Prof. Jean-Marc Moschetta Professor of Aerodynamics Department of Aerodynamics Energetics and Propulsion ISAE BP54032, 31055 Toulouse Cedex France
[email protected]. Hovering Capabilities of Fixed-Wing Micro-Aerial Vehicles Jean-Marc Moschetta1, Institut Supérieur de l’Aéronautique et de l’Espace, Université de Toulouse, France and Sergey V. Shkarayev2, Department of Aerospace and Mechanical Engineering, University of Arizona, Tucson, Arizona
Fixed-wing micro air vehicles (MAV) are very attractive for outdoor surveillance missions since they generally offer better payload and endurance capabilities than rotorcraft or flapping-wing vehicles of equal size. They are generally less challenging to control than rotorcraft in outdoor environment and allow for a dash capability to escape enemy attention. On the other hand, they usually fail miserably to perform vertical take-off and landing (VTOL) and sustain stable hover flight which proves to be crucial for urban surveillance missions including building intrusion. The present paper investigates the possibility to improve the aerodynamic performance of fixed-wing MAV concepts so as to allow for true hovering capabilities and still maintain high cruise speed for covertness. Several combinations of rotors and fixed-lifting surfaces were tested, analyzed and compared. First, a tandem-rotor biplane MAV configuration was designed and tested as a result of different biplane powered configurations. A low-speed autonomous fixed-wing MAV was fabricated and flight tested to perform multi-tasking outdoor surveillance missions. Secondly a side-by-side comparison of a tiltwing and a tilt-body powered configuration with a pair of counter-rotating motors in tractor configuration with MAV configurations were carried out. The tilt-body configuration was shown to be more suitable for MAV applications with higher hovering performances. Second, a coaxial tilt-body concept based on coaxial motors and contra-rotating propellers inspired from the Convair XFY1 “Pogo” experimental aircraft was designed and tested. A wind tunnel test was carried out to fully characterize the aerodynamic performances of the coaxial tail-sitter configuration, named Vertigo. An autonomous version was developed in order to autonomously perform transitions between horizontal and vertical flight. A smaller 300-mm span version, called mini-Vertigo, was designed and fabricated based on a series of wind tunnel tests using miniaturized coaxial-rotor propulsion set. Autonomous altitude hold and attitude stability augmentation were then achieved using specific control laws adapted for the Paparazzi autopilot system. Thirdly, a new no-through-shaft coaxial-rotor configuration has been proposed in order to enhance the prototype ruggedness through an embedded spherical structure made of carbon rods. It is believed that such a crash-proof VTOL MAV, called Cyclope, can be very attractive for the use of MAV systems in real operations and allows for further size reduction such as for Nano Air Vehicles applications. Current prospects include both further wind tunnel tests using new high-precision micro sting-balances on coaxial-rotor tail-sitter MAVs and the development of control laws to autonomously perform transition flights. 1
Professor of Aerodynamics, Department of Aerodynamics, Energetics and Propulsion, ISAE BP54032, 31055 Toulouse Cedex, France, jean-
[email protected]. 2
Associate Professor, Department of Aerospace and Mechanical Engineering, The University of Arizona, PO Box 210119, Tucson AZ 85721-
00119,
[email protected].
1st US-Asian Demonstration & Assessment of Micro-Aerial & Unmanned Ground Vehicle Technology, 10-15 March, Agra, India
Hovering Capabilities of Fixed-Wing Micro Air Vehicles Jean-Marc Moschetta Institut Supérieur de l’Aéronautique et de l’Espace, Toulouse, France Sergey Shkarayev University of Arizona, Tucson, Arizona
Motivation Typical MAV mission profile Phase 1
Phase 2 Loiter
Dash Phase 3 Hover
Launch
Building intrusion Phase 4 March 10-15, 2008
MAV’08
2
MAV design challenges Rotorcraft MAVs
Fixed Wing MAVs MinusKiool (2003)
Br2C (2007)
20 cm, 60 g 36 cm, 490 g
Poor image capturing • Maneuverability in confined environment •
Challenging to control • Fast horizontal flight •
• VTOL Fixed-Wing MAV – High speed for covertness – Low speed for clear images – Limited electric consumption March 10-15, 2008
MAV’08
3
A low-speed in-flight adaptive wing MAV Camber changes h = 3-9% with synchronous change of reflex hi = 1-3%
Test flights were conducted with Pitot tube sensor With camber increases from 3% to 9% minimal speed decreases by about 100% March 10-15, 2008
MAV’08
4
A low-speed tandem wing MAV
• Tandem-wing MAV • Gimbaled 2-axis camera
TYTO Moschetta, Thipyopas, J Aircraft, vol. 44, 2007 March 10-15, 2008
250g - 40 cm MAV’08
5
New MAV concepts Tilt-rotor
Tilt-body Tilt-wing March 10-15, 2008
MAV’08
6
Experimental setup • Wind tunnel
l
– Test section 45 x 45cm – Velocity 2-20 m/s
Pivot Point
r
Aft Strut always perpendicular to the model longitudinal axis
• Measurements
Left Strut Right Strut
LC5 LC3
– Improved 5-component balance :
Adjustable plate controlled by motor to change AoA. LC6 LC2
• force 0.004 N • moment 0.002 N.cm
Free Stream Velocity
LC4 LC1
Motor
Ground LC1 – 4 are fixed onto the ground, LC5 - 6 are supported by adjustable plate and always rotate with the model
Kochersberger, AIAA paper 2005-4759 Suharyono, Aerospace Sc. & Tech., 2006 March 10-15, 2008
MAV’08
7
Future MAV wind tunnel (2009) 6-component sting micro-balance: 8-mm diameter, 100g max load,
Variable-pitch fan
Test section 120 x 80 cm Contraction ratio 10:1 Turbulence 0.3% Speed 0 - 25 m/s Test section length 2.5 m Length 19 m Width 4.5 m Height 7 m March 10-15, 2008
MAV’08
8
Tilt-wing vs. tilt-body 30°
• Speed 0-15 m/s • tilt angles 0°, 30°, 60°, 90° • PWM signal 40, 60, 80, 100% • Struts corrections (AoA, speed)
60°
Tilt-body (0°) March 10-15, 2008
MAV’08
9
α TW = α TB = α Tilt-body
S Λ , 2 2
S, Λ
LTB = qS ×
(Di )TB
=
πΛ 2
(α − α 0 )
1 2 qS πΛ × (α − α 0 ) 4
March 10-15, 2008
Tilt-wing
LTW = 2 q
(Di )TW MAV’08
=
S πΛ (α − α 0 ) = 1 × LTB × 2 4 2
1 1 2 qS πΛ × (α − α 0 ) = × (Di )TB 8 2 10
LTW = LTB = W Tilt-body
S Λ , 2 2
S, Λ
LTB = qS ×
πΛ 2
Tilt-wing
(α TB − α 0 ) = W
LTW = 2 q
S πΛ (α TW − α 0 ) = W × 2 4
α TW = 2α TB − α 0
(Di )TB
=
1 2 qS πΛ × (α TB − α 0 ) 4
March 10-15, 2008
(Di )TW MAV’08
=
1 2 qS πΛ × (α TW − α 0 ) = 2 × (Di )TB 8 11
Tilt-wing vs. tilt-body Downward force
100
90
Lift = Weight Drag = Thrust
80 70 60
tilt body
80
tilt body
PWM Signal (%)
Tilt angle or angle of attack (deg)
tilt wing
tilt wing
50
40 30 20
60
40
20
10 0
0 0
5
10
15
20
0
Velocity (m/s)
10
15
20
Velocity (m/s)
Tilt-wing requires more power near hover
Tilt-wing may benefit from vortical lift effects March 10-15, 2008
5
MAV’08
12
Tilt-wing vs. tilt-body 0,001 0 0
5
10
15
20
-0,001
tilt wing
dM(CG)/dAoA
-0,002
tilt body
-0,003 -0,004
• « CG » located at 10% chord ahead of the AC at 16 m/s • Each gradient calculated around equilibrium points • Longitudinal stability greatly reduced at low speeds
-0,005 -0,006 -0,007
• Tilt-wing / tilt-body MAV configurations very comparable
-0,008 -0,009
Velocity (m/s)
March 10-15, 2008
MAV’08
13
A coaxial tail-sitter mini-UAV • Tilt-body • Coaxial rotor in tractor configuration • Autonomous transition flight • rotor 500 mm • weight 1.6 kg (incl. IMU, GPS, Modem…) Convair « Pogo » XFY-1
ISAE « Vertigo » mini-UAV
(1954)
(2006)
March 10-15, 2008
MAV’08
14
Wind tunnel measurements 1,2 1 0,8 0,6
CL
0,4 0,2 0 -0,2
Stall angle
-0,4 -0,6 -20
• Sting-mounted model
0
20
40
60
80
AoA°
• Speed 0, 5, 10, 15, 20 m/s
• strong vortical effects on wing-alone tests
• AoA -10°to 90°, sideslip
• CD0 = 0.054, L/Dmax = 4
• flap efficiency, throttle
• CL > 0.8 at AoA 15°-55°
March 10-15, 2008
100
MAV’08
15
Coaxial tail-sitter 0,1 0,05 0
CM@CG
-0,05 -0,1 -0,15 -0,2
stall angle
-0,25 -0,3 -20
0
20
40
60
80
100
AoA°
• Stabilizing vortical effects (low aspect-ratio) • Constant pitching moment gradient after stall (CoP) March 10-15, 2008
MAV’08
16
14
• Balanced lift and drag
10
V(m/s)
Powered tests
12
(V, AoA, throttle)
8 6 4 2 0 0
20
30
40
50
60
70
80
90
100
Ao A °
1,8
0,18
1,6
0,16
1,4
• Equilibrium found throughout transition
0,14
Aircraft horizontal force Aircraft vertical force
Forces (daN)
1,2
0,12
Propeller horizontal force Propeller vertical force
1,0
0,10
Aircraft pitching moment Propeller pitching moment
0,8
0,08
0,6
0,06
• Aircraft pitching moment almost constant in transition (flap efficiency)
0,4
0,04
0,2
0,02
March 10-15, 2008
0,0 0
10
MAV’08 20 30
40
50
60
70
80
90
17 0,00 100
Pitching moment (daN.m)
• Propeller-alone tests carried out to evaluate propellers contribution during transition
• Propeller-induced noseup pitching moment (side force)
10
14
Propeller-wing interaction
12
exp.
V, m/s
10
T
+
model
8 6 4
α V
2
2w VR
αS
0 10
20
30
40
50 60 AoA, deg
70
80
90
McCormick
w w0
4
3
w V w + 2 cos α + w0 w0 w0 2
2
V w0
2
= 1
V w VR w = +2 cos α + 2 sin α w0 w0 w0 w0
2
w w0 sin α w0 VR
α S = sin −1 2
Acknowledgement: Boris Bataillé & Damien Poinsot, ISAE & ONERA, Toulouse March 10-15, 2008
MAV’08
18
Simulation of transition flight 20
14
15
12
10
Delta m, deg
10 Velocity, m/s
AoAw, deg
15
8 6 4
5
20
30
40
50 60 AoA, deg
70
80
Effective wing AoA
90
0 10
5
V Vr
2 0 10
10
20
30
40
50 AoA
60
70
80
90
0 10
20
Relative wind speed
30
40
50 60 AoA, deg
70
80
Flap deflection
• Unstalled wing (stall angle 23°) • Flap efficiency throughout transition Poinsot et al., VTOL conference, RAeS 2008 March 10-15, 2008
MAV’08
19
90
Mini Vertigo
Dual motor-propeller
50 g
3-cell Li-Poly batteries
60 g
Speed controller & receiver
16 g
Gyros & servos
22 g
Airframe 44 g ------------------------------------------------Total weight 192 g
Fiberglass, 1 ply/side
2 Bolts, D2 Plastic insert
Fiberglass 2 plies/side
Motor mount Carbon tube 6 mm 46
Battery cell RC Receiver
200
Speed controller
Carbon tube 3 mm
Servo
March 10-15, 2008
300
Elevon
Gyro-stabilized prototype (2007) MAV’08
20
Scaling effects MAV
Size (cm)
Shkarayev et al., J Aircraft (submitted), 2007
Rec
Λ
S (dm²)
m (g)
m/S (g/dm²)
Vertigo
65
16 x 104
1.80
24
1600
67
Mini-Vertigo
30
4 x 104
1.91
5
160
32
• Mini-Vertigo used high-pitch counter-rotating propellers (low RPM) • Power loading (PL) includes motor efficiency MAV
Drotor (mm)
T (N)
DL (N/m2)
P (W)
PL (W/N)
Pind (W)
Vertigo
500
15.7
80
165
10.5
90
Mini-Vertigo
140
1.55
101
45
29.0
8.2
March 10-15, 2008
MAV’08
21
Conclusions • Tilt-body MAV configurations combining
tandem/coaxial rotors and fixed wings are promising candidates for complex outdoor/indoor surveillance missions in autonomous mode • Tilt-body better than tilt-wing MAV • Coaxial tail-sitter allows for balanced transition
March 10-15, 2008
MAV’08
22
Current prospects: nanorotors
• Weight decreases with size as
W = 17.884 d 1.535
• FoM and power loading decrease with size (XROTOR) March 10-15, 2008
MAV’08
Acknowledgements Liu Zhen, PhD candidate, ISAE, Feb. 2008 23
Current prospects: autonomy
• Step 1: altitude/attitude hold • Step 2: autonomous transition Acknowledgements Roman Krashanitsa, Univ. Arizona, 2007 March 10-15, 2008
MAV’08
24
Current prospects : improved airframe Vision (35 cm, 450 g)
• Embedded coaxial fixed-wing • No through-shaft system • Roll & fly
MAV’07, Toulouse, France, sept. 2007 Acknowledgements Dominique Bernard, ISAE, 2007 March 10-15, 2008
Frankfurt, Germany, oct. 2007 MAV’08
25
Thank you
Cyclope (20 cm, 100g)
March 10-15, 2008
MAV’08
26
