Click on Picture to enlarge LoFLYTE

+ Larger Font | - Smaller Font

 

A Test Aircraft that Learns by Doing Unveiled in Wisconsin

 

Click on Picture to enlarge

NASA and the U.S. Air Force unveiled a jet-powered aircraft equipped with state-of-the-art flight control technologies at a briefing in Oshkosh, Wis. on August 2, 1996. The 8-foot 4-inch aircraft was built to demonstrate a computerized flight control system that learns as it flies -- especially important for the demands of ultra high speed flight.

The "LoFLYTE" aircraft has been developed by Accurate Automation Corporation in Chattanooga, Tenn., for NASA and the Air Force. Technologies being implemented in the LoFLYTE program could eventually find their way into commercial, general aviation, and military aircraft.

The experimental LoFLYTE aircraft will be used to explore new flight control techniques involving neural networks, which allow the aircraft control system to learn by mimicking the pilot.

The model is a Mach 5 waverider design which is a futuristic hypersonic aircraft configuration that actually cruises on top of its shockwave. Waverider aircraft powered by airbreathing hypersonic engines, would fly at speeds above Mach 4, LoFLYTE represents the first known flying waverider vehicle configuration, but in upcoming flight tests at NASA's Dryden Flight Research Center in California it will be flown at subsonic speeds to explore take-off and landing control issues.

The remotely-piloted aircraft has been designed to demonstrate that neural network flight controls are superior to conventional flight controls. Neural networks are computer systems that actually learn by doing. The computer network consists of many interconnected control systems, or nodes, similar to neurons in the brain. Each node assigns a value to the input from each of its counterparts. As these values are changed, the network can adjust the way it responds.

The LoFLYTE aircraft's flight controller consists of a network of multiple-instruction, multiple-data neural chips. The network will be able to continually alter the aircraft's control laws in order to optimize flight performance and take the pilot's responses into consideration. Over time, the neural network system could be trained to control the aircraft. The use of neural networks in flight would help pilots fly in quick-decision situations and help damaged aircraft land safely even when controls are partially destroyed.

The main objective of LoFLYTE is to demonstrate the ability of such a flight control system that learns through experience, said Robert Pegg of Langley's Hypersonic Vehicles Office. In addition to experimenting with neural networks, the flight of the model is also key as a low-speed demonstration of a hypersonic vehicle. "We're very interested in both outcomes, both the neural net technology and the flight characteristics," Pegg said.

"We see a big advantage to using this type of control system in a hypersonic vehicle," Pegg said. "At those high speeds, things happen so quickly that the pilot cannot control the aircraft as easily as at subsonic speeds."

The initial configuration for the aircraft was developed at NASA Langley and then Accurate Automation Corporation integrated the neural network technology into the Langley design. Successful tests of the waverider concept in Langley's 12-foot Low-Speed Wind Tunnel and 30- by 60-foot Full Scale Tunnel preceded the development of this model aircraft.

The construction of the model was completed at SWB Turbines of Appleton, Wis. This company provided the small turbine engine that powers the model. The shell of the model was made at Mississippi State's Raspet Flight Research Laboratory and then shipped to SWB Turbines so that the radio control gear and the engine could be installed.

The waverider was chosen as the testbed for the neural networks because the configuration has an inherently high hypersonic lift-to-drag ratio. If neural networks can control this "worst-case scenario" configuration, then they should be able to handle any other desired configuration. The waverider configuration was also chosen because it allows for long hypersonic cruise ranges of up to 8,000 miles. At an altitude of 90,000 feet the Mach 5 waverider would be able to fly at a rate of one mile per second.

The program contracts are being administered through the NASA Langley Research Center in Hampton, Va. and the Air Force Wright Laboratory in Dayton, Ohio, under the Small Business Innovative Research (SBIR) program.

Pegg also added another positive implication that LoFLYTE could have. "We want to make the public aware that the government is getting a good return on its SBIR-invested money," he said. "We hope this project will help us further demonstrate to the public that the SBIR program is a viable investment for the American people."

All Star Network

 

 

Click on Picture to enlarge

LoFLYTE

In 1995, Accurate Automation Corp. (AAC) was selected by the NASA and the AFRL (Air Force Research Lab) to build the airframe and control system for a sub-scale research UAV called LoFLYTE (Low-Observable Flight Test Experiment). LoFLYTE was to test the low-speed handling characteristics of an airframe optimized for hypersonic flight at Mach 5+, and to develop an automatic stabilization and control system for the vehicle. The LoFLYTE aircraft flew for the first time on 16 December 1996.

The LoFLYTE UAV is a highly-swept flying-wing delta with twin vertical fins, and is powered by a small turbojet engine in a ventral duct. It has a retractable tricycle undercarriage for conventional take-off and landing. The airframe is a so-called "waverider" shape, which can use its own shockwave at hypersonic speeds for optimized lift and drag characteristics. The aircraft is normally flown by a ground-based pilot, but is also equipped with a GPS-based navigation system for autonomous missions. The most important component of LoFLYTE is its Neural Adaptive Controller (NAC) flight control system, also designed by AAC. It is a "neural network" consisting of many interconnected computer "nodes" with a dynamic and self-adapting control logic. This network is supposed to be able to "learn" to keep the aircraft automatically stable similar to the way a human brain learns things by dynamically changing the properties of the connections between its neurons. The basic idea of the LoFLYTE program was that if such a neural network can learn to handle a hypersonic airframe at low speeds (which is one of the most difficult aerodynamic control tasks), it can do that for any other configuration as well. It was actually planned to remove the vertical fins once the flight control system was able to keep the aircraft completely stable.

The LoFLYTE flight tests were conducted by the USAF's 419th and 445th Flight Test Squadrons at Edwards AFB. The first flight had uncovered basic control problems, and flight testing did not resume before June 1997. For the rest of that year, the basic handling characteristics were established, which showed that a Mach 5+ shape could indeed take off and land on a runway at "normal" speeds. These initial tests were not yet flown with the NAC, and used a conventional computerized stabilization and control system instead. Actual flight tests using the NAC began in December 1997 and continued into 1998. The flights also included tests which were to show the ability of the NAC to handle changed airframe configurations and (simulated) damage to the control surfaces. It must be noted that LoFLYTE was never intended to fly at anywhere near the speed suggested by its waverider shape.

From that point on, the history of the LoFLYTE project becomes a bit unclear. NASA's and the USAF's involvement with LoFLYTE probably ended at some time in 1998. However, the tests must have been reasonably successful, because NASA awarded AAC a follow-on contract for the HyFLYTE UAV, later called the "X-43A-LS" (LS = Low Speed). This vehicle was actually built and flown, and tested the low-speed handling of the X-43A Hyper-X Mach 10 configuration using the NAC flight control system. The LoFLYTE itself, of which AAC built a total of three examples, was apparently still used for general research into neural network control systems as late as 2002. Whether this research included actual test flights is unclear.

Specifications

Note: Data given by several sources show slight variations. Figures given below may therefore be inaccurate!

Data for LoFLYTE:

Length	2.54 m (8 ft 4 in)
Wingspan	1.58 m (5 ft 2.2 in)
Height	0.61 m (2 ft)
Weight	32 kg (70 lb)
Speed	465 km/h (290 mph)
Ceiling	?
Endurance	?
Propulsion	SWB Turbines SWB-3 turbojet; 0.156 kN (35 lb)

Military Rockets & Missiles

 

View a movie of LoFLYTE

 

 

Flights Demonstrate Airworthiness of "Waverider" Shape

NASA Fact Line  July 1997

NASA's LoFLYTE™ Program Flown
 

Click on Picture to enlarge

During the week of June 23, 1997, LoFLYTE™ made three successful flights at Edwards Air Force Base under the direction of the 445th Flight Test Squadron. In the next stage of the program, a neural network flight control system will be installed and tested.

The LoFLYTE™ (Low Observable Flight Test Experiment) program is a joint NASA Langley Research Center/U.S. Air Force Research Laboratory ground and flight test program that has resulted in a demonstration aircraft. LoFLYTE™ passed an important milestone in 1997 with a series of successful flights. This was the first flight of a true waverider aircraft configuration. The LoFLYTE™ prototype, a 100-inch-long jet-powered remotely-piloted vehicle (RPV), has demonstrated the subsonic airworthiness of the "waverider" shaped aircraft.

The LoFLYTE™ program is designed to provide a technology testbed for many emerging aerospace technologies with initial emphasis on neural network controls. The LoFLYTE™ jet completed its first flight on December 16, 1996, at Mojave Airport after completing design, airworthiness, and flight safety reviews required by NASA and the U.S. Air Force. LoFLYTE™ is flown at Edwards Air Force Base in California. Flight testing using neural network controls will begin in late 1997. This small jet-powered aircraft will demonstrate neural network control and sensing technology. The LoFLYTE™ concept was first tested at NASA Langley Research Center as a wind tunnel article with 191 runs in both the 12-Foot Low Speed Tunnel and 30 x 60 foot wind tunnels. The present LoFLYTE™ shape is the same size as the wind tunnel article. In addition, the Naval Postgraduate School in Monterey, California, did water tunnel flow visualization tests and tested a 72-inch-long drop model.

Click on Picture to enlarge

LoFlyte banks after take-off

Accurate Automation Corporation, of Chattanooga, Tennessee, was selected as the contractor for LoFLYTE™ under the NASA and U.S. Air Force Small Business Innovation Research (SBIR) Programs. The LoFLYTE™ RPV will eventually become an unmanned autonomous vehicle (UAV). The shape of LoFLYTE™ is based upon a high lift/drag Mach 5 configuration. The actual shape takes advantage of engine/body integration and was derived from a Mach 5 conical flowfield. The LoFLYTE™ vehicle demonstrates clearly how rapid prototyping can build flight-quality hardware inexpensively. It also features onboard subsystems, including an advanced real-time airborne data acquisition and control system with 16 channels of analog sensor input as well as 14 channels of control telemetry, GPS for position, retractable landing gear, video and spread spectrum communications to the ground.

Click on Picture to enlarge

The LoFLYTE™ Mobile Ground Control Station provides the flight test team with all of the facilities needed to conduct flight testing, including all aircraft maintenance and safety equipment, computers for telemetry and data recording, and a weather station.

LoFLYTE™ will eventually fly the Accurate Automation Neural Network Processor and Neural Air Data Subsystem. This 72 pound aircraft is powered by a 38 pound thrust jet engine built by SWB Turbines. An advanced engine controller will be tested for future use with this miniature JP-8 fueled engine.

Some of the technologies that will be tested with the LoFLYTE™ aircraft include:

• Rapid Aircraft Prototyping and Design Concepts
• Aircraft Instrumentation
• Fault Diagnosis and IsolationTechniques
• Real-time Airborne Data Acquisition, Control System and Video
• Miniature Spread Spectrum Telemetry
• Antenna Placement
• EMI Minimization
• Tiperons
• Neural Network Flight Controls
• Interface for the Neurocontrol With Flight and Propulsion Control
• Neural Air Data Subsystem for Determining Angle of Attack, Sideslip, and Velocity
• Advanced Nozzle Concepts
• Various TitaniumAlloy Parts With Subsequent Non-destructive Evaluation
• Adaptive Compensation for "Pilot-induced Oscillations"
• Trajectory Control
• Advanced Landing Concepts

 

Neural Network Flight Controls

 

Click on Picture to enlarge

LoFLYTE™ flying over the pilot's position on the flight of June 23, 1997.

The LoFLYTE™ Neural Network Flight Control System is an important advance in aerospace technology because of the adaptive nature of the control system. The controller is designed to learn as it flies, so the control system, not the pilot, determines the most effective commands to give the plane for a particular situation.

Click on Picture to enlarge

Accurate Automation's Neural Network Processor.

During normal flight, the neural controller will use the data it receives from the telemetry system to compute the most efficient flight characteristics and adjust the control surfaces accordingly. However, where the neural control system has an enormous advantage over traditional control systems is during abnormal and unexpected flight conditions. For example, if the control system determines that the rudder is not responding, it will adjust quickly to control the aircraft using the remaining flight surfaces. Neural network control is necessary in hypersonic vehicles where the center of gravity of the vehicle can change significantly throughout the flight. The neural network can adjust to changing flight conditions faster than a human pilot, greatly enhancing the safety of the aircraft.

The neural network control system, designed and manufactured by Accurate Automation Corporation, is based on the companyÕs successful Neural Network Processor (NNP®), also funded under the SBIR program. The NNP® is a multiple instruction/multiple data (MIMD) system that can be used in personal computers as well as aircraft.

The LoFLYTE™ fiber-optic "Fly-by-Light" communications offer lighter weight, increased transmission capability and safety from electrical system short circuits and EMI problems.

The LoFLYTE™ telemetry captures the data from the instrumentation in real-time and displays it for operational decisions during flight and transmission to remote sites.

 

Future Version Of LoFLYTE

Once testing of the 100-inch version of LoFLYTE™ is concluded, a larger transonic version may be developed to explore supersonic flight characteristics of the waverider shape. This larger version will include a unique hypersonic flowpath configuration. Other hypersonic aerodynamic shapes will be tested using the LoFLYTE™ subsystems.

An advanced engine controller is being developed for ramjets under NASA Lewis Research Center's SBIR program. This controller will be tested on the LoFLYTE™ vehicle with the current turbine.

An advanced auto landing system is being developed for NASA Ames Research Center, under SBIR, that may be tested with LoFLYTE™.

 

Small Business Innovation Research

 

The objective of NASA's Small Business Innovation Research Program is to stimulate technological innovation in the United States by using small business, including minority and disadvantaged firms, to help meet Federal research and development needs.

This is the 15th year of the NASA SBIR program, which allocates 2.5 percent of NASA's research and development budget to support SBIR.

NASA Langley Research Center, the U.S. Air Force and LoFLYTE™ were awarded the Small Business Administration's National Tibbetts Award in 1996.

 

Glossary Of Terms

 

Click on Picture to enlarge

One of 191 wind tunnel tests conducted at NASA Langley Research Center to test the airworthiness of the LoFLYTE™ waverider shape.

Conical flowfield. The cone-shaped shockwave generated by a supersonic or hypersonic vehicle.

Hypersonic. Operation at Mach numbers exceeding 4.

Neural network. A class of computational methods that loosely imitate the function of the brain. Among the benefits of neural networks are that they learn from experience, can generalize from their data set, are fault tolerant, and can exploit parallel systems for rapid processing.

Pilot-induced oscillations. A condition of aircraft uncontrollability caused when pilot's intent and the control system get out of sequence, forcing the plane to swing back and forth, often with disastrous results.

Supersonic. Operation between Mach 1 and Mach 4.

Waverider. A type of supersonic or hypersonic aircraft where the vehicle takes advantage of the shockwave flowfield, instead of cutting through it, increasing lift and reducing drag.

 

Accurate Automation Corporation

NASA Langley Research Center

 

 

USE YOUR BROWSER "BACK" BUTTON TO RETURN TO PERVIOUS PAGE