Aerospace Technology Development

Safety Flight Control Software Tested

NASA IS TAKING ANOTHER STEP TOWARD its goal to reduce commercial aircraft accident rates by a factor of five over the next 10 years by being the first to apply experimental "neural network" software in a safety-related environment. The new "smart" software will enable pilots to control and safely land disabled airplanes that have sustained combat damage or encountered major systems failures.

Neural network software has the ability to "learn" by observing patterns in the data it receives and processes, and then to perform different tasks in response to new patterns. The Intelligent Flight Control System (IFCS) employs experimental neural network software developed by computer scientists at NASA's Ames Research Center, Moffett Field, California, and the Boeing Company's Phantom Works division, St. Louis, Missouri.

Dryden Flight Research Center, Edwards, California, conducted flight evaluations of a preliminary version of the new "smart" software using a highly modified F-15 aircraft. The tests at Dryden demonstrated how the development version of the neural network software, pretrained to the F-15's aerodynamic data base and operating with a newly developed adaptive controller, can correctly identify aircraft stability and control characteristics. It then can immediately adjust the control system to maintain the best possible flight performance. About 16 flights were flown over a four-week period.

In its flight control application, the neural network software program compares the pattern of how the aircraft is actually flying with the pattern of how it should fly. These patterns are based on preprogrammed aeronautical equations, or control laws, that define how the airplane flies. If there is a mismatch caused by equipment failures, combat damage or other reasons, the aircraft's flight control computer uses the new neural network programming to "relearn" to fly the plane with a new pattern six times every second.

Using its on-line learning capability, the neural net software would identify that something has changed, then reconfigure the flight control computer system to adapt to those changes, making the failure or damage almost "transparent" to the pilot. To enable the pilot to maintain or regain control, it may change the way the remaining functional control surfaces and systems are used to compensate for the loss of the inoperative or damaged surfaces or equipment.

Future versions of the software could be developed for use in new airplanes that have digital fly-by-wire flight control systems, a requirement for the IFCS software. The system also has potential application to NASA's proposed Mars aircraft concept. These software versions will have even faster self-learning capability.

Proving that the neural network software works could lead to other uses, such as in power plants, automobiles and other less complicated systems to avoid potential disasters after equipment failures.

For more information, contact Chuck Jorgensen at Ames Research Center.
Call: 650/604-6725, Fax: 650/604-3594, E-mail: cjorgensen@mail.arc.nasa.gov Or contact Mike Thomson at Dryden Flight Research Center.
Call: 661/258-3097, E-mail: Mike.Thomson@dfrc.nasa.gov
Or contact James M. Urnes, Sr., at Boeing Phantom Works.
Call: 314/234-3775, E-mail: james.m.urnes-sr@boeing.com
Please mention you read about it in Innovation.

 

A finalist for a Discover Award,Transportation category, smart plane software tested aboard this modified F-15 is designed to "learn" how to fly a crippled airplane to help its pilots land it safely.