Integrated MAV Systems:Hovering:Rotary-Wings & Flapping-Wings

Alfred Gessow Rotorcraft CenterUNIVERSITY OF MARYLAND

1st US-European Micro-Aerial Vehicle TechnologyDemonstration and Assessment, Germany

September 20, 2005

Inderjit ChopraAlfred Gessow Professor &

Dirctor Alfred Gessow Rotorcraft Center(chopra@eng.umd.edu)

• Design Requirements• No dimension exceeds 15 cm (6 inch)• Gross takeoff weight 100 grams• Loiter time of 60 minutes• Payload capacity of at least 20 grams

• Additional considerations• Minimum mechanical complexity• Fully autonomous (out of sight operations)• All weather operations• Low production cost• Rapid deployment• Low detection

Micro Air Vehicles: Definition

Hover

Small

Novel

• Military• Surveillance missions (over the hill

and confined areas)• Infrared images of battlefields and

urban areas (around the corner)• Mine detection in war zone

• Civil• Biological/chemical agent detection• Agriculture Monitoring• Communication Nodes/GPS• Traffic monitoring (long endurance)• Counter-drug operations

Urban MAV Missions•Monitoring Traffic Flow•Surveillance Imagery

MAV Applications

• Increasing terrorists and Urban warfare threats• Miniaturized Sensors: Availability• Expanded capability of data acquisition, analysis and

transmission (IT & wireless technology)• Micro actuators and multifunctional smart materials• Potential for long endurance systems• Low cost systems

(can be organic with a soldier)• Increasing focus on biologically-

inspired flight systems

Micro Air Vehicles: Key Drivers!!

MAV Weight Breakdown Foch (NRL)

• Airframe ~ 21%• Engine ~ 11%• Battery ~ 30%• Payload ~ 21%• Avionics etc ~ 17%

10-3 10-2 10-1 1 10 10010-8

10-6

10-4

10-2

1

102

104

Wing Span or Rotor Diameter [m]

Mas

s [k

g]

Rotorcraft

Birds

Insects

Rotaryseeds

UAVs

Some Perspective on ScaleSome Perspective on Scale

Reynolds NumberReynolds Number

Aerodynamic ScaleAerodynamic Scale

ρ = air density

c = chord

V = velocity

µ = fluid viscosity

ρµcV

ForceViscousForceInertial =

MAVs operate in the very low Reynolds number flight regimeMAVs operate in the very low Reynolds number flight regime

Reynolds Number

Gro

ss W

eigh

t (lb

s)-

Adapted from:

McMichael, J. and Francis, M.,

“Micro Air Vehicles – Toward

a New Dimension in Flight”,

DARPA, 1997.

Aerodynamic EnvironmentAerodynamic Environment

106104

1021

10-2

10-4

Sender

Pioneer

Cessna 150

F/A- 18 Hornet

C-5 GalaxyMAVs

MAVs

Less than 6”

3 104

105

106 10 7

108

Dragonfly

Hummingbird

10

Reynolds Number EffectReynolds Number Effect

Max Lift to DragRatio

Max Lift Max Drag

Reynolds Number EffectReynolds Number Effect

Reference Reynolds Number Reref=105

Profile DragRe>105 Cd=Cd0refRe<105 Cd= Cd0ref (Reref/Re)1/2

CLmin drag( )CLmin drag( )

ref

=CL min power( )

CLmin power( )ref

=Reref

Re

14

If Reynolds number is reduced from 105 (Reference) to 103,profile drag increases 10 times its reference value andlift coefficient for minimum drag or power has to increase over 3 times its reference value (say CLref=.8, CLnew =2.5)

Most Small Insects100 < Re < 1000

Large Insects to Small Birds1000 < Re < 15000

BirdsRe >15000

•Delayed stall•Wake capture•Rotational circulation

•Dynamic stall•Delayed stall•Wake capture

•Bound circulation•Quasi-Steady mechanisms

Biological Lifting Mechanisms at Low Re

MICRO HOVERING AIR VEHICLES

• Non-Hovering Vehicles: Fixed-wing based

• Hovering Vehicles: Rotor Based• Single main rotor (with & without tail rotor)• Ducted fan rotor• Co-axial rotor• Tiltrotor, tiltwing, quadrotor, hybrid systems• Revolutionary designs

• Hovering Vehicles: Flapping-Wing Based• Bird-flight based• Insect-flight based (Efficiency at small scale?)

• Hovering Vehicles: Reaction Based(power intensive)

Micro Hovering Air Vehicles:Rotor-Based

MICOR (University of Maryland)MICOR (University of Maryland)

15 cm (6”)15 cm (6”) dia dia coaxial 2coaxial 2--bladed rotorsbladed rotors

Weight~100 g, Payload ~10g8% camber circular arc airfoilsRe.75R ~20,000Endurance ~ 10 minutesFixed pitch, variable speedrotors (feedback on lower)

Swashplate controls only lower rotor

First generationNo lateral control

Second generationLateral control implemented

Using swashplate

(Video)

QuickTime™ and aYUV420 codec decompressor

are needed to see this picture.

Rotor system

MotorVanes foranti-torque

Yaw control surfacesServos

Rotor diameter 27 cm

Battery pack

Main rotor- Two bladed teetering - Pitch flap coupling(δδδδ3 angle of -45°)

-Servo paddles

Swashplate Control- Longitudinal- Vertical- Lateral- Pitch- Roll

Vanes (feedback)- Anti-torque- Yaw control Electronics

3 micro-servos Receiver, brushless motor controllers

Main rotorStabilizer bar

Motor

Anti-torque vanes

Protective ring

Weight breakup

20 . 6.5Swashplate

40 . 13Electronics28 . 9Rotor system

58 . 19Motor (brushless DC)55 . 18Battery (700 mAh Li-Poly)

106 . 34.5Structure

307 . 100Total

Weight (gms) %

Component

QuickTime™ and aYUV420 codec decompressor

are needed to see this picture.

Thrust load cell

Torque Sensor

Hall effect sensor

Measurement of HoverPerformance:

•Thrust•Torque•Rotational speed

Hover test stand

FMCTCP

Inverted rotor

FM =Ideal Power required to hover

Actual Power required to hover

Rotor Hover Test

Figure of Merit

Blade Airfoil Variations

Baseline

Twisted

Tip-Taper

Planform-Taper

Planform-Taper

Camber Distribution Planform Distribution

Experimental Results

4500RPM

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0 0.005 0.01 0.015 0.02 0.025CT

FM Twisted 8%

Untwisted 8%

NACA 0012

Flat plates

3500RPM

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0.45

0 0.005 0.01 0.015 0.02 0.025CT

FM Twisted 8%

Untwisted 8%

NACA 0012

Flat plates

Maximum FM at 4500 RPM is 0.43 with twisted 8% camber bladesFM of full scale helicopters ranges from 0.7 to 0.85

FM: Figure of Merit

Sharpened Leading-Edge Airfoils

• Sharp leading-edge increases FM

• Smaller rise in FM for cambered airfoil

Sharpened LE can improve airfoil performance

7.0% camber 7.0% camber

FM

CT/σσσσ0 0.04 0.1 0.14 0.2

0

0.1

0.2

0.3

0.4

0.50.55

7.0% camber

with LE camber sharpened LE

Flat plate

Flat plate with sharpened LE 15º

CT/σσσσ

Blade Tip Design

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 0.005 0.01 0.015 0.02

CT

FM

Prandtl’s Tip

Rectangular Tip

Improved rotor performanceby modifying tip shape

Thrust/Power of MAV

Better power-loading curve for cambered blades

Higher power-loading at reduced disk loading

Planform variation has small effect 0 5 10 15 20 25 30 35 400.05

0.1

0.15

0.2

0.25

Disk Loading [N/m2]

Pow

er L

oadi

ng [N

/W]

7.0% camber at x/c=1/2 and sharpened LESharpened LE flate plate 15oSharpened LE flate plate 15o with 2:1 tip taper

Camber FM =.53

Flat Plate FM=.40

Flow Visualization

7% camber, 2.75% thickness with sharpened LED=6” 2-bladed rotor, 3600 RPM, Re=36.8*103

Main Vortex

VortexSheet

Rotor Plane

Wake Obstruction

MainVortex

Strong tip vortices

High induced velocities in tip region

Vortical shed wake obstruction increases DL and lowers FM

Rotating-Wing MAV Performance

ProfileEffects

InducedEffects

•Better designs may come through careful aerodynamic optimization•Gains may not come through improvements in airfoils alone•Performance goals met through understanding of flow physics•Induced and profile effects have strongly interdependent effects

Improvement of hover efficiency using duct around the rotor (plus safety protection of rotor)

Figure of Merit M: Hover Efficiency is defined in terms thrust production per unit input powerFor present designs: M is less than 0.5Goal:Increase M over 0.8

Rotor Hover Efficiency

Shrouded-Rotor Concept

Key Design Parameters

• Expansion ratio/Diffuser angle– Want this to be as large as

possible for best performance

• Inlet lip radius– Incoming flow forms a suction peak

on the inlet lip; cause of thrust augmentation

• Blade tip clearance– Proximity of shroud wall reduces strengthof blade tip vortices; reduces blade tip losses

Experiment: Model Configurations

Test stand

Electric motor

Rotor by itself Rotor with shroud attached

Rotor inside shroud, not connected to

shroud

Experiment: Thrust Ratio vs. Total Power

Thrust Coefficient, CT Thrust Ratio, Ttotal / Tfree

0 0.005 0.01 0.0150

0.01

0.02

0.03

Total Power Coefficient, CP

LR13-D00 LR09-D20-δδδδ

LR06-D10 LR09-D20

Free Rotor

0 0.005 0.01 0.0151.2

1.4

1.6

1.8

Total Power Coefficient, CP

LR13-D00

LR09-D20-δδδδ

LR06-D10 LR09-D20

Increase lip radius: Increase thrust

Decrease tip clearance: Increase thrustDiffuser angle: Thrust increases with small angle <100

PowerPower

0 0.2 0.4 0.6 0.8 1

-0.5

0

0.5

1

1.5

2

2.5

r / R

v / v

iShrouded Rotor Wake:

Effect of Blade Tip Clearance

δδδδtip = 0.1%

δδδδtip = 0.8%

δδδδtip = 1.6%

Isolated rotor

LR13-D10, 31% l/Dt

(z/R = 0.62)

As tip gap decreases, wake profile approaches to uniform

Shrouded-Rotor

Inlet Diffuser

0.00

0.02

0.04

0.06

0.08

13579111315

Tap #

Pre

ssur

e C

oeff

icie

nt, -

Cp

Rotor Plane

40o

30o

20o

10o

Optimized Configuration: 13% lip radius, 100 diffuser angle and 72% diffuser length results in 95% increase in thrust for same power

Challenge: Structural weight of shroud must be less than lift augmentation plus lower performance degradation in forward flight

Rotary Wing Micro Air Vehicles:

Unconventional Configurations

Problems with Conventional Rotary-Wing Systems

• Poor hover efficiency at small scales– Mainly due to low Re– Profile power dominates

• Maximum Figure of Merit (FM) < 60%– Full scale helicopters have FM~80%

• Unconventional rotor-based configurations- Cycloidal propulsion- Whirling arm rotor- Other configurations?

power Actualpower Ideal=FM

Cycloidal Propulsion System

•Pitch angle of each blade is varied sinusoidal in each revolution•Thrust vector changes with pitch amplitude and phase

MAV Vehicle Model

Cycloidal Rotors

Electronics Package

Fuselage/Landing Gear

Electric Motors

Battery

Two Cycloidal Rotors rotating in opposite directions

Cycloidal Propulsion System

•Good Points:Ability to change direction of thrust instantly up to 360 deg

Good maneuverability

High thrust possibility

•Weak Points:•Limited previous work

•Complex airflow

•Weight of mechanism

Cycloidal Propulsion System

•Challenges:To achieve a higher figure of merit with low structural weight

Simplify pitch changing mechanism

Optimize BladesThrust

0 400 800 1,200 1600

20

40

60

80

100

120Theoretical, 25 degreesTheoretical, 10 degreesExperimental, 25 degreesExperimental, 10 degrees

Rotation

Whirling Arm Rotor

!"

•Good Points:Reduced induced power due to simulated forward flightPotential for higher figure of merit

•Weak Points:•Complex airflow•Asymmetric lift distribution•Complexity of 2 rotors

Whirling Arm RotorV∞∞∞∞

V∞∞∞∞

Ω

Induced

Profile

Total“ Power Bucket ”

µµµµ

P/Ph

#! $

0 0.1 0.2 0.3 0.4 0.5

! %

&'%(

$%

!! %

•• Develop innovationsDevelop innovations rotor designs to increase hover rotor designs to increase hover

figure of merit (~0.8)figure of merit (~0.8)

•• Requires a good understanding wake phenomena and Requires a good understanding wake phenomena and

unsteady aerodynamicsunsteady aerodynamics at low Reynolds numbersat low Reynolds numbers

-- DevelopDevelop comprehensive CFD toolscomprehensive CFD tools detailed detailed testingstestings at low at low

Reynolds numberReynolds number flightsflights

MAV RotaryMAV Rotary--Wing Aerodynamics: Wing Aerodynamics: Summary ConclusionsSummary Conclusions

Micro Hovering Air Vehicles:Flapping-Wing Based

Mechanism of Flapping-Wing FlightInsects vs Birds

What can be learned from nature?

Bird Flight

Down Stroke: power stroke, outer and faster moving parts, also moving forward especially towards the end of stroke, At the end of stoke, the wing first rotate upward from the shoulder while the elbow is relaxed so the outer wing bends down as well as rotate to present least resistance to forward motion, During the down stroke, primary feathers are held close, perfect airfoil

Up Stroke: move up not to produce unnecessary drag and lose lift involving acomplex bending and twisting motion, Approximately halfway through the upstroke , the outer wing is moved up and back at a very rapid rate with outerfeathers separated, this reduces drag and produce extra lift, During up stroke the primary feathers get separated (slots)

Take off and landing: forward speed low, fast flapping, high amplitude, wings are spread wide

QuickTime™ and aMicrosoft Video 1 decompressorare needed to see this picture.

Adaptive Morphing Vehicle Geometries Inspired by Nature

Active Wing Tips for Roll Control

Aeilerons for Maneuvering

)*)* $$

+*+*

ROTATIONAL phaseROTATIONAL phase –– when the wings when the wings

rapidly rotate and reverse directionrapidly rotate and reverse direction

The wing stroke of a hovering insect is divided into four The wing stroke of a hovering insect is divided into four kinematic stages:kinematic stages:

,*,* -.-.

/*/* ..

TRANSLATIONAL phaseTRANSLATIONAL phase –– when the wings when the wings

sweep through the air with a high pitch anglesweep through the air with a high pitch angle

1 2 3

4 5

stroke plane

wing path

net force

wing section

downstroke

upstroke

Insect Flight FundamentalsInsect Flight Fundamentals

Rotational Lift producedduring pronation and supination by Magnus effect

Wing quickly rotates, using the shed vorticity of previous stroke to create lift: wake capture

Unsteady Mechanisms

Delayed Dynamic Stall: During translation phases, intense leading edge vortex stabilized by radial flow

Leading edge vortex

Wingmotion

StartingVortex

wing section

Aerodynamic Net-force

Birds vs Insects

<10,000>10,000Reynolds No.

Modest, tilting body and stroke plane

High, wing morphingSpeed

Quite commonVery rareHovering

High >50HzModest <10 HzWing frequency

Rigid wing, base motionActive wing morphingMorphing

UnsteadyLift enhancement

Quasi-steadyDrag-reduction

Aerodynamics

.1m and less0.15 to 3mSize

Less than .2g20g to 15 kgWeight

InsectBirdFunction

This mechanism is insect based flapping wing, passive pitch, biThis mechanism is insect based flapping wing, passive pitch, bi--stable stable device that is capable of replicating the complex kinematics of device that is capable of replicating the complex kinematics of insect insect wings in hoverwings in hover

Wings

Brushless motor4:1 gearbox

Crossedroller slide

Scotch yoke

Flap bearingassembly

Pitch bearingAssembly

Pitch actuators

Rotating diskassembly

Insect-Based Flapping-Wing Mechanism

Experimental Setup

QuickTime™ and aCinepak decompressor

are needed to see this picture.

QuickTime™ and aCinepak decompressor

are needed to see this picture.

Experimental Setup

Miniature Bending Beam Load-cell

Scaled-up Fruit Fly Wing Mounted on the Load-cell

Test Results & Validation

Wing III produces larger thrust from rotational circulation & improves prediction

Average ThrustPitching axis

0.5c

0.5c

0.2c

Mass: 1.8 grams

Mass: 1.1 grams

Mass: 1.1 grams

4 6 8 100

1

2

3

4

5

6

Frequency (Hz)

Thrust (grams)

Stroke : 80o

Pitch : +30o/-30o

Analysis (Wing III)

Analysis (Wing II)

Experiment (Wing II)

Experiment (Wing III)

3 11

Power Measurements

Wing IIIStroke : 800

Pitch: Downstroke 450, Upstroke -450

Thrust decreases rapidly at high frequencyWing III too heavy for high frequency operations

Flapping Frequency (Hz)

Thrust (grams)

9 10 11 121

2

3

4

5

6

7

9 10 11 120.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

Flapping Frequency (Hz)

Power (Watts)

Power = flap vel. x horizontal force= aero power + inertial power

Light Wings

Wing III (1.3 grams) : too heavyLighter wings with composite frames were built

Results: Thrust increases marginal

Wing III

Wing VII

Wing IV

Wing IX

Wing VIII

1.3 grams

0.49 grams

0.86 grams

0.56 grams

0.61 grams

RC Microlite

Light Composite Wings

Visualization at 50% Span for Half Cycle

0o180o

Wing “Starting” Vortex

• Formation of LEV occurs during translation – form of dynamic stall• Powerful starting vortex at trailing edge until mid-stroke, with time-varying strength

The images represent 0 to 60 deg of wing cycle0 deg 13 deg

25 deg 40 deg 60 deg

Dynamic Stall Vortex (LEV)

• LEV: grows in size gaining more energy as the translation continues• LEV “spills” from leading edge (dynamic stall)• A secondary LEV is clearly visible• “Wagner” starting vortex, even though diffused, is present until mid-stroke

The images represent 65 to 90 deg of wing cycle

65 deg 75 deg 80 deg

90 deg 97 deg

Multiple LEVsImages represent 107 to 153 deg of flapping wing cycle

• LEV starts shedding (clearly a form of dynamic stall)• A new LEV forms before the first LEV reaches mid-chord• Presence of multiple vortices on top of the wing – can be a source of enhanced lift• Occurence depends upon the operating Re and the span location

107 deg 117 deg

137 deg 153 deg

Flow Developments During Supination

The images represent 157 to 196 deg of wing cycle

• Secondary LEV starts shedding • Shed wake vortices are clearly visible – complex 3-D flow field

157 deg 173 deg 180 deg

188 deg 196 deg

Schematic of Flow Structures

A strong leading-edge vortex during early part of stroke and gains energy from shed wake and then shed from LE

Second LE vortex forms before shedding of first from trailing-edge

Spanwise flow is of same order as tip speed, not strong enough to stabilize le vortex

Multiple vortices that enhance lift

Complex wakes: numerous vortices over the surface Complex wakes: numerous vortices over the surface

and welland well--structured wakes below the wingsstructured wakes below the wings

Unsteady aerodynamic phenomena dominantUnsteady aerodynamic phenomena dominant

Understanding of wake phenomena and modeling Understanding of wake phenomena and modeling

lackinglacking

-- Requires detailed experimental testing and Requires detailed experimental testing and

comprehensive CFD modeling (prediction tools)comprehensive CFD modeling (prediction tools)

FlappingFlapping--Wing Aerodynamics: Wing Aerodynamics: Summary ConclusionsSummary Conclusions

• Torque applied by an eddy current brake• Signal to speed controller varied to maintain constant

motor RPM• Voltage, current, torque, RPM are measured• Efficiency is

• Losses in speed controller are included in measurements

Electric motor testing

power Electricalpower Mechanical=η

Measured motor efficiency

0

10

20

30

40

50

60

0 5 10 15 20 25 30 35 40

Mechanical Power, Watts

Efficiency, %2000 rpm

3000 rpm

4000 rpm5000 rpm

0

0.1

0.2

0.3

0.4

0.5

0.04 0.08 0.12 0.16CT /σσσσ

FMMech FM

Elec FM

Measured Figure of Merit

Motor efficiency is approximately 50 %

Soft HingesFluidic Sensor

Piezoelectric Directional Fibers

Nanofiber Composites

Lithium Ion Battery

ElectrodeFiberElectrode

Multifunctional Wing

Multi-Functional Structure•Loads Carrying•Energy storage•Energy harvesting•Communication

Micro Hovering Air Vehicles:Power Issues

Hovering MAV

1

10

100

10 100 1000 10000 100000 1000000

Storage Efficiency Qr (W-Hr/Kg)

Pow

er P

lant

Eff

icie

ncy ηη ηη

pwr

(%)

JP10

Batteries

Model Aircraft Engine(378g; Methanol)

DARPA DMFC Objective

Endurance 1.0 min100 min

10 min1000 min

χf = 0.2

ηr = 0.4

drotor = 10 cmM = 100g

+−

=

fdiscr

Rpwr Lg

Qχ

ρηητ1

112

Challenges

IC Engines• Efficiency is THE critical parameter

– Scales unfavorably with decreasing size:» Increased storage volume leads to increased thermal losses» Thermal coupling to structure increases volume required for

complete combustion

• Reaction rate– Scales unfavorably with increasing storage efficiency (Qr)

» More energy dense fuels tend to be heavier (higher MW) and therefore react more slowly requiring larger volumes

– Can be slowed dramatically via heat loss to structure

• Mixing– Hydrocarbon fuels can be difficult to atomize and mix with air in small

volumes

• While the micro-engine projects seem promising, practical implementation in MAVs is still years away

– Enormous manufacturing challenges– Very inefficient– Governing physics at small scale not well understood

• Small hobby engines offer an interim solution– Plentiful and cheap– Problems:

» Quantitative performance data not available» Would be nice to have a reliable scaling relationship that designers of

MAVs could use to estimate how power output and fuel efficiency changes with engine size

» Noise level too high» Present efficiency level inadequate - but can be improved with modest

investment in research

Problem vs. Opportunity

OS OS

OS

OS

Performance Scaling

Reliable scaling relationships for efficiency do not exist!

1

10

100

0.01 1 100 10000 1000000

Mass (kg)

Effi

cien

cy (%

)

Estimated from Manufacturer's data

Measured

?

What is required?

• Fundamental investigation of physical processes impeding the miniaturization of heat engines– Thermal losses– Fluid-structure coupling– Flame broadening– Micro-scale mixing

• Quantification of performance of existing engines– Power output– Fuel efficiency – Scaling of performance with size– Quieting

Enable high-performance UAVs of the future

Understand what can be done TODAY

Micro Hovering Air Vehicles:Navigational Issues

Goal:Goal: Examine application of insect navigational strategies in the deExamine application of insect navigational strategies in the design sign of autonomous navigational system (outof autonomous navigational system (out--ofof--sight) for an MAV for sight) for an MAV for collision avoidance, terrain following and landing, insect collision avoidance, terrain following and landing, insect ocellusocellus--based based sensors for roll stabilization and panoramic vision systemsensors for roll stabilization and panoramic vision system

Insect vision and navigational systemInsect vision and navigational systemGood Points: fast, reliable and accurateGood Points: fast, reliable and accurateWeak Points: Not a simple reflexive system, simple nervous systeWeak Points: Not a simple reflexive system, simple nervous systemm

Insect Navigation

On-board stabilization

• Inertial Measurement Unit (IMU), GPS• Interface sensors and actuators with

microcontroller• Implement open and closed loop control• Hardware testbed: implement yaw

stabilization• Improving the sensor output (filter)• Implement control/semi-automation on MAV

Requires comprehensive understanding and Requires comprehensive understanding and

modeling of insectmodeling of insect--inspired visual guidanceinspired visual guidance

Development of algorithms for collision avoidance Development of algorithms for collision avoidance

and panoramic vision system in single and swarm and panoramic vision system in single and swarm

flight modeflight mode

Requires a set of light weight sensors including Requires a set of light weight sensors including

miniaturized camera for outminiaturized camera for out--ofof--sight flight guidancesight flight guidance

Insect Navigation: Summary Insect Navigation: Summary ConclusionsConclusions

Insects Birds Airplanes

•Airfoil very thin flat•Modest active camber•Sharp leading-edge• L/D~2-6

•Moderately thick•Flat, modest camber and twist• Rounded leading-edge• Max camber at leading-edge• L/D~10

• Thick airfoil• Rounded leading-edge• Sharp trailing-edge• Max camber at mid-chord• L/D>10

Comparison: Insects, Birds & Airplanes

Hovering MAV Goals

Micro-engineBatteryPower

SwashplatelessSwashplatePrimary

All DOFVerticalControl

10 km50 mRange

>20 g0Payload

<100 g>100 gGross Weight

Vision-basedNoneAutonomous

20 m/s3 m/sSpeed

.85.5Hover FM

60 Minutes<10-15 MinutesEndurance

Expected LevelCurrent LevelItem

• Aerodynamics at low Reynolds numbers• Laminar flow modeling and viscous drag prediction• Flow separation control (passive or active, synthetic jets, blowing,suction, etc.)• Unsteady aerodynamics and wakes modeling

• Efficient power/propulsion at small scale• Modeling of phenomena associated with energy conversion devices• Energy Density efficiency prediction

• Structures and Materials• Modeling of nonlinear coupled motions• Morphological shape changes

• Stability and Navigational tools• Insect-inspired navigational strategies• Algorithms development for flight• Panoramic view from limited vision• Modeling Collision avoidance• Swarm mode of flight

• System Integration• Comprehensive aeroelastic models• Electronics Miniaturization• Flight testing and validation

10-4

10-3

10-2

10-1

1

101

102

103

104

105

106

Reynolds Number103 104 105 106 107 108

Micro Air Vehicles(Less than 6”)

MAVsGro

ss W

eigh

t (kg

)

Sender

F-18

Modeling Challenges and Technical Barriers

Propulsion and Power- micro engine- Thermo efficiency- Efficient fuel- Energy Storage

Maneuvering Capability- control surfaces- distributed sensors and actuators

Lightweight Wing Structures- Active shape deformation- Wing morphing

Sensing and Navigation- miniature electronics- Mems & insect based Low Reynolds Number Flow

- delayed Stall- flow control- wake capture

Biomimetic Kinematics- actuation (thorax)- efficiency- scaling

Flight Inspired by Nature at Low Re

AcknowledgementsGraduate StudentsMat TarascioBeerinder SinghtJason PereiraBen HeinEric ParsonsBeverly BeasleyJaye FallsNitin GuptaPeter CoppBrandon FitchettMoble BenedictA. Abhishek

Faculty ColleaguesNorm Wereley (Aero)Darryll Pines (Aero)Jayant Sirohi (Aero)Paul Samuel (Aero)Anubhav Datta (Aero)Gordon Leishman (Aero)Roberto Celi (Aero)Jim Baeder (Aero)Ben Shapiro (Aero)Ella Etkins (Aero)Chirs Cadou (Aero)Fred Schmitz (Aero)Marat Tishchenko (Aero)Bala Balachandran (Mech)Elisbeth Smela (Mech)Satinder Gupta (Mech)Rama Chellappa (EE)Reza Ghodesi (EE)Shivjumar (NCA&T)Srini Srinivasan (ANU)

SponsorARO (Gary Anderson)NRTC (Yung Yu)