Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING...

10
MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907 [email protected] ABSTRACT Using a dynamically scaled robotic wing, we studied the aerodynamic torque generation of flapping wings during roll, pitch, and yaw rotations of the stroke plane. The total torque generated by a wing pair with symmetrical motions was previously known as flapping counter-torques (FCTs). For all three types of rotation, stroke-averaged FCTs act opposite to the directions of rotation and are collinear with the rotational axes. Experimental results indicate that the magnitude of FCTs is linearly dependent on both the flapping frequency and the angular velocity. We also compared the results with predictions by a mathematical model based on quasi-steady analyses, where we show that FCTs can be described through consideration of the asymmetries of wing velocity and the effective angle of attack caused by each type of rotation. For roll and yaw rotations, our model provided close estimations of the measured values. However, for pitch rotation the model tends to underestimate the magnitude of FCT, which might result from the effect of the neglected aerodynamics, especially the wake capture. Similar to the FCT, which is induced by body rotation, we further provide a mathematical model for the counter force induced by body translation, which is termed as flapping counter-force (FCF). Based on the FCT and FCF models, we are able to provide analytical estimations of stability derivatives and to study the flight dynamics at hovering. Using fruit fly (Drosophila) morphological data, we calculated the system matrix of the linearized flight dynamics. Similar to previous studies, the longitudinal dynamics consist of two stable subsidence modes with fast and slow time constants, as well as an unstable oscillatory mode. The longitudinal instability is mainly caused by the FCF induced by an initial forward/backward velocity, which imparts a pitch torque to the same direction of initial pitch velocity. Similarly, the lateral dynamics also consist of two stable subsidence modes and an unstable oscillatory mode. The lateral instability is mainly caused by the FCF induced by an initial lateral velocity, which imparts a roll torque to the same direction of initial roll velocity. In summary, our models provide the first analytical approximation of the six-degree-of-freedom flight dynamics, which is important in both studying the control strategies of the flying insects and designing the controller of the future flapping-wing micro air vehicles (MAVs). INTRODUCTION Recent studies on the tuning dynamics of animal flight [1, 2] showed that during low-speed yaw turns (rotation about the vertical axis such as saccade), flapping wing fliers ranging in size from fruit flies to large birds are subject to substantial passive damping through an aerodynamic mechanism termed flapping counter-torque (FCT). As an inherent property, FCT helps the flapping-wing fliers to slow down body rotation during rapid maneuvers and thus reduces the required active torque produced by asymmetries of wing motion. As a trade- off, however, flapping-wing fliers must overcome extensive aerodynamic damping (a result of FCT) to accelerate or to initiate a maneuver [1]. Not only was the passive damping found crucial during fast yaw rotations, simulation results [1] suggested that it is also present during roll and pitch maneuvers. In flying animals, measurements of body kinematics showed that most yaw turns are accompanied by substantial change in roll angular velocity [3]. Even at low speed maneuvering or hovering, most flapping wing flies perform banked turns which involve rolling. Furthermore, during escape or tracking flight, rapid reorientations of roll and pitch angles (causing reorientation of the net aerodynamic force vector) are essential for fliers to achieve fast maneuvers [4, 5]. 1 Copyright © 2010 by ASME Proceedings of the ASME 2010 Dynamic Systems and Control Conference DSCC2010 September 12-15, 2010, Cambridge, Massachusetts, USA DSCC2010- Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm

Transcript of Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING...

Page 1: Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering,

MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS

B. Cheng X. Deng

School of Mechanical Engineering, Purdue University, West Lafayette, IN 47907

[email protected] ABSTRACT

Using a dynamically scaled robotic wing, we studied the aerodynamic torque generation of flapping wings during roll, pitch, and yaw rotations of the stroke plane. The total torque generated by a wing pair with symmetrical motions was previously known as flapping counter-torques (FCTs). For all three types of rotation, stroke-averaged FCTs act opposite to the directions of rotation and are collinear with the rotational axes. Experimental results indicate that the magnitude of FCTs is linearly dependent on both the flapping frequency and the angular velocity. We also compared the results with predictions by a mathematical model based on quasi-steady analyses, where we show that FCTs can be described through consideration of the asymmetries of wing velocity and the effective angle of attack caused by each type of rotation. For roll and yaw rotations, our model provided close estimations of the measured values. However, for pitch rotation the model tends to underestimate the magnitude of FCT, which might result from the effect of the neglected aerodynamics, especially the wake capture.

Similar to the FCT, which is induced by body rotation, we further provide a mathematical model for the counter force induced by body translation, which is termed as flapping counter-force (FCF). Based on the FCT and FCF models, we are able to provide analytical estimations of stability derivatives and to study the flight dynamics at hovering. Using fruit fly (Drosophila) morphological data, we calculated the system matrix of the linearized flight dynamics. Similar to previous studies, the longitudinal dynamics consist of two stable subsidence modes with fast and slow time constants, as well as an unstable oscillatory mode. The longitudinal instability is mainly caused by the FCF induced by an initial forward/backward velocity, which imparts a pitch torque to the

same direction of initial pitch velocity. Similarly, the lateral dynamics also consist of two stable subsidence modes and an unstable oscillatory mode. The lateral instability is mainly caused by the FCF induced by an initial lateral velocity, which imparts a roll torque to the same direction of initial roll velocity. In summary, our models provide the first analytical approximation of the six-degree-of-freedom flight dynamics, which is important in both studying the control strategies of the flying insects and designing the controller of the future flapping-wing micro air vehicles (MAVs).

INTRODUCTION Recent studies on the tuning dynamics of animal flight [1,

2] showed that during low-speed yaw turns (rotation about the vertical axis such as saccade), flapping wing fliers ranging in size from fruit flies to large birds are subject to substantial passive damping through an aerodynamic mechanism termed flapping counter-torque (FCT). As an inherent property, FCT helps the flapping-wing fliers to slow down body rotation during rapid maneuvers and thus reduces the required active torque produced by asymmetries of wing motion. As a trade-off, however, flapping-wing fliers must overcome extensive aerodynamic damping (a result of FCT) to accelerate or to initiate a maneuver [1]. Not only was the passive damping found crucial during fast yaw rotations, simulation results [1] suggested that it is also present during roll and pitch maneuvers. In flying animals, measurements of body kinematics showed that most yaw turns are accompanied by substantial change in roll angular velocity [3]. Even at low speed maneuvering or hovering, most flapping wing flies perform banked turns which involve rolling. Furthermore, during escape or tracking flight, rapid reorientations of roll and pitch angles (causing reorientation of the net aerodynamic force vector) are essential for fliers to achieve fast maneuvers [4, 5].

1 Copyright © 2010 by ASME

Proceedings of the ASME 2010 Dynamic Systems and Control Conference DSCC2010

September 12-15, 2010, Cambridge, Massachusetts, USA

DSCC2010-4234

Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm

Page 2: Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering,

Tsf

btdoaaham[difv

seomdt[efia

irpcpbfhccf

M

E

dsasatw

Thstafli

botradyopairaehoa mi[8dais flave

simexof madythe[1esflainsan

invroprcoprboflahocacofre

MA

Ex

deshanseactorwi2.

heabigh

odanynpprsp

eroovsaig].

amc

apeh

muxpf wan

yner1]seapse

nd

vell,

resomrovodapovalcomee

A

xp

escheend ns

ctinrqin.5

erebilht

Sdy nslnamospeodveracght

Tmpcruppiic

Iulerwnenamref]. enppiect

c

Ies, pse

mpvi

dy ppivercumpedo

AT

peT

cret a

song

qung 4

efolizt.

Simr

latmsiteedynrinccat r

Thpinucin

cle

Inslatrim

wineuvmfo

Hntiint on

In tigpint

pardet

inr c

ulapreom

TE

erThrib

Pan or,g

ue hcm

forzat

mrotioicte

ed nangadre

herngciang es.

setiomengveic

oreHoal

ng flnt

thgaitct ree trang coateehm

ER

rimhe bedPE

a aonse

holm

re,tio

milaotaoncs.

amg [de eflre

g, alf

.

ectonen

g erac e owl r

figtro

hiatechm

e tman

cone hen

fl

RIA

mec

d ETaspann enld

m

, aon

aratins . Ito[6

mi[7

ilecefoin

fli

t fns ntskiabmre

weroflighol.

s e t

h, math

matnslco

ndith

nsflig

AL

encoo

inT-Gpend th

nsoder

2

aen

r tio

It o 6]ics]. is ct oren atogh

fliw

s inbilmoesev

oleigh

ht .

stthan

athheithlatouitihesivgh

L

ntorn G ect

whe orr 2.

eroof

to n,misth

]. s dFina

e,allo ht

ighwitw

nemlitodul

vere ihtdy

tuhe ndhemir hemtio

untione veht

A

tard

F(

t rwi

wr ((F.54

odf

th,

mays he

Pdu

Funv

an al rth

t a

htth

witmty dellter, int, yn

dya

d mp

montenss

e ady

AN

al in

Figthratinwi(AFig4

dyro

hethy cl

e Pruriurtvea

a crohean

t wh cth

mata

ls ed

in t

ana

y, erya

matpre

matnsr-s itaanyn

ND

senatg. hictiog ng

ATg. cm

ynoll

e ahe

aleafl

reintheersaercootae nd

waco

tetican

d it theandam

wroawticedtic, fois abnana

D

ette

1cko h

g TI N

1m

naml

aere alarlig

eving erseroomatiuni

asomethcs

nd asini

e d

mic

wedy

w cadiccath

orcin

ilialyam

M

tue s1Aknof

howN

1Am) w

mian

roaeso

r tghiofo

rmly

odympionndin

s smp

he

sssn is aa

cs

e uynro

al ctial hecenvityytimic

ME

upsysA.esf 8ld

werNAA)w

icnd

odero thht ousformoy ryn

prenade

t

shputeristaumane

era cs i

usnaotm

iom

e e vey icacs

ET

p st T

ss 8.

derre

AN. T

wer

c md

dynrod

batd

s rwrerenaehal ersthe

owta

red

abmn evrocois

se amtatmons

mom(F

estd

als.

TH

emTho

5 r)

e mNOThre

mpi

nady

bet rdir

wwae, elaamhen

astae

wnatiod thili

medo

viddy

omi

amic

tioods de

meFCtigdel a

HO

msheof (i. meO-he im

ecitc

amyn rerew

ardit

atemic

nsandand

n onlo

he ityd opdeyn

mpim

a dc dondewelch

CFgaeriap

OD

s ae

0inThea-18e wm

chch

minaprla

ector

d fw

edc sivd ndes

tonaoc

ky.

nen

entnapre

mp

dydans ls

wits haF)ateivapp

D

anw

0.0clhe

asu8,w

mm

hanh

c amre

atitiorkfliwad tcovetr

dinsig

o bal uskeO

nen-t

ameh

por

ynaam

o fth fo

ani.

edatro

S

ndwin06lude ur A

winme

nian

damiceseveon

k ighas toone urannggn

beflustseyOuegloth

micherta

ammpof foe

or isT

d uiv

ox

d wng6 idiin

redApngrs

ismng

amc ene n

hhtsh

o tnstunns

g n

e iuis

y ur lig

oophacsensan

mipinf thor exam

Thusivesxim

wigs iningnstd peg ased

msgle

mpd

nt aia

hat ihothtrandslaofof

inid [1top

gip

at asivnt

icnghee

xpeaerm he ins. ma

ina

nchg taw

ex,and

s e

pidam

airs

ands s ow

he ain

deratif f b

nhed

10o preibl

mb

anveto

alg (e eaerrooff

ng

ati

g areh)th

antwit

, Nnd in

this

ingmanspd

diwn

fntrstioth

bi

erdyn0],

aevlemabond e o

lly(astchrimodyf fli

FInio

ke ) whetanth NCth

n a

has

g mpindpee

valiffn tfort otanaheio

renna, aachvio arod

sana

y ss troh meynwgh

FCn on

kinfa

wie lnea

Chea t

at im

(fin

d edvalsoferthrwonandal e -in

ntamanhioub

rgidy strna

ana

sca okty

enna

whht

CTso

nemasitheneo

a s, Ue gtan

camp

frong im

d aryo re

hatwan tdidfl

ns

tlymind ev

us odin

vrucalyal

calre

ke ypntaamhict dT asumof

mshih ngousixUSgenk

anpo

om

mpwyin

entt tardthingdegfligsp

y uics

avinw

dynalvectytlyz

leesu

ppeal

micchdyanm

f t

matioa

gthus x-cSAeak (

n ort

m capo

wilngsht fthed e g grgh

pir

uns macng

woy llyeloturticze

d ulplae o

rc

h ynnd

m, the

ticone

lehsfo

coA)

arb(4

actan

FCauortll g hofroe vflo

reeht red

nstmtivg ork

vy ocracale f

rolt anofresdaw

namd F

we

c peden

s oorom) abo46

chnt

Cusetainli

owomtuelligf es

dd

tamet

veth

k vest

cital l fli

obof

nef suam

we m

FCwesi

pad ngof rcempat

ox c

hiet i

T)ed

antndin

wnm urnlocghaes odym

ablthe mheo

elotabtiedymig

bof F

e. ro

ultmp

dmicCFe ix

arafr

gthf the potta

(m

evin

) cd t

dunean

thnici

ht eroofynmic

leodme

on ocbles yn

moght

otiFCW

otats.pindecs F m

px-d

amroh hean

onach2.

m w

ve n f

cabto

cearth

hangitydyod

f fnamcr

bds

mooh

itile

pna

odet

ic CT

Weat Wng

esci

mprode

memof

e gndenhe.5wi

pfla

auby o e rlyha

at g y, yndyfremro

bas [du

obhoie

spl

amelst

wT)e fioWg crin

modov

egr

etem

f ged nt ed4 id

paap

usebf

a y at duraw

nayneed

mica

se[9]ulse

oves sylay

micintab

win) ufu

on We

duribn ndevidre

erpl21

eartofo

d tc

de

sspp

edboflidw

uratewhamnamdo

cs ae

ed] aaterver

aysty cs

ngbi

ngunurt

aa

urbe neelsde

ee-

rs la1 rborqforo m

sivpin

d bodighdrawit

thrine ohicmicmom

ori

d oantioverinantem

as og olit

g tndetheanalsrin

aeas te -o

arstcm

boxqurceth

m 4

veng

bydyhtagthhengofchcsic

m,ofal

onndonedngndmanofofty

toererndsongas

ar-toa

of-

reicmx,uee-he

41

e g

y y t g h e g f h s c , f l

n d n d g d

m n f f y

o r r d o g s -o a -

e c

m , e -e

cv

Fpkrodk(Baxcorep

S

psokrorom(iwwwam

ropwh(Fhfrca

flth

cmvis

Figlan

kinota

defkinB)xiomedlan

St

proup

orikinobob

moin

winwewinandme

otpitwinhorFi

horfroconadv

flahe

m sco

g. 1ne

nematifinnem) - s,

mpd ane

ro

odpeigineboboosn ang

erengd ea

tattchngrizigriz

omntva

appeir

hos

1. e. (maioned

ma(D(C

ponarre (u

okT

dueriin

emotiotit ia cg e g

asu

Inteh yg zo. 2zo

m tinan

Inpir e

igsit

Sc(Aaticn ad aaticD)C)neowun

keTo uctim

nalmaic ic incopacawsuur

n d yss

on2Bonre

nunce

n ingeff

ghty

cheA) A

cs anas cs ) T) pnt

ws nde

e kse

tiomplly

aticww

nseooairal

werubed

ath

s, tr

ntaB ntaes

uoue

eg

ffe

=

emA

angl

thof

Thepitts o

rer

kieeonpoy cswiwinec

ordr cure

btrd i

all hean

rokal an

al st,usan

acfr

ect

=

matdy

arele he f te wtchofepno

ne tn oose

sys inng

ctsdinarulae mracin

oe sndke

pnd(F as snd

chrets

15

ticynae s

anthewih f wpreo r

emtheofedymw

ng.g s dnarouatmecte

n th

of strd yes plad Fianstrd s

h squo

52

c vamsp

nge leingax

wineserot

me f ad tmm

was. beduateuntedeaedhe

f troya

sanCg.

ndrosta

seue

on

2

viemicec

gleeftg kxisng enttati

maefa thms Wecurie ndd. asud e o

thkeawstane). . 2

d okeart

et enth

cm

ew calcifi

(n b

ft akins,

vt tion

atiffewe

metth

Wecauinfr

d tT

urt

oi

e e pw ar, F

2Dtoeste

oncyhe

m

ofllyiednobetandneman

velthen)

icec

winan

trihee us

ng ramthThredthel.

thplzs

rtean

FoD)o s, d

of y e t

m lat

f wy sd

ot twed rmandoce c.

s ct ngngicaenpeseh

mhrehed e

hrlans, ed nd

or . Neswth

e(

tor

lot

wincabysheerigati

d (cityco

aofg gual

n per w

hovme ee e g

bm

reeneFi

wd yaNsta

we he

exp)

rq

ong

ngaley howen ghtic (Dy imp

anf tpa

ulaw

plrfowiveatp

gry m

e e aigwst

awotabp

e r

pe) aqu

g)

g ked th

wnth

t was

D) indpo

ndthaiar wilayorminger tta

priravp

mea

inab

g. ithtow te bliplarot

erane

)

kinro

he n hhe winsym

yduone

d ee ir

vinyemg-[

acinvi

plaas

ndbo1Ah

opproth

ishayta

imnd pr

fi,

nemobo

inhew

ngmm

yawuceen

exst(o

veg

edmed-w12chcita

aysur

deoutAa

piotahah

yedtio

mea

ro

illed

maotinsterewings mew ed nts

xptroorlom

d td

win2,

hedipaatiyinre

epet e

A). a ingatat

sd on

enanodu

edden

aticc wtane). ng aretraxbo

peokr tocimothth

ng1

d alon

ngem

eneaFtig tioth

simtwn a

ntsguuc

d ns

c pwinta

Tche priexisy

of

erike toityotihroheg i3]to

l ana th

me

ndachFoltea

onhemiwoat

s wulcti

wsit

painganeThehopla

es s. row

imp my iooue in].

o taxal he

ent

denh

or edat n, t wilao sth

wario

witty

arag weoe

ordayduT

otain

meplamo

onsugh

ente

Bthesa

e wts

ntofro

d-sa

thwinar syhe

we r von

th=

mwiousin

d ayeduriTheatiog

enaneaf s h xp

eraBahe s iandw

t sf iolstr

anhe ngi

yme t

sve. W

h m=

eteth s sstaandd inge onve

ntne’asuth(Ft

peac

asese

ind

winfr

seitsl rootst

g win

mmthi

syseloW

m

ersth

strantd wnd

g tgr

n oelo

al’s ur

he Fitheerictieden

n thin

ngro

etss pan

okthtrowiti

meird

stoc

We

min

s ahe roktanwi

divthereeof oci

l pro

re s

g.e imiond onshene

g km

s prnd

ke erok

wasiaetrd

emcitfi

ne

andstr

keneingvide ren thity

prot

tstr. 2d

mensonore sertkin

m

ofrind p

pr kes sl ricst

maty irs

era

d rro a

eoug vduarot

ae

y r

roattherok2A

dynents an mr),strtianeth

f ncpiplatil

e pstac

catro

at(

st

al

rotkeangus vealltatiarrstrrel

octioe keA)nats arme, troal emhe

exciptcanlteplaaron

al ok

tic

fi

o

tate pglean

eloly ionrowroklat

ceonF

e p). am

we eatheok

tmae

xppach needanrtendw

ke

cal)

ixe

oil

tioplae ng

ocionnswsketiv

dn oFCplT

miwn

asue

ke-torati

t

peal a

re, d-sneeddit

win.

llytoed

l )

onsane

gleityn t a

s re pve

uonCTlan

Thicaithegurto

-prqicstot

eriax

rotp

stre wd imtiong

y o d t

(k).

s oe fr

e oy the

aborep

plato

ren thT),ne

he allh gliredorqplaques ta

imxetat

pasrowamong s

vinth

kin

of tfram

,of

e aoutpre

aneo t

eshe, we ir

lya

igd qu

anesin

al

mees tioss

okeas

mpuns str

varnve

ne

theme aat

appt: ese, the

s e tweines

y sa gib

touee s on t

t

nt(ronine aul

dro

rieves

an

em

e se. antta

pa(Benan

e s

toe

ntosuscsi

bleores frofthto

ts,ronsng

palwlsidu

oke

edsting

ma

strWd ack. T

araB) nt nd str

orqfi

o tultacalinge quo

ramf te rq

, ll , t

g tplawaiveuries

d tiggu

ati

rokWin

thk Th

aturothth

rok

quirsthanleglfoue

of mthai

qu

wx

ththanayelins i

that

ula

ic

ke ng he is

he us. oll he he ke

ue st

he nt ed le or es a

me he ir

ue

we xs, he he ne ys ly ng in

he te ar

2 Copyright © 2010 by ASME

Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm

Page 3: Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering,

v0wwfv(

Faarsms

T

tsataadb

wfdnwddv

omSdi

ve0.4wiwhflave(5

Figangandresstromastro

Th

thestrantypasandimbe

whfudenowidimdimve

of maSpdoinc

elo45ithhiapeloº/s

g. gled sulokarkok

he

e ronalpeym

nd m

e w

henc

enoninmm

elo

f aanpeowcr

oc5 Hh ichppiocstr

2.e 60

ltinke kinke

eoTgkelye mo

menwr

erectisi

n-dng-menmenoc

Dattnneciwnrea

cityHth

h in

cityro

. W

0 sngpl

ng pl

orThgeoe

ysiom

outnsrit

e ioty

dim-flnsns

city

Dutacnerifinwas

y z his

isng y

ok

Wi(b

sing frlanth

lan

rethe omp

is,f

mettesiotte

on y, m

flasiosioy

urckr ica

wasin

atws s frfr

ke)

ingblu

nfrone.he ne.

ticm

mela, wrotri

er wonen

o

menappononin

ink o

wal

ardng

t 8wit

raty

frero) i

g kue

m . Tle.

cameetranewotiew

naln a

of

nspinlenlen t

ng of

witly

d vg t

8ºth anyp

equominc

kine) , rro

Thad

alecrice’e ats

winl

as:

(gis

siongesesth

rf thth

y, veth

º s0

ngpicue

m 2cr

nean

resotae

din

echc ’s

dtioo

ngtim:

ges wong s s e

rolhe

h rtheloe

s-1 0.0e caen2ºrem

mnd speatiso

ng

eshan

asro

deronof gs m

eow

nal

sc

st

ll e ire

he ocin

a05

oal ncyº sm

atiro

ectionolid

ed

stinisyotriv

n. (w(F

me

isom

winl a

secetro

roin

espa

citnn

and in

of f

y s-1

men

icsotativns d bdg

imsm

ymtatveF

wFi

s menganm

econtok

otnnepeangty ne

d ncf

foat (5nt

s iativel

abblae.

mam mmtioedForing.

intr

g lng

mponterke

taterectgua

r

vacreflar t 05ºts.

in ionly. boackTh

ato

meond ar

ng. 1(

nstriclengulitndr p

tior (t ulaat (l

arem

appin

0.4º/s

thn a(B

outk lhe

tioof etrn. a y)

1D=

tanc) ngulatu

d o

pla

on(letoarth

lef

riemepins4 str

he anB-Dt thline d

onF

rieBm

yawv

D)=

ntan

gthar ude

mof an

n, efto r vheft)

edeningseH

rok

ronglD)hene das

nsFCes

Bamaw

vel [

tanngh,

ve

mop

ne

tht) th

vee l) w

d fntsg ct

Hzke

oboe ) T rode

she

s oCTs oaseath

rlo1,)

negle

ve

omprfr

hew

hoelolefw

flas. fr

t (e)

oti

Traollened

ofT cofedherotci, 2an

eoe

eloan

meresram

wwinoseocft

win

appT

freflRto

ic (

ajel (

notd b

f fcaf wd oemtaity2].nd

ouofis

ocndenssm

winge ityh

ng

piThequlig

Re o

w(grect(B)eslac

flaanwion

maatiy . Td

s f s mcitd

nt ur

me,

ingg a

oy

halg’s

ine ue

gh=14

winreetor),

s thck

apn binn aticonmThit

datmty

ore, a

g arof arlf s a

g Ren

ht = 84º

ng en)riepihe

k li

ppbeng a can,

maghes

dratta

meao

frof

e, an

ve t

roo

an

frReyncy

[80º s

ex) v

es oitch

wine

pine e

kq

al t

gne in

st

Φ

agackanof ref

nd

velbo

theounof ng

reqyny 14

00)s-1

xpvaof h

wine i

ngexkinqu

mthenitnstro

Φ

g-fk

n cf tquw

loote ndthle

quno

w4]) a1 (

peraryf th(C

ng ind

g xpne

uasmoe tustaok

fo

chthue

win

cith o

d hee o

ueoldwa. an(3

rimyinhe C),

chdic

cplaemsi-odF

udanke

orc

hohe enng

iis

itya

outh

e sof

endsas

Wnd5º

meng

re ahocat

coainma-stdeFCe

ntae-a

ce

ordw

ncyg is s s

y aaltte

he strf a

cy nf

Wed vº/s

ntas

esuandordtes

ouneatitel

CTb

anav

e [1d wiy,

at

str

aner

er r

roatt

y fnufroe vastr

s. s sultd yd, ws th

ned icseadof

T betneover

co15lein

arethrok

ndred

wolketa

frumomth

ariro

(Asinanyawwihe

teb

s dyf artwoura

oe],

enng,

eahe ke

d ed

will e ck

ommbm heiedoke

A)nusnt ww ithe o

erbyca

y F

risweusge

eff

gt,

a, y

e a

effin

ingaxplk.

m er3

end e)

Psowi(D

h thrie

r-ty cauaeCseeens Fed

fic

th

yaan

ffen ag xilanS

0r a300n th

) w

PloidingD)heent

tocouseeroT s n FCd v

cieis

, a

awng

ectan(

is ne

Sim

0.1as0 fi

he wi

otss g k a

e fitat

rqonedodffrth

CTva

,

ens

an

w gle

tivn oFiine, m

15soto

xea

ith

o65

kinxeilletio

qunsid bdyfofroheT aalu

,

ntsfl

ndis ia

e.

veopignd

thmila

5 Hoco edangh 2

of 5 cnemes edon

ueideby

ynr

ome iat ue

,

s uii

s is

ang

e appg. duchear

Hzcia

9d gu2º

strcosmaof

d cof

esery

name

m inn

e

asids

gu

anpo

1ceererly

z tate90

thulaº s

roksatif thcycf th

s rinth

macth

nnenonca

(

(

s d/a

tharthth

ula

ngsiB

es eby,

toed0,hears-1

ke

icsheclehe

ngheicchheern-an

(1

(2

aairherehehear

lete

B).a

byit

o d , e r 1

e

s e e e

g e c h e r -n

)

2)

a r e e e e r

e e . a y t

inaBinasa(f

as

wfr

aqma(aas(passA

wfrptoitcd(s

rodwtop

T

b

ndandBenc

a lim

aerfo

at tr

whfra

attqumiangan

atttrpo

atttrtr

Ap

whfrapitorts

comdowse

otdirwoo

pre

Th

bo

dud

ecacrela

miroor

nook

heam

acariddgunteacokosacokok

pp

heamtchrqu

imw

ee

tatrecord

thev

he

dy

ucer

auea

argilaody

d

onke

reme

Dckrtedlulaerckkesteckkeke

pen

remeh ueins

mpawns

re

Ftioctdshe

vio

eoS

y

esredusasegear ynet

n-e i

e e.

Duk er le,arriok des.erik des.e lnd

e e.

toe starstres

Furonios, e ou

oreSim

ro

s adue esr to

natai

diis:

uriars

, r or)du. Siodu. Tleadix

Nordianre ro

su

rthn, onthro

us

etmiota

anuc

ts w

mo yamils

im:

ingretrmve) uriSir)urThadx A

Noquiffnta

toklts

heths

heotsi

ticilaat

n ucesthwimaya

mics,

me

g e rokmidelohainim h

rinheds A

teue

ffeanth

ke s)

ermheofreatim

caar tio

ups e ithagawc re

en

is

palkeddocal

ng milhange c

toA):

ise te oerenehe

to.

me sf re itio

mu

al to

on

pwtht

h tgniw rto

efe

si

s

itcltees dlecitlf

tlaralfg tcoo

s thoveneou

to

morstrrois onula

eo t

n,

wahetotthituro

orqer

on

th

chere

(e tyo

therlyf othollth

thatve

nceustog

rero

otan

n ati

esthbo

arde ta

he udotaqur to

na

he

h reddoto

y of e y, ofe leche

het wer e s orqiv

e, okatio axio

tihe od

d van

al an

deatiueo

al

e

rod oro inthveit

f td

ctip

e pweabeva

quve

oke-iore

xison

ma

dy

veng

ang oio

e (ro

ti

ro

otaa

rsad

nduheent ithedoiv

pit

pie

a wetal

ue a

our-avn ess re

maer

y t

elogleaeglofn,

(Foll

im

ol

atisyal douce ntrine rs

ve tch

itcob

witwue

pa m

r ve(cul(A

es

atirotra

oce roe f , tCl r

|

me

|

ll

ionym

toorscestradust

sale

h F

chbtin

weee prm

mercoltaAsul

iodyan

cito

odoffotheTrot

|

e

|

a

n,mmo saes troal uctrol

effFC

h atai

ng enc

romor

moag

ollannnlts

onynnsl

ty of dynf a

orce ) ta

,

ang

, wmem

al: a

okto

ceokto

fecCT

anins

n andure

odgelinnt nes .

nsnala

aa

naattcekiop

ati

t

gu

wietr

midF

ankeo s

keo ct T

ngnedstrdonnuce i

deled neto

ex.

oamatio

at attamta

e inppon

the

ula

inricddFi

n pma

e pmo(f

|

guld rokownoctiin

l F

earorqx A

of mic

on

thtacmiackth

nempon,

e

ar

ng cadleiguppl

midd

plamidof fo

lathkewnot iontu

inFCr wquA

fc dn

he ckic k

hanmosi

A

av

r

valle, . pwandd

dowanddthr

ar hee, nsbn

uit

ndCTwiue

A),

ladaal

rk

f[1n atite

Ap

ve

ve

velly m1

wanedlewnnedlehide

ve p

sstrbe

btiv

dicT ith

e aa

apamlso

rigo

fo15

thtice tpp

era

el

low

miCare e nw

e ae s eta

velpitin

roko

beve

catach aban

ppmpo

ghf

orc5],he

c atoen

ag

oc

ciwid

C).rd ananwaananasai

lotc

nckeobetw d

tectsthou

nd

pinpinim

ht hth

ce the asyo tnd

ge

cit

itywithddl

v

ndndar

nd ndsyils

ocich e e btawede

s s heut

d t

ngngmp

hahe

e heoymthedix

d

ty

y ah le Spved d rd

id yms, r

ityFitan

ainee

esc

thst r

t ath

g g (pa

alfe ace iutme x A

v

y

anrt

peelorethvncth

mmre

y FCt indneencr

hatrirotanyhis

co(i.art

f oo

ctintermdiA)

val

in

ndreo ecocedhe velcrhemeefe

inCTis d edn ip

at icttaty

s

ou.e.ts

ofutin

nner

metire):

lu

n

d espv

cifcityduc

mloreae metrer

n T b

dup

d. th

pti

ftlytioaxis

un., a

f tter

ng erwtriec

ue

th

effpevenficy cemiociasmryto

thbydepsA

heion

fory onxi

s c

ntFC

an

ther

r wwinie

cti

o

he

ffeectntcal

aes idityes

midy oo p

hey pe

strAs e n

r op

n ais co

teC

n a

e (ron

wings

ion

ov

e

ectt trallyat

tddly s tddovpi

e avenro aupo

eppaxp

on

er-T)asy

strign ingg. len

er

st

tivto

al,y,

tthele atth

dleveritc

stvendkea ps

of

acpoxiser

nsi

-fo) iym

trogh

tg T

eadof

r o

tro

veo , v

tthee t

t the e r och

troeradene re

strp

chosis)rpist

orinm

okht)thepr

Thd f r

on

ok

e ath

vethe a

to the

atoon

h r

okagntthesroitc

h ite. Iente

rcndum

ke ) e rohetoro

,

ne

,

ke-

anheenthe

vang

ve ano nerot

ke-gint ohesulkech

tye Inndnt

ceuc

met

pww

odereo aota

e w

-p

ngle trap

venglvendogldo

e wta

,

-pngonerelt,e h

ypto

n odict

escetry

plawinwi

ducefoa nati

(

wi

pla

lefoalpitntle ntorle orwiatio

plag tn tefo, a

FC

e o tothcuw

ed y

annginceorneio

(3

in

(4

an

ooul ttctrao

trarsa

orsainon

(

anththorw

anCT

othhe

ulawit

bo

ne g. ng es re et

on

)

ng

4)

ne

of ur to ch al of al al of al

ng n,

(5)

ne he he re

we nd T

of he er ar th

by on

f

f

f

)

f

3 Copyright © 2010 by ASME

Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm

Page 4: Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering,

tdF((d

rofNaicpaHaos

Sd

fsptiapobsgcmbsvsacnaa

se

thedirFC(F(de

resof foNoanin copaaeHoaeonste

Stdu

foustuprtheinsanplaorbostrgrcemobysixvesixanconoaealo

, staeq

e reCTFC

)eri

spfrwot

nd g

omaraeroow

eronlyea

taur

unudrese se

ngan

rigodroraventod

y tx elox nd omoneroon

atequ

wectT,

CF), piva

pew

watab

gempamodweody ady

abrin

Anddiesem

ectlenegindy

kevittedetheeq

ockio

mpnlinodng

)e, at

wintiow). piat

ctwiardbly

enplemedynevdyna y

bilngA d iesnt

mat

e) e fn fr

e tyr

elee Nqucityinoriplene

dyng o) a

ttio

ngonwe

Titctio

tivind/y,

nerexetena

vernaf

m

ityg deins t cthborefralo

frapl

y (o

edN

uaty

nemieeteeanaoranth

on

g kn oe tThchon

veng /ba inonra

x feramr, amflig

mo

y het

n [abco

hemodelaamocamlan(Fof d aNew

tio(

mnte ar amr and heyns.

kinofterhe h (ns,

ely

acn nl

al, funs.

mih

migh

ode

dovai17boonmdyatimecatme

neFig

gas wton

,attat

so

miaro

hy .

nef trmst

(, r

y, a

ckth

ly h

ncT

icshoicsht el

deveile7]ounte

maty ive (tede (e g. gr

atons, tictiosyoric ouhe

c

emtrametro)

ref

ware

kwhea

hoctiThs ows n

ls

eriered] aut exticor

ve (d (fr3

rava ron w

,c eonystrdi

funadca

maanedok, afer

whea

ware cas owiohe o

w re

ne[1

ivr

d dan

lxt, carieto,

a,

ram3).virig-E

wi, eqn teinfornd d an

atinsld tkeanr t

era rdcufu

wense

f l

emear15

vat

dend lina

al eno t,

at , m. TtygiEuith

)qu(m

narrcth

anb

icslathe

e-andto

re

d, urrunevs o

effre

latmar h5,

tiv

sc[

neanm

nteth

th

e,Thy d

uleh t)

uat,

m ryceshengbe

s, tioe navd yo A

anla

renncveroffecevteainho16

ve

cri10

ear a

moede ,

he,

, bheisb

er traastio o

y s e bglee

wonne

veryawAn

ndatenttior, f tctvoeransov6]

es

ip0]rizanoded who

e w

bue ds bo

eans uon

,of d(Xboe co

whn. et raw

nn

d ert son

trats olval s uve c

s a

pti. Rze

nalel wor

)w

ut dispdyqu

nslun

ns , fdifX,od

on

hicInfo

agew (nex

ralstuns

anof

vina

unr

can

an

onReed lyo

ithriz) in) w

istpey, ualankd

effe, Ydy

ns

chn fored

x

l, uds o

nslf fngannecon

nd

n ea

yticof h zois

ng h

wittanec

iatiatiokndes)

equereY,ado

sid

h rac

rcd F

) B

is

andy of an

latfog ndexponst

d

ofadhocaFa

onts f

bhavth nciftsionon

nowscreuaen, Zaxo de

recce FCa)

s t

ndw

f thndtio

orww

d plndtil

li

f lderoval FCa ftafixbavinth

cefie cnsnaw

crielaatintiZ)esn

ere

sucoch

CFaxear

the,

d vwehed onwa

winv

lorditl a

n

linr cvein

CTfix

al sxeasenghe b

ed cos oal nsibiatioial) s [

noted

ultrdhaFsesre

e n

vee ce g

naarng

verretioap

ea

necaerinv

T axestr

ed e g e obe

bomofves. intivnsl an[1t a

d

ts daans fs og

no

ertcoge

al vrd/gs rtied.onpp

ar

earan ingveanedror(Fthor

etwby

mpf melW

ng ve s e

nd7]afse

inancngeforof giv

|

on

ticoneo ave/bh

ica. I

ns;ply

riz

riza

g estnd d pokrelFiherigwey plemolocWi

thti

equd ]. ffeep

n cee r t

f thven

n-d

cansiomarelo

bachaal In; ty.

ze

zeals

ftigF

pike latg.

e sginee

eteoticitithheto inuatoB

ectpar

a e wastrahen

di,

al ide

mere occkav

n tth

ed

edso fligaFCtcpltiv. 3samn

en .

e iotyh e bth

nclatiorqBet tra

newis fan

e sby

im

ver

etre

citkwe trth

her

d f

d fre

ighatiCFch lave3)mloth

. dy

on y (thboheluioqu

ecath

ate

etithflansstry:

mea

elr fricexty

wab

ranhisre

fli

fligefhton

F. Wa

anee t).

me ocheIfyn[

(heode E

udensueau

he ely

t dh aplarok:

nsanlocforc axpy aardbeenss sfo

ig

ghfert n Wane (toHo

cate f tna18,

e adyEaess es use

ay

drath

ppatike

siond circanec

and tenslastuore

gh

ht r tdis

We ng(F t

Heoritewth

am8]

ady par

s i(L

e aer

w

aghepinone p

on

itice ngcte

nd trn atiude

ht

dto

dyns g

file

Figtheereiend

winhe mic

, ,

ddporth

twnc

(Lboro

wit

g o t

ngnspl

na

escleedwanst

iondypr

d

dyo [nagi

firse g. e e nta

ngi

csw

itiosih-fwco, od

odyth

opterg cs alan

al

s, oe

e od winnstuns

y, re

dy

yna11

amivest

3bowtatat g bins a

whi)

ioitifix

welorpM

dyyn

h

pprmcoalone

fir

reefof to

ngla

udis w

ev

n

am1]

mienc

3).od

we tiothbaseareic) aon onxelvpo

M, y pnaap

pominounone f

,

,

rs

esfficf ao g katiiea

weio

am

mi fcs

n, con

(f Tdy

donhe asecte

ch an

on ed

ve ora

Npoampp

senont

ngfra

,

st mre

speci

attb

kiiond

affe cou

m

icsfors. b

nsfreThy wden w

ce t dec

nd of (

d f

atN)osimicpr

edoloter

g tam

,

mepecientacbe inen b

fecco

us

mic

s r r

Ibassidee

he w

efiwcen

aboes

comaa,

fraco

tin), itic op

d togr-ftheme

,

momprectintck

memon

byct

onq

cs

careInsedee bstithneithntndodscrm

angano

amounga

iostpr

o gyfoe re (

(

(

mesivets k mman

y t nsiqua

s

anelan ed er botroh e h terd dyrib

mprguot,

meupg ctn tearia

thy oorcro(fo

(6

(7

(8

mensenel

moratth[7th

ideas

n batethoth

odokththth

r oth

y berisulathe e, ple

thtin(

adate

heofceollor

6)

7)

)

ntntly.

;reiche7].heersi-

beedheonhedykeheheheofheis

edsearer)a

edheng

,dyed

e f e l r

t t .

; e c e . e r -

e d e n e y e e e e f e s d e r r ) a d e g , y d

lithafudqsb

awdN(Fropsso

NSd

T

ndb

inhe

appfunderquta

be

anawitdimNoFiot

plaimta

on

NoStader

The,

nonden

od

neaeoprncrivan

abifo

althmeotaigtatanmpabi

F

n-abiriv

e s, n-dnotdy

Tar

oryroctivantiiliou

Byt

h fenab. tio

ne.pliili

FC

-diilitvat

sta

dimtesle

To izy. xiionatiityityun

Baticflinsibly3)on. ifiity

CT

imty tiv

abi

mes teng

ze

imnsivy y dnd

secaghioy, ); n a

Wficy d

T a

TA

men

ves

 

ilian

ensthegth

ant

Amas eso

dein

edal ht

onath

arWeatdean

AB

ns

s

 

ty ndsioe h a

nath

A ati

os of ern

d oe

t dalaxherou

We tioerind

BLD

io

dd onno

as

alyhe

cng

of mt

iv[1

onestdynl fxisereun

onivF

LEDIM

ona

der

aliona f

yze

crug

tmuthevat17

n timnafors efndne

n, vatFC

E M

al

riv

ize-dfra

ze equcththeule tiv

7].

thmaamrm

ford egw

tivCF

1.EN

vatared

dimact

quciahe e ltim

ve

heatmims

re

glewhve

F m

ANS

A

4

tivre

bymentio

thatal ap

iplmoes.

e dioicss o( a

echices mo

ANSI

An

2

2

1

4

1

4

vesn

y ns

on

hetio

aerpelieoti. A

deons dof

a r(

ctechar

od

NAIO

nal

2

2 4

16

16

4 R

s on

sioof

e on

stroertedioA

eris duf s

) ro

edh rede

ALYON

lyt

4 R

R6Φ

R

n-d

onaf w

nns te

odyturd on

co

ivo

urista

dota

) d

ise els

LYTNA

tic

Φ

Φ

, dim

al win

neao

ep ynrb

bv

om

vedof inabdoatipth

s exs.

TIAL

cal

R

me;

vang

arof

nabedbyvamp

d t

ng biloeiopluheju

pe

ICL S

l e

d

R

d

R

aens; walu

g le

r-hf m

inamd y aripl

Fthhitys n

use usec

CAST

esti

α

α

d

dd

R

ansiowhueeng

homn

micmt

iablet

FCe

hovy na ae

sticte

AL TA

im

α

α

d

d

d onahers, gth

ovemot

tc mothblte

CTs

vede

noaroa effified

EAB

mat

c

s

|

α

alire

h (

ertiothfoote les d

T staer.er

ot oulin

ffeied

d t

ESIL

tio

cos

sın

|

α

|

ize

R(

Fcbcldciva

r on

he orctio

s)de

anab. Trivc

unne

ectd o

STLIT

on

s

n

s

α

⁄ar

ed e

R i/

Figcoobocenlocdocenis velang

)

fln

ceonco, sc

ndbilTavacoid eat

ibe

TIMTY

ns

s

sın

re

bequis

)

g. ordyntecatt).ntesplocguara

liguli

es n oras

cri

d ity

abativin

ar o

in e

MAY D

sın

ın

n

ın

co

, n

by uaas

).

3.din

y er ted. er peccit

ulare axe

ghusiinanv

rres ip

FCy leve

nci

trof

tze

ATDE

n

os

non

als spe

Anafrof

d bThof

cifty ar dees,

ht ineandarespwtio

CFd

e 1esid (

ranfther

TIOER

n-d

toec

A sateramf gbehef gfie

cove

efin, re

dng arid triapo

weon

F de1 ss wde (nsth

he o

ONRI

c

s

dim

o 2t

che fmegraeloe grad omeloneesp

dys

iztoabonelln

mrisuwi

w

slahi

D(e

NSIV

os

sın

me

2Φrat

hemfrae avow

diavby

mpocied pe

ynsm

zatorqblendl aof

mova

umithwi) atis De.g

S VAT

s

n

ens

ΦnRtio

maam

is ity

w tist

vityy ponityab

ect

namatioques dinasf t

odati

mmh nithcoio

tisg.

OTI

sio⁄R o

atie

oy (thetany a

.neny cbouiv

amallonue

(ngs th

deiv

manoh or

on tr

scu,

OFV

on,

. an

c v(ori(ree wncean. Tntscomut ely

micl n s (i.

g obis

elsvesarionax

rreo

anus

F NVE

nali T

nd

vie,

igied wie

nd Ths (mpth

y.

cspep

as.ed

bt m

s, s izenzexiespof nsssi

,

NOS

ize

he

ew, inado

ingb

thhe (pohe

, erprs a., distaime

wa

eseris pothlaio,

ON

edan

e s d

w o

ateot)g bbethe

tra,

one

wrtuocan

sstuineth

weasss tro

onheatin.

N-

d bndupde

of t, ed), batwwan

,ent

,

weurbcenalstaur

ninho

e dsothev

nde ion. )

-

by d perno

th

d wh

aseee

winnsla ts

e ba

esslyabrbngod

deocie n

val

ds stn Ob

rscote

e b). at hice (en ng ati

)(

,

catis

ytibilan

g td c

eriiatnolu(to

tro

Othas

cries

boTt

ch(bl

tbaion a,

, a

caio

icalitncthca

ivteon

ues

o okfohese

ariptth

odyThethe

h isluetheasenaand

and

an on is al ty ce he an

ve ed n-s. ) a

ke or er ed

, re t + he

y e e s e e e l d ,

d

4 Copyright © 2010 by ASME

Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm

Page 5: Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering,

am

Lw

w

a

lmtprpddlt

R

Rt

p=wclgr(c

anmo

Lawr

wh

are

linmotheplarefpladedylatthe

RE

Roto

pit= wicolefgered(Fco

nd om

astrit

he

e

neome anfean

eriyntere

ES

oborq

tc0.ith

omft endu

FCour

me

tlytte

ere

c

earme

pneereneiva

namrast

S

boqu

RT

h,.3h

mp(ieruc

CTrs

then

y, en

e,

co

rizen

proe ene fatmal tru

U

otue

RoTh, a5 a

parinnratcedT) se

he nts

tn a

on

zentsodis

ncefrativic(d

uc

L

tices

olhe anH

a risneted.co

A

s a

theas:

nst

ed s dus e am

vecs de

ctu

T

c ws

l ti

ndHz

stsoerd Tou

of

As

rear

e :

tan

do

ucta bo

mes i

escure

S

w

roim

d yz, tatonr) w

bThun

f p

ssu

efere:

li

nt

dyf t

pode

(intcre o

win

otmeya

ticn. Wwbyhisntepre

um

fer:

ne

ts

yninopldy(F

(intoribof

ng

tate cw

c W

winy s aered

mi

ren

ea

o

amnef lany fFin

o bef s

g

tioco

w a st

Witngthasr tdic

in

nc

ari

,

of

miert

innefraig

Tlo

ed sy

ex

onouax= troth

g (hesymto ct

g

ce

ize

f

ictianee am. Taon

byst

xp

n ursxes

2okth

(bl rm

ted

ho

ed

a

b

cs,a erto

me3)

abgi

bytem

pe

ses d

23ºkehelurig

mmthed

ov

va

d

an

bo

, abtiaf e ), leitu

y m

er

s duº/s

e e rue gh

mee ro

ve

alu

eq

,

d

dy

boa

swt

e ud

m

rim

ofurstrplrocu

ht etr

doll

eri

ue

qu

,

y

ou

(symwitthe1

din

ma

m

f trinrola

oll ur(

ry dirl

in

es

ua

m

isut

mth en),

na

atr

e

thngokane

rorv(ole

recto

g

s

atio

m

s th

mmt

n

al

rix

nt

he g rke,e ot

vesuteactiorq

re

o

on

mom

bohe

ame

heusw(

x

ts

mrol, R(

tats, teradsioqu

efe

f

ns

m

ode banetre sin

we(de

s o

mell RenotioFr) s t

on ue

fer

t

s o

men, dybo

ndry)sang

e es

.

on

earo

e =o onigw

too

e b

ren

the

a

of

nt

y od

d ). amg ca

scr)

n

asuot= r

n, g. w

o aof by

nc

e

and

f m

ts m

dy

Nme

thanribd

fl

urat70

rolth4)ina cf ty

ce

a

d

mo

ois

ma a

Noe ohen bedy

a

edtio00ll he) a

ng coturE

c

ae

ot

ofs tssax

aotore

dedyna

p

d on0)

re tar

(onrn

Eqn

con

ero

tio

f thes, xesarte riees

ded bam

pi

ton a. T

rotorre (re

ntinninn.

nd

od

on

,

ine

s e

tenstiecobymi

in

orqareThtarqenednung. 3

dit

dyn

n i

nesy

(z

thntaimouy ic

ng

que he

atiquenhd uog (3

tio

na

in

ertys,

zerhatati

maup

s,

g c

uesh

e toneshacu

ou(Fis

on

am

.

n m

tiate

, rot ionatiple

a

co

s hoton)s ganur

us Figs

ns

mi

ma

a, em

o, byn on

e

as

ou

arowrq

gencervto

g. al

:

ic

at

m m,

sy a

nsth

i

un

rownquis

eneed

vesota

4lso

f

tri

stma

, sin

cas s he

illu

nt

ounn iue per

d, s, al 4Ao

,

fo

ix

tatmatan

ncchthofe

us

er

ndin ppl

ratbuFr

A, s

orc

f

te tri

nd )

ce hohef h

str

r-

d F

proloteutFigolt

sho

ce

for

ix

), t

ose sst

ho

ra

thFigodtted t thg.ll theow

es

rm

vo

thesinstrtabove

) te

heg. duedbyhe 4toe wn

(

a

(

m

(

(1

ecof

e ngrobier

aed

r4

ctd y e o4)orqti

n

(9)

an

10

ar

1

12

ctothar

g okelit

rinanb

rol4 (tiofothon) quimfo

)

nd

0)

re

,

1)

2)

orhereis

ae-tyngndby

ll,(onorheneis

uemeor

d

)

e

)

r e e s

a -y g d y

,

n r e e s e e r

cFbythth

fumpEptorowtosm

sp

comFigby yawhahe

funmeparEqpreo ol

whortr

mi

ympro

mg.

twat e r

nceara

qnedthll herquokid-

mod

mpa4

thew t

thro

Inctiasuam. 4

dichat

Freueke-s

PD

mmdu

ar4Ae orhell

n ionur

me4.ctet o

FCe e ie ttr

PitDumetuct

risA)

Frqere

a

Fn edte Medof

CTthis theok

tcuritritio

sonis

FCquee ixi

Figo

d ersMd rf thT hhe

ae ke

chingicon

n)s a

CTesis is.

g. ofF

s orroheham

sslif

e,

rg

c wn a

). ab

T ms an

.

5f FCanrell e masmosoft as

rop

wab

Tbomarno

5,s

CTndovto

m aomcifo

s o

otapi

winbou

Thutode re

, strT d veor

meaa meiatforob

attc

ngut

he t 6debes

wrok

hcl

er,rquasdentercebse

tioch g kt t

s6%el otsu

we kehalo, aueurelnt d e er

on

kithe

str%

(Ath

ulta

pe as oseas e brelay

owmrve

n roinee

roo

Ac

an

plfa

elys

byed yeof witmay

ed

otaemro

okof nn

clont

otfrea y

shoy to

edf ath y d i

atimaoll

ke-thneosto

t eq

sm

owthene

d parmlihin

ioatl a

-avhe exse or

th

qustrmawne e (pemiftav

n p

onican

vep

x Ato

rqu

heenroatcn FC(sea

m it fvepre

, cs nd

erapeaA)o ue

e nconchinC

solak is fore aev

thty

agak), ze

e a

stcy g

hen FT lidoa

rca vvio

hethayaw

FctoyfrleamcthTflis(Atop

gek r

therab

tro

ls Fim

d gocat ce,vaou

e at w

Figcouorq

yawframeft

andmincurhe

Theflaps A)orq

pla

F(AFfrRe(Baisawlip

ed roheo

bo

okanlinth

igmogrcum

, waluus

lee

w a

g. ursquw met ad rnurvee oe ppsh). qu

ane

FigA)

FCfreqReyeacB)

as s s

arewitine

pre

volle (F

out

ke-ndnehe. 4odreeurmaweuest

efenax

4sesues

(e.

andredus es,ord

ppinhow

Tue e (

g. 5) T quynch ) Ta sh sth es)edi

val tstFit a

-ad eare 4A

delenrriaxe

e htu

ft nsues

4. s s a(C)

Td rd cle

, adinpreng wn

Thegeno

5. Ta

uennol

frThefuow

shoth

). ict

aluortroigan

avea

r th

A,l (n)ng

ximexhigudi

aurs.

Aduabo)

Therigcureft)andnatedi

con be beno r

StTheas ncyldsfreqe s

uncwnowheThion

uerqok. 4

ny

eran

dhe, t(d. Ng muxpghie

anre T

Aeuriouax

e ghtrve) ad ttesictoubybla

nerrol

troe a

y s nqustrctin. wn

ir he ns

e queke-4B

y a

ragngdeeorthe

dotNoa

umpeches

nd a

Thu

eroingut xetort wesarethes rtedunty aacratl r

okesfis

numuenrokonThinlire

s b

(fe. -aB ax

gegulepree tteot

aftm.ct

er (F

ra us

odyg ros rq

win. Te se frepd tera dck tionrot

e-atro

funs mbncyke-n ohe n fineed by

filA

aveanis

edlaren

etiti

edtabter. t thth

Fig

rigsy

s i

ynro

oll o

quengThshfilpre

ror-todocu

n tati

avoknct

shbey -avof

mfilearlinEq

leAlernd p

d r ndcam

d gblr tBha

hang.

ghymin

namoll

(Aof es gs ae towlleeseollorq

otteurwion

verke-tiohor ais

vean

meledr neqn

ed soragd per

rov

denal

me grey,thecat n 4

ht mm

n F

miro

A)thg

aretotwnd entl qued rve

withn)

ragav

on owasss arangasd fit

es rn. 4

go, geCrp

ollvencec

ee, thhe caath

4 i

mFig

c ota), pheene staln bgrt tto

ue g

es h a.

gedver

own,socals

ageulurblttinrep

4.

gra

edC),pe

l eloceestcouenhem

auftehein

wmet

g.

tatipit

e nershol toby reetheorq

(Fgre

rea

d rragof , ciaso ed ar

redlacngpr

rees p

d p innd

Foce timur

n) e mmiseer

e o[1

wintri6

toriontchstr

ratoworq

sen eirqueFCeenepsta

rolgeflanate

prov

d tck gs res

enprpinddi

Ccit

omrseismide r thon15

ngic 6,

rqun. h (roktedwnquolicy

r ae CTn creati

ll Fd lapnd ed plooll veltorcy(b

sen

n retcdicu

T ty.on

mate

s smead-sthhe

ne 5])

gs t

w

ue (A

(Bke

d bn buesid yc

aveb

T) cuseic

FCr

ppit

wottFCocrquycblant t

cydi

ch caula

a.

tesof

simasstrhe e min).

hto

we

A)B), e-pby

by s (r

gcleeray m

urvent

st

CTrolingthe

withedCTcityuecleackthe

ycicta

atiar

asTbos f tmiur

rokr

mn t

harqp

tim)-(

aplay tblrig

gres

aget

modve t ttro

T. ll g e h

d. T y s s k e

cleteaninr t

s Thotbthilarekero

midth

avquplo

meC)

andanethelueghteenonesthedelin

theoke

e, ed nd ng to

a he th by he ar ed e, ll

d-he

ve ue ot

e ) d e e e t n n . e l n e e

5 Copyright © 2010 by ASME

Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm

Page 6: Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering,

owdTddrat(d6cNqgp

odps7utam

(ape

onwiduThduduredantha(fidu6)chNequgrpr

of depitstr7Auntheaemo

(leanpren

nlyinuriheuriuridu

nd atirsuri.

hanevuarreered

f septcroA)nde

eroos

eftnd rodnh

y ng ine ininuc

mt pst inInngverteendic

Fstrenh ke

). der

faodst

YD

ft)

duan

thang FCg g

cedmipit

hg

n gerther

n, ct

Figrondro

e

reaildynof

YaDu

wre

ucnc

he ndd

CTtth

d idtch

haldth

e oher sF

tio

g.ke

denotfrT

stlunaf t

awurwiedcedced

tid

dowT thehedudlh lf

dorhe ofelestr

Figon

7e ncatreqThtimureamth

w iningdud d

imcown

me e surle toorso

f pesrog. n (

7 fr

cytioquhemae mihe

rong g

ucebdu

meomnsmof

serint

orqf alothpiss,oke

6do

shreqy onuee ateoficsv

oty

(bedbyur

e cmpstrod

firecongo quthl theitc, es

6),ot

hoquof

n (en

e f ts ar

tayabl

d trin

coparodestong t

vuehe to er chth

s s, atte

owuef (Fncy

thththocria

tiawlue

dthng

ouareokeel t hndth

vene is

bmt

thestial

ed

ws encFCigy hehehe ccati

ow e

duhe g d

urse te p

had he nts r

blumitwtore illthg

thcyCT

g. is

eoe m

pcuion

n rocu

urino

do

se th(

prealfhaot

trareueidd

wo rqobl ghogre

hey T 7B

s boremeprerrns

oturngouow

ohe

edf alfthal)edue cdlq

qubsgivugee

e ano

B)beetieaedrins o

tatrvg utewn

of to=

dico

f oher). uccule qu

ue servegh

en,

stndon), etticaasudicngof

tioves

der nst

thorq= 0ctsf ofr tH

ceurv

suar

crve h , F

rod n athteralurcti

g af p

ons, do

(tro

hequ0.3s d

f utw

Hoed vetrrte

comveriit

Fig

okananher reioat pit

n, F

ow(riok

e mue35th

doupwoow

oe, okermd ses g.

ke-ngnge rfi

d n th

tch

thFigwnigke

me g5 Hhaw

psto qwevonlFiker s

mpat

e tti

. 6

-agugurelittpvm

heh

heg. nstght

a

meageHat

wntroquvelyige (strartortoim6).

aveulalalatedre

valmie sto

e 8

trot)

and

asene

Hz,t

stokuaer

y dg. (firoredrq

o ame.

erar ar tiod edluighstaorq

to8) ok

wd

surer, thetrokeartr, tdu6)irs

okd qua e c

ragvv

onby

dicue.htarqu

orqi

kewire

F(AfuRvshthqufipr

Fcopanan(rgrcythprm

rerat

e ok (erth

uri), st esto

ue suco

gevelvelnshy

cti. At rrt ue

qus

e; ined

igA)un

Reyelohoheiuaittire

igouitcndndredreycheire

mo

ed tio

p

ke (mr she inganh

s, o

aubou

edlolohiaonAreof

e a

ueeni

ng duc

g. ) Snctiynoc

owir adringdic

g. ursch d dd rd men

cleir dicde

pon

=pit

(midstre

g nd

halthth

asbsturs

d pociociip a qnss suf

are

e nhin

(ce

7. Strion

nolcity

wn

ratgscti

6.ses

rodowredmin cs avcte

el i

pitn d= tch(dddroexpved lf hehoymtanse

pitityityb

qus, w

ulteae o

prha

n (reed

Srokn dsy. inlin

tic. ion

. s otawnd inucuo

veredis

tchdu23h

dordleokepeenit o

ereosemnti

tchy. y betua

wilt facob

roanc

edd d

Strke-of

s nT

n ne

c R

ns

Aeab

ation scu

us urvn rag

d p

h uri3ºto

rsae eserintr

isof e e

mmtias

h Aistwdr

hll fro

ch bs

odunceon

d du

rok-avf numhefilar (

Redby

erobouonstrurvbl

vesth

gepit

toinº/soralto

s (imras thisu

metal d

FAlts werathobomh

er

ucedntcu

uri

ke-veflamb

e mlle

dod y E

odut

n. Trokvelues, he

es tch

orqng strrqul to d(vmeal tslhes nundtrip

dif

FCthst

eenticowe m harv

ced dtrauring

-avraappbemeed

(botte

liEq

dynp

Thkeses. e) an

e ov

h t

quu

rokuetodoenentoig

e rnodeie

pitffe

CThoutilln c

wedt

lfed

ed duastrvg

veagepin

er) eas

bblaedineqn.

napit

he s a

Tar

nd orvertor

ueupskee o orsntrnt o mghredo er es tchere

T augl pifu

evdisthef-sd (

burt, esup

raed ng

asu

blaack

d es . 5

amtchto

areThre

trdinr orqu

e fstre, ismsarare

mihtlyd cona

h en

asghvaitcuner

scue str(F

byrin

ts, ps

gepi

g fandureackk

br

5.

ic h

orqe she

shthenaonue

foroR

s midal)al esuiddy cuonnoat F

nt

ah talchncr, usun

rokFig

y ngthF

str

ed itcfred d

k

blarep

ta

quesho

tohoe fatene e b

or okRe

endd), touldlenurnso r

dCfr

a fthlidh Ftio

ssenmkeg.

thg uhe Figro

pch equ(Btocylin

ackpre

toraxees owotawnfil

es st

by

a e =nhdlebu

o mltsle nhrvidrodiT

fro

fuhe d FCon

gedmoe, 6

heup

g.ke

pitcF

ueB) orqyclnesk ese

rques du

wn al n blle

retro

th

sto

= 7hae) utms sst

hae,

derotaff(

om

unl

duCTn grd odw

6).

e ipstto 8e.

ch CT

encan

queless)

cen

ue d

uribyto

byd ep

okehe

ino t70an

at iidshtro

an, Fraatifer(som

ctinurT a(F

realade

wh

introorq8)T

FT cy nges s w

cunt

tduingy borqy s

grpree. e F

ngth00ceanit ddhookceFigabionrenolith

tionearinanFiatlateeleher

nneokqu) Th

FCas(

gula

wian

urvth

timuring ubluqusolree

eseTh

FC

leat

0).edndislewkeedg.len.ntidhe

onarngndg.lyer,edre

erkeueis

his

CT.s a(orlarareithnd

ve)he

mengupue

uesliden

enthe

CT

e t . d d s e

w e d . e . t d e

n r g d .

y ,

d e

r e e s s

. a r r e h d ) e

e g p e s d n t e T

aetoagFreclivb

Fli

F

v1a

ined

s2csaa

asyentor

avegreFCes

cycin

velby

Figine

Fli

val)

are

n eigdim

ub2),conub

andacc

ymtirrqueree

CTsucl

nealoE

g. 9es

ig

luan

e s

T

-1

Egeme

bs, ansbsd co

mmre ueragen T ultaesar

ociEq

9. re

ghN

uesndsu

TA

1.4

Wqnnven

Tsidansissid

om

ms

e gec

acans, lyity

qn

Stepr

ht Nos) d

um

AB

48

Win. vensi

Thdendstede

mp

etstrby

edcyctsnt F

y y . 2

trores

son

afr

mm

LE

ith1

ecio

e encd aenenc

pa

tryroy

d vcls t

Figin(F2.

okesen

ta-d

areruma

E

h t1 to

on

cea nt ce, ni

y loke

Evale,c

torg. ncFig

e-ant

abdime cit ari

2

-1

this

orsal

loe mslw

e w

ied

lee

Eqalu, Folrq8

creg.

avth

bilmecafl

ize

. N

.3

he s cs cl (

onmlowmwd

ead(F

qn.ueFilli

queBea 9

erhe p

liten

alcflyed

NO

39

stcaca

ngiod

owith

mowhi

dsFi. e iiginee a

ase9)

agpre

tynscuy md i

ON

taalcan

)

itude

w h deicby

s tig1 is . eaa

anesa

geded

y aiolamin

N-

-

abicun th

an

ude (st

te h y

to . isa

8Carlarod

s an

d ydic

at onateor

n T

-D

-1

iliulahend

din(mabthco

a8Cs abC)ly ouCw

nd

yawctio

hnaledrpTab

DIM

.5

ityateend n

namoblee orrea

a cC;al

bou). w

unC).wit

c

w on

hol

d bhobl

M

8

y dedn bno

al ode

prreesu

co; lsoutS

wid . Uth

clo

FCs b

ovs

baolle

EN

ded, tbeon

de su

presulfo

onttho t

Simithth

Unh os

CTby

vestaselog2

NS

-

erith

e on-d

dy1)ubrevspoltsorw

tinhe sh19mih hensthel

T. y E

riabedgi

2.

SI

-0.

ivhenobdi

yn), bsvions w

nut

ho9%ilath

e suhely

ReEqn

nbild oica

IO

.5

vatn

btam

naan

sidiound

iwar

uotimow% ar hero

urpe

m

efen.

glitonal

ON

6

tivthain

men

mn deus

ds n rd

oumewn

ot

e olprist

ma

fer 2.

ty n th

a

NA

vehe nens

micun

encs

t

d

Fc(ptt

s e n of to yal isitroatc

to

hean

AL

0

es e

edsio

cs nsceC

to a (b

Figcou(Apitthethe

yacofo

f tr

awaninokch

o th

de ad

L S

0.8

kig

d aon

stae mCF

a

ba

g. ur

A)-(tche se l

awouor theol

w nd

ngkehe

he

derank

ST

87

knogenaccna

cabm

FDan piac

8.rse(Ch (streg

w ur

ce ll a

d ly

e s

e le

rivnalkin

TA

7

ownvco

al (

onble

moD

iitc

ckw

Aes C) (Brokgen

Fse

compan

axip

y, tfrth

eg

valyne

AB

wnvaord(

nse oder

in-chw

Aerd

toB), ke nd

FCe meandis itcthrehe

gen

atiytiem

IL

-0

n,aludi

sisose res-p

h war

roddurorq

anplof

CTof

mpakd

bch

he qu

e t

nd

ivca

mat

LIT

0.

tueing)

st sci(msuph

drd

dyrinquendlanf F

T pf ar

k ypibeh m

ueth

of

esal tic

TY

19

thees glei

illmoulthasdod)

ynang es

d yneFig

prthrisyaitcecax

meeneo

f F

s e

c

Y D

9

e an

ly.ig

oflatodts seow

am

ayawe frg. 4

rodhe soawchauxeeancyor

Fig

(stda

DE

synd. Ten

f tode

[ c

wntr

micya

abow

fram4 f

dup

n)w h Fuses suy et

g.

(wtimat

ER

-0

ysd cTanv

aorye 3[1co

n an

c tawou(C

mefor

ucpr). toFC

se (

urea

tic

5

witmata.

RI

0.9

tecoabval

a y m3)9]

ou(u

ns

torw ut C)e. r d

ctioedT

orCTth

(fiedanca

fo

thati. T

IV

92

emorrblelu

fm), ].

uplup

sla

rquroro

) aR

det

ondicThrquT,heilld ynd l

or d

h ioTh

VAT

2

m mree 3ues

fasmod

w

linp)ati

ue otaoll axe

Reftai

n ct

he ue terledya

es

de

nnshe

TI

mesp3 ls.

st de

whThng

ion

timatio

(Aes ferils

ovted

se thee d awan

sti

tai

nos (e r

V

-1

matpolis

e (hiche

g m

n

meonA)o

r to.

ved str(fe isg

w ngim

ils

on(Tre

VE

1.3

trionsts

st(mche of

mo

e n. ),

of o

er y

rokfillys

greFCgu

ma

s. R

nzeTabsu

S

36

ixndis t

tabmoh a

ff otiw

ayawkele

yawn

eeCT

ulaate

Re

erblult

6

x inth

blodarfas

iowit

an w e-ed w

no en T ar es

ed

ro le ts

ng he

le de re st

on th

6 Copyright © 2010 by ASME

Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm

Page 7: Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering,

decaying magnitudes. The unstable oscillatory mode corresponds to a near out-of-phase coupling of and (phase difference is larger than 125º); therefore, in large part of an oscillating cycle, the insects pitch up (down) while moving forward (backward), also refers to [19]). Intuitively, for the in-phase coupling of initial and , pitch FCT and the pitch torque generated by FCF ( ) act together to reduce the angular velocity , resulting in stable subsequent motions. For the out-of-phase coupling of and , however, the pitch torque induced by FCF ( ) acts in the same direction as , thus magnifying the pitch angular velocity and causing the instability (see Discussion). Slow subsidence (mode 3) corresponds to a damped ascending/descending motion resulting from the FCF in a vertical direction.

TABLE 3 DIMENSIONAL AND NON-DIMENSIONAL EIGENVALUES

Longitudinal Lateral Mode 1 Mode 2 Mode 3 Mode 4 Mode 5 Mode 6

-39.15 14.17 ± 30.62i -5.11 -143.02 9.12 ±

34.85i -43.30

-0.18 0.067 ± 0.14 i -0.024 -0.67 0.043 ±

0.16i -0.20

Eigenvalues (dimension of Hz) are non-dimensionalized by dividing the flapping frequency . The inverse of Re( ) and Re( ) indicate the dimensional and non-dimensional time constants of the corresponding mode.

Similar to the longitudinal dynamics, the lateral dynamics consists of two subsidence modes with relative fast (mode 4) and slow (mode 6) convergences and an unstable oscillatory mode (mode 5). The fast subsidence mode corresponds to a highly damped yaw rotation (the larger value of means less wing-beat time to reduce the initial disturbance, Table 3) with an out-of-phase coupling of and . The slow subsidence mode has a similar out-of-phase coupling of and (with difference magnitudes), but with less damped yaw rotation. The unstable oscillatory has a near in-phase coupling of and

. Intuitively, for the out-of-phase coupling of initial and , roll FCT and the roll torque generated by FCF (

) act together to reduce the angular velocity , resulting in stable subsequent motions. For the in-phase coupling of and , however, the roll torque induced by FCF ( ) acts in the same direction as , thus magnifying the roll angular velocity and leading to an unstable subsequent motion.

DISCUSSION

Effects of unsteady aerodynamics on flapping counter-force/torque production

In the FCT models (Eqns. 1 to 5), we considered only the translational aerodynamic force resulting from the delayed stall and ignored other aerodynamic mechanisms [20]. Previous works show that most of those unsteady aerodynamic mechanisms alter the aerodynamic force transients near stroke

reversals. For example, the Kramer effect generates rotational forces before or after stroke reversals, depending on the phase of wing rotation, but wake capture occurs immediately after the stroke reversals [15].

As indicated by the plots of instantaneous torques about roll, pitch, and yaw axes (Figs. 4, 6, and 8), the peaks of roll and yaw torques occur near the middle stroke where their moments of arm are maximized. Pitch torque, however, reaches peak between the stroke reversal and the middle stroke. Therefore the unsteady aerodynamic effects, especially the wake capture, are most likely to affect pitch torque transients at the two quarter strokes after stroke reversals (ventral to middle and dorsal to middle strokes, Fig. 6). The roll and yaw torques, however, might be less affected by those unsteady effects. This might explain the discrepancy between the measured and the predicted pitch FCT (solid and dotted green, Fig. 6), especially at the two quarter strokes after the stroke reversal.

Flapping counter-torque and stability derivatives during free flight In the current study, we assumed a constant angular velocity during rotations around each principal axis of the stroke-plane frame and measured the corresponding FCT over one wing stroke. In free flight, however, the insect is expected to experience a time-varying angular velocity on non-principal axes. Previous simulation results suggest that a rotation about a non-principal axis would also yield an FCT that linearly depends on both stroke frequency and angular velocity. However, it will be very difficult to obtain mathematical models for such rotations.

Furthermore, how FCTs depend on the accelerations is still undetermined. Although the FCT and FCF models can be used to estimate the first derivatives with respect to time (e.g., ), we are unable to estimate the second derivatives (e.g., ). For fixed and rotary aircraft, such stability derivatives are important under some flight conditions (short-period mode, for example), in which the time constant associated is so short that the acceleration becomes critical to aerodynamics [17, 21]. However, it is still reasonable to assume that the first derivatives dominate the aerodynamics in most flight conditions [17, 21].

It should also be pointed out that by applying the small perturbation theory to approximate aerodynamic torques, we also neglected the nonlinear terms with higher orders, such as

, , , , and . The estimation of these terms can be directly obtained in the FCF or FCT model. For example, it can be shown that and are equal to ρ L α and ρ L α , respectively.

Aerodynamic damping during non-hovering flight By assuming the hovering condition in our study, we

consider that the effect of body velocity on flapping-wing aerodynamic model is negligible; therefore the previous quasi-steady aerodynamic model [15] is applicable. However, in non-hovering flight, i.e., forward flight with high advance ratio, the

7 Copyright © 2010 by ASME

Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm

Page 8: Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering,

aerodynamic model needs revision to give better estimations. Unfortunately, the understanding of the flapping-wing aerodynamics in non-hovering flight is now very limited. With respect to the forward flight, Dickson and Dickinson (2004) studied the effect of advance ratio on revolving wings and modified the quasi-steady aerodynamic model. This result can be directly applied to estimate the FCF in forward and backward translations based on Eqn. 6, with the revised lift and drag coefficients (by taking account of tip velocity ratio ). Similar to forward flight, vertical and lateral flights are very likely to have non-negligible effects on the aerodynamics. For example, in rotary wing aircraft a vertical descent changes the aerodynamics by reducing the downward momentum produced, and it greatly affects the lift force [17, 21]. A large descending velocity can even lead the flight into a so-called vortex-ring state, which is characterized by unstable flow conditions with erratic lift variations. Therefore an understanding of the aerodynamics, as well as the measurements of the force coefficients during non-hovering flight conditions, is desirable for further investigation of insect flight dynamics. However, as an approximation of the flight dynamics, our models provided an applicable analytical tool, which is important to designing the controllers for the flapping-wing MAVs, as well as to understanding the control strategies adapted by flying insects. For example, in the aircraft controller design, the uncertainties resulted from the neglected aerodynamics can be simply viewed as a disturbance term, which can be compensated by appropriate robust control strategies.

ANNEX A A mathematical model of FCT for yaw rotation has been

described in previous studies [1, 2]. In this appendix, we provide a derivation of FCT models for roll and pitch rotations. For simplification, wing deviation is not considered in the current study. First, with a standard blade-element model, the lift and drag of a single flapping wing at a particular instant in the wing stroke cycle [15, 22] are:

L , (A1) D , (A2)

where is the magnitude of the wing tip velocity at non-dimensional time , is the non-dimensional second moment of wing area, and are instantaneous lift and drag coefficients [15] as functions of effective angle of attack , is fluid/air density, is wing length, and is mean chord length. Roll, pitch, and yaw torques around the center of gravity are then calculated as:

cos , (A3)

sin , (A4)

, (A5)

where is stroke position and is the normalized center of pressure on the wing. We assume that the center of pressure is located at 70% along the wingspan at any instant of time [12].

Note that roll and pitch torques are determined only by lift force ( L), but yaw torque is determined only by drag force ( D). Each moment of arm is determined by the orthogonal distance between the center of pressure and the corresponding principal axis ( , , and ). Notably, it can be shown that equations A3-A5 describe the torque productions regardless of the location of center of gravity, as long as we assume that it is below the wing base and along the yaw axis .

Roll rotation Rotation around roll axis xs adds a downward velocity

(always normal to the stroke plane) to the left wing while increasing its effective angle of attack . On the contrary, however, it adds an upward velocity to the right wing and reduces its effect angle of attack. Therefore we write the lift of the left wing and the right wing as:

sin , (A6)

sin , (A7)

where is the geometric angle of attack determined by wing kinematics, is the angle between the stroke plane and total wing velocity , and and are magnitudes of wing tip velocity as a result of wing flapping and roll rotation, respectively:

and cos , (A8 and A9)

where is the non-dimensional angular velocity of the wing, and are wing-flapping amplitude and frequency, is

the roll angular velocity of the stroke plane, and is stroke position. Furthermore, is given by:

arctan . (A10)

The magnitude of wing velocity can then be written as:

. (A11)

Collectively, the roll torque of the left and right wings are calculated as:

sin cos

, (A12)

sin cos . (A13)

We then approximate as:

| . (A14)

The total roll torque of a wing pair is:

|sin cos Φ , (A15)

8 Copyright © 2010 by ASME

Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm

Page 9: Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering,

wT

Ms

P

pa((hoiiw

Ssw

T

wT

Ntt

a

l

whTh

Msy

Pi

pitatt(d(mhaof is inswr

Sistrwi

Th

whTh

Nethethe

a c

lef

hihe

Morym

itc

tctaor

midalff u

strit

mroin

hei

hehe

exe e

co

ft

iche a

remm

chS

h ckrsaddf ouploantte

milke

ng)

e as

ereere

xt psy

Wom

Fa

h av

eovme

h Simro

k aldleof st

ocantaen

lare ) i

angi

e efo

witcym

Wemp

Fuand

iser

veetr

romot

l te

f dtroatan

n a

rilp

is

ngliv

for

wech

mm

e ple

urtd

s thrag

er,ry

otmilatat

toto

dooktedneas:

ly,law

leven

re

e ah tme

|

thete

theri

hege

, fo

tatartiois

o o owke d eou:

, wanewri

bn b

w

aptoet

|

hene

erig

e Fed

itfor

tior tons mdo

wn(va

us

we,itt

beby

iwe

pporqri

s

n st

rmght

FCd v

t ir l

onto

n. Fe

midorstveat s

whe tten

twy:

s e h

proquc

in

hatro

mot

CTva

is lif

n o rFo

enhddrsatroen

li

enthen

we

thha

oxuest

avok

orew

T alu

eft a

roorhadleal okntr

thft

n te as

een

heave

xims tro

veke:

e, win

foue

eaan

ollr eane s

ke ralhe

o

thin

s

n

e pe t

maat

ok

e th:

itng

or o

asynd

l reitnc

sstr

(ml te of

he ns

th

ar

pith

att t

ke

he

|

t igs

r rov

y d d

rotheedtr

rokmto

df

wsta

he

rct

itche

te twp

e a

s

is h

roler

tdra

otaerd rokke

midm

dora

wian

e s

tan

chp

wopos

av

ın

eahav

ll r o

|

to ag

atir l

dkee),ddmidrss

si

ngta

s

str

n

h aitc

o isit

ve

n

asve

roon

sg p

ionef

due) , a

dleddsalsin

in

g ane

sin

ro

anch

intio

era

sy e

otne

sepr

n,ft

urina

ane tdlel ng

iseo

n

oke

ngh t

nston

ag

s

tosy

tatw

ın

e rod

, wornganndtoe h

gle

s lou

e

gulto

si

sin

tanns

s

ged

ın

o ym

tiowin

tdu

wr rg ndd i vst

hae

lous

pl

larq

in

n

ntw

sin

d p

n

semm

onng

thauc

.

we rigt

d tit vetroalf

w

ocali

lan

ar qu

ts wit

n

pi

eem

n ag s

co

at cti

nghhethis

entokfwin

ateift

ne

vues

oth

itc

e tet

at str

os

bon

noht e

he s rtra

ke)ong

edft

e a

vels:

, f

h r

ch

thatri

a ro

s

ben,

owwfs

real ). f g

d of

an

lo

ati

res

h t

atic

nok

ecw

w winfirseedu

sTt

(i

atf

nd

,

ci

as msp

tor

Φ

t dw

nonke

cauwe

dengst

ecouc

strThethi.e

t ta

d t

,

ity

inme,pec

Φ

rq

Φ

duwi

n-is

use h

erg tt onceroker

he e.,

thsi

the

y

n , wct

Φ

que

uring

-ds:

se ha

rivthha

nded keref

,

e in

e

of

Ewt to

e

.

ingg

dim

Φ

oave

ve e al

d d

e) fostle

vngl

to

f

Eqnwhi

o

.

of

.

g m

me

Φ

ofe

teff ha

dua

oretroeft

venle

ota

th

si

sin

n.ichth

.

f a

pmo

en

f

theffeo

aluriande wokt

.

ntre w

al

he

in

n

(A

. Ah

he

a w

itcoti

nsi

.

th

e ecof f ngd wh

ke w

.

rawi

.

w

s

A

Ac m

w

chion

io

.

he

Fcti

do

g thhe

win

al in

win

str

A2

14or

mi

win

h rns

ona

g

FCivedoof th

heenpl

ng

hng

ng

ro

1

4, rreid-

ng

ros;

al

ge

CTe ow

uhee fn tla

g)

hal(

g

oke

an

aes-s

g p

otath

ti

(A

eo

(A

T an

wnupe sfirtheane

c

(

lf (i.e

ve

e

nd

ansptr

pai

atihe

im

A

om

A

dungnstpstserste we, can

(A

oe.

(A

el

(A

p

d A

nd onok

(A

ir

(A

ionere

me

16

me

17

urgletrotro

ecot hw

n

A1

of ,

A

oc

A2

pla

A2

sndke

A

o

A2

n ef

e

6)

tr

7)

rine ookokonha

winthb

18

thle

19

cit

20

an

.

22

sumd te:

23

ove

24

thfor

.

)

ric

ngofkekendalfnghebe

8)

heeft

9)

ty

0)

ne.

,

2)

mto

3)

er

4)

here

.

c

g f e e d f g e e

e t

y

.

,

m o

)

r

e e

se

p(fth

Flavthtr(n

F

baa

wtr

Fnv

Atwp

Bp

Mfo

ymexp

profohe

Figate

velohe ranne

Fo

be appa w

whra

Fronegver

Anw

po

Bypai

Mofor

mpe

ovorwe s

g. eraocb

nslga

orw

dpl

wi

heans

omgars

ndo si

ρ

y air

orrw

mmec

Svidwastr

Aal citybodlatativ

wA

dirlyiing

resla

m atia

d ti

tio

ρ

avis

reowa

metcte

Simdearro

1. (By (dy tiove

waA dreing

e at

Fivfo

theinon

vers:

ovard

tried

mie ard/oke

SB), (so

tron,e si

rdde

ectng ca

ρ

tio

Figve or

enstn:

ra

ved/b

icd:

ilaa d/be-

Schan

oliran aign

d/eritlyb

an

ρSC

on

g.dfo

n wtan

co

agi

er, ba

c to

ar de

bac-pl

hemndid nsas ns

/bivy blan b

CD

n, w

Adurfor

ρ

wnt

os

in

wac

or

toerckla

mad vfolatin

) o

acvat

obadbe

D α

iswh

Arinrw

SC

ρ

e s

Φ

ng

wekw

rq

o iv

kwane

ativeror tiondiof

cktiobt

de e w

α

s hi

1Ang

wa

CD

ρSC

suo

Φ C

m

e cwa

que

Fvatwae f

ic rticup

onaicafo

kwontaiel

wr

thich

A,g dard

α

CD

umof

CD

mo

caar

e g

Ctio

ardfr

vcapstal ate

orc

wan oinlerit

heh

, wdod m

α

D α

m f t

D α

ore

an rd

ge

T ond)am

iewal ttrove

ed ce

arof

nedm

tte

e is

wowm

α

thtim

α

e t

Φ

eam

en

mn o, pme

w traokeelo

bpr

rdf Fd,

menen

ts d

we wnot

hem

c

th

Φ

asmo

ner

maofpie.

oanse, ocby rod

d tFC

bnta

tipde

cnstio

e dme

cos

han

siloti

ra

athf Fitc

of sladaityth

duc

traCFbat tas:

p fin

cantron

drc

s

n

α

ly io

ati

heFCch

FCatioashy. hecti

anF asethe

vne

n ok

n).

ragco

on

α

ons

ρ

on

AemCFh (

CFonheFC eion

nsinedeo

veed

ske. T

g orr

c

ne

c

obts:

ns

ANmaF (la

F pns. ed CFenhn.

slndd ory

lod a

seee aTh

(ore

co

e w

cos

ta

CL

s a

NNatifoate

prRfo

F (ha

atducony,

ocias

e anher

of sp

s

wi

s

ain

L α

aro

NEicaor er

odRedor (b

anc

tiocen th

2

ity:

thndre

f apo

in

n t

α

ou

0.

EXal b

ral

ducd adoluecem

oned p

he

y

.

had pefo

2

a son

ng

the

c

un

.

X m

bodl),

ctiarrowe me

n d bpree i

r

at poor

2

sindi

.

st

e

cos

nd

Bmody, a

ionrow

wnsarent

byevin

ra

inosire,

2

ngin

tro

ch

s

d th

B ody an

n wsstrrrot

y fviost

atio

n iti,

gleg

ok

ha

he

deltr

nd

dus drokow(p

fooutan

o

bive

|

e wt

ke

an

e r

lsran

y

uridenke)s) os

rwus nta

i

bace

|

wto

e, t

.

ge

ro

, insya

ingno), a

siti

war

an

in

ckdu

|

win

th

e o

oll

inslaaw

g foteanctsive

ardesne

,

n

kwur

|

ngan

he

of

.

a

n thati

w (

fore dnd s oe

d/sueou

,

fo

warin

g) n

n

f l

and

hiion(v

rwdire

gopsig

/baultus

or

ardng

ali

net

lif

d

is n

ver

warecre

ppogn

acs s d

rw

d g u

,

lode

t d

ft r

pi

Aal

rti

rd/tioenosi

ns)

ckfr

dr

war

mup

.

ngen

dr

re

itc

Aploica

/baonn aite a

k mrorag

rd

mopst

g nti

ag

esu

ch

ppongal

acs o

arre tand

moomg

d/b

otitro

ica

g o

ult

h a

(

eng l)

kwof rowto d

otm

ac

ba

onok

al

of

tin

ax

(A

ndthax

waf thwsthre

tio[7cti

ack

n ke

a

f a

ng

xes

A2

dixhexe

ardhe s dhe edu

on7]in

(

kw

(

e (

(B

(

axst

(

a w

(

g f

(

s a

25)

x e res

d (w

denb

uct

n c.

ng

(B

wa

(B

(v

B3

(B

xistro

(B

wi

B

fro

(B

ar

)

wro

o

(Awinnotodtio

caBo

B1

ar

B2

iic

3)

B4

s aok

B5

in

6)

om

B7

re

we ll

of

A), ng te

dy on

an y

on

1)

rd

2)

is ce

)

4)

at ke

)

ng

)

m

7)

f

9 Copyright © 2010 by ASME

Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm

Page 10: Mathematical Modelling of Near-Hover Insect Flight Dynamicsxdeng/DSCC.pdf · MATHEMATICAL MODELLING OF NEAR-HOVER INSECT FLIGHT DYNAMICS B. Cheng X. Deng School of Mechanical Engineering,

Note that the above equations also apply to the forward translation.

Lateral translation Similar to forward translation, the instantaneous drag

acting on a wing during lateral motion can be written as:

ρSCD α 2 , (B8)

where is the tip velocity ratio in lateral translation, which is defined as:

1 , (B9)

where = 0 and 1 for the left and right wings, respectively. We can see in Fig. A1B that during a lateral motion to the right (left), the net drag of a wing pair along the pitch axis is always pointing to the left (right) at any instant of time (i.e., an enhanced drag to the left (positive ) and decreased drag to the right (negative ) during downstroke at ventral half of the stroke plane, Fig. A1B). Collectively, we have

∆ Φ α sin , (B10)

and the stroke averaged value is:

∆ Φ α sın . (B11)

Moreover, the change of lift because of lateral motions is:

∆ ρ CL α sın . (B12)

Vertical translation A vertical translation changes the effective angle of attack

at a wing section by:

arctan , (B13)

where is the angle between total wing velocity, and the stroke plane is the spanwise location of the wing section. The magnitude of total wing velocity is:

. (B14)

Thus the sectional lift force during ascent is given by:

d sin

. (B15)

Furthermore, the change of the sectional lift force resulting from vertical velocity is:

Δd sin ,

sin |

. (B16)

Integrate the above equation over an entire wingspan, and we have the net lift change of a wing pair:

Δ Φ

|sin . (B17)

The stroke averaged value is:

Δ Φ

|sın . (B18)

ACKNOWLEDGMENTS This work was supported in part by NSF Grant#0545931.

REFERENCES [1] B. Cheng, S. Fry, Q. Huang, and X. Deng, "Aerodynamic damping during

rapid flight maneuvers in the fruit fly Drosophila " Journal of Experimental Biology, vol. 213, pp. 602-612, 2009.

[2] T. L. Hedrick, B. Cheng, and X. Deng, "Wingbeat Time and the Scaling of Passive Rotational Damping in Flapping Flight," Science, vol. 324, pp. 252-255, April 10, 2009.

[3] S. N. Fry, R. Sayaman, and M. H. Dickinson, "The aerodynamics of free-flight maneuvers in Drosophila," Science, vol. 300, pp. 495-498, Apr 2003.

[4] C. P. Ellington, "The Aerodynamics of Hovering Insect Flight .6. Lift and Power Requirements," Philosophical Transactions of the Royal Society of London Series B-Biological Sciences, vol. 305, pp. 145-181, 1984.

[5] A. R. Ennos, "The Kinematics and Aerodynamics of the Free Flight of Some Diptera," Journal of Experimental Biology, vol. 142, pp. 49-85, Mar 1989.

[6] C. T. David, "The relationship between body angle and flight speed in free-flying Drosophila," Physiol. Entomol, vol. 3, pp. 191-195, 1978.

[7] W. B. Dickson and M. H. Dickinson, "The effect of advance ratio on the aerodynamics of revolving wings," Journal of Experiment Biology, vol. 207, pp. 4269-4281, November 15 2004.

[8] H. Buelthoff, T. Poggio, and C. Wehrhahn, "3-D Analysis of the Flight Trajectories of Flies (Drosophila rnelanogaster)," Z. Naturforsch, vol. 35c, pp. 811-815, 1980.

[9] M. Sun and J. K. Wang, "Flight stabilization control of a hovering model insect," Journal of Experimental Biology, vol. 210, pp. 2714-2722, Aug 2007.

[10] G. K. Taylor and A. L. R. Thomas, "Dynamic flight stability in the desert locust Schistocerca gregaria," Journal of Experimental Biology, vol. 206, pp. 2803-2829, Aug 2003.

[11] X. Y. Deng, L. Schenato, W. C. Wu, and S. S. Sastry, "Flapping flight for biomimetic robotic insects: Part I - System modeling," IEEE Transactions on Robotics, vol. 22, pp. 776-788, Aug 2006.

[12] S. N. Fry, R. Sayaman, and M. H. Dickinson, "The aerodynamics of hovering flight in Drosophila," Journal of Experimental Biology, vol. 208, pp. 2303-2318, Jun 2005.

[13] J. H. Marden, "Maximum Lift Production During Takeoff in Flying Animals," Journal of Experiment Biology, vol. 130, pp. 235-258, July 1, 1987.

[14] R. Dudley, The biomechanics of insect flight: Princeton University Press, 2000.

[15] M. H. Dickinson, F. O. Lehmann, and S. P. Sane, "Wing rotation and the aerodynamic basis of insect flight," Science, vol. 284, pp. 1954-1960, Jun 1999.

[16] S. P. Sane and M. H. Dickinson, "The aerodynamic effects of wing rotation and a revised quasi-steady model of flapping flight," Journal of Experimental Biology, vol. 205, pp. 1087-1096, Apr 2002.

[17] B. Etkin and L. D. Reid, Dynamics of Flight: Stability and Control. New York: Wiley, 1996.

[18] R. M. Murray, Z. Li, and S. S. Sastry, A Mathematical Introduction to Robotic Manipulation: CRC, 1994.

[19] M. Sun and Y. Xiong, "Dynamic flight stability of a hovering bumblebee," Journal of Experimental Biology, vol. 208, pp. 447-459, Feb 2005.

[20] S. P. Sane, "The aerodynamics of insect flight," Journal of Experimental Biology, vol. 206, pp. 4191-4208, December 1 2003.

[21] R. W. Prouty, Helicopter Performance, Stability, and Control. Boston: PWS Engineering, 1986.

[22] S. P. Sane and M. H. Dickinson, "The control of flight force by a flapping wing: lift and drag production," Journal of Experimental Biology, vol. 204, pp. 2607-2626, 2001.

10 Copyright © 2010 by ASME

Downloaded 29 Sep 2012 to 128.46.184.175. Redistribution subject to ASME license or copyright; see http://www.asme.org/terms/Terms_Use.cfm