here - Quaternion Aerospace

Transcription

Multidisciplinary Design for Flight Test of a Scaled Joined
Wing SensorCraft
Jenner Richards1 and Afzal Suleman2
University of Victoria, Victoria, BC, Canada
Tyler Aarons3 and R. Canfield4
Virginia Polytechnic Institute and State University, Blacksburg, Virginia, USA
A 1/9th scale test article is required to investigate the aeroelastic and gust response of the
Boeing Joined Wing SensorCraft concept and validate existing analytical models. The
remotely piloted vehicle must exhibit equivalent aeroelastic properties while also being
designed for flight worthiness. This poses unique challenges that must be addressed using
multidisciplinary design approach. Analytical methods are used across disciplines including
structures, propulsion, stability, control, simulation and flight testing. Design choices are
validated using a combination of software simulations, hardware in the loop testing and
preliminary flight tests using a several low risk flight test vehicles. Flight testing of the joined
wing concept will be performed using a rigid prototype. Future work includes re-winging
the test vehicle with aeroelastically scaled lifting surfaces in order to achieve the desired
aeroelastic characteristics.
Nomenclature
AFRL
AVL
CAD
CCMS
CG
CL
CNC
DoF
EDF
EOM
FE
FTP
GLA
GVT
HALE
HASC
INS
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
=
Air Force Research Laboratory
Athena Vortex Lattice
Computer Aided Design
Collaborative Center for Multidisciplinary Sciences
Center of Gravity
Lift coefficient
Computer Numerically Controlled
Degree of Freedom
Electric Ducted Fan
Equations of Motion
Finite Element
Flight Test Program
Gust Load Alleviation
Ground Vibration Test
High Altitude, Long Endurance
High Angle of Attack Stability and Control
Inertial Navigation System
1
Ma Sc. Candidate, Dept of Engineering, University of Victoria, PO Box 3055, V8W 3P6, Member AIAA
Principle Investigator, IDMEC-IST, Lisbon, Portugal, AIAA Associate Fellow
3
MS Candidate, Aerospace and Ocean Engineering Dept, Virginia Tech, 24061, Member AIAA
4
Professor, Virginia Tech Aerospace and Ocean Engineering Dept, Blacksburg, VA 24061, Associate Fellow AIAA
2
1
American Institute of Aeronautics and Astronautics
ISR
JWSC
MC
NAST
PDE
R/C
RPV
TOGW
UAV
VLM
=
=
=
=
=
=
=
=
=
=
Intelligence Surveillance and Reconnaissance
Joined Wing SensorCraft
Model Center
Nonlinear Aeroelastic Simulation Toolbox
Partial Differential Equation
Radio Controlled
Remotely Piloted Vehicle
Takeoff Gross Weight
Unmanned Aerial Vehicle
Vortex Lattice Method
II. Introduction
I. H
IGH altitude, long endurance (HALE) unmanned aerial vehicles (UAVs) are capable of providing
revolutionary intelligence, surveillance and reconnaissance (ISR) capabilities over vast geographic areas when
equipped with advanced sensor packages. The aeroelastic responses, specifically aft wing buckling and gust load
response, associated with the Joined Wing SensorCraft (JWSC) have been demonstrated and investigated in
numerous computational and wind tunnel studies1-3; however, these phenomena have never been successfully tested
in a flight test program. The Air Force has determined that an aeroelastically scaled Remotely Piloted Vehicle
(RPV) provides a low cost and effective way to investigate these non-linear aeroelastic responses. By
experimentally demonstrating, investigating and measuring these responses, future flight test programs will be able
to test active aeroelastic control and gust load alleviation systems to reduce the structural and aerodynamic effects of
the nonlinear responses. In addition, the testing of the RPV will serve to validate the aeroelastic scaling methods
employed and support management planning for future tests of SensorCraft technologies.
III. Background
The Sensorcraft is a concept initiated by the Air Force Research Laboratory (AFRL) to serve as a nextgeneration, HALE reconnaissance system. Traditional aircraft design typically sees an airframe being designed first
and a selection of sensors chosen to integrate into that platform. With the Sensorcraft project, this process is
reversed and an aircraft platform is specifically designed around an optimized collection of sensors. This has
resulted in a departure from conventional configurations as is apparent with the Boeing Joined Wing Concept4
shown in Figure 1.
Figure 1. Joined wing SensorCraft Concept4
The joined-wing configuration offers many potential benefits over conventional aircraft designs. Foremost is the
ability to incorporate conformal radar antennae in the fore and aft wings to provide persistent 360 degree, foliage
penetrating surveillance. Among other potential benefits are reduced induced drag, a stiffer and potentially lighter
wing structure, and direct lift/side force control. These benefits do come at a cost however. Previous computational
studies of joined-wing aircraft configurations have shown the importance of geometric nonlinearity due to large
deflections and follower forces that may lead to buckling of the aft wing1,5. This potential buckling represents a
2
unique and challenging aeroelastic design problem. The non-linear behavior, which is a result of the joined wing
configuration and advanced, lightweight structural design, could be removed by strengthening the wing to a point
where these non-linear behaviors vanish; however, this would result in large penalties in aspect ratio and structural
weight, greatly reducing the performance of the aircraft. To avoid these penalties, nonlinear aeroelastic design,
analysis and testing are required to ensure that the Joined Wing SensorCraft is able to sustain the nonlinear
responses required to complete the proposed ISR mission. An aeroelastically scaled RPV provides a low cost and
effective way to investigate these non-linear aeroelastic responses.
IV. Program Overview and Approach
The aeroelastic scaling and flight testing of the JWSC is the topic of an ongoing international collaboration
between Virginia Tech and The University of Victoria (Canada). The program leverages the resources of
Quaternion Engineering Inc. (Victoria, Canada) as well as the AFRL/ Virginia Tech/ Wright State Collaborative
Center for Multidisciplinary Sciences (CCMS) at Virginia Tech. This ambitious program has two primary
objectives. The first is to develop a cost-effective, aeroelastically scaled, RPV to investigate the known
geometrically nonlinear behavior associated with the JWSC configuration. The second objective is to design and
perform flight test experiments to demonstrate, measure and characterize the nonlinear behavior in flight.
The full scale flight point of interest corresponds to a fully fueled take off condition at sea level. The airforce
requests that a 1/9th scale RPV be built with equivelent scaled physics. Table 1 summarizes both the full scale and
the corresponding reduced scale test point while Figure 2 illustrates the size of the full scale and RPV airframes.
Table 1. Summary of full scale and RPV test points
Span (ft)
Flight Speed (kt)
Test Point Weight (lb)
Altitude (ft)
Full Scale JWSC
150
168
155337
Sea Level
RPV
16.4
56
204.5
Sea Level
Figure 2. Relative size of full scale JWSC and RPV
Due to the unique JWSC geometry, the relative large scale of the RPV and the challenge of fabricating
aeroelastically scaled, flight worthy components, the JWSC Flight Test Program (FTP) is divided into two distinct
phases: a flight worthiness program and an aeroelastic response program. The flight worthiness program involves
the conceptual design of a rigid RPV (one with equivalent rigid body dynamics, i.e. preserved aerodynamics, overall
mass and moments of inertia, but no aeroelastic scaling) including construction methods, flight test instrumentation
3
selection and integration, control system tuning and flight test program development. This also includes the
construction and flight testing of several preliminary models to determine flying qualities and trim requirements.
During this program, preliminary aeroelastic scaling, which represents the beginning of the aeroelastic response
program, will also be performed. Once the rigid RPV flight tests are completed, construction will begin on the
aeroelastically scaled RPV and a second flight test program will be developed. The project will culminate in the
completion of the flight test program of the aeroelastically scaled RPV.
A major challenge posed is from the broad range of complex, and often conflicting, design requirements.
Balancing these requirements involves analysis, optimization and testing across several disciplines including
structures, propulsion, stability, control, simulation and flight testing. A flowchart presenting the breakdown of
disciplines involved is presented in Figure 3.
Figure 3. Flowchart of multidisciplinary approach
An additional challenge is presented because the flight test article also must be designed for flight worthiness.
This must include allowances for fuel, avionics, sensors, landing gear etc., while also providing acceptable flying
characteristics for the ground based pilot. In addition, the aircraft must be fully instrumented in order to accurately
capture the aeroelastic response in flight in accordance with an incremental flight test approach. The remainder of
this paper serves to summarize the work performed across each of the above mentioned disciplines in order to
produce an aeroelastically scaled RPV, and supporting flight test program. Note that details of all five RPV aircraft
in use or in development for this program, including pictures and specifications, are presented in the Appendix for
reference.
V. Structures
The following sections give a brief description of the scaling parameters used to map the full scale test point to
the RPVs scale. Detailed descriptions of both the initial Rigid RPV as well as the methods used to optimize the
RPV’s structure in order to accurately reproduce the desired aeroelastic responses are also included.
A. Scaling the Full Scale Test Point for RPV Tests
Both the rigid and aeroelastically equivalent models are constrained by a set of scaling parameters, used to map
the full scale test point, stiffness, mass and geometry to the reduced scale test point. These scaling parameters are
derived from a chosen set of governing equations. Due to the limited time and resources, some assumptions are
applied to simplify the scaling process. In the present case, a simplified physics model is chosen- the small
disturbance, linear potential partial differential equations (PDE):
4
[
̈
]{ }
̈
[
]{ }
[
][
]{ }
(1)
Where
X
Ѳ
Mij
Kij
b
Qij
Vector of translational degrees of freedom
Vector of rotational degrees of freedom
Block matrix terms in mass/inertia matrix
Block matrix terms in stiffness matrix
Reference length
Block matrix of aerodynamic terms
The choice of scaling, based on the linear PDE equations above, allows the elimination of several constraints that
would otherwise be present if a more complex procedure was chosen (i.e. one based on the Navier Stokes
equations). For instance, the equations are based on thin airfoil theory which allows an arbitrary airfoil thickness to
be assigned, assuming the mean camber line is maintained. Also, if inviscid incompressible flow is assumed, the
equations of motion (EOM) will not constrain the Mach number or Reynolds number.
While limited, these physics are adequate for a low-cost exploration of flight mechanics. When the EOM
describing this system are non-dimensionalized, they yield a set of parameters that are used for scaling the baseline
aircraft. Table 2 below summarizes the calculated scaling parameters.
Table 2 – Summary of Scaling Parameters
Scaling parameter
Length Scale
Velocity-Ratio
Air Density Ratio
Mass-Ratio
Frequency-Ratio
B. Rigid Prototype
The Rigid prototype (refered to as Rigid RPV for remainder of the paper) employs a composite construction
created using twelve separate computer numerically controlled (CNC) machined female moulds. These moulds are
used to lay up the skin of the aircraft, and a rib and spar design, shown below in Figure 4, is used to provide
structural stiffness.
5
Figure 4 . Internal Structure of the Rigid RPV
As mentioned previously, the Rigid RPV is designed to only reproduce the aerodynamics, overall mass and
moments of inertias of the full scale JWSC. The aerodynamic matching is achieved by preserving the outer mould
line of the original geometry, whereas the overall mass and inertias are matched through the addition of trimming
masses. Careful attention is paid to the location of these trimming masses so that both overall mass and inertias are
matched. In order to achieve this, four areas are chosen as locations for the trimming weights such that their effects
on the principle moments of inertia ( ,
, and
) are uncoupled. The locations of the trimming masses are
depicted in Figure 5.
The first location is chosen within the fuselage. Since it is nearer the center of gravity, weight added here has
little effect on any of the principle moments of inertia, and is only used to augment the takeoff gross weight
(TOGW) of the aircraft. Next, a compartment is chosen within the wing near the joint. Addition of weight here has
the greatest effect on the rolling and yawing moment, while largely uncoupling the pitching moment of inertia. (This
location is also desirable structurally since the main wing is supported here by the aft wing). A compartment is
chosen in the aft wing/tail boom intersection due to its larger volume and large effect on the yaw and pitch moments
with the final location in the nose of the aircraft to help trim the aircraft longitudinally.
Figure 5. Location of Trimming Weight Bays (1 in fuselage, 2 in wings, 3 in tail and 4 in nose)
A finite element (FE) model has been developed to represent the geometry as modeled and is used to determine
parameters such as material thicknesses, ply build-up etc. Analysis is done using two-dimensional shell elements.
Point masses are used to represent components such as engines, fuel tanks, landing gear etc. These point masses are
then coupled to the geometry using beam elements. An initial set of analyses includes a 3-g landing, 1-g
deceleration due to parachute recovery and a 4.5-g maximum aerodynamic loading where trimmed flight wing6
loading results from the Vortex Lattice Method (VLM) analyses are scaled by 4.5 and distributed on the FE model.
Figure 6 shows the vortex lattice model that was used to calculate the aerodynamic loads on the aircraft.
Figure 6. Vortex Lattice Model Used to Calculate Aeroedynamic Loads
Since the required scaled maximum takeoff weight is quite high for the test point (204.5 lb), the first RPV has
room to be structurally overbuilt. Once testing is complete on the Rigid RPV, structural optimization will be
performed to save weight for the second, aeroelastically scaled RPV. All bulkheads and ribs are assigned material
properties of woven 0o/90o s-glass/epoxy laminate with 0.12 in thickness. Spars are assigned properties of
unidirectional carbon/epoxy with a thickness of 0.39 in for forward spars and 0.20 in for rear and boom spars. All
skin material consists of 0.08 in of 0o/90o s-glass placed at a 45o bias along the span. Parts are sized such that a
minimum safety factor of 2 is maintained to account for uncertainties in manufacturing. Future structural testing of
finished rigid RPV components may lead to structural optimization to save weight; however, minimizing structural
weight is not a driving design consideration as ballast is needed to achieve the high test point weight.
When designing the detailed parts in the computer aided design (CAD) environment, special attention is paid to
the ease of construction and assembly. All rib and spar elements are planar and can be machined from laminate
sheets using a CNC router table. When possible, locating features are built into the parts themselves. For instance,
all of the bulkhead pieces have grooves designed into them so they interlock, making the structure stronger and
easier to assemble, as shown in Figure 7. In addition to the detailed parts themselves, special tooling (such as jigs for
building the wing assembly) have been designed. This will greatly speed up the construction process and improve
the accuracy of the overall aircraft geometry.
Figure 7. Exploded View of Fuselage Bulkheads
7
C. Aeroelastic Scaling
Once testing of the Rigid RPV is complete, the next step is to fabricate and test an aeroelastically scaled model
(refered to as Aeroelastically Scaled RPV for remainder of the paper). Aeroelastic scaling for a flight model is a
difficult task, and it presents numerous challenges over scaling of a wind tunnel model including structural
discontinuities (wing joints, access panels, etc.), added masses due to supporting systems and space considerations
for components and fuel. All of these considerations must be included while still ensuring an accurate aeroelastic
response.
The most accurate method for matching the full scale JWSC mass and stiffness distributions is to simply scale
down the exact geometry. Unfortunately, this is not practical in the case of the JWSC for several reasons. For one,
the exact internal structure has not been released. In addition, manufacture of scaled components may be very
expensive and in some cases impossible due to their small resulting sizes. Accordingly, an alternative method for
matching the model’s mass and stiffness distribution to that of the full scale aircraft has been developed.
The mass and stiffness distributions are not available to directly define the structural properties of the
SensorCraft; however, mass normalized modal shapes and natural frequencies of the aircraft are given. From the
governing linear elasticity equations it can be shown that any model with the same scaled mass and stiffness
distribution, over the same scaled geometry, will result in the same modal shapes and frequencies as those of the full
sized aircraft (after appropriate scaling). Additionally, if the aerodynamics correspond then the aeroelastic response
will be equivalent.
By designing a model with similar modal response (mode shapes and scaled frequencies) to the full scale
aircraft, similarity of the mass and stiffness distributions are achieved independently of internal structure. This
allows a simplified internal structure to be employed. The requirements of the structure are to allow a proper
stiffness distribution while still fitting into available space, as well as not using up too much of the available mass.
Once this is achieved, concentrated masses are added to attain the desired mass distribution.
The scaling process employed is similar to that performed in the previous work 6-10. This method is applied to the
joined wing using an optimization framework shown in Figure 8 below which is built using Phoenix Integration’s
Model Center 9.0 Software.
Figure 8. Optimization Framework for Mode-Shape Matching
Figure 8 shows two separate routines, the first involves matching a set of scaled deflections due to a known load,
in order to achieve the desired stiffness distribution. The second routine tunes the mass distribution, of the
stiffness matched model, by tuning a distributed set of point masses placed throughout the structure. Figure 9
below shows preliminary results of the process applied to the joined wing to match the first six mode shapes of the
baseline aircraft using a low fidelity, one dimensional beam FE model. The model, which was created for use the
Nonlinear Aeroelastic Simulation Toolbox (NAST11), is shown in Figure 10. The goal of using this preliminary,
8
low fidelity model, is to get an initial approximation of the structural properties that serves as a starting point for
subsequent optimizations of higher fidelity models.
Figure 9. Comparison of Baseline Modal Results
(Points) and Results of Optimized Beam Model
(Lines)
Figure 10. Beam Model Used for Initial Scaling
Study
Further work is being performed to apply these methods using a higher fidelity, FE model similar to that shown
in Figure 11.
a)
b)
Figure 11.
a) Full Scale Aircraft Response and b) Matched Reduced Scale Displacements
VI. Propulsion
The following section outlines the process for determining the required thrust, specifying suitable engines and
integrating them into the aircraft. A brief description of several challenges are also discussed including the high fuel
weight required and its effect on stability.
A. Engine Sizing
Engine selection and sizing is performed based on the military specifications for stall speed, take off distance
and climb requirements (MIL-C5011A). These calculations use aircraft drag polar data obtained through wind
tunnel tests performed at NASA Langley on a 12ft span model 3. The maximum required thrust of approximately
90lb calls for the use of two JetCat P200 Turbine Engines embedded within the fuselage. Using the experimental
drag data and manufacturer’s fuel consumption data for the chosen engines, a typical mission, as shown in Figure
12, uses approximately 46 lbs of fuel12.
9
Figure 12. Typical Flight Test Mission Profile
In order to simplify the design of the aeroelastic equivalent aircraft, no fuel will be stored in the wings. This
requires that the main fuel compartments be placed in the fuselage as far aft as possible in order to have minimal
impact on the center of gravity (CG) location throughout flight (the CG location required for stability is located near
the rear of the center-body as shown in Figure 13 below). While the aircraft is capable of carrying more than the 21
kg of fuel, the location of the tanks in the center-body limits the fuel weight that can be stored without affecting the
overall desired mass distribution. If too much of the payload weight were added in fuel (contained in the centerbody), there would be insufficient mass left to distribute throughout the aircraft’s wings in the form of tuning masses
(required to have proper mass distribution for equivalent aeroelastic response).
Figure 13. Final Component Space Reservations
In any flight test situation the goal is to maximize the time on station in order to gather the largest amount of data
possible while minimizing the risk posed during takeoff and landing. An additional challenge posed by a remotely
piloted vehicle is in the positioning of the aircraft on station for gathering meaningful data while still allowing the
pilot the appropriate line of sight. The relatively high specific fuel consumption of JetCat turbine engines poses a
large challenge for the above reasons as well as the added weight and CG shift. Future work is planned that will
investigate these effects using a series of bench tests of the engines as well as a flight simulator testing.
B. Other considerations
The inlet and hot gas passages within the fuselage have been directly scaled for the RPV with the exception of a
slight narrowing of the section around the turbines. This portion has been narrowed to create the required 0.75 in
10
clearance around its perimeter of each turbine to allow air passage for cooling. Figure 14 shows a cross section of
the modified fuselage ducting.
Figure 14. Fuselage Cross-section Showing Inlet and Exhaust Routing
The engine is placed near the rear of the pass-through for two reasons. The first is to directly route the hot
exhaust out the fuselage minimizing heat transfer, and the other reason is to increase the cooling airflow around the
engine. Placing the engine at the point where the duct begins to constrict sets up a venturi effect which helps to draw
cooling air around the outside of the engine. This is especially important when the aircraft is stationary, as in taxi
and startup.
Due to the danger posed by ingesting foreign objects into the turbine, mesh strainer screen is placed across the
inlet to the fan opening as shown below.
Filter Screen
Figure 15. Strainer Screen to Prevent Ingestion of Debris
VII. Stability and Control
A detailed analysis of joined wing flight dynamic stability has been performed using High Angle of Attack
Stability and Control software (HASC13) and Athena Vortex Lattice code (AVL14) models. These initial
investigations were performed to determine aerodynamic coefficients and static stability characteristics. These,
along with the scaled aircraft moments of inertia, were used to determine overall dynamic stability. Results are
graded using Military Specifications for Flying Qualities (MIL-F-8785C) and show that while most of the flight
modes display good (level I) flying qualities, the Phugoid mode has marginal (level II) qualities while the Dutch roll
mode is worse still, being slightly unstable. (see Figure 16). An additional area of concern is a potential lack of yaw
control authority as the present configuration does not have a rudder surface.
11
Figure 16. Plot Showing Stability Modes for Baseline Configuration
In order to address these challenges, three proposed modifications to the initial prototype are being investigated.
The first solution, depicted in Figure 17a, involves the addition of a vertical tail surface aft of the tail boom which is
fitted with a control surface. Figure 17b shows the addition of a conventional tail boom to the initial flight test
article. This would add additional pitch authority from an elevator surface mounted to the new horizontal surface,
while also adding the of the vertical tail. The final solution, shown in Figure 17c, involves scheduling the existing
surfaces such that they offer a degree of yaw authority. One proposed control surface schedule includes the mixing
of differential deflection on the outboard ailerons and outboard elevators creating a “split flap” type effect for yaw
control; however, the use of half of the elevator as a yaw control device could lead to unsatisfactory pitch control. A
second proposed control surface schedule includes the mixing of differential deflection on the outboard ailerons and
inboard ailerons. While this scheduling would create improved yaw control authority, roll authority comes into
question during high sideslip maneuvers.
a)
b)
c)
Figure 17. Proposed Modifications for Increased Stability/Yaw Authority
HASC and AVL analyses of the first two modifications (a and b above) show an improvement in overall
dynamic stability. Both configurations result in improved Dutch roll mode stability (Level II) while the second
configuration also results in a slight improvement in the phugoid mode (but still Level II). In addition, both
configurations result in acceptable levels of yaw authority. Of the three modifications, the third configuration is the
most desirable since the external aerodynamics, and therefore aeroelastic response, is not affected; however, as
mentioned above, the use of the ailerons and/or outboard elevators for yaw control could result in diminished pitch
and/or roll control authority. Efforts are being carried out to investigate potential control surface schedules using
flight simulators (see Section VIII for more information).
In addition to simulation, three reduced complexity JWSC RPV models are being created to experimentally
validate proposed control surface schedules. Each aircraft is of increasing complexity and better represents the
actual aircraft geometry. The two completed models are shown in Figure 18 below (see Section IX and the
Appendix for more information regarding the reduce complexity models and current testing).
12
Figure 18. Reduced Complexity Models used for Investigation of Stability and Control
VIII. Simulation
Simulation for the JWSC FTP is broken into three categories: 6 Degree of Freedom (DoF) flight simulation,
hardware in the loop simulation and pilot training. Each category is described in the following section.
C. 6 Degree of Freedom Flight Simulation
The 6 DoF simulator currently in use is based on the Matlab Aerospace Blockset and the Unmanned Dynamics
Aerosim v1.2 Blockset15. The simulator models the aircraft using a set of first order terms. The model is been altered
to allow non-linear aerodynamics to be modeled through the use of multi-dimensional lookup tables. This allows
coefficients to be input for various angles of attack and sideslip angles.
A parametric model is developed using Phoenix Integration’s Model Center Software (MC) 16. This model
includes the vortex lattice software, AVL14. The MC model runs the AVL analysis through an array of angle of
attack and sideslip values. It then parses all the output data and compiles the coefficient arrays into a Matlab m-file.
This resulting m-file is in a format that is required by the simulator to initialize an aircraft model. The end result is
that when modifications are made to the aircraft’s configuration, the new aerodynamics will be simulated in the
flight simulator in a matter of minutes. This will allow rapid investigation of design changes by both the engineer
and the pilot.
The simulator calculates the aircraft states and then outputs this to an open source flight simulator, FlightGear
for visualization15. A custom block has also been created and incorporated that allows the pilot to control the aircraft
using a standard radio controlled (R/C) transmitter attached to the computer through the “training” port. Figure 19
shows a typical flight simulation with the Simulink model, custom gauges and FlightGear all running
simultaneously.
Figure 19. 6 Degree of Freedom Flight Simulator
D. Hardware in the Loop Simulation
The simulator is taken one step further as a hardware in the loop setup has recently been acquired that allows the
aircraft autopilot to be incorporated into the simulation. The Micropilot 2128 THWIL (True Hardware In the Loop)
system uses data acquisition hardware to interface the 2128 autopilot within the simulation. The hardware in the
loop setup electrically stimulates all sensor outputs using analog to digital conversion, signal conditioning and pulsewidth-modulation boards. This allows a higher level of fidelity by more closely replicating in-flight conditions while
13
on the ground. Micropilot supplies their own Simulink model to achieve this, but this has been modified to replace
their physics model with the one developed by the authors.
E. Pilot Training in RealFlight G5
The unique configuration and flight characteristics of the JWSC makes flying the RPV a challenge for even the
most experienced pilots. The 6 DoF simulator previously discussed addresses this problem by allowing the pilot to
fly the simulator using a conventional R/C transmitter. Unfortunately, a gap between the simulation and reality still
exists because the FlightGear visualization program was designed for manned aircraft simulation, and accordingly
the flying sites and pilot spawn locations (where the user can view the aircraft from) are not ideal for RPV flight
simulation. In order to address these problems and further reduce the gap between simulation and RPV flight, a
model was created for use in RealFlight G5, the most advanced R/C flight simulator in Knife Edge Software’s
widely used R/C flight simulator line17. A screenshot of the JWSC RPV in flight in RealFlight G5 is shown in
Figure 20.
Figure 20. Screenshot from RealFlight G5
As RealFlight G5 was designed with an R/C pilot and modeler in mind, it offers the ability to rapidly modify
aircraft configurations including the incorporation of hundreds of pre-programmed, commercially available R/C
components such as servos, motors, propellers, Electric Ducted Fans (EDFs), turbines, etc. Accordingly, RealFlight
provides an ideal first look into the feasibility and applicability of proposed changes to the RPV and its components.
Additionally, the photo-realistic recreation of R/C flight is a valuable tool when designing and preparing for the
Rigid RPV and Aeroelastically Scaled RPV flight test programs. Unfortunately, the 6 DoF physics model used in the
RealFlight G5 software is proprietary and currently unavailable, and accordingly all in flight responses of the JWSC
RPV are also tested using the 6 DoF simulator.
IX. Flight Testing
Flight Testing of the JWSC presents two unique but interlinked goals. The first goal is that both the Rigid RPV
and the Aeroelastically Scaled RPV are flown safely, and successfully through all required flight test maneuvers.
The unconventional configuration and control surface layout of the JWSC, coupled with the large scale and
complexity of the RPVs make this a significant challenge. The second goal is that all data required to validate the
aeroelastic scaling and measure the non-linear, post-buckled aeroelastic response is successfully acquired during the
FTPs. The design and testing of an instrumentation system for each FTP as well as the incorporation of these
systems into each RPV present another challenge.
At the onset of this program, it was decided to initially address each of the above mentioned goals individually
through the use of reduced complexity flight tests. These tests, which include the fabrication and flying of sub-scale
JWSC models for configuration testing as well the use of conventional “trainer” style, high wing aircraft for
instrumentation testing, allow for an incremental increase in complexity. The overall aim is to merge the lessons
14
learned from both the configuration and instrumentation/ data collection testing into the final RPV, thereby
increasing the chances of a safe and successful FTP. In addition, preliminary flight testing allows for the pilot’s
perspective to be included in the design and optimization of the RPV.
A. Reduced Complexity JWSC Models
This section details current progress in the series of three reduced complexity JWSC models. The purpose of
each model is presented along with important design characteristics. For more details on these models, including
complete design specifications and pictures see the Appendix.
1. Flat Plate Foamie
The first sub-scale model created is a reduced scale (b=66 in) flat plate foam aircraft with a flying weight of 2.2
lbs. This aircraft is designed to serve as an experimental platform to test and validate proposed control schemes as
well as address the yaw stability and control authority questions. The model, shown in flight in Figure 21, is
constructed of 0.25 inch flat extruded polystyrene fanfold insulation underlayment board with carbon fiber arrow
shaft spars. The streamer shown in Figure 21 is added to aid the pilot in referencing the aircraft during flight. The
unique, diamond shape of the JWSC proved to be difficult to track, especially during turns, due to the similarities
between the profile of the aircraft from the front and side when viewed from a distance.
Figure 21. Flat Plate Foamie in flight during a flight test.
When designing the model, a balance was struck between the simplicity and operability of the model and the
desired accuracy of the representation of the flight dynamics of the full scale RPV. The reduced scale features
dihedral, anhedral and sweep angles representative of the full scale geometry; however, the camber and complex
spanwise twist distribution present in fore and aft wings was omitted due to the selected material. Additionally, the
turbine propulsion system was replaced with a simplified, high thrust-to-weight ratio propeller setup for ease of
operation and maintenance. Finally, the laminar flow airfoils selected for the full scale RPV are replaced by flat
plates due to the small scale of the model and the chosen materials.
A series of flight tests were completed to first establish an acceptable CG location for the model and trim the
aircraft. Once this was completed, remaining flight tests were used to test the proposed control surface layout for
the model, shown in Figure 22, assess control authority, investigate yaw stability and gain pilot experience flying a
joined wing aircraft. These flight tests yielded several results. The first and most important result is that the aircraft
was able to maintain stable, controlled flight with the proposed control surface layout. Yaw control authority is
minimal; however, this is not a problem for controlled flight. One major problem encountered was the sensitivity of
the aircraft to the thrust line of the motor. Approximately eight degrees of declination is required on the motor
mount to allow for stable, level flight. Without this declined thrust line, the aircraft pitches up oncontrollably when
power is applied. A second major problem is the large angle of attack required to maintain level flight and the
tendency of the aircraft to tip stall in sharp turns. This second problem was primarily caused by the low flight speed
afforded by the propulsion setup chosen for this model, but the absence of the spanwise twist distribution could have
also played a large factor.
15
Elevtor
Aileron
Rudder
Flaperon
Figure 22. Control surface layout for Flat Plate Foamie
In summary, the Flat Plate Foamie provides a simple, first attempt at flying a JWSC model. The lessons
learned and the problems encountered were used to design and test the second model, the EDF Twist Corrected
Foamie.
2. EDF Twist Corrected Foamie
The second sub-scale model created, the EDF Twist Corrected Foamie, was designed and built to correct the
shortcomings of the Flat Plate Foamie. The model, shown in flight in Figure 23, is constructed of extruded
polystyrene foam wrapped in fiberglass with unicarbon strips to provide stiffness to the lifting surfaces. When
designing the model, the two primary goals were to include the proper twist distribution in both the fore and aft
wings and to include an EDF propulsion system.
Figure 23. EDF Twist Corrected Foamie in flight during a flight test.
Preliminary flight testing has shown a marked improvement over the Flat Plate Foamie. As with the Flat
Plate Foamie, the aircraft is able to maintain stable, controlled flight with the chosen control surface layout. Yaw
control authority is minimal; however, this does not prove to be a problem for controlled flight. The pilot
commented that the rudder is not required for acceptable handiling qualities; however, the necessity of the surface in
a crosswind landing situation is still in question.
The EDF propulsion system created two positive results. The first is higher flight speeds than with the Flat
Plate Foamie, and accordingly the danger of tip stalling in turns is greatly reduced. The pilot commented that as
long as the aircraft maintains speed through turns and long shallow turns were taken as opposed to abrupt high bank
angle maneuvers, the aircraft behaves favorably with predictable behavior. The second positive result of the EDF
propulsion system is the removal of the necessity for a declined thrust line. Moving the propulsion system to the
16
center of gravity as well as the inclusion of the proper twist on the fuselage and aft wings corrected the pitch up
tendency of the Flat Plate Foamie.
One major problem that was encountered was the sensitivity of the aircraft to pitch up and lose control at low
speeds just after takeoff. Upon launch (which for this aircraft is a hand launch), pitch input resultes in a sharp nose
up tendency, which may blanket the aft wing control surfaces or cause the aft wing to stall. Even with the
application of power and down elevator input, the aircraft is unable to recover from this attitude. This problem is
easily avoided by waiting for the aircraft to accelerate while flying level before trying to climb; however, it brings
calculated rotation speed and takeoff speed into question. Future testing will include the incorporation of various
landing gear configurations to test takeoff conditions.
3. Mini SensorCraft
The final iteration of the reduced complexity flight tests consists of a directly scaled, 37.5% model of the JWSC
RPV (b=73 in) powered by twin 2.75 in diameter EDFs. To create this model, the female molds created for the full
scale RPV are scaled to 37.5% of the RPV mold size. By using these molds, the outer mold line of the Mini
SensorCraft will be identical to that of the full scale RPV, thus providing the best reduced-scale model possible.
The molds are currently being machined for the Mini SensorCraft, and fabrication will be started immediately upon
the completion of the molds. For more details on the Mini SensorCraft and the proposed design, see the Appendix.
B. Instrumentation/ Data Collection Testing
In addition to the three reduced complexity JWSC models for configuration testing, instrumentation and data
collection testing is being completed using a commercial “trainer” style, high wing aircraft. The Senior Telemaster,
shown in Figure 24 is an extremely effective trainer with docile flying qualities. The large size (94 in wingspan)
allows sufficient payload capacity for the autopilot, onboard instrumentation and an oversized fuel tank.
Figure 24. Senior Telemaster Trainer Used For Instrumentation and Data Collection
The autopilot that is being tested is a MicroPilot 2128LRC dual modem system. By using an autopilot as the
primary datalogging and data acquisition system on board of the aircraft, many key sensors are already integrated
into a convenient package. Aircraft altitude, attitude, airspeed, groundspeed and servo commands (servo
deflections) are just several of the key parameters that the MicroPilot is able to log. The MicroPilot also has built in
gyros/accelerometers which allow for the incorporation of stability augmentation, if this is deemed necessary for the
JWSC. Finally, the MicroPilot also incorporates a backup transmitter in the event of signal loss between the pilot
and the aircraft on the primary transmitter.
Flight testing to date has included data logging and reduction, video overlay and camera stabilization as well
as autonomous operation and stability augmentation. Current tests focus on development and optimization of flight
test procedures including flight maneuver testing and custom instrumentation integration and testing. A flyover
during an instrumentation flight test is shown in Figure 25 and the ground station used for flight test operations is
shown in Figure 26.
17
Figure 25. Senior Telemaster in flight during
an instrumentation flight test
Figure 26. Ground station used for data
collection
C. Proposed Rigid Flight Test Program
The results from the reduced complexity flight tests and the instrumentation and data collection flight tests will
be used to plan the rigid FTP. The goal is to have the lessons learned from all of the preliminary flight testing
promote a safe and successful flight test program for the full scale Rigid RPV.
The Rigid RPV FTP, for the airworthiness program, will begin with ground tests to ensure all instrumentation
and the onboard control system are functioning correctly. This will include static structural loadings to validate
structural design and a bifilar pendulum test to validate scaled moments of inertia. This will lead into taxi tests and
eventually short “trimming” flights to ensure that the aircraft has satisfactory stability and adequate control in roll,
pitch and yaw. The final phase of the Rigid RPV FTP will consist of maneuvering and dynamic characterization.
The overall goal of the Rigid RPV FTP is to validate the flightworthiness of the aircraft, test the onboard
instrumentation system, gain pilot experience and lay the groundwork for the future Aeroelastically Scaled RPV
FTP.
D. Proposed Aeroelastic Scaled Flight Test Program
The Aeroelastically Scaled RPV FTP, for the aeroelastic response program, will begin with ground tests to
ensure all instrumentation and the onboard control system are functioning correctly. This will include the same
static structural loading and bifilar pendulum tests that were used in the Rigid RPV FTP, but will expand to include
ground vibration testing to validate the matched mode shapes. The initial flights will be identical to those in the
Rigid RPV FTP including taxi tests and short “trimming” flights. The final phase of the Aeroelastically Scaled RPV
FTP will include the high risk maneuvers including control surface excitation for low frequency aeroelastic
response, an on-board excitation system for high frequency response (if needed), and wind-up turns for static
nonlinear aeroelastic response.
X. Conclusion
The aeroelastic scaling and flight testing of a 1/9th scale JWSC is the topic of an ongoing international
collaboration between Virginia Tech and The University of Victoria (Canada). A multi staged flight test program
has been adopted which involves the fabrication and testing of a series of aircraft with increasing complexity. This
paper has outlined the development of this program including the design of the Rigid RPV, the testing of an
autopilot system using a commercially available R/C trainer, two reduced complexity JWSC models and numerous
flight simulators. Future work involves the fabrication and testing of a reduced scale, Mini Sensorcraft as well as the
1/9th scale Rigid RPV. Upon the completion of the FTP for the Rigid RPV , the airframe will be modified with
flexible, aeroelastically scaled fore/aft wings and boom to produce the Aeroelastically Scaled RPV. This program
will culminate with the flight testing of the Aeroelastically Scaled RPV.
18
Appendix
This appendix serves as a reference on each of the six aircraft involved in this study. Photographs (or CAD
renderings) each aircraft are presented along with design specifications and a description of key design features.
A. Senior Telemaster
Figure A-1. Senior Telemaster Used For Development of Autopilot and Instrumentation Systems
Figure A-2. Senior Telemaster During Autonomous Flight
Description
Due to the unconventional configuration and unknown flying
qualities of the JWSC, provisions for an autopilot are included to allow
for stability augmentation. In addition, the autopilot allows for safety
features such as an electronic fence, return to home and the ability to
release a parachute etc. In addition, the autopilot offers a convenient way
to both log and send telemetry of the aircraft states in real time as well as
offering 16 analog to digital inputs for items such as rpm sensors, strain
gauges etc.
Before integrating and operating the autopilot within the JWSC, it
was decided to test the setup using a commercially available trainer. The
Senior Telemaster exhibits docile flying characteristics while offering
sufficient cargo space and payload carrying capabilities.
Flights have been performed under both manual and autonomous
modes and a basic flight test program performed in order to characterize
the aircraft and tune the PID control loops. Future tests will include the
testing of instrumentation systems that are slated for the Rigid RPV and
Aeroelastically Scaled RPV flight test programs.
Figure A-3. Mobile Command Center
Senior Telemaster Specifications
General
Span
Length
Wing Area
TOGW
Construction
Airframe
Stiffeners
Propulsion
Engine
Propeller
Thrust
Other
Stability
Augmentation
Telemetry
Other
19
94 in
64 in
1330 in2
19 lbs
Balsa/Monokote
Carbon Fiber
reinforcements
Saito 1.8 ci 4-stroke
APC 16 x 8
14 lbs
Micropilot 2128LRC
Autopilot
Aircraft states
Onboard video with
telemetry overlay,
stabilized camera
gimbal
B. Flat Plate Foamie
Figure A-4. Flat Plate Foamie
Figure A-5. Underside of Aircraft Highlighting
Carbon Fiber Stiffeners
Description
The Flat Plate Foamie, shown in in Fig. A-4, represents the
lowest complexity model of the JWSC used in this program, and
also the first to be designed and flown. When designing the model,
a balance was struck between the simplicity and operability of the
model and the desired accuracy of the representation of the flight
dynamics of the full scale RPV. The Flat Plate Foamie features a
flat plate, extruded polystyrene airframe stiffened with carbon fiber
arrow shafts, as shown in Fig. A-5. Dihedral, anhedral and sweep
angles match full scale geometry; however, the complex spanwise
twist distribution present in fore and aft wings was omitted as this
was impossible to replicate with commercially available fanfold
foam without the inclusion of custom, composite stiffeners.
Additionally, the turbine propulsion system was replaced with a
simplified, high thrust to weight ratio propeller setup for ease of
operation and maintenance. Figure A-6 shows the Flat Plate
Foamie in flight during a flight test.
Figure A-6. Flat Plate Foamie in Flight
Flat Plate Foamie Specifications
General
Length
Span
TOGW
Construction
Airframe
Stiffeners
Control
Transmitter
Reciever
Servos
Propulsion
Motor
ESC
Propeller
Battery
20
42.25 in
66 in
2.2 lbs
0.25 inch extruded
polystyrene
Carbon Fiber arrow shafts
Futaba T10CAP
Futaba R319DPS
7 x Hextronik HXT 900
TowerPro BM2409-18T
TowerPro MAG8 25 Amp
GWS 10 x 4.7
Turnigy 1000mAh 3 Cell
LiPo
C. EDF Twist Corrected Foamie
Figure A-7. EDF Twist Corrected Foamie
Figure A-8. Aft Wing in Jig Under Vaccum to Create
Spanwise Twist Distribution
Description
The EDF Twist Corrected Foamie, shown in Fig. A-7, represents
a marked improvement over the Flat Plate Foamie. As with the Flat
Plate Foamie, dihedral, anhedral and sweep angles are
representative of the full scale geometry; however, for the EDF
Twist Corrected Foamie the airframe is fabricated using extruded
polystyrene wrapped in a fiberglass/ expoxy composite with
unicarbon strips are used as stiffeners in the lifting surfaces and tail
boom. Wing jigs, such as the one shown in Fig. A-8 and Fig. A-9,
were used to create the complex spanwise twist distribution present
in fore and aft wings. A 2800 kV brushless outrunner motor
spinning a 6 blade, 2.75 in diameter EDF was used to power this
model. Figure B-4 presents the EDF Twist Corrected Foamie in
flight.
Figure A-9. EDF Twist Corrected Foamie in Flight
EDF Twist Corrected Foamie Spec.
General
Length
Span
TOGW
Construction
Airframe
Stiffeners
Control
Transmitter
Reciever
Servos
Propulsion
Motor
ESC
Propeller
Battery
21
42.25 in
66 in
2.6 lbs
0.28 inch extruded polystyrene
3.2 oz/in2 fiberglass/epoxy
composite
Unicarbon strips
JR 1221
JR 12X
2800 kV Brushless Outrunner
45 Amp
6 Blade, 2.75 in diameter
Turnigy 2450 mAh 4 cell LiPo
D. Mini Sensorcraft
Figure A-10. Mini Sensorcraft
Figure A-11. Relative Size of Mini Sensorcraft
Compared to Rigid RPV
Description
The Mini Sensorcraft is similar in size to both the Foamie
Sensorcraft and the Twist Corrected Foamie but employs the exact
outer mould line of the full scale aircraft, including accurate airfoils
and twist distribution. This variant will be the most accurate
reduced complexity model before the final Rigid RPV and
Aeroelastically Scaled RPV.
The aircraft will be fabricated using CNC machined female
moulds. The outer skin will consist of carbon/epoxy composites and
two part polyurethane foam will be expanded within the fuselage
and wings. The aircraft will use two 2800 kV brushless outrunner
motors spinning 6 blade, 2.75 in diameter EDFs in order to more
accurately reproduce the scaled thrust to weight of the given test
point.
Additional testing of landing gear configurations, control surface
scheduling, stall characteristics and aircraft trim will be determined
through flight testing.
Figure A-12. Assembly of Mini Sensorcraft Used
for Glide Tests
Mini SensorCraft Specifications
General
Length
Span
TOGW
Test Point
Construction
Airframe
Stiffeners
Control
Transmitter
Reciever
Servos
Propulsion
Motor
ESC
Propeller
Battery
22
42.25 in
66 in
3.6 lbs (Estimated)
10.8 lbs
2 part PU foam core expanded
into 3.2 oz/in2 fiberglass/epoxy
composite skins
Unicarbon
JR 1221
JR 12X
2 x2800 kV Brushless
Outrunner
2 x 45 Amp
2 x 6 Blade, 2.75 in diameter
2 x 2450 mAh 4 cell LiPo
E. Rigid RPV
Figure A-13. Internal Structure of Rigid RPV
Figure A-14. Female Fuselage Moulds
Figure A-15. Cross Section of Aft Wing
Description
The Rigid RPV is a 1/9th scale prototype of the JWSC. All
external geometry is preserved along with the overall scaled mass
and moments of inertia. The structure consists of a main fuselage
or center body made up of a carbon/epoxy skin stiffened with a
series of laminate bulkheads. The fuselage fabricated using a set of
CNC machined moulds as shown in Figure A-14 above.
The fuselage has space reservations for all avionics, upwards
of 40lb of fuels, two turbine engines and possibly retractable
landing gear. The forward wings consist of a CNC molded
composite skin with conventional rib and spar internal structure.
The wing box consists of a fore and aft shear web connecting the
spar caps consisting of uni directional carbon. The main wing is
attached to the fuselage using two carbon wing joiner tubes with a
similar system used to join the fore/aft wings. The rear wings are
constructed using a foam core with a fore and aft composite shear
web connecting top and bottom uni-carbon spar caps (see Figure
A-15). The internals consist of a conventional rib and spar
structure with the outer skin being constructed using CNC
machined female moulds.
While not aeroelastically equivalent to the full scale aircraft,
the Rigid RPV plays a critical role in the flight test program. It
serves as an important stepping stone towards the final aircraft, one
which allows the authors to determine and test required supporting
systems while also training pilots and other support crew.
Rigid RPV Specifications
General
Span
Length
Wing Area
TOGW
Construction
Airframe
Scaling
Propulsion
Engine
Thrust
Other
Stability
Augmentation
Telemetry
Other
23
16.4 ft
10.6 ft
15.0 ft2
205 lbs
Carbon/ epoxy composite
buildup
Exact geometry (OML), rigid
body dynamics, aerodynamics
2 x JetCat P200sx
104 lbs
MicroPilot 2128LRC
Aircraft states, engine
monitoring, fuel flow, strain
gage measurement
On board video with telemetry
overlay, parachute recovery
system
F. Aeroelastically Scaled RPV
Figure A-16. Aeroelastically Scaled RPV
Description
The Aeroelastically Scaled RPV is the final aircraft and
will involve re-winging the fuselage of the Rigid RPV with
aeroelastically equivalent fore/aft wings and boom. However,
additional work is required to redistribute weight in the
fuselage in order to achieve the overall required mass
distribution of the aircraft.
The goal of the aeroelastically flight testing stage is to
determine validate the scaling work and therefore
instrumentation is required to measure mode shapes and
frequencies on the ground as well as in flight. This testing will
involve instrumenting the wing for initial ground vibration tests
as well as data logging and telemetry in flight. Several methods
are being considered for in flight measurements including
stereoscopic cameras to record wing deflections.
Aeroelastically Scaled RPV Specifications
General
Span
Length
Wing Area
TOGW
Construction
Airframe
Scaling
Propulsion
Engine
Thrust
Other
Stability
Augmentation
Telemetry
Other
24
16.4 ft
10.6 ft
15.0 ft2
205 lbs
Carbon/ epoxy composite
buildup
Linear aeroelastic equivalence
2 x JetCat P200sx
104 lbs
MicroPilot 2128LRC
Aircraft states, engine
monitoring, fuel flow, strain
gage measurement, structural
deformations
On board video with telemetry
overlay, parachute recovery
system, stereoscopic camera
system for measuring wing
deflection
Acknowledgments
AFRL funding, the support of MSTC Director, Dr. Raymond Kolonay, project managers Dr. Ned Lindsley, Dr.
Max Blair, and Dr. Pete Flick is gratefully acknowledged as is the guidance and support of Dr. Craig Woolsey, Dr.
Rakesh Kapania, the Virginia Tech CCMS and José Costa.
References
1
Blair, M., Canfield, R. A., and Roberts Jr., W., “Joined-Wing Aeroelastic Design with Geometric Nonlinearity,” Journal of
Aircraft, Vol.42, No.4, 2005, pp. 832-848.
2
Bond, V., “Flexible Twist for Pitch Control in a High Altitude Long Endurance Aircraft with Nonlinear Response,” Ph.D.
Dissertation Prospectus, Department of the Air Force, Air Force Institute of Technology, Wright-Patterson AFB, Dayton, OH,
2007.
3
The Boeing Company. “Selected Results from Test of Model on Sting”, Presentation Excerpts, NASA Langley, Hampton
VA, slides 2-5.
4
Lucia, David J., “The SensorCraft Configurations: A Non-Linear AeroServoElastic Challenge for Aviation,” AIAA Paper
2005-1943, 2005.
5
Roberts, R. W., Jr., Canfield, R. A., and Blair, M., “Sensor-Craft Structural Optimization and Analytical Certification,”
AIAA Paper 2003-1458, April 2003.
6
Pereira, P., Almeida, L., Suleman, A., Canfield, R., Bond, V., and Blair, M., “Aeroelastic Scaling and Optimization of a
Joined-Wing Aircraft,” AIAA Paper 2007-1889, April 2007.
7
Blair, M., Garmann, D., Canfield, R., Bond, V., Pereira, P., and Suleman, A., “Non-Linear Aeroelastic Scaling of a JoinedWing Concept,” AIAA Paper 2007-1887, April 2007.
8
Bond, V. L., Canfield, R. A., Cooper, J., and Blair, M., “Scaling for a Static Nonlinear Response of a Joined-Wing Aircraft,”
AIAA Paper 2008-5849, September 2008.
9
Bond, V. L., Canfield, R. A., Suleman, A., and Blair, M., “Aeroelastic Scaling for Verification and Evaluation of Geometric
Nonlinearity on a Joined-Wing Aircraft Model,” AIAA Paper 2008-6073, September 2008.
10
Richards, J., Suleman, A., Canfield, R., and Blair, M., “Design of a Scaled RPV for Investigation of Gust Response of
Joined-Wing Sensorcraft,” AIAA Paper 2009-2218, May 2009.
11
Cesnik, C., Shearer, C. and Su,W., Nonlinear Aeroelastic Simulation Toolbox (NAST) [Software], version NAST-R2-0209-2008, 2008.
12
Jetcat Model Engines Germany, Jetcat Products/Turbojets/P200, http://jetcat.de/jetcatturbinen/strahlturbinen.htm, 2009.
13
NASA Langley Research Center, High Angle of Attack Stability and Control 95 [Software],
Version 95, 1995
14
Drela, M. and Youngren, H., Athena Vortex Lattice (AVL) [Software], http://web.mit.edu/drela/Public/web/avl/ , version
3.27 ed., 2009.
15
Unmanned Dynamics, LLC, Aerosim Blockset [Software], version 1.2 ed, 2004
16
Phoenix Integration, Model Center [Software], Version 9.0, http://www.phoenix-int.com/software/phx_modelcenter.php,
2008.
17
Flight Gear Flight Simulator [Software], Version 2.0, http://www.flightgear.org, 2008.
25

here - Quaternion Aerospace

Transcription

Similar documents

Please use one form per client, send the filled booking form back to

For more information, contact Kevin Halpin | khalpin@elitetest.com

KING AIR C90GTx - Beechcraft

FLIGHT SAFETY Technology and the Human Factor

Wherever you are. Whenever you need it.

TerraMap-3D® TerraMap-3D®

UAV Systems and Security

aerocet 6650 installation pricing

SD Flight Tracker

PDF to - CTR-FS