MIT AATT Model Project

BDT: Banc De Test

(Last Update: NP 02/96)

1. Primary Model Category

Simulation tool which generates aircraft trajectories to test automated conflict resolution algorithms.

2. Summary

The Banc de Test tool (BDT) was developed at Centre d'Etudes de la Navigation Aérienne (CENA) as a support tool in the AGACER project (Algorithmes Génétiques Appliqués au Contrôle En Route). The main process of BDT uses aircraft flight plans and simplified dynamics to generate trajectories in a given airspace. It can be used alone to detect and count conflicts (i.e. horizontal or vertical separation violations), or used as a testbench for an independent conflict resolution module.

Competing models include RAMS,TAAM, and ASIM.

3. Input Requirements

  • Location of navigation beacons in the airspace;
  • Basic aircraft performance data for each aircraft type;
  • Flight plans containing a list of navaids and the requested flight level.

4. Outputs

The standard outputs of the model are:
  • Departure time, arrival time and delay for each flight.
  • Number of airplanes in the air and their altitudes, at 5 minutes intervals.
  • For each conflict: the miss distance, the aircraft involved and their positions, speeds, altitudes,...just before and just after the separation violation.
Additional code has been written to extract higher level conflict information and statistics from these standard outputs, but it is not well documented and it is not easy to use.

5. Major Assumptions

Aircraft trajectories are simplified:
  • Aircraft climb directly to the cruise flight level at a constant speed and rate of climb;
  • Airspeed and altitude are constant during the cruise segment;
  • Aircraft descend directly to their destination at a constant speed and rate of descent; Terminal areas and airport capacity are not modeled.
Trajectory uncertainties are modeled as uncertainty on ground speed and climb/descent rates.

Additional assumptions are usually made in the conflict resolutions modules.

6. Computational Characteristics

The source code was written in C and was available for the evaluation. It is well structured and well documented.

Hardware requirements:

  • platform: the evaluation was done on a Sun Sparc 5.
  • operating System: Unix.
  • memory: 32 Mb of RAM, 10 Mb of hard drive.
Software requirements:

GNU C compiler.

Documentation:

No formal user’s guide is available. This is mostly a research tool. A document describing the structure and functions of the model was used for the evaluation. However this document does not contain complete information on configuration files and input/output formats.

User interface:

Very limited. The program runs as a batch file.

Typical run time:

A few minutes without conflict resolution. Considerably more with conflict resolution, depending on the algorithm. Typical algorithms include Genetic Algorithms.

7. Startup Effort

The program is easy to run in less than a week.

8. Modularity and Flexibility

The source code is modular and well organized. This is an open-architecture software. Different conflict resolution schemes can be selected without modifying the source code. The source code was made available to MIT.

9. Status

The model is continually evolving. It is currently used by French civil aviation research groups to test several automated conflict resolution schemes (including optimization by Genetic Algorithms).

10. Extent of model validation

Unknown.

11. Principal applications

The principal application of this model is currently the evaluation of tactical conflict resolution algorithms.

12. Availability

Upon request to Jean-Marc Alliot, CENA, FRANCE.

13. Information for model evaluation

Source code, simulation runs, and discussions with Jean-Marc Alliot, Centr d'Études de la Navigation Aérienne.

14. Contact points

Jean-Marc Alliot
Centre d'Études de la Navigation Aérienne
7 avenue Édouard Belin
31055 TOULOUSE CEDEX
Phone: 011-3362-17-4054
Email: alliot@pc-allt.eis.enac.dgac.fr

15. Summary Evaluation

BDT is a modular program which allows to test new automated conflict resolution schemes at the tactical level. However, it is not a system-wide model, and it could not be readily used to validate Air Traffic Control concepts (e.g. Free Flight). In particular, it does not presently take into account Air Traffic Flow Management, terminal areas, airport capacities and weather.


Back to completed models page