[33] The control system expertise that Ames engineers were beginning to acquire went beyond variable stability aircraft. Using the new capabilities available with electronic systems, efforts were expanded into the areas of guidance, control, and displays, and a number of aircraft were adapted for those purposes (table 6). A particular demand for this technology at the time involved precision tracking of a target aircraft. In order to understand the contributions of the aircraft's response to precision tracking, a series of flight tests was performed in which the tracking performance of two straight-wing fighters, the P-51H and the F8F-1, was compared with that of two swept-wing candidates, the F-86A and F-86E. Each aircraft used a fixed gunsight. George Rathert and Burnett Gadeberg were the principal engineers responsible for these tests; the results were reported in reference 76.

Sometime later, a lead-computing sight was evaluated in the F-86D in another development effort and test carried out by Gadeberg and Rathert. It was observed that the pilots were able to compensate for a wide range of stability and control characteristics without any variation in tracking performance when using this sight (refs. 77 and 78). Until this time, another aircraft had acted as the "target" for these tracking tests. As a way of cutting costs and improving flight safety, Ames engineers developed a method of simulating the maneuvering target aircraft using equipment on the tracking aircraft itself. In this case, the pilot tracked a spot of light projected onto the windscreen as described in reference 79 and shown in figure 73. The spot....

[34] ....would move in the same manner as the image of a maneuvering target; the main drawback of this simulation was that the tracking pilot could not perceive the "attitude" of the target, and could not anticipate the next maneuver. Brian Doolin and G. Allan Smith were in charge of this part of the program, and Fred Drinkwater was the pilot. This target simulator was eventually adapted for simulations of the guidance of a radio-controlled missile, the Bullpup, with Joseph Douvillier and John V. Foster responsible for its development (ref. 80). The TV-1 (a Navy trainer version of the P-80) aircraft served as the test bed for this simulator, and the simulator proved so successful that it was eventually used to train Navy pilots in Bullpup operations. Fred Drinkwater again was the principal pilot for these evaluations. As a consequence of this success, the team developed a target simulator for the Air Force's E-4 radar scope presentation fire-control system. ²¹ Foster led the effort, including the installation on the F-86D. Flight results showed that the target simulator duplicated the attack phase for an actual airborne target, and further that it might be useful for pilot training as well (ref. 81).

Control systems were eventually developed to allow remote piloting of one aircraft from another. ²² This was first demonstrated when an SB2C-5 was remotely flown....

[35] ....from an F6F-5, in a project led by Howard Turner and John White. The mother aircraft (the F6F-5) was flown by Rudy Van Dyke. The beep control proved generally satisfactory for remote-control flight tests, including takeoff and landing, except for rollout after touchdown in crosswinds (ref. 82). Although the Navy lost interest in this kind of remotely flown vehicle, the control system installed in the SB2C-5 was adapted to another test. In this case, the aircraft was flown as a sort of "radar- controlled" interceptor; the pilot visually tracked the target with a periscope, and the aircraft would in turn respond to the motions of the periscope to track the target. This system was tested in simulation before flight. Howard Turner, William Triplett, and John White carried out the experiment (ref. 83).

The F-84F aircraft was lent to Ames for a brief time for fixed-sight tracking tests. Although documentation is sketchy, it appears that the F9F-8 Cougar aircraft was at Ames to evaluate the application of an A-1 sight to a "toss-bombing" technique. The F-102A and F-106A tests that involved Ames consisted of evaluations of the fire-control and auto-maneuvering systems. The fire-control system used in the F-106A was designated MA-1. One of the F-102A aircraft flew with an adaptive control normal acceleration command system that was able to maintain consistently satisfactory response characteristics from landing approach to low supersonic speeds at altitude (ref. 84).

Research on artificial vision for landing marked the beginning of display work at Ames. This project involved the evaluation of a television display of the forward scene in the R4D-6 with variations in field of view. The experiment was carried out by Bernard Kibort and Fred Drinkwater (ref. 85). (Similar conceptual ideas are being explored currently for application to a next-generation High Speed Civil Transport.) Years later (1981), in a program of major significance, a head-up guidance display was demonstrated at Ames in a Boeing 727-100 airplane operated by the FAA. This research was motivated by a series of wind-shear-induced landing accidents in the mid-1970s and became a key element in a joint NASA-FAA investigation of the use of head-up displays for landing approach.

The display concept, conceived and developed by Richard (Dick) Bray (ref. 86), consisted of a flightpath-centered, pursuit-tracking presentation in which the primary controlled element represented the direction of flight of the aircraft. The aircraft was flown by directing the flightpath symbol at outside references such as the intended touchdown point or, for instrument flight operations, at appropriate guidance elements in the display. An example of the HUD image during a visual approach can be seen through the windscreen of the B-727 in figure 74. This concept underwent extensive development by Bray in the flight simulators at Ames and proved to be successful, in part because of the availability of inertial measurements that were sufficiently accurate for the pilot to directly observe and control the flightpath. Earlier HUD tests, carried out on one of the Center's C-8A Buffaloes, had been unsuccessful because the attitude sensors were not sufficiently accurate and because it....

....was necessary to rely on angle-of-attack sensors to derive flightpath information. The display has subsequently been developed by industry for application to commercial transports and has been certificated by the FAA for operation to low-visibility minimums. Federal Express adopted the display for its cargo operations, and several airlines, including Alaska, Southwest, Delta, and United, now have or are preparing to put this display into service. It has been developed further in simulation and successfully demonstrated in flight experiments for V/STOL fighters and transports and for rotorcraft by Vernon (Vern) Merrick and Charles (Charlie) Hynes. Gordon Hardy made significant contributions to this work, both from his background as a test pilot and as an engineer, including participation in the FAA flight program to certify the display for commercial operation. For his contributions to flight safety, Dick Bray received the Adm. Luis de Florez Air Safety Award in 1984.

Research in inertial navigation was pursued to provide the improved accuracy demanded for operations, particularly in the terminal area, including automatic landings. These efforts grew out of Stanley Schmidt's research in application of the Kalman filter to inertial navigation and to its implementation in airborne computers. The system was initially flown on the Convair 340 in tests conducted at the White Sands Missile Range in a cooperative program between Ames and the NASA Manned Spacecraft Center and the Army Instrumentation Directorate at White Sands. This [37] navigation scheme was of particular interest to the space shuttle program and to White Sands for their tracking systems. ²³ Leonard McGee led the program and Gordon Hardy and Glen Stinnett carried out the flight operations. Accuracies sufficient to meet stringent automatic landing requirements were demonstrated and reported in reference 87. Investigations of time-constrained area navigation (4D RNAV) for short takeoff and landing (STOL) aircraft were also pursued on the CV-340. The objective of this research was to reduce the amount of airspace used for STOL operations compared with their conventional transport counterparts. This work validated Heinz Erzberger and Homer Lee's concept for the on-board computation of trajectories that would produce the desired time of arrival at the final approach fix. The 4D RNAV was tested extensively in piloted simulation and then implemented in a digital avionics system (STOLAND) developed for eventual use in an Ames STOL research aircraft. In the flight program, led by Lee with Hardy as the project pilot, time of arrival within 5 seconds of the target was typically achieved. Results also pointed out steps in design of the guidance algorithms that were required to reduce workload for manual control (ref. 88). Erzberger summed up the status of the on-board flightpath generation techniques in reference 89. Later, applications were made to STOL aircraft operations, including flight tests on the QSRA conducted by Charlie Hynes and Erzberger.

In conjunction with this area of research, it should be noted that an integrated digital flight management, guidance and navigation system was developed by an industry team from Honeywell and King Radio under the direction of George Callas and Dallas Denery and demonstrated on a Cessna 402B for general aviation applications (ref. 90). In addition, Charles Jackson conducted tests on DME/DME (distance measuring equipment) navigation and touch-panel displays on this aircraft. Also, in the early 1990s, precision landing guidance research using satellite-based navigation was pursued on the Beech Super King Air by David McNally and Russell Paielli with Rick Simmons as project pilot. The navigation system was interfaced with the aircraft's approach guidance and autopilot system, and coupled approaches were flown to altitudes of 50 feet. Data that quantified the accuracy of the differential global positioning system and landing guidance algorithms were obtained from these flights and, in cooperation with Stanford University, demonstration programs were performed by American and United airlines to assess their readiness for commercial operations (ref. 91).

In order to carry out research in guidance and control, it was necessary at some point to obtain credible mathematical models of the aircraft's response characteristics for use in analyzing the system designs. This required either measures of the aircraft's open-loop response to control inputs or identification of the important aerodynamic force and moment characteristics that determine the aircraft's response. Wind tunnel data could provide reasonable estimates of static stability derivatives such as angle-of-attack or directional stability coefficients, but few facilities were available from which rotational stability derivatives such as pitch, roll, or yaw damping from ground-based....

...tests could be obtained. Thus, flight test methods were developed to extract this information during dynamic maneuvers of the aircraft. ²⁴ Early efforts at Ames involved the work of William Triplett in the 1950s to acquire overall measures of the aircraft's frequency response from which transfer functions relating the aircraft's important state variable to its individual controls could be defined (ref. 92). Dallas Denery later made contributions to the identification of the aircraft's individual stability and control characteristics (ref. 93). Rodney Wingrove and Ralph Bach devised methods to extract the best estimate of the aircraft's motions in the presence of uncertainties owing to measurement inaccuracies or external disturbances to the aircraft (ref. 94). The aircraft state estimation methods were ultimately used extensively in aircraft accident and incident analyses by the National Transportation Safety Board, particularly for determining the wind environment in which an aircraft was operating. In recent years, Mark Tischler returned to the use of frequency-response methods to extract transfer functions and stability and control derivatives from flight records and has developed an analysis program (ref. 95) that has found wide use in the aircraft industry and military services.

A number of the aircraft that were used in this research are shown in figures 75-88. Flight research personnel in 1959 are shown in figure 89, and their counterparts in guidance and navigation are shown in figure 90.

[41]

 

Figure 90. Guidance and navigation personnel circa 1969. Front row: Dick Kurkowski, Michele Hilliard, Brent Creer, Grace Webster, Fred Edwards. Second row: Rod Wingove, Bedford Lampkin, Armando Lopez, Del Watson. Third row: Fred Boltz, Clark White, Gordon Hardy, Don Smith. Back row: Hank Lessing, Dallas Denery, Dick Acken, Bob Coate.

 

---

21. John V. Foster 1998: personal communication.

22. Howard Turner 1998: personal communication.

23. Dallas Denery 1998: personal communication.

24. Dallas Denery 1998: personal communication