings and recommendations of the second phase are currently being
finalized. The third phase, which has commenced, is exploring the
implementation of open architecture techniques for airport and
airspace simulation by examining how to implement controller and
pilot behavior modules being developed under a related research
project funded by NASA using SIMMOD PRO!. This phase of the
research will define common constructs for these and other future
modules, and compare these constructs to those adopted in other
airspace simulation models that currently use an open architecture
approach, such as the FAA National Airspace System Simulation Model
or the Euro control Reorganized ATC Mathematical Simulator.
Probabilistic Weather Forecasts and Decision Models to Support
Ground Delay Program
Planning at San Francisco Airport
Sponsoring Agency: AUA
NEXTOR Team: M. Ball, UMD
Industry Partner: NCAR
Uncertainty related to both air traffic demand
and the capacity of airspace and airport resources represent very
significant challenges to effective air traffic flow management.
One of the principal objectives of the collaborative decision making
(CDM) effort has been to improve information accuracy, which, in
turn, reduces the level of demand and capacity uncertainty.
Recently, a prototype forecast product was installed at SFO. This
product specifically addresses forecasting fog burn-off times. In
this project, NEXTOR is developing methods for producing probability
distribution functions for the airport acceptance rate from this new
forecast product. It is also investigating and testing methods for
integrating these distribution functions into the CDM decision
support models.
|
| |
|
|
|