Engineering topics
Publications and source records attributed to Allen, Michael J..
The present invention provides a practical method for UAVs to take advantage of thermals in a manner similar to piloted aircrafts and soaring birds. In general, the invention is a method for a UAV to autonomously locate a thermal and be guided to the thermal to greatly improve range and endurance of the aircraft.
The first phase of the Autonomous Airborne Refueling Demonstration (AARD) project was completed on August 30, 2006. The goal of this 15-month effort was to develop and flight-test a system to demonstrate an autonomous refueling engagement using the Navy style hose-and-drogue air-to-air refueling method. The prime contractor for this Defense Advanced Research Projects Agency (DARPA) sponsored program was Sierra Nevada Corporation (SNC), Sparks, Nevada. The responsible flight-test organization was the NASA Dryden Flight Research Center (DFRC), Edwards, California, which also provided the F/A-18 receiver airplane (McDonnell Douglas, now The Boeing Company, Chicago, Illinois). The B-707-300 tanker airplane (The Boeing Company) was contracted through Omega Aerial Refueling Services, Inc., Alexandria, Virginia, and the optical tracking system was contracted through OCTEC Ltd., Bracknell, Berkshire, United Kingdom. Nine research flights were flown, testing the functionality and performance of the system in a stepwise manner, culminating in the plug attempts on the final flight. Relative position keeping was found to be very stable and accurate. The receiver aircraft was capable of following the tanker aircraft through turns while maintaining its relative position. During the last flight, six capture attempts were made, two of which were successful. The four misses demonstrated excellent characteristics, the receiver retreating from the drogue in a controlled, safe, and predictable manner that precluded contact between the drogue and the receiver aircraft. The position of the receiver aircraft when engaged and in position for refueling was found to be 5.5 to 8.5 ft low of the ideal position. The controller inputs to the F/A-18 were found to be extremely small
The first phase of the Autonomous Airborne Refueling Demonstration (AARD) project was completed on August 30, 2006. The goal of this 15-month effort was to develop and flight-test a system to demonstrate an autonomous refueling engagement using the Navy style hose-and-drogue air-to-air refueling method. The prime contractor for this Defense Advanced Research Projects Agency (DARPA) sponsored program was Sierra Nevada Corporation (SNC), Sparks, Nevada. The responsible flight-test organization was the National Aeronautics and Space Administration (NASA) Dryden Flight Research Center (DFRC), Edwards, California, which also provided the F/A-18 receiver airplane (McDonnell Douglas, now The Boeing Company, Chicago, Illinois). The B-707-300 tanker airplane (The Boeing Company) was contracted through Omega Aerial Refueling Services, Inc., Alexandria, Virginia, and the optical tracking system was contracted through OCTEC Ltd., Bracknell, Berkshire, United Kingdom. Nine research flights were flown, testing the functionality and performance of the system in a stepwise manner, culminating in the plug attempts on the final flight. Relative position keeping was found to be very stable and accurate. The receiver aircraft was capable of following the tanker aircraft through turns while maintaining its relative position. During the last flight, six capture attempts were made, two of which were successful. The four misses demonstrated excellent characteristics, the receiver retreating from the drogue in a controlled, safe, and predictable manner that precluded contact between the drogue and the receiver aircraft. The position of the receiver aircraft when engaged and in position for refueling was found to be 5.5 to 8.5 ft low of the ideal position. The controller inputs to the F/A-18 were found to be extremely small.
Understanding the wing twist of the active aeroelastic wing (AAW) F/A-18 aircraft is a fundamental research objective for the program and offers numerous benefits. In order to clearly understand the wing flexibility characteristics, a model was created to predict real-time wing twist. A reliable twist model allows the prediction of twist for flight simulation, provides insight into aircraft performance uncertainties, and assists with computational fluid dynamic and aeroelastic issues. The left wing of the aircraft was heavily instrumented during the first phase of the active aeroelastic wing program allowing deflection data collection. Traditional data processing steps were taken to reduce flight data, and twist predictions were made using linear regression techniques. The model predictions determined a consistent linear relationship between the measured twist and aircraft parameters, such as surface positions and aircraft state variables. Error in the original model was reduced in some cases by using a dynamic pressure-based assumption. This technique produced excellent predictions for flight between the standard test points and accounted for nonlinearities in the data. This report discusses data processing techniques and twist prediction validation, and provides illustrative and quantitative results.
A guidance and control method was developed to detect and exploit thermals for energy gain. Latency in energy rate estimation degraded performance. The concept of a UAV harvesting energy from the atmosphere has been shown to be feasible with existing technology. Many UAVs have similar mission constraints to birds and sailplanes. a) Surveillance; b) Point to point flight with minimal energy; and c) Increased ground speed.
Thermals caused by convection in the lower atmosphere are commonly used by birds and glider pilots to extend flight duration, increase cross-country speed, improve range, or simply to conserve energy. Uninhabited Aerial Vehicles (UAVs) can also increase performance and reduce energy consumption by exploiting atmospheric convection. An autonomous soaring research project was conducted at the NASA Dryden Flight Research Center to evaluate the concept through flight test of an electric-powered motorglider with a wingspan of 4.27 m (14 ft). The UAV's commercial autopilot software was modified to include outer-loop soaring guidance and control. The aircraft total energy state was used to detect and soar within thermals. Estimated thermal size and position were used to calculate guidance commands for soaring flight. Results from a total of 23 thermal encounters show good performance of the guidance and control algorithms to autonomously detect and exploit thermals. The UAV had an average climb of 172 m (567 ft) during these encounters.
Thermals caused by convection in the lower atmosphere are commonly used by birds and glider pilots to extend flight duration, increase cross-country speed, improve range, or simply to conserve energy. Uninhabited Aerial Vehicles (UAVs) can also increase performance and reduce energy consumption by exploiting atmospheric convection. An autonomous soaring research project was conducted at the NASA Dryden Flight Research Center to evaluate the concept through flight test of an electric-powered motor-glider with a wingspan of 4.27 m (14 ft). The UAV's commercial autopilot software was modified to include outer-loop soaring guidance and control. The aircraft total energy state was used to detect and soar within thermals. Estimated thermal size and position were used to calculate guidance commands for soaring flight. Results from a total of 23 thermal encounters show good performance of the guidance and control algorithms to autonomously detect and exploit thermals. The UAV had an average climb of 172 m (567 ft) during these encounters.
Large birds and glider pilots commonly use updrafts caused by convection in the lower atmosphere to extend flight duration, increase cross-country speed, improve range, or simply to conserve energy. Uninhabited air vehicles may also have the ability to exploit updrafts to improve performance. An updraft model was developed at NASA Dryden Flight Research Center (Edwards, California) to investigate the use of convective lift for uninhabited air vehicles in desert regions. Balloon and surface measurements obtained at the National Oceanic and Atmospheric Administration Surface Radiation station (Desert Rock, Nevada) enabled the model development. The data were used to create a statistical representation of the convective velocity scale, w*, and the convective mixing-layer thickness, zi. These parameters were then used to determine updraft size, vertical velocity profile, spacing, and maximum height. This paper gives a complete description of the updraft model and its derivation. Computer code for running the model is also given in conjunction with a check case for model verification.
A viewgraph presentation on autonomous soaring flight results for Unmanned Aerial Vehicles (UAV)'s is shown. The topics include: 1) Background; 2) Thermal Soaring Flight Results; 3) Autonomous Dolphin Soaring; and 4) Future Plans.
Flight testing of the 14ft span CloudSwift UAV was conducted during the summer of 2005. Test maneuvers included aircraft checkout, Piccolo gain tuning, FTS range tests, and thermal soaring research flights.
The Active Aeroelastic Wing research program was a joint program between the U.S. Air Force Research Laboratory and NASA established to investigate the characteristics of an aeroelastic wing and the technique of using wing twist for roll control. The flight test program employed the use of an F/A-18 aircraft modified by reducing the wing torsional stiffness and adding a custom research flight control system. The research flight control system was optimized to maximize roll rate using only wing surfaces to twist the wing while simultaneously maintaining design load limits, stability margins, and handling qualities. NASA Dryden Flight Research Center developed control laws using the software design tool called CONDUIT, which employs a multi-objective function optimization to tune selected control system design parameters. Modifications were made to the Active Aeroelastic Wing implementation in this new software design tool to incorporate the NASA Dryden Flight Research Center nonlinear F/A-18 simulation for time history analysis. This paper describes the design process, including how the control law requirements were incorporated into constraints for the optimization of this specific software design tool. Predicted performance is also compared to results from flight.
The Active Aeroelastic Wing airplane was successfully flight-tested in March 2005. During phase 1 of the two-phase program, an onboard excitation system provided independent control surface movements that were used to develop a loads model for the wing structure and wing control surfaces. The resulting loads model, which was used to develop the control laws for phase 2, is described. The loads model was developed from flight data through the use of a multiple linear regression technique. The loads model input consisted of aircraft states and control surface positions, in addition to nonlinear inputs that were calculated from flight-measured parameters. The loads model output for each wing consisted of wing-root bending moment and torque, wing-fold bending moment and torque, inboard and outboard leading-edge flap hinge moment, trailing-edge flap hinge moment, and aileron hinge moment. The development of the Active Aeroelastic Wing loads model is described, and the ability of the model to predict loads during phase 2 research maneuvers is demonstrated. Results show a good match to phase 2 flight data for all loads except inboard and outboard leading-edge flap hinge moments at certain flight conditions. The average load prediction errors for all loads at all flight conditions are 9.1 percent for maximum stick-deflection rolls, 4.4 percent for 5-g windup turns, and 7.7 percent for 4-g rolling pullouts.
Successful flight-testing of the Active Aeroelastic Wing airplane was completed in March 2005. This program, which started in 1996, was a joint activity sponsored by NASA, Air Force Research Laboratory, and industry contractors. The test program contained two flight test phases conducted in early 2003 and early 2005. During the first phase of flight test, aerodynamic models and load models of the wing control surfaces and wing structure were developed. Design teams built new research control laws for the Active Aeroelastic Wing airplane using these flight-validated models; and throughout the final phase of flight test, these new control laws were demonstrated. The control laws were designed to optimize strategies for moving the wing control surfaces to maximize roll rates in the transonic and supersonic flight regimes. Control surface hinge moments and wing loads were constrained to remain within hydraulic and load limits. This paper describes briefly the flight control system architecture as well as the design approach used by Active Aeroelastic Wing project engineers to develop flight control system gains. Additionally, this paper presents flight test techniques and comparison between flight test results and predictions.
Understanding the wing twist of the active aeroelastic wing F/A-18 aircraft is a fundamental research objective for the program and offers numerous benefits. In order to clearly understand the wing flexibility characteristics, a model was created to predict real-time wing twist. A reliable twist model allows the prediction of twist for flight simulation, provides insight into aircraft performance uncertainties, and assists with computational fluid dynamic and aeroelastic issues. The left wing of the aircraft was heavily instrumented during the first phase of the active aeroelastic wing program allowing deflection data collection. Traditional data processing steps were taken to reduce flight data, and twist predictions were made using linear regression techniques. The model predictions determined a consistent linear relationship between the measured twist and aircraft parameters, such as surface positions and aircraft state variables. Error in the original model was reduced in some cases by using a dynamic pressure-based assumption and by using neural networks. These techniques produced excellent predictions for flight between the standard test points and accounted for nonlinearities in the data. This report discusses data processing techniques and twist prediction validation, and provides illustrative and quantitative results.
A relatively unexplored method to improve the endurance of an autonomous aircraft is to use buoyant plumes of air found in the lower atmosphere called thermals or updrafts. Glider pilots and birds commonly use updrafts to improve range, endurance, or cross-country speed. This report presents a quantitative analysis of a small electric-powered uninhabited air vehicle using updrafts to extend its endurance over a target location. A three-degree-of-freedom simulation of the uninhabited air vehicle was used to determine the yearly effect of updrafts on performance. Surface radiation and rawinsonde balloon measurements taken at Desert Rock, Nevada, were used to determine updraft size, strength, spacing, shape, and maximum height for the simulation. A fixed-width spiral path was used to search for updrafts at the same time as maintaining line-of-sight to the surface target position. Power was used only when the aircraft was flying at the lower-altitude limit in search of updrafts. Results show that an uninhabited air vehicle with a nominal endurance of 2 hours can fly a maximum of 14 hours using updrafts during the summer and a maximum of 8 hours during the winter. The performance benefit and the chance of finding updrafts both depend on what time of day the uninhabited air vehicle is launched. Good endurance and probability of finding updrafts during the year was obtained when the uninhabited air vehicle was launched 30 percent into the daylight hours after sunrise each day. Yearly average endurance was found to be 8.6 hours with these launch times.
The Active Aeroelastic Wing airplane was successfully flight-tested in March 2005. During phase 1 of the two-phase program, an onboard excitation system provided independent control surface movements that were used to develop a loads model for the wing structure and wing control surfaces. The resulting loads model, which was used to develop the control laws for phase 2, is described. The loads model was developed from flight data through the use of a multiple linear regression technique. The loads model input consisted of aircraft states and control surface positions, in addition to nonlinear inputs that were calculated from flight-measured parameters. The loads model output for each wing consisted of wing-root bending moment and torque, wing-fold bending moment and torque, inboard and outboard leading-edge flap hinge moment, trailing-edge flap hinge moment, and aileron hinge moment. The development of the Active Aeroelastic Wing loads model is described, and the ability of the model to predict loads during phase 2 research maneuvers is demonstrated. Results show a good match to phase 2 flight data for all loads except inboard and outboard leading-edge flap hinge moments at certain flight conditions. The average load prediction errors for all loads at all flight conditions are 9.1 percent for maximum stick-deflection rolls, 4.4 percent for 5-g windup turns, and 7.7 percent for 4-g rolling pullouts.
Successful flight-testing of the Active Aeroelastic Wing airplane was completed in March 2005. This program, which started in 1996, was a joint activity sponsored by NASA, Air Force Research Laboratory, and industry contractors. The test program contained two flight test phases conducted in early 2003 and early 2005. During the first phase of flight test, aerodynamic models and load models of the wing control surfaces and wing structure were developed. Design teams built new research control laws for the Active Aeroelastic Wing airplane using these flight-validated models; and throughout the final phase of flight test, these new control laws were demonstrated. The control laws were designed to optimize strategies for moving the wing control surfaces to maximize roll rates in the transonic and supersonic flight regimes. Control surface hinge moments and wing loads were constrained to remain within hydraulic and load limits. This paper describes briefly the flight control system architecture as well as the design approach used by Active Aeroelastic Wing project engineers to develop flight control system gains. Additionally, this paper presents flight test techniques and comparison between flight test results and predictions.
Neural networks were used to model wing bending-moment loads, torsion loads, and control surface hinge-moments of the Active Aeroelastic Wing (AAW) aircraft. Accurate loads models are required for the development of control laws designed to increase roll performance through wing twist while not exceeding load limits. Inputs to the model include aircraft rates, accelerations, and control surface positions. Neural networks were chosen to model aircraft loads because they can account for uncharacterized nonlinear effects while retaining the capability to generalize. The accuracy of the neural network models was improved by first developing linear loads models to use as starting points for network training. Neural networks were then trained with flight data for rolls, loaded reversals, wind-up-turns, and individual control surface doublets for load excitation. Generalization was improved by using gain weighting and early stopping. Results are presented for neural network loads models of four wing loads and four control surface hinge moments at Mach 0.90 and an altitude of 15,000 ft. An average model prediction error reduction of 18.6 percent was calculated for the neural network models when compared to the linear models. This paper documents the input data conditioning, input parameter selection, structure, training, and validation of the neural network models.