TransLab is a Laboratory  of Computational Model for Air Transportation in the Dept. of Computer Science of the University of Brasilia - UnB with the support from the Brazilian Science and Technology Committee - CNPq. The main research interest is Artificial Intelligence in Air Traffic Management (ATM).

P003.gif (2881 bytes)

 P003.gif (2881 bytes)

1. Objective

In TransLab, some projects have been developed to form a new solution for Air Traffic Flow Management (ATFM) problem. The main objective of our research is to improve computational efficiency for ATFM by combing computing methods with multi-agent techniques. The research is supported by CNPq and in the cooperation with ITA, Poli/USP and CINDACTA I of Brazilian Air Force. The TransLab is also supported by Atech.

  • Project of CNPq/UnB  (proc. 306065/2004-5): "Inteligente Computation Method to Balance the Communication among Multi-Agents in Real Time Traffic Synchronization", 2005 - 2011.
  • Project of CNPq/UnB (proc. 4739462009-2): "Evaluation and decision support module (MAAD): computational tool applied to Air Traffic Flow Management", 2010 - 2012.
  • Project of FINEP/ATECH/UnB (proc. 2680/09):"Take-off Management System at Airports - SIGEP",  2010-2012.
  • Project of CNPq/UnB (proc. 485940/2007-8): "Interface of Inteligente Computation Method to Balance the Communication among Multi-Agents in Real Time Traffic Synchronization", 2007 - 2009.
Computer - UnB
Transport - UnB
CEFTRU - UnB
University - UnB
ITA
INPE

 

Publications

Artificial Brain Book

Neural Networks Course
Thesis of Graduate Students
 
ITS Conference

 




2. Team

  • Dr. Li Weigang, Coordinator of TransLab, Associate Professor of CIC/UnB;

  • Dr. Alba Cristina Magalhães Alves de Melo, Researcher, Associate Professor and Coordinator of PPGI/UnB;

  • Maj Av Antonio Marcio Ferreira Crespo, Coordinator (Operation), Chief of ACC-BS, CINDACTA I, Graduate student of PPGI/UnB;

  • Alessandro Leite, Researcher, Graduate student (Dr.) of PPGI/UnB;

  • Junior Arruda, Researcher, Graduate student of PPGI/UnB;

  • Victor Rafael R. Celestino, Invited Researcher, Graduate student (Dr.) of PPGT/UnB;

  • Leandro Cruciol, Researcher, Graduate student of PPGI/UnB;

  • Darcy Zheng, Researcher, Graduate student of PPGI/UnB;

  • Cicero Almeida, Researcher, Graduate student (Dr.) of PPGI/UnB;

  • Vitor Filincowsky, Researcher, Graduate student of PPGI/UnB.

     Consultant

  • Brig. Av Carlos Vuyk de Aquino, CISCEA.

  • Dr. Cláudio Jorge Pinto Alves, Professor of IEI/ITA.

  • Dr. Antonio Pedro Timoszczuk, Researcher of Atech.



3. Research idea

  • The tactical Air Traffic Flow Management (ATFM) has a direct impact on the Air Traffic Control (ATC) activities. The most appropriate configuration suggests the implementation of a decision making process in which an automated agent provides suggestions to human agent. There are metrics to evaluate the efficiency air traffic flow control measures generated by a multi-agent automated system using reinforcement learning algorithms. However, the implications of the ATFM on the air traffic control (a critical and complex activity) and on the controllers’ workload, recommend the adoption of decision supporting ATFM computational tools which take into account the Airspace Complexity (or ATC Complexity) as a critical criterion for the evaluation of the scenario. This research has the objective of presenting a methodology for the evaluation of the ATFM scenario which aggregates the Airspace Complexity criterion. The new evaluation function is proposed to determine the awards in reinforcement learning. The research includes a case study which focused on the air scenario observed in the Brazilian airspace. The comparative results of the application of the scenario evaluation function for each criteria: aircraft counting and ATC complexity are promising. It is possible to verify that the scenarios with the same number of aircraft presented a state value with variations about 10% between these two functions (with or without Airspace Complexity). These variations have a remarkable influence on the reinforcement learning performance and, consequently, on the suggestions that the decision making supporting system offers to the traffic flow manager.

4. Publications

WEIGANG, L. ; Alves, D. P. ; DIB, M. V. P. ; Crespo A. Intelligent  computing methods in Air Traffic Flow Management. Transportation Research. Part C, Emerging Technologies, ISSN: 0968-090X, v. 18, p. 781-793, 2010.

Weigang, L., Bueno B. de Souza, Antonio M. F. Crespo, Decision support system in tactical air traffic flow management for air traffic controllers, Journal of Air Transport Management, ISSN: 0969-6997v. 14, pp. 329-336, 2008.

Marcos Vinicius Pinheiro Dib, Li Weigang, Alba Cristina Magalhães Alves de Melo, Approach to Balance the Communication among Agents for Traffic Synchronization, IEEE Latin America Transactions, Vol. 5, pp. 338-345, 2007.

Li Weigang, Marcos Vinicius Pinheiro Dib, Alba Cristina Magalhães Alves de Melo, Claudio Jorge Pinto Alves, A Method to Balance the Communication among Multi-Agents in Real Time Traffic Synchronization, in the Journal of Brazilian Air Transportation, Vol. 2, No. 2, 63-82, 2006.

Daniela Pereira Alves, Li Weigang, Bueno Borges de Souza, “Uso de aprendizagem por reforço e controle em nível meta para melhorar a performance da comunicação em gerência de tráfego aéreo”, em Anais do V Simpósio de Transporte Aéreo - SITRAER, Li Weigang, Erico Santana (Eds), pp.337-344, Brasília, 2006.

Li  Weigang, Claudio J. P. Alves, and Nizam Omar, “An expert system for Air Traffic Flow Management”. Journal of Advanced Transportation”, ISSN: 0197-6729, Vol. 31, pp. 343-361, 1997.

Li  Weigang, Claudio J. P. Alves, and Nizam Omar, “Knowledge-Based System For Air Traffic Flow Management: Timetable Rescheduling And Centralized Flow Control“, In the Applications of Artificial Intelligence in Engineering VIII, Vol. 2, pp. 655-670,  Computational Mechanics Publications & Elsevier Science, Paris, 1993.

  5. Events

VIII Simpósio de Transporte Aéreo - SITRAER, Sao Paulo, 2009.

VII Simpósio de Transporte Aéreo - SITRAER, Rio de Janeiro, 2008.

I Fórum dos Aeroportos de Excelência da América do Sul - FAEXAS,  Brasilia, 2006.

 

 


 


Address

University of Brasilia - UnB
Department of Computer Science - CIC
Caixa Postal 4466
70919-970 Brasilia - DF
Brazil 

Tel: (061) 31076379; Fax: (061) 3273.3589

Initial publication : 09-15-95; Last updated: 12-09-2010.