Department News
Control in Information Rich World: Application to Air Traffic Control
Seminar Date
2005-07-27
Author
빈종훈
Date
2005-07-21
Views
1985
1. 제 목 : Control in Information Rich World: Application to Air Traffic Control
2. 연 사 : Prof. Inseok Hwang
School of Aeronautics and Astronautics Purdue University, West Lafayette, IN 47907, USA
3. 일 시 : 2005년 7월 27일 (수) 오후 4:00 - 5:00
4. 장 소 : 서울대 301동 117호
5. 내 용 :
The continued growth of air travel and recent advances in new technologies for navigation, surveillance, and communication have led to proposals by the Federal Aviation Administration (FAA) to provide reliable and efficient tools to aid Air Traffic Control (ATC) in performing their tasks. The current ATC has a rigid and centralized structure: aircraft fly along predefined airways between waypoints, and ground controllers direct aircraft using radar track and flight information from plan view displays and voice communication over radio channels. New technologies such as the Global Positioning System (GPS) for navigation and Automatic Dependent Surveillance-Broadcast (ADS-B) for communication, will enable automation of some of the ATC functions. These functions include multiple-target tracking and identity management and aircraft’s intent inference for air traffic surveillance, and conflict detection and resolution between aircraft for air traffic control. A set of algorithms and tools with provable properties has been developed: these algorithms may be used either in a fully autonomous way, or as supporting tools to increase controllers' situational awareness and to reduce their work load. In this presentation, four problems frequently encountered in air traffic surveillance and control will be discussed.
The first topic is a multiple-maneuvering-target tracking and identity management algorithm which can keep track of maneuvering aircraft in noisy environments and of their identities. Secondly, an intent inference algorithm that can estimate pilot’s flight intent is discussed. This inferred information can be used to conformance monitoring in which ground controllers or pilots to see if aircraft follows ATC regulations and flight plans. After 9.11, this functionality has become more important than ever. Thirdly, I discuss a hybrid probabilistic conflict detection algorithm between multiple aircraft which uses flight mode estimates as well as aircraft current state estimates. The proposed algorithm is based on hybrid models of aircraft, which incorporate both continuous dynamics and discrete mode switching. As such, the algorithm can provide advantages over existing conflict detection algorithms which use current state estimates from continuous dynamics only. Finally, an algorithm for multiple (greater than two) aircraft conflict avoidance is presented. The algorithm is based on a closed-form analytic solution. Using this analytic solution, I develop a multiple-aircraft conflict resolution protocol: a simple rule which is easily understandable and implementable by all aircraft involved in the conflict, and which provides guarantees of safety. I present simulation results using a dynamic aircraft model for various multiple aircraft conflict scenarios derived from actual air traffic data (Enhanced Traffic Management system data).
6. 약 력 :
Present - Assistant Professor, School of Aeronautics & Astronautics, Purdue University,
West Lafayette, Indiana, USA.
2004- Ph.D. the Department of Aeronautics and Astronautics, Stanford University
1994-97 instructor, the Department of Aerospace Engineering, Korea Air Force Academy, Korea
1994- M.S. Aerospace Engineering, Korea Advanced Institute of Science and Technology (KAIST)
1992- B.A. Aerospace Engineering, Seoul National University, Korea
7. 문 의 : 기계항공공학부 기 창 돈 (☏ 880-1912)
2. 연 사 : Prof. Inseok Hwang
School of Aeronautics and Astronautics Purdue University, West Lafayette, IN 47907, USA
3. 일 시 : 2005년 7월 27일 (수) 오후 4:00 - 5:00
4. 장 소 : 서울대 301동 117호
5. 내 용 :
The continued growth of air travel and recent advances in new technologies for navigation, surveillance, and communication have led to proposals by the Federal Aviation Administration (FAA) to provide reliable and efficient tools to aid Air Traffic Control (ATC) in performing their tasks. The current ATC has a rigid and centralized structure: aircraft fly along predefined airways between waypoints, and ground controllers direct aircraft using radar track and flight information from plan view displays and voice communication over radio channels. New technologies such as the Global Positioning System (GPS) for navigation and Automatic Dependent Surveillance-Broadcast (ADS-B) for communication, will enable automation of some of the ATC functions. These functions include multiple-target tracking and identity management and aircraft’s intent inference for air traffic surveillance, and conflict detection and resolution between aircraft for air traffic control. A set of algorithms and tools with provable properties has been developed: these algorithms may be used either in a fully autonomous way, or as supporting tools to increase controllers' situational awareness and to reduce their work load. In this presentation, four problems frequently encountered in air traffic surveillance and control will be discussed.
The first topic is a multiple-maneuvering-target tracking and identity management algorithm which can keep track of maneuvering aircraft in noisy environments and of their identities. Secondly, an intent inference algorithm that can estimate pilot’s flight intent is discussed. This inferred information can be used to conformance monitoring in which ground controllers or pilots to see if aircraft follows ATC regulations and flight plans. After 9.11, this functionality has become more important than ever. Thirdly, I discuss a hybrid probabilistic conflict detection algorithm between multiple aircraft which uses flight mode estimates as well as aircraft current state estimates. The proposed algorithm is based on hybrid models of aircraft, which incorporate both continuous dynamics and discrete mode switching. As such, the algorithm can provide advantages over existing conflict detection algorithms which use current state estimates from continuous dynamics only. Finally, an algorithm for multiple (greater than two) aircraft conflict avoidance is presented. The algorithm is based on a closed-form analytic solution. Using this analytic solution, I develop a multiple-aircraft conflict resolution protocol: a simple rule which is easily understandable and implementable by all aircraft involved in the conflict, and which provides guarantees of safety. I present simulation results using a dynamic aircraft model for various multiple aircraft conflict scenarios derived from actual air traffic data (Enhanced Traffic Management system data).
6. 약 력 :
Present - Assistant Professor, School of Aeronautics & Astronautics, Purdue University,
West Lafayette, Indiana, USA.
2004- Ph.D. the Department of Aeronautics and Astronautics, Stanford University
1994-97 instructor, the Department of Aerospace Engineering, Korea Air Force Academy, Korea
1994- M.S. Aerospace Engineering, Korea Advanced Institute of Science and Technology (KAIST)
1992- B.A. Aerospace Engineering, Seoul National University, Korea
7. 문 의 : 기계항공공학부 기 창 돈 (☏ 880-1912)