0% found this document useful (0 votes)
49 views8 pages

Four-Bar Linkage Mechanism For Insectlike Flapping Wings in Hover: Concept and An Outline of Its Realization

Uploaded by

Hosein Pahlavan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
49 views8 pages

Four-Bar Linkage Mechanism For Insectlike Flapping Wings in Hover: Concept and An Outline of Its Realization

Uploaded by

Hosein Pahlavan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

Four-Bar Linkage Mechanism for

Insectlike Flapping Wings in


Hover: Concept and an Outline of
Rafał Żbikowski
Department of Aerospace, Power, and Sensors, Its Realization
Cranfield University (RMCS Shrivenham),
Swindon SN6 8LA, England This paper describes the concept of a four-bar linkage mechanism for flapping wing micro
air vehicles and outlines its design, implementation, and testing. Micro air vehicles
(MAVs) are defined as flying vehicles ca. 150 mm in size (handheld), weighing 50–100 g,
Cezary Galiński and are developed to reconnoiter in confined spaces (inside buildings, tunnels, etc.). For
OBRUM, Ul. Toszecka 102, 44-101 Gliwice, this application, insectlike flapping wings are an attractive solution and, hence, the need
Poland to realize the functionality of insect flight by engineering means. Insects fly by oscillating
(plunging) and rotating (pitching) their wings through large angles, while sweeping them
forward and backward. During this motion, the wing tip approximately traces a figure
Christopher B. Pedersen eight and the wing changes the angle of attack (pitching) significantly. The aim of the
Department of Aerospace, Power, and Sensors, work described here was to design and build an insectlike flapping mechanism on a 150
Cranfield University (RMCS Shrivenham), mm scale. The main purpose was not only to construct a test bed for aeromechanical
Swindon SN6 8LA, England research on hover in this mode of flight, but also to provide a precursor design for a future
flapping-wing MAV. The mechanical realization was to be based on a four-bar linkage
combined with a spatial articulation. Two instances of idealized figure eights were con-
sidered: (i) Bernoulli’s lemniscate and (ii) Watt’s sextic. The former was found theoreti-
cally attractive, but impractical, while the latter was both theoretically and practically
feasible. This led to a combination of Watt’s straight-line mechanism with a drive train
utilizing a Geneva wheel and a spatial articulation. The actual design, implementation,
and testing of this concept are briefly described at the end of the paper.
关DOI: 10.1115/1.1829091兴

1 Introduction Insects fly by oscillating 共plunging兲 and rotating 共pitching兲 their


This paper describes the concept of a mechanism for micro air wings through large angles, while sweeping them forward and
vehicles realizing insectlike flapping based on a four-bar linkage backward. The wingbeat cycle 共typical frequency range: 5–200
and outlines its design, implementation, and testing. This work is Hz兲 can be divided into two phases: downstroke and upstroke 共see
part of a larger research effort at Cranfield University 共RMCS Fig. 1兲. At the beginning of downstroke, the wing 共as seen from
Shrivenham兲 aimed at developing a flapping-wing micro air the front of the insect兲 is in the uppermost and rearmost position
vehicle 关1,2兴. with the leading-edge pointing forward. The wing is then pushed
The paper is organized as follows. Section 1.1 defines and pro- downward 共plunged兲 and forward 共swept兲 and rotated 共pitched兲
vides the motivation for the problem and gives the necessary in- continuously with considerable change of the angle of attack. At
formation about micro air vehicles and insect flight. The design the end of the downstroke, the wing is twisted rapidly, so that the
specifications for the mechanism are listed, with annotations, in leading-edge points backward, and the upstroke begins. During
Sec. 1.2. The mechanism concept based on the classical Watt’s the upstroke the wing is pushed upward and backward and rotated
straight-line mechanism is presented in Sec. 2 and shown to again, which changes the angle of attack throughout this motion.
be a feasible solution. Our design according to that solution is At the highest point, the wing is twisted again, so that the leading-
summarized in Sec. 3, while Sec. 4 gives a brief account of the edge points forward and the next downstroke begins.
mechanism implementation and testing. The conclusions follow in Insect wing flapping occurs in a stroke plane that generally
Sec. 5.
remains at the same orientation to the body and is either horizon-
1.1 Motivation and Background. Micro air vehicles tal or inclined 共see Fig. 1兲. In forward flight, the downstroke lasts
共MAVs兲 are defined as flying vehicles ca. 150 mm in size 共hand- longer than the upstroke because of the need to generate thrust. In
held兲, weighing 50–100 g and are being developed to reconnoiter hover they are equal, resulting in the wing tip approximately trac-
in confined spaces 共inside buildings, tunnels, etc.兲. This requires ing a figure eight 共as seen from the insect’s side兲. The work pre-
power-efficient, highly manoeuvrable, low-speed flight with stable sented here focuses on hover.
hover. Such performance is routinely exhibited by flying insects Since each half cycle starts from rest and comes to a stop, the
and, hence, the focus on emulating insectlike flapping by engi- velocity distribution of the flapping is nonuniform, making the
neering means. A detailed discussion of the future utility of MAVs resulting airflow complex. It is also unsteady, i.e., the aerody-
and the advantages of considering insectlike flapping-wing pro- namic force varies in amplitude and direction during each wing-
pulsion have been presented elsewhere 关3,4兴. Here we summarize
beat cycle. The variability of the force is compounded by the
the relevant basics of insect flight, emphasising flapping-wing ki-
strong influence of the viscosity of air 共due to the small scale兲 and
nematics and touching on the complex aerodynamics involved.
significant interaction of the wing with its wake 共due to hover兲.
The details are explained elsewhere 关5兴; it is only noted here that
Contributed by the Mechanisms and Robotics Committee for publication in the
JOURNAL OF MECHANICAL DESIGN. Manuscript received September 8, 2003; re- the expected aerodynamic loading will be time varying, with mag-
vised May 26, 2004. Associate Editor: G. R. Pennock. nitude peaks at the end of each half cycle.

Journal of Mechanical Design Copyright © 2005 by ASME JULY 2005, Vol. 127 Õ 817

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


Fig. 1 Typical motions of insect wing in hover. Insect body is orientated almost vertically, while
the wing tip traces a flat figure-eight around the stroke plane. The stroke plane is inclined by the
angle ␤, typically ␤É15 deg.

1.2 Design Specifications. Not only are insectlike aerody- • Biomimetic extraction. The mechanism should realize ideal-
namics quite complex, but their observation and measurement in ized insectlike flapping, implemented by engineering means.
nature is very challenging 关6 –9兴. This motivated the development • Test bed for aerodynamic and mechanical aspects of flapping.
of aerodynamically scaled flapping mechanisms, most notably El- The mechanism’s main function is to foster research on aero-
lington’s flapper 关10,11兴 and Dickinson’s Robofly 关12兴. These de- mechanical aspects of insectlike flapping in hover, but it
vices allowed remarkable progress in gathering experiment data should also provide a precursor design for the final platform.
on insectlike aerodynamics 关13,14兴. However, they tend to be • Durability versus weight. The need for a robust mechanism
bulky constructions not convenient for developing into light- for prolonged testing takes precedence over weight optimiza-
weight, 150 mm versions suitable for a future flapping-wing MAV. tion.
Also, while they emulate the correct Reynolds number, their • Planar mechanism with an articulation adding the third di-
Strouhal number is different, which is a disadvantage in the in- mension to the motion. This decouples the realization of the
vestigation of unsteady aerodynamics. wing tip envelope from the rest of the wing motion.
Ellington’s flapper is an electromechanical model of Manduca • Four-bar linkage realization. Start with the simplest planar
sexta hawkmoth, an insect with a wingspan of 10 cm. The model mechanism.
has a 1 m wingspan, has ten times the length and breadth of the • Stroke plane orthogonal to the fuselage. The wings should
moth, and operates at 1/100 of the original insect frequency of 26 flap in a plane perpendicular to the long axis of the fuselage
Hz. A gearbox and a system of bevel gears, driven by several for ease of testing and handling 共see Fig. 1兲.
servomotors, was used. Dickinson’s Robofly is an application of • Rotary DC motor propulsion. For ease of use and power ef-
Ellington’s approach to aerodynamic scaling of the fruit fly ficiency.
Drosophila melanogaster, an insect whose typical size is 2.5 mm. • Low cost. Where possible, the use of off-the-shelf parts
Robofly has a 60 cm wingspan, typically flaps five times per sec- should be considered, drawing on the clock- and watchmak-
ond, and is immersed in two tons of mineral oil. Six computer- ing and hobby industries.
controlled motors specify the three rotational angles of each wing.
The wings are equipped with sensors for measuring instantaneous 2 Idealized Wing Tip Kinematics: Watt’s Sextic
aerodynamic forces.
On the other end of the scale, Fearing 关15,16兴 aims at building As outlined in Sec. 1.1, the trace of the tip of an insect’s wing
an insectlike flapping robot weighing 1 g and having 25 mm wing seems to be a figure eight, when projected onto the plane of the
span. This approach is based on MEMS technology, as the ex- animal’s longitudinal symmetry. This is only approximately true,
pected forces 共and payloads兲 are below 1 g. as the available data for Diptera flies 关17兴 show that the trace
In contrast to these efforts, the flapping mechanism described tends to be more irregular. The kinematic data from biology are
here was designed to the following specifications: uncertain due to experimental difficulties and a great variety of
insect flapping styles. This variety has aeromechanical conse-
• Miniature mechanism, no MEMS. The final vehicle is to be quences, which are not fully understood, so that it is not possible
built on the 150 mm scale with 50–100 gram weight, so the at present to determine the ‘‘best’’ motion envelope. Thus, there
power required 共a few watts兲 necessitates conventional me- exist no reliable reference kinematics that could be used as input
chanical engineering. to a standard dimensional analysis. Instead, an idealized figure
• Fuselage 150 mm long and 25 mm in diameter. The mecha- eight was adopted as a starting point in order to allow a systematic
nism must fit into a 150 mm⫻25 mm cylinder, the MAV and controlled study of the influence of the kinematics on aerody-
dimensions compatible with the size of the human hand. namic forces. For simplicity, the resulting kinematics had to be
• Design wingbeat frequency of 20 Hz. This flapping frequency implementable with a four-bar linkage. This naturally led to con-
should generate the necessary lift while ensuring minimal sidering the linkage configurations realizing: 共i兲 Bernoulli’s lem-
aeroacoustic signature. niscate and 共ii兲 Watt’s straight-line mechanism. A spherical four-
• Fixed motion. The mechanism should implement predefined bar linkage was also considered because the wing motion is
共as opposed to adjustable兲 kinematics of hover, so that the essentially spherical. This kind of design should be theoretically
motion of each wing should be the mirror image of the other. possible 关18兴. However, there was a concern about the manufac-

818 Õ Vol. 127, JULY 2005 Transactions of the ASME

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


共see Fig. 2兲. This mathematical singularity can lead to surprisingly
diverse coupler curves 关24 –26兴, but can be avoided in implemen-
tation with the use of elastic elements 共see Fig. 3兲, as explained
below.
An ideal mechanism would experience infinite forces and ac-
celerations in the locked position 关Fig. 3共a兲兴, a kinematic conse-
quence of the mathematical singularity. A physical mechanism
with realistic linkage assembly tolerances has some freedom of
motion in the neighborhood of the locked position 关Fig. 3共b兲兴.
Thus, without friction, the inertia of the driven rocker and the
coupler would be large enough to overcome the transition prob-
lem. If friction is present, two springs attached to the end of the
driven rocker are necessary to run the mechanism properly 关Fig.
3共c兲兴. Unbalanced tension of the springs provides the force driving
the rocker through the locked position. The mechanism may still
suffer from the transition problem if the crank of the driving link-
age is too long or if distance CD between axes of the crank and
the driving rocker is incorrect 关Fig. 3共d兲兴. To prevent this phenom-
enon, an elastic coupler 关27兴 was introduced between the driving
linkage and the drive train 关Fig. 3共e兲兴. These concepts were tested
in a workbench demonstrator 共see Fig. 4兲, and experiments did not
reveal any excessive forces in the resulting mechanism. This
mechanism can also run in the resonance mode if the characteris-
tics of the springs and mass distribution are appropriately chosen.
The insect thorax-wings system indeed implements mechanical
resonance; the elastic element is the so-called resilin, which lines
the interior of the thorax. The use of resonance is a power-efficient
way of generating a periodic, reciprocal motion.

3 Summary of Mechanism Design


Section 2 focused on achieving the required wing-tip kinemat-
ics while ignoring the wing orientation. The orientation 共pitch兲
determines the effective angle of attack, a critical parameter in
establishing lift and drag. Thus, Watt’s mechanism of Section 2
has to be endowed with an articulation, which will not only trans-
mit the figure eight motion to the wing, but also rotate it according
Fig. 2 Transitional character of Watt’s straight-line mecha-
nism and the related families of Watt’s sextics for varying to the required pitch program.
lengths of the links It follows from the description of the insectlike kinematics in
Sec. 1.1 that there are two phases in each half cycle: translational
共wing moving forward or backward兲 and rotational 共at the end of
each stroke兲. In order to investigate the distinct aerodynamic con-
turing tolerances necessary to create such a mechanism on the tributions of each phase, the angle of attack should be constant
required scale. Finally, it was decided that the cost of achieving during translation and rotate through 180 deg during the flipover.
the required tolerances was not justifiable at this point of research, Thus, the theoretically ideal pitch program requires an intermittent
but a spherical four-bar linkage could be considered again in the rotational motion with reversal, a variation of dwell motion. Fi-
future. nally, the planar action of the linkage 共plus rotation兲 has to be
The crank-rocker, non-Grashof, four-bar linkage with propor- transformed into a spherical movement of the wing.
tions 1:&:1 has Bernoulli’s lemniscate as its coupler curve 共关19兴 Hence, the complete mechanism consists of three main sys-
p. 79; 关20兴, Chap. 11兲. However, its implementation within the tems: 共i兲 a four-bar linkage figure eight generator, 共ii兲 two wing
geometric constraints of a 150 mm⫻25 mm cylinder 共fuselage兲 is articulations for adding the third dimension to the motion 共Fig. 5兲,
very difficult. Such difficulties are absent for the classical straight- and 共iii兲 a Geneva wheel for rapid pitch reversal 共Fig. 6兲.
line mechanism of Watt 共e.g., 关19兴, p. 79; 关20兴, Chap. 3兲. This The dimensions of the four-bar linkage in 共i兲 were determined
double-rocker four-bar linkage produces families of curves of the under two main constraints: biomimetic fidelity of the figure-of-
figure eight type according to the link proportions chosen 共see Fig. eight shape and fuselage size 共a 150⫻25 mm cylinder兲. A geomet-
2兲. Also, a horizontally orientated linkage generates the coupler ric analysis of Watt’s sextics 共Fig. 2兲 allowed the meeting of con-
curve orthogonally to its base, so it can be housed in a cylindrical straint 1 by a judicious choice of link proportions, resulting in a
fuselage and drive the wings in a vertical plane, as required. Fur- realistic figure eight. It was also required that the sweep angle
thermore, the links will not make excursions beyond the housing change at least from ⫺45 to ⫹45 deg, as this is typical for insects.
due to the double-rocker configuration. Due to the conflicting demands of biological fidelity and engi-
While figure eight lemniscates are quadrics, the straight-line neering practicality, the design process was performed in an itera-
mechanism curves are sextics and do not admit particularly con- tive loop. There were two main phases of this process. In the first
venient parameterizations 关19,21兴. Still, a closed-form expression phase the geometrical aspects were considered within the dimen-
is available 关22兴, and a formula for the corresponding spherical sion constraints of the assumed fuselage. In the second phase, the
curve can be derived. A more important issue is that the linkage kinematics, dynamics, and stresses for the whole mechanism from
thus configured is transitional 共关23兴, Chap. 7兲. This means that it Phase 1 were calculated. The cycle Phase 1→Phase 2→Phase
has an uncertainty configuration when transitioning between a 1 . . . was continued until the geometric and strength requirements
Grashof and non-Grashof kinematic chain. In other words, when were satisfied.
the coupler is momentarily collinear with another link, it can ei- The coupler link was realized as a tube with two ‘‘ears’’ 共see
ther continue its previous motion or suddenly reverse the direction Fig. 5兲, so that the ball-tipped wing axles could slide in and out of

Journal of Mechanical Design JULY 2005, Vol. 127 Õ 819

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


Fig. 3 Forcing the desired transition of Watt’s mechanism during the fli-
pover „see Fig. 2…: elastic elements replace the singular „infinite… forces with
finite ones. Notation: F1 first spring tension, F2 second spring tension, Fd
driving force, Md driving moment, ␻ driven rocker rotation velocity and r
driving crank radius.

the tube 共see Section A-A in Fig. 5兲. In this way, the planar
figure-eight curve generated at the center of the coupler link was
transformed into a spherical figure-eight at the wing tips. Hence,
the dimensions of the coupler were established first by choosing
the tube’s size to provide ⫾45 deg sweep angle for the assumed
axle dimensions. Then two ears were added symmetrically to the
central tube 关see Fig. 5共a兲兴. Their lengths were minimized within
practical limits, so that the remaining links could also have small
absolute dimensions. This was done iteratively, as the choice of
the length of the coupler and the shape of the figure-eight deter-
mined the sizes of the rocker and the base. Moreover, the dimen-
sions of these two links were limited by the fuselage, so the rocker
and the base sizes had to be revisited and the process repeated
until a satisfactory solution was found. The results of this geo-
metrical analysis were then the starting point for a dynamical/ Fig. 4 A workbench demonstrator of the transition-free Watt’s
stress analysis. In this analysis, velocities, accelerations, and mechanism, conceptualised in Fig. 3; note the springs and the
forces were calculated for a given linkage configuration and the elastic coupling with the drive train

820 Õ Vol. 127, JULY 2005 Transactions of the ASME

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


Fig. 6 Flapping mechanism based on the four-bar linkage
figure-eight generator: the drive train with Watt’s linkage and a
Geneva wheel for rapid pitch reversal, and the kinematic
Fig. 5 Flapping mechanism based on the four-bar linkage scheme
figure-eight generator. Section A-A shows the heart of the
mechanism: note the wing axles „tipped with balls… moving in a
common tube at the center of the coupler: „a… implementation
of the coupler link, „b… wing articulation: general view; and „c… wheel is then transferred to the crank and rocker of the four-bar
wing articulation: cable action. linkage mechanism in order for the crank to perform a 180 deg
rotation when the Geneva wheel moves. The crank pulls the cou-
pler and the rocker forward or pushes backward when rotating.
The rocker is connected to the axle and two double equal arm
design frequency. Stresses were calculated and analyzed for all levers, to which pitch-controlling cables are also attached. The
parts. If the stresses were excessive, the parts’ dimensions were assembly consisting of the rocker, the axle, and both levers is
modified, and the process was started again. called the rocker assembly in Fig. 6.
The wing articulation 共Figs. 5共b兲 and 5共c兲兲 consists of an axle, a
rocker, and a yoke. The yoke and the rocker enable two rotations
about orthogonal axes driven by the figure-eight generator. The 4 Implementation and Testing
axle is tipped with a ball that is connected to the four-bar linkage Despite recent progress 关9,13,14兴 in experimental elucidation of
coupler center in order to transfer the figure-eight motion to the insect flight, its theoretical underpinnings are not very well under-
wing. The wing is attached to the opposite side of the axle. There stood. However, the existing empirical evidence enables system-
is a cable wound around the axle center. Variable tensions of both atic aerodynamic modeling of the complex flow involved 关5兴. A
ends of the cable enable pitch control. Cable ends pass along the preliminary version of such an aerodynamic model was used in
yoke axis and are attached to the rockers of the Geneva wheel the lift-force prediction for the current mechanism design with the
mechanism controlling the pitch reversal schedule. This design assumption that the wing is rigid.
enables pitch to be decoupled from other motions, so that it can be Inspiration for this simplified approach was drawn from the
precisely controlled. modeling of other unsteady aerodynamic systems, such as heli-
The required pitch program consisted of long periods with con- copter blades 关28兴 and the vortical lift of delta wings 关29兴. Unlike
stant angle of attack, punctuated by rapid pitch reversals at both helicopters 共or delta wings兲, insect wings operate in a laminar
ends of the wing beat. This type of kinematics is best put into 共turbulence-free兲 and low Reynolds number 共significantly viscous兲
practice by dwell mechanisms of which the Geneva wheel mecha- regime, so that turbulence and air compressibility can be ignored.
nism is a natural choice. The output of the Geneva wheel is zero This simplified the analysis considerably. However, flapping
rotation for most configurations of the driving wheel and rapid wings do generate strong shed vortices, have varying speed
rotation for remaining configurations. This output has to be con- through a wingbeat, and varying angle of attack, all of which
verted into linear reciprocal motion to drive the cables in the complicate the analysis.
articulation. Therefore, the pitch reversal generator consists of a An analytical, modular model was developed to account for the
Geneva wheel mechanism, gears, and a crank-and-rocker mecha- main elements of this unsteady flow, i.e., 共i兲 quasisteady kinematic
nism 共see Fig. 6兲. The crank of the figure-eight generator drives effects 共‘‘frozen’’ at each time兲, 共ii兲 vortical lift 共as on delta wings
the Geneva wheel, resulting in the angular velocities of the crank using the Polhamus model 关29兴兲, 共iii兲 unsteady kinematic effects
and the Geneva wheel driver being equal. The Geneva wheel 共modeled using the Wagner function 关30兴兲, 共iv兲 shed wake effects
driver has two arms, so that two rotations of the driven wheel are 共modeled using the Küssner function and corrections to the Wag-
generated during one cycle of the driver. Output from the Geneva ner effect 关30兴兲, and 共v兲 added mass effects 共noncirculatory lift

Journal of Mechanical Design JULY 2005, Vol. 127 Õ 821

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


due to acceleration of the surrounding air by the wings 关31兴兲.
Given the required wing kinematics and geometry, all lift compo-
nents are calculated and summed up to give the total lift. The
results are shown in Fig. 7.
The aerodynamic loading analysis described above allowed
stress analysis in the mechanism. However, the actual loading
calculation, performed with Dynamic Designer Motion 共a part of
the COSMOS WORKS software兲, was not straightforward: the transi-
tional character of the figure-eight linkage results in a mathemati-
cal model with two singularities. In simulation, this caused exces-
sive accelerations and inertial loads in the neighborhood of locked
positions, and the simulation was aborted. An appropriate integra-
tion step length had to be selected, so that the beginning and end
of the step were sufficiently distant from the singularities. This
approach allowed the simulation to run smoothly, but left unre-
solved the real value of loadings in the neighborhood of the
locked positions. As a remedy, a simulation study was undertaken
to elucidate relations between the main contributing factors and
the loads to minimize the stresses.
Due to uncertainties concerning aerodynamic and inertial load-
ings, it was not possible to calculate the stresses precisely. How-
ever, an FEM parametric study did allow to estimate the mecha-
nism safety and to optimize component shapes, so that stress
concentrations were avoided.
The assembled mechanism is shown in Fig. 8. The test program
began as soon as the necessary parts were manufactured and the
mechanism assembled. The first set of tests focused on selection
of suitable cables in the spatial articulations, which would assure
reliable pitch reversal. Several specimens made of steel and nylon
were tested. The steel wires were too rigid and did not have suf-
ficient fatigue strength, and therefore a nylon line with the diam-
eter of 0.4 mm was selected.
The second batch of tests was aimed at verifying that the kine-
matics of the mechanism were implemented correctly. This was
done with a high-speed, digital camera operating at the rate of 500
Fig. 7 Predicted aerodynamic loading

Fig. 8 View of the complete mechanism with wings

822 Õ Vol. 127, JULY 2005 Transactions of the ASME

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


Acknowledgments
This work was supported by NATO Advanced Scientific Fel-
lowship 28/B/01/PL and the UK Engineering and Physical Sci-
ences Research Council 共EPSRC兲 through Grants No. GR/
M22970 and No. GR/M78472. The last project was cofunded by
the UK Ministry of Defense. Contributions from the US Army
Contract No. N62558-03-M-0030 are also acknowledged. The
high-speed camera used for capturing the wing motion was spon-
sored by DARPA 共Sam Wilson from the Tactical Technology Of-
fice兲 and loaned to us by NASA Langley Research Center 共Dy-
namics and Control Branch; special thanks to Marty Waszak兲.

References
关1兴 Żbikowski, R., 1999, ‘‘Flapping Wing Autonomous Micro Air Vehicles: Re-
search Pro-Gramme Outline,’’ 14th International Conference on Unmanned
Air Vehicle Systems 共Supplementary Papers兲, pp. 38.1–38.5.
关2兴 Żbikowski, R., 1999, ‘‘Flapping Wing Micro Air Vehicle: A Guided Platform
for Micro-Sensors,’’ Proc. of Royal Aeronautical Society Conference on Nano-
technology and Microengineering for Future Guided Weapons, pp. 1.1–1.11.
关3兴 Żbikowski, R., 2000, ‘‘Flapping Wing Technology,’’ European Military Rotor-
craft Symposium, Shrivenham, UK, March, pp. 1–7.
关4兴 Żbikowski, R., 2002, ‘‘Aerodynamics: Red Admiral Agility,’’ Nature 共Lon-
don兲, 420共6916兲, pp. 615– 618.
关5兴 Żbikowski, R., 2002, ‘‘On Aerodynamic Modelling of an Insect-Like Flapping
Wing in Hover for Micro Air Vehicles,’’ Philos. Trans. R. Soc. London, Ser. A,
360共1791兲, pp. 273–290.
关6兴 Willmott, A. P., and Ellington, C. P., 1997, ‘‘Measuring the Angle of Attack of
Beating Insect Wings: Robust Three-Dimensional Reconstruction From Two-
Dimensional Images,’’ J. Exp. Biol., 200共21兲, pp. 2693–2704.
关7兴 Willmott, A. P., and Ellington, C. P., 1997, ‘‘The Mechanics of Flight in the
Hawkmoth Manduca Sexta, I. Kinematics of Hovering and Forward Flight,’’ J.
Exp. Biol., 200共21兲, pp. 2705–2722.
关8兴 Willmott, A. P., Ellington, C. P., and Thomas, A. L. R., 1997, ‘‘Flow Visual-
ization and Unsteady Aerodynamics in the Flight of the Hawkmoth, Manduca
Sexta,’’ Philos. Trans. R. Soc. London, Ser. B, 352共1351兲, pp. 303–316.
关9兴 Srygley, R. B., and Thomas, A. L. R., 2002, ‘‘Unconventional Lift-Generating
Mechanisms in Free-Flying Butterflies,’’ Nature 共London兲, 420共6916兲, pp.
660– 664.
关10兴 van den Berg, C., and Ellington, C. P., 1997, ‘‘The Vortex Wake of a ‘Hover-
ing’ Model Hawkmoth,’’ Philos. Trans. R. Soc. London, Ser. B, 352共1351兲, pp.
317–328.
关11兴 van den Berg, C., and Ellington, C. P., 1997, ‘‘The Three-Dimensional
Leading-Edge Vortex of a ‘Hovering’ Model Hawkmoth,’’ Philos. Trans. R.
Soc. London, Ser. B, 352共1351兲, pp. 329–340.
关12兴 Dickinson, M. H., Lehmann, F. O., and Sane, S. P., 1999, ‘‘Wing Rotation and
the Aerodynamic Basis of Insect Flight,’’ Science, 284, pp. 1954 –1960.
关13兴 Ellington, C. P., van den Berg, C., Willmott, A. P., and Thomas, A. L. R., 1996,
‘‘Leading-Edge Vortices in Insect Flight,’’ Nature 共London兲, 384共6610兲, pp.
626 – 630.
关14兴 Birch, J. M., and Dickinson, M. H., 2001, ‘‘Spanwise Flow and the Attachment
of the Leading-Edge Vortex on Insect Wings,’’ Nature 共London兲, 412共6848兲,
pp. 729–733.
关15兴 Fearing, R. S., Chiang, K. H., Dickinson, M. H., Pick, D. L., Sitti, M., and Yan,
J., 2000, ‘‘Wing Transmission for a Micromechanical Flying Insect,’’ Proc. of
IEEE International Conference on Robotics and Automation, ICRA ’00, IEEE,
Fig. 9 Sequence of frames from a high-speed camera movie of New York, Vol. 2, pp. 1509–1516.
关16兴 Avadhanula, S., Wood, R. J., Campolo, D., and Fearing, R. S., 2002, ‘‘Dy-
the mechanism in action, illustrating the stroke reversal of the
namically Tuned Design of the MFI Thorax,’’ Proc. of IEEE International
wing Conference on Robotics and Automation, ICRA ’02, IEEE, New York, Vol. 1,
pp. 52–59.
关17兴 Ennos, A. R., 1989, ‘‘The Kinematics and Aerodynamics of the Free Flight of
Some Diptera,’’ J. Exp. Biol., 142, pp. 49– 85.
frames/s for flapping frequencies between 6 and 7.5 Hz. The ki- 关18兴 Gupta, K. C., and Ma, R., 1995, ‘‘A Direct Rotatability Criterion for Spherical
Four-Bar Linkages,’’ J. Mech. Des., 117共4兲, pp. 597– 600.
nematics were indeed correct, and an example illustrating the 关19兴 Hartenberg, R. S., and Denavit, J., 1964, Kinematic Synthesis of Linkages,
stroke 共pitch兲 reversal is given in Fig. 9. McGraw-Hill, New York.
关20兴 Dijksman, E. A., 1976, Motion Geometry of Mechanisms, Cambridge Univer-
sity Press, Cambridge.
关21兴 Rutter, J. W., 2000, Geometry of Curves, Chapman & Hall/CRC, Boca Raton.
5 Conclusions 关22兴 Wunderlich, W., 1978, ‘‘Approximate Optimization of Watt’s Straight-Line
It has been shown that it is feasible to realize analytically trac- Mechanism,’’ Mech. Mach. Theory, 13共2兲, pp. 155–160.
关23兴 Hunt, K. H., 1978, Kinematic Geometry of Mechanisms, Clarendon Press,
table, insectlike flapping kinematics of hover with a four-bar link- Oxford.
age and a spatial articulation. Both of these elements require elas- 关24兴 Torfason, L. E., and Ahmed, A., 1978, ‘‘Double Points of a 4-Bar Linkage,’’
tic parts for practical functioning, which can be implemented Mech. Mach. Theory, 13共6兲, pp. 593– 601.
easily and robustly. The overall concept is quite simple and trans- 关25兴 Hernandez, A., Amezua, E., Ajuria, M. B., and Llorente, J. I., 1994, ‘‘Multiple
Points on the Coupler Curve of Transitional Four-Hinge Planar Linkages,’’
parent, but requires further optimization. In its present form, it is Mech. Mach. Theory, 29共7兲, pp. 1015–1032.
adequate for its purposes as a test bed and a precursor design and 关26兴 Pennestri, E., 1998, ‘‘The Transition Curve of the Planar Four-Bar: An Ana-
can be designed and manufactured without too much difficulty. lytical Approach,’’ Mech. Mach. Theory, 33共8兲, pp. 1293–1299.

Journal of Mechanical Design JULY 2005, Vol. 127 Õ 823

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use


关27兴 Saggere, L., and Kota, S., 2001, ‘‘Synthesis of Planar, Compliant Four-Bar Wings Based on a Leading-Edge-Suction Analogy,’’ NASA Tech. Note TN
Mechanisms for Compliant-Segment Motion Generation,’’ J. Mech. Des., D-3767.
123共4兲, pp. 535–541. 关30兴 von Kármán, T., and Sears, W. R., 1938, ‘‘Airfoil Theory for Non-Uniform
关28兴 Leishman, J. G., 2000, Principles of Helicopter Aerodynamics, Cambridge Motion,’’ J. Aeronaut. Sci., 5共10兲, pp. 379–390.
University Press, Cambridge, England. 关31兴 Sedov, L. I., 1965, Two-dimensional Problems in Hydrodynamics and Aerody-
关29兴 Polhamus, E. C., 1966, ‘‘A Concept of the Vortex Lift of Sharp-Edge Delta namics, Interscience Publishers, New York.

824 Õ Vol. 127, JULY 2005 Transactions of the ASME

Downloaded From: http://mechanicaldesign.asmedigitalcollection.asme.org/ on 01/28/2016 Terms of Use: http://www.asme.org/about-asme/terms-of-use

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy