This photo, taken in 1949 in the Engine Research Building at NACA's Lewis Flight Propulsion Laboratory, is an early analog computing machine.
This 1951 photo shows one of the early computers used by NACA. The equipment is located at the NACA's Lewis Flight Propulsion Laboratory.
Wind tunnel facilities often employed groups of women whose job it was to perform mathematical calculations relating to airflow. They were euphemistically referred to as "female computers."
Two people working with IBM type T04 electronic data processing machine used for making computations for aeronautical research.
Melba Roy headed the group of NASA mathematicians, known as "computers," who tracked the Echo satellites in the 1960s. The group's computations helped produce the orbital element timetables by which millions could view the satellite from Earth as it passed overhead.
Computer Update Equipment (CUE) provided automatic distribution of flight-plan data to air traffic controllers beginning in the early 1970s.
The Cray 2 Supercomputer showing its water cooling towers installed at the NASA Ames Research Center in 1986.
The advent of computer technology offered a way to transform the capabilities of air traffic control. In 1967, IBM delivered a prototype computer to the Jacksonville Air Route Traffic Control Center. The software written for the ensuing National Airspace System En Route Stage A project contained more instructions than any previous computer program.
The National Full Scale Aerodynamic Complex (NFAC) 40 x 80-foot Wind Tunnel Data Acquisition System at the NASA Ames Research Center, 1990.
Three views of an SST configuration drawn entirely by a computer. Such analytical techniques proved inadequate for "off design" portions of the SST mission, such as stability under various aerodynamic forces. Wind tunnel testing was necessary.
The Cray Y 190A Supercomputer at the NASA Ames Research Center, 1990.
Computers in Aviation
Just as computers have affected virtually every aspect of modern life, from medicine to sports to education, they have also had a major impact on aviation. Computers are now used in all parts of aviation. They are used to design airplanes, to control them in flight, and to ensure that they reach their destinations safely and (more or less) on time.
Determining when computers first took flight depends in part upon one's definition of a computer. A little more than a decade after the Wright brothers flew at Kitty Hawk, the brilliant U.S. engineer Elmer Sperry adapted gyroscopes—which consisted of spinning weights that maintained a specific orientation—to electric and pneumatic control systems connected to an airplane's flight controls. This device, soon named an autopilot, could hold a plane level and on a specific course when the pilot took his hands off the controls. Over the next several decades, Sperry and others continued to perfect this technique. The famous Norden bombsight used aboard B-17 and B-29 bombers during World War II was a calculating device connected to an autopilot and controlled the airplane and held it steady when the bombs were released. These devices were all types of computers, although extremely primitive even by the standards of only a decade or so later.
Aircraft during the 1950s and early 1960s also carried analog computers as part of their radar equipment. These were used to provide targeting information for guns and missiles. The Heads Up Display (HUD) that projected information onto a piece of glass in front of the pilot relied upon computer input to help the pilot aim his guns or select his weapons.
Computers at the National Advisory Committee for Aeronautics (NACA) first used calculating machines in the 1930s to aid researchers in their work, sometimes to perform complicated calculations of airflow over airfoils. Furthermore, wind tunnel facilities often employed groups of women, who were sometimes referred to as "computers," whose sole job it was to perform mathematical calculations concerning airflow.
By the 1950s, as IBM developed better calculating and tabulating machines for office use, more and more powerful computers were pressed into service to assist in number crunching the results of wind tunnel tests and in trying to predict some of the results before actual models were placed into wind tunnels. Computers and wind tunnels both had an impact on each other: More powerful computers allowed designers not only to process wind tunnel test results better and faster, but to determine some of those test results before a model was even built, and wind tunnel data allowed designers to develop better programs for their computers to predict airflow. Computers allowed aircraft designers to narrow their research and test fewer designs in wind tunnels than before. By the 1970s, an hour of wind tunnel testing could cost thousands of dollars, so designers wanted to gain as much data about their aircraft as they could before they ever put a model in a wind tunnel.
By the 1980s, computers had become so powerful that for some applications, they actually began replacing wind tunnels entirely. This saved tremendous amounts of money. Aeronautical engineers began developing advanced computer programs to conduct computational fluid dynamics (CFD) experiments. This demand played a major role in pushing the development of new, powerful, so-called "supercomputers" capable of conducting millions of calculations per second. By the 1990s, computers had all but replaced wind tunnels for aeronautical research. Furthermore, because powerful computers and CFD programs could be bought by any large company (unlike wind tunnels, which could not be bought from a supplier), commercial aircraft designers no longer had to build models to fly in government wind tunnels.
An important early computer, not only for aviation but for computers in general, was the Whirlwind computer started at the Massachusetts Institute of Technology (MIT) in 1944. Whirlwind was a flight simulator. It was the first computer to respond immediately to actions taken by its operator. Previous computers simply took inputs and then made calculations and eventually produced an output, sometimes hours later. But Whirlwind responded in "real time."
As previously noted, the first computers to fly were primitive mechanical devices used to control planes in flight. Airplane and missile designers kept improving these systems, which enabled them to do new things. The German A4 (V-2) rocket of World War II used an early computer control system. The Canadian CF-105 Arrow interceptor airplane, which flew in March 1958, was the first aircraft to use an analog computer not as an autopilot but as a means of improving the flyability of the aircraft. The Arrow's computer was used to reduce the plane's tendency to yaw back and forth in flight. The Apollo Lunar Module also used an analog computer flight control system and other U.S. spacecraft such as Mercury, Gemini, and Apollo all had computer flight control systems.
The General Dynamics (now Lockheed-Martin) F-16, which entered service in the late 1970s and has been built in large numbers, was the first operational jet fighter to use an analog flight control system. The pilot steers the rudder pedals and joystick, but these are not directly connected to the control surfaces such as the rudder and ailerons. Instead, they are connected to a "fly-by-wire" flight control system. Three computers on the aircraft constantly adjust the flight controls to maintain the aircraft in flight and reply to the commands from the pilot. The F-16 is inherently unstable by design, meaning that it would fly out of control if the computers failed (which is why there are three of them). The designers made it unstable in order to improve its maneuverability. The computers constantly readjust the flight surfaces to keep the plane flying. Initially, pilots often referred to the F-16 as "the electric jet." But computer control systems have become so common that they are no longer unusual.
Similar computers were also used on the Space Shuttle and the F-117 stealth fighter. In 1972, NASA tested a modified Navy F-8 Crusader with a digital fly-by-wire system, which has replaced analog systems for most applications. Today fly-by-wire control systems are common on all advanced fighter aircraft. The F-18, F-22, the Joint Strike Fighter, the Eurofighter Typhoon, the Swedish JAS 39 Grippen, and the French Rafale all use flight control computers.
Although the U.S. Air Force made a major step to adopt computers in its airplanes in a major way in the 1970s, it took more than a decade before commercial airliners began to adopt them. The MD-11, a highly modified version of the popular DC-10, was the first commercial aircraft to adopt computer-assisted flight controls. The Airbus A340 also adopted them. Fly-by-wire has many advantages for small, nimble aircraft but fewer clear advantages for large, slower ones that rarely reach the edges of their performance envelope. Because of this, aircraft designers disagree on how much control should remain with the pilot and how much should be given to a computer.
The next major area for computers to conquer was the design of aircraft. Designing a plane such as a large passenger airplane is an immensely complex job involving thousands of engineers and ultimately hundreds of thousands of pages of blueprints. Keeping track of all the changes and making sure that workmen are using the latest blueprints and not outdated ones are major tasks in themselves. In the 1980s Boeing decided to build a new large passenger jet and the program's managers made the radical decision to design the plane entirely on computers, without using traditional paper blueprints. The result was the Boeing 777, which first flew in 1994 and many people referred to as "the first Twenty-First Century Jet."
The latest application of computers is for what is called "the Airborne Internet." Airplanes will be connected by radio and satellite link to a global information system that will provide them with information on the weather and aircraft in their immediate vicinity, as well as their flight plan.
The ultimate application of computers may be to take over control of flying completely. Already planes can be flown over long distances, with multiple course changes, entirely by computer. They can also take off and land automatically (although current flight rules prohibit this). It is not a far step to completely automatic operation and pilots may become unnecessary.
--Dwayne A. Day
Sources and further reading:
Beniger, James R. The Control Revolution. Cambridge, Mass.: Harvard University Press, 1986.
Ceruzzi, Paul E. Beyond the Limits. Cambridge, Mass.: MIT Press, 1989.
Eppler, Richard, Airfoil Design and Data. New York: Springer-Verlag, 1990.
Fallows, James, Free Flight: Inventing the Future of Travel. New York: Public Affairs, 2001.
Phillips, Don. "FAA Sees Future in New Satellite Guidance System." The Washington Post, June 18, 2001, p. 1.
Reed, Fred. "The Electric Jet." Air & Space, December 1986, 42-48.
Sabbagh, Karl. Twenty-First Century Jet. New York: Scribner, 1996.
Siuru, Bill, and Busick, John D. Future Flight. Blue Ridge Summit, Penn.: TAB Books, 1994.
Tomayko, James E. Computers Take Flight – A History of NASA's Pioneering Digital Fly-by-Wire Project. NASA SP-2000-4224. Washington, DC, 2000.
Tomayko, James E. Computers in Spaceflight. NASA Contractor Report-182505, Washington, DC, March 1988. Also at http://www.hq.nasa.gov/office/pao/History/computers/Compspace.html
Waldrop, M. Mitchell. "The Origins of Personal Computing," Scientific American, (December 2001) 84-91.