Comparison of centralized and decentralized conflict resolution strategies for multiple-aircraft problems

Under the concept of Free Flight, future air traffic operations may permit the transfer of responsibility for various tasks, including conflict detection and resolution, from ground-based air traffic controllers to the flight crew. While aircraft-based conflict resolution is attractive from a user-preferences standpoint, the degree to which it can efficiently resolve complex conflicts among several interacting aircraft remains a significant issue. The objective of this study is to compare the performance of a decentralized conflict resolution strategy against that of a benchmark centralized conflict resolution strategy. A generalized methodology for such a comparison does not exist; therefore two qualitatively different conflict scenarios, characterized by complex interactions among multiple aircraft, were constructed to conduct simulation studies using an idealized model. Preliminary investigations indicate that a decentralized resolution strategy can successfully solve complex multiple-aircraft problems in real time, albeit with some performance degradation relative to a centralized strategy. Research Scientist, Automation Concepts Research Branch; Mail Stop 210-10; E-mail; kbilimoria@mail.arc.nasa.gov. Associate Fellow, AIAA. f Software Specialist, Raytheon ITSS. Research Assistant, Dept. of Aeronautics and Astronautics. Associate Professor, Dept. of Aeronautics and Astronautics; E-mail: feron@mit.edu. Senior Member, AIAA. Copyright © 2000 by the American Institute of Aeronautics and Astronautics, Inc. No copyright is asserted in the United States under Title 17, U.S. Code. The U.S. Government has a royalty-free license to exercise all rights under the copyright claimed herein for Governmental purposes. All other rights are reserved by the copyright owner. Introduction The advent of new technologies for Air Traffic Management (ATM) has the potential to change the operational paradigm of the air transportation system by delegating more responsibility to the flight deck for airplane trajectory management functions; this includes conflict detection and resolution (CDR this is an idealized representation of a situation where several free aircraft are simultaneously attempting to arrive at some terminal area entry point. The second scenario features two closelyspaced traffic streams (each consisting of several aircraft) intersecting at a common point; this is an idealized representation of aircraft flying along intersecting free routes in en route airspace. The performance of a reactive decentralized resolution method that provides sequential, pairwise conflict resolutions based on a geometric optimization approach10 is compared against that of a proactive centralized resolution method that provides a single coordinated conflict resolution based on a global optimization algorithm.5 Although determining a fully coordinated optimal solution to conflicts involving multiple aircraft is computationally challenging, the centralized conflict resolution method described in Ref. 5 is able to compute these solutions, which are used as benchmarks for the best performance that can be achieved. The performance of decentralized strategies is then compared against these benchmark solutions. This paper is organized as follows: First, the mathematical models used for aircraft trajectories and conflict avoidance are presented. The centralized and decentralized conflict resolution methods used in this study are then outlined. Next, performance measures for conflict resolution are defined, and two conflict scenarios are described. Finally, a comparison of the performance of the decentralized conflict resolution strategy relative to the centralized conflict resolution strategy is presented, along with some general observations about their characteristics.