Publications

In the design of a tracking filter for air traffic control (ATC) applications, a maneuvering aircraft can be modelled by a linear system with random noise accelerations. A Kalman filter tracker, designed on the basis of a variance chosen according to the distribution of the potential maneuver accelerations, will maintain track during maneuvers and provide some improvement in position accuracy. However, during those portions of the flight path where the aircraft is not maneuvering, the tracking accuracy will not be as good as if no acceleration noise had been allowed in the tracking filter. In this paper, statistical decision theory is used to derive an optimal test for detecting the aircraft maneuver: a more practical suboptimal test is then deduced from the optimal test. As long as no maneuver is declared, a simpler filter, based on a constant-velocity model, is used to track the aircraft. When a maneuver is detected, the tracker is reinitialized using stored data, up-dated to the present time, and then normal tracking is resumed as new data arrives. In essence, the tracker performs on the basis of a piecewise linear model in which the breakpoints are defined on-line using the maneuver detector. Simulation results show that there is a significant improvement in tracking capability using the decision-directed adaptive tracker.

This report investigates the requirements imposed on a surveillance system for supporting independent approaches to closely spaced parallel runways. Based on a proposed procedure for monitoring aircraft approach paths and controlling deviations from proper approach paths, the required spacing between runway centerlines is derived as a function of surveillance system characteristics and other parameters. Potential trade-offs between the surveillance system characteristics are then investigated to determine whether the DABS sensor might be utilized for position measurement and/or communication in such a surveillance system. The results indicate that the required runway spacing is more sensitive to delays and data update intervals than to position measurement accuracies, and that, if DABS is to perform the communication function in the system, it should probably be used for position measurement as well.

A concept formulation study of the control aspects of the fourth generation air traffic control system is presented. The results of this study are not strongly influenced by present-day equipment. They are influenced by certain aspects of present airspace utilization and procedures which appear necessary for the design of an effective system. The inputs to the control system design include the fourth generation air traffic demand, characteristics of fixed elements (types of aircraft, etc.), and disturbances such as weather effects. The control system which has been formulated includes flight plan generation, flow control, conformance monitoring, and collision avoidance as control functions. A baseline control system is given as a first iteration of the fourth generation system. The baseline system is defined by classifying types of airspace, conformance requirements, and required segregation of classes of flight paths. The airspace is divided into three categories: positive control air space containing only controlled aircraft, controlled (mixed) air space containing both controlled and cooperative aircraft, and uncontrolled airspace containing uncontrolled aircraft. Cooperative aircraft must be able to accept IPC co-ands as well as simplified flight plans when flying in high density retied air space. The surveillance, navigation, and communications systems complete the interacting parts of the control system. Candidate fourth generation system concepts ranging from the completely tactical to the highly strategic have been described both in this report and elsewhere. In order to characterize a proposed concept we have drawn up a list of decisions which we find must be made in the course of a flight. We then consider where these decisions are made and thereby characterize the system. The feasibility of generating conflict free flight plans is investigated with the aid of analytical models. A consideration of the factors which influence the flight planning process is presented. Use is made of a generally accepted traffic density model for the 1995 time period. The expected number of conflicts for selected routes and the distances required to resolve conflicts are evaluated. The use of aircraft performance characteristics in evaluating the effectiveness of conflict resolution maneuvers is discussed. The level of conformance necessary for conflict free flight plans is determined for each maneuver. For cases in which the required conformance was unrealistically high, it was determined that providing velocity structure in high density airspace permitted a decrease in conformance requirements. Factors which directly influence the capability of aircraft to conform to flight plans in a strategic system as well as the relevant technology areas peculiar to the implementation of conflict free flight plans are considered. The conclusions reached during this study are followed by recommendations for future work in specific areas.