ALBERT

Description: Most decisions made in the cockpit are related to safety, and have therefore been proceduralized in order to reduce risk. There are very few which are made on the basis of a value metric such as economic cost. One which can be shown to be value based, however, is the selection of a flight profile. Fuel consumption and flight time both have a substantial effect on aircraft operating cost, but they cannot be minimized simultaneously. In addition, winds, turbulence, and performance vary widely with altitude and time. These factors make it important and difficult for pilots to (a) evaluate the outcomes associated with a particular trajectory before it is flown and (b) decide among possible trajectories. The two elements of this problem considered here are: (1) determining what constitutes optimality, and (2) finding optimal trajectories. Pilots and dispatchers from major u.s. airlines were surveyed to determine which attributes of the outcome of a flight they considered the most important. Avoiding turbulence-for passenger comfort-topped the list of items which were not safety related. Pilots' decision making about the selection of flight profile on the basis of flight time, fuel burn, and exposure to turbulence was then observed. Of the several behavioral and prescriptive decision models invoked to explain the pilots' choices, utility maximization is shown to best reproduce the pilots' decisions. After considering more traditional methods for optimizing trajectories, a novel method is developed using a genetic algorithm (GA) operating on a discrete representation of the trajectory search space. The representation is a sequence of command altitudes, and was chosen to be compatible with the constraints imposed by Air Traffic Control, and with the training given to pilots. Since trajectory evaluation for the GA is performed holistically, a wide class of objective functions can be optimized easily. Also, using the GA it is possible to compare the costs associated with different airspace design and air traffic management policies. A decision aid is proposed which would combine the pilot's notion of optimality with the GA-based optimization, provide the pilot with a number of alternative pareto-optimal trajectories, and allow him to consider unmodelled attributes and constraints in choosing among them. A solution to the problem of displaying alternatives in a multi-attribute decision space is also presented.

Description: Most decisions made in the cockpit are related to safety, and have therefore been proceduralized in order to reduce risk. There are very few which are made on the basis of a value metric such as economic cost. One which can be shown to be value based, however, is the selection of a flight profile. Fuel consumption and flight time both have a substantial effect on aircraft operating cost, but they cannot be minimized simultaneously. In addition, winds, turbulence, and performance x,ary widely with altitude and time. These factors make it important and difficult for pilots to (a) evaluate the outcomes associated with a particular trajectory before it is flown and (b) decide among possible trajectories. The two elements of this problem considered here are (1) determining, what constitutes optimality, and (2) finding optimal trajectories. Pilots and dispatchers from major U.S. airlines were surveyed to determine which attributes of the outcome of a flight they considered the most important. Avoiding turbulence-for passenger comfort topped the list of items which were not safety related. Pilots' decision making about the selection of flight profile on the basis of flight time, fuel burn, and exposure to turbulence was then observed. Of the several behavioral and prescriptive decision models invoked to explain the pilots' choices, utility maximization is shown to best reproduce the pilots' decisions. After considering more traditional methods for optimizing trajectories, a novel method is developed using a genetic algorithm (GA) operating on a discrete representation of the trajectory search space. The representation is a sequence of command altitudes, and was chosen to be compatible with the constraints imposed by Air Traffic Control, and with the training given to pilots. Since trajectory evaluation for the GA is performed holistically, a wide class of objective functions can be optimized easily. Also, using the GA it is possible to compare the costs associated with different airspace design and air traffic management policies. A decision aid is proposed which would combine the pilot's notion of optimility with the GA-based optimization, provide the pilot with a number of alternative pareto-optimal trajectories, and allow him to consider un-modelled attributes and constraints in choosing among them. A solution to the problem of displaying alternatives in a multi-attribute decision space is also presented.

Description: NASA Lewis Research Center's Icing Research Tunnel (IRT) is the world's largest refrigerated wind tunnel dedicated to the study of aircraft icing. In the IRT, natural icing conditions are duplicated to test the effects of in-flight icing on actual aircraft components and on scale models of airplanes and helicopters. The IRT's ability to reproduce a natural icing cloud was significantly improved with the recent installation of a new spray bar system. It is the spray bar system that transforms the low-speed wind tunnel into an icing wind tunnel by producing microscopic droplets of water and injecting them into the wind tunnel air stream in order to accurately simulate cloud moisture.

Description: IUE low dispersion observations were made of the T(sub c)-deficient peculiar red giant (PRG) star, HDE 332077, to test the hypothesis that T(sub c)-poor PRG's are formed as a result of mass transfer from a binary companion rather than from internal thermal pulsing while on the asymptotic red giant branch. Previous ground-based observations of this star indicated that it is a binary, but the secondary star was too massive for an expected white dwarf. A deep, short wavelength prime (SWP) exposure was needed to search for evidence of an A-type main-sequence companion. We obtained a 120 minute LWP exposure (LWP 23479), followed by a collaborative 1230 minute SWP exposure (SWP 45113). These observations were combined with our earlier IUE and optical data on this PRG star to model the spectral energy distribution of the system.

Description: A new spray bar system was designed, fabricated, and installed in the NASA Glenn Research Center's Icing Research Tunnel (IRT). This system is key to the IRT's ability to do aircraft in-flight icing cloud simulation. The performance goals and requirements levied on the design of the new spray bar system included increased size of the uniform icing cloud in the IRT test section, faster system response time, and increased coverage of icing conditions as defined in Appendix C of the Federal Aviation Regulation (FAR), Part 25 and Part 29. Through significant changes to the mechanical and electrical designs of the previous-generation spray bar system, the performance goals and requirements were realized. Postinstallation aerodynamic and icing cloud calibrations were performed to quantify the changes and improvements made to the IRT test section flow quality and icing cloud characteristics. The new and improved capability to simulate aircraft encounters with in-flight icing clouds ensures that the 1RT will continue to provide a satisfactory icing ground-test simulation method to the aeronautics community.

Description: Major upgrades were made in 1999 to the 6- by 9-Foot (1.8- by 2.7-m) Icing Research Tunnel (IRT) at the NASA Glenn Research Center. These included replacement of the electronic controls for the variable-speed drive motor, replacement of the heat exchanger, complete replacement and enlargement of the leg of the tunnel containing the new heat-exchanger, the addition of flow-expanding and flow-contracting turning vanes upstream and downstream of the heat exchanger, respectively, and the addition of fan outlet guide vanes (OGV's). This paper describes the rationale behind this latest program of IRT upgrades and the program's requirements and goals. An overview is given of the scope of work undertaken by the design and construction contractors, the scale-model IRT (SMIRT) design verification program, the comprehensive reactivation test program initiated upon completion of construction, and the overall management approach followed.

Description: This paper describes mainly the 1997 version of the Solar EUV Rocket Telescope and Spectrograph (SERTS-97), a scientific experiment that operated on NASA's suborbital rocket flight 36.167-GS. Its function was to study activity on the Sun and to provide a cross calibration for the CDS instrument on the SOHO satellite. The experiment was designed, built, and tested by the Solar Physics Branch of the Laboratory for Astronomy and Solar Physics at the Goddard Space Flight Center (GSFC). Other essential sections of the rocket were built under the management of the Sounding Rockets Program Office. These sections include the electronics, timers, IGN despin, the SPARCS pointing controls, the S-19 flight course correction section, the rocket motors, the telemetry, ORSA, and OGIVE.