Professional Documents
Culture Documents
EDFPROP
EDFPROP
DOI 10.1007/s10846-009-9368-0
Received: 1 February 2009 / Accepted: 1 August 2009 / Published online: 17 September 2009
© Springer Science + Business Media B.V. 2009
Abstract Contrary to the manned tailsitter aircraft concepts, which have been
shelved and forgotten after mid 1960’s, the unmanned versions of these concepts
have become popular. Since, tailsitter type UAVs combine both vertical takeoff
and landing (VTOL) operation and relatively high speed cruise flight capabilities
which provide manifest advantages over the other VTOL aircraft concepts, including
helicopters and organic air vehicles (OAVs). However, there is no mini class tailsitter
UAV with efficient high speed cruise flight capability. This paper presents the
design methodology and optimization of ITU Tailsitter UAV concept with hybrid
propulsion system approach to fill that gap. The initial design and analysis show the
advantageous performance over other mini-class VTOL UAVs.
1 Introduction
Tailsitter UAVs combine vertical takeoff and landing (VTOL) operation and rela-
tively high speed cruise capabilities in a single airframe and such a concept provides
manifest advantages over the other VTOL aircraft concepts including helicopters
and organic air vehicles (OAVs). As a result of the increasing demand for efficient
and silent UAV concepts which require no regular “runway” for urban-civilian
applications, the design of ITU-BYU tailsitter concept is tailored towards city and
urban operations with possible autonomous recharging capability to allow 24 hour
on demand reconnaissance and surveillance for various usage areas from traffic/law-
enforcement to border patrol.
The design and development of manned tailsitter concepts had begun in the
beginning of 1950s and many experimental aircraft were built and tested during
the period between 1950 and 1960s. The pilots of such manned aircraft were in
charge of the aircraft’s attitude control by only looking at the displays and sensing
the behaviors of the aircraft in an upside position with unfamiliar control inputs
compared to the helicopter pilots. Moreover, such manned aircraft had control
problems especially during landing and hover-to-cruise transition phase, because of
having no stability augmentation system to help decrease test pilots’ extremely high
workload. Therefore, none of the experimental tailsitter concepts have demonstrated
an apparent advantage over helicopters or fixed wing aircraft. As a result of several
accidents mainly due to the high workload of the test pilots, the development
of manned tailsitter projects were suspended after mid 1960s. However, with the
help of the advances in both hardware and software based technologies [4, 6, 14],
the distinct tailsitter concept was finally successfully developed to take its place
among the other VTOL UAVs. Likewise, there are many already mature or under
development VTOL concepts. At the present time, OAVs are the most popular
ones which can be classified under tailsitter concepts. Allied Aerospace’s iStar series
Organic Air vehicles (OAV) and Honeywell’s electric powered mini class OAV are
the foremost examples with a wide range of application around the world. On the
other hand, although the OAVs show efficient static-low speed flight due to the
shroud and duct around the propeller blades, such ducted fan UAV concepts still
have some aerodynamic problems, especially in efficient forward flight regime and
hoverto-cruise transition phase. In addition, OAVs with internal combustion engines
also suffer from high levels of noise during operationwhich making these types of
tailsitter UAVs unsuitable for “silent intelligence” in urban operational areas. Other
examples of mini to mid size tailsitter concepts are T-Wing and Heli- wing. T-Wing
was developed at University of Sydney at 2000s [21], and Heliwing was developed by
Boeing. However, both of these similar concepts are not suitable size wise for urban
applications because of their noisy internal combustion engines. Besides, there are
also several micro class tailsitter UAV concepts such as Brigham Young University’s
tailsitter UAV [13] and University of Arizona’s coaxial propeller driven UAV [18].
Nevertheless, such micro size UAVs are not suitable for carrying a “useful” payload
and for service in severe weather conditions. Two of the discussed tailsitter concepts
are seen in Fig. 1.
Since 2001, related to the recent progress in Lithium based batteries, aircraft
designers have started to consider electric powered aircraft concepts [16]. Moreover,
advances in brushless DC electric motors have accelerated the development period
of such kind of aircraft. Selection of an electric propulsion system for ITU Tailsitter
was a major pre-design choice and brings with it the advantages of low noise levels,
easier maintenance and the unique capability to autonomously recharge the units
from base landing stations. However, for an electric powered vehicle within the mini-
UAV class, this unique capability calls for a trade-off between speed limited high
power propeller configuration and the power limited high speed electric ducted fan
(EDF) system. In this study for mini class UAVs, design optimization and an intuitive
thrust-power-airspeed trade-off approach, which leads to a hybrid “propeller-ducted
fan propulsion system” design that can achieve maximum horizontal flight time and
J Intell Robot Syst (2010) 57:505–529 507
a b
Fig. 1 a T-wing concept demonstrator from University of Sydney [1], b micro tailsitter UAV
prototype from University of Arizona [18]
maximum range for the ongoing ITU Tailsitter Project, is presented. As a result, the
hybrid propulsion system, consisting of both propulsion systems, was decided to be
used in order to design an “all flight regime” efficient aircraft which meets the desired
requirements.
Due to the aim of designing an efficient tailsitter UAV, propulsion system has the
highest priority among the design requirements. Since electric powered propulsion
system has advantages over internal combustion engines in terms of maintenance
and noise level; it has been chosen as one of the requirements. Propeller driven
system supplies high thrust to power ratio for VTOL operations. However, the thrust
is rapidly decreased as the incoming airspeed is increased. Hence, the performance
is decreased at high speed and the system becomes insufficient for cruise flight. On
the other hand, Electric EDF systems are capable of producing the same thrust with
a higher thrust to power gradient than the propeller systems. Although the power
consumption is increased, EDF system’s thrust is necessary only for long range cruise
operations.
In ITU Tailsitter, a folding propeller system, located on the nose of the aircraft, is
used for hovering, vertical take-off, vertical landing and low-speed transition mode,
whereas an EDF system, which is placed between the stabilizers, is used for level
and high speed flight modes where the propeller folds onto the fuselage in order
to reduce drag when it is turned off. In addition, to calculate the approximate
empty weight, instead of the classic method of empty weight fraction, an “aircraft-
based” weight modeling and optimization study have been conducted so as to see
the most efficient design possible. Initial system performance analysis with candidate
propulsion units has shown that up to 40m/s cruise speed and maximum 120 min of
flight endurance can be achieved while carrying 1 kg of payload in 10kg of flying
system with 3 min of vertical take-off and landing operation duration—a distinct
performance in comparison to the same class rotary-wing and OAV alternatives.
In the following sections, a trade off analysis is presented where the hybrid-dual
propulsion approach with a qualitative analysis of the selected propulsion systems,
508 J Intell Robot Syst (2010) 57:505–529
is described. This is followed by the design overview and the design and design
optimization approach. Finally, the control methodology and the results are denoted.
2 Trade-off Analysis
The information given in this section is for clarifying the facts about both type of
propulsion systems. General characteristics of both propeller and EDF systems are
presented based on the wind tunnel test and analytical calculations.
of available performance data about propellers [5, 19, 23], these propellers are mostly
used on commercial or military manned aircraft. On the other hand, there are no
sufficiently and systematically catalogued propeller performance charts for use on
small scale UAVs, except for some test results [3, 15].
As mentioned before, in ITU tailsitter UAV, the propeller system is considered
to be used as main lift generating device during VTOL operations. Thus, to get the
highest specific thrust (T/P) value in static and low speed climb/transition phases, a
propeller with the largest plausible diameter and a relatively low pitch value should
be selected. This is because, propellers with high pitch value are designed for high-
speed applications and a high percentage of propeller blades are stalled during low
speeds and static condition. In light of these considerations, a RASA brand 28.5 × 12
size (28 in. of diameter and 12 in. of pitch) propeller is selected. However, instead of
wind tunnel tests, which give unsatisfactory data due to the experiment room size of
ITU wind tunnel, an analytical calculation method is employed.
To derive the characteristics of the selected propeller analytically, physical
modeling of the propeller’s physical variables, which are pitch distribution, chord
distribution and propeller’s airfoil data in each section, is defined. After completing
the modeling study, JavaProp software [9], which is written by Martin Hepperle,
is used to analyze the propeller. JavaProp is a java based application which uses
blade element theory as background theory and the validation [10] of JavaProp
software has been written by Dr. Martin Hepperle with the help of comparison of
the calculations using NACA’s test results [22].
specific thrust value is a function of advance ratio that is also a function of angular
and airspeed. So, three dimensional complex surface geometry analysis should be
conducted in order to see and quantize the performance of both of the propulsion
systems. However, with the help of the scaled specific thrust method, which is
structured during the design phase of ITU Tailsitter UAV, the three dimensional
specific thrust determination problem has scaled down to two dimensional problem.
The results are plotted as seen in Fig. 3.
Moreover, the calculation method of scaled specific thrust for both EDF and
propeller propulsion systems is seen in Figs. 4 and 5. Note that, for the selected
“propeller based” propulsion systems, thrust and power coefficients can be written as
a quadratic function of advance ratio. Therefore, the letters a1 , b 1 , c1 and a2 , b 2 , c2 ,
which are seen in Figs. 4 and 5, are the coefficients of the quadratic equations of EDF
and propeller systems respectively.
In Fig. 6, there are two regions, which are valid for our consideration. The first
region is zero speed, which is simulated as hovering maneuver; the second region is
from the black vertical line, which intersects the stall speed (16 m/s) on the x axis,
to the desired maximum airspeed, which is 35 m per second. The pink area between
“very-low” speed (0–5 m/s) and stall speed is intentionally left blank. This is because
the calculation of exact and optimized transition maneuvers will be done after exact
modeling of the aircraft by flight tests.
A more detailed account of the procedure is as follows; the propulsion system only
overcomes parasite and induced drag from the stall speed to the maximum airspeed.
However, the aircraft’s weight is added to drag component, which propulsion system
must overcome during the transition maneuvers. Therefore, the accurate transition
phase/airspeed is investigated after calculating the optimized transition maneuvers.
As seen from Fig. 6, each of the propulsion systems with any given airspeed condition
results in different revolution per minute and advance ratio values. Therefore, the
Fig. 3 Scaled specific thrust comparison of the selected propeller and EDF systems
512 J Intell Robot Syst (2010) 57:505–529
Fig. 4 Scaled specific thrust calculation methodology for EDF propulsion system
comparison between the propulsion systems can be made outright in Fig. 6. The
EDF system is about 64 times more efficient than the propeller system for the whole
flight regime from stall to maximum airspeed. Moreover, for static case analysis,
specific thrust for both propeller and EDF systems are calculated as 0.1052 and
0.0122. Hence, depending on the static case calculations (for hover and low speed
Fig. 5 Scaled specific thrust calculation methodology for propeller propulsion system
J Intell Robot Syst (2010) 57:505–529 513
Fig. 6 Scaled specific thrust graphic for propeller and EDF propulsion systems in both hover and
cruise conditions
climb) it is seen that the propeller system is about 9 times more efficient than the
EDF system. However, due to angular velocity restriction, the EDF system can not
produce adequate thrust as the propeller system.
As a result, the breakthrough advantage of the hybrid propulsion approach is
evident in Fig. 6.
The design philosophy developed hinges on obtaining the maximum possible payload
capacity while achieving both high T/W ratio for VTOL, maneuverability and low
energy demand per unit operation time (i.e. a low power demand for enhanced en-
durance). To do this, an aircraft, which has a relatively high cruise speed with vertical
take off and landing capabilities, has been delineated. In addition, restrictions placed
by the city operational environment were reflected via area and volume limitations
before starting the design. In this section, the optimization design methodology for
the ITU Tailsitter UAV is summarized.
As indicated in Fig. 7, the design methodology approach consists of three main
phases. These phases are inputs, basic calculations and optimization. Input part
includes two sub-parts The first sub-part is called component based constant weight
inputs, which includes the weights of the components that are fixed for the design
process. The second sub-part is design inputs including wing span, wing loading,
maximum take off weight and the horizontal tail arm. In basic calculations part, drag
coefficient, empty weight and battery weights are calculated. Optimization part is
used to get most desirable design within our criteria and constraints.
4.1 Inputs
it’s generally hard to find sufficient data to to carry out empty weight prediction
for relatively novel concepts. Consequently, a quasi exact weight prediction method
has been developed to calculate a “more accurate” empty weight value in terms of
aircraft size and total weight for ITU Tailsitter.
Empty weight of the aircraft is composed of four main weight components as seen
in Eq. 1 and these are described in detail in the following subchapters;
Wempty = W f uselage + Wwings + Wstab ilizers + Wstructural (1)
Fuselage Skin Weight Modeling Before starting the calculations, composite fabri-
cation method had been selected. As mentioned previously, during cruise flight, the
folding propeller folds onto the fuselage to reduce the drag and then the EDF system
is activated. Moreover, in “cruise to hover” transition phase, EDF system is shut
down and the propeller system is reactivated in order to perform a power efficient
landing. It is important to note that, unfolding process of the propeller requires a
symmetrical fuselage shape to prevent the propeller from damaging the fuselage
skin. Therefore, in light of this prediction, a symmetrical airfoil, NACA 642-015, was
selected to shape the fuselage geometry. The relation between fuselage surface area
and fuselage weight was derived as a next step. The measurements on the selected
fuselage airfoil show that the surface area is about 0.303 m2 for 1 meter of fuselage
length. Therefore, the function for the fuselage weight can be written as given in
Eq. 2 in terms of surface area and composite skin’s surface weight density:
W f uselage = 0.303·l f uselage ·ρskincomposite (2)
It should be noted that, l f use is the fuselage length in meters and ρskincomposite is
the composite skin’s surface weight density in N/m2. However, fuselage length in
Eq. 2 is not a sufficient parameter for the optimization problem since fuselage length
depends on horizontal tail arm (lht ) and root chord of the wing (Croot ) which are
variables of the optimization problem. Hence, fuselage length is expressed in terms
of these variables as given in Eq. 3:
1
l f uselage = lnose + Croot + lht (3)
4
The variable lnose , which is determined as a constant value of 0.4 meters, represents
the distance, which is assigned by considering propeller’s clearance, between the
leading edge of the wing root and nose of the fuselage. Therefore, the equation of
fuselage weight is derived and seen in Eq. 4.
1
W f uselage = 0.303·ρskincomposite · 0.342 + Croot + lht (4)
4
Wing Weight Modeling During the calculation of the wing weight function; wing
airfoil, taper ratio, wing loading, wingspan are the main parameters. Hence, the
calculations have been started with wetted surface area of the wing. Thus, for a meter
of chord length, the wetted surface area is calculated as 2.06633·b m2 , where b is the
wingspan. Following these, the wing skin weight can be expressed as given in Eq. 5:
Wwingskin = 2.06613·c¯wing ·b ·ρskincomposite (5)
It should be noted that, the production method and the material that is planned to
be used for both the fuselage and the wing are same. Moreover, the weight of wing
516 J Intell Robot Syst (2010) 57:505–529
spar, which is made of hollow carbon tube, must be obtained to derive the entire and
accurate wing weight. Since the deflection at the tip of the wing is a crucial parameter
for the wing stiffness, the tip deflection is restricted as 5 percent of the wingspan in
order to be able to sustain 2.5g of loading. According to previous experiences, this
loading condition can be modeled by applying a force equal to half the weight of the
aircraft to the wing tip when the fuselage is stationary. 2.5g load case for the half wing
can be simulated as the tip deflection of a cantilever beam loaded with a concentrated
force at the free edge. The deflection is then expressed as given in Eq. 6 in detail:
W0 b 3
F·l 3 · 2
δ = 0.05·b = = 2 (6)
12·E·I 12·E·I
For the equation given above, E is the modulus of elasticity of spar material and I
represents moment of inertia for a circular wing spar section. The moment of inertia
can then be calculated in order to find the inner diameter of the spar, which is a
hollow carbon tube. Note that, the outer diameter of the spar is restricted as given in
Eq. 7:
t
φouterspar = cwingtip · − 0.003 (7)
c
Where cwingtip represents the wing tip chord, ct represents thickness ratio of wing
airfoil and the numerical value, 0.003, represents the thickness of skin composite
material. Hence, the moment of inertia can be calculated as given in Eq. 8 and the
inner diameter of the spar can be expressed as seen in Eq. 9. Finally, the weight of
the carbon spar can be calculated in Eq. 10, given as follows:
W0 b 3
4 4 · 2
I= φouterspar − φinnerspar = 2
(8)
4 12·E· (b ·0.05)
3 1
4
4
4· W20 b2
φinnerspar = − φouterspar (9)
12··E· (b ·0.05)
2 2
Wspar = · φouterspar − φinnerspar ·b ·ρspar (10)
4
For the equation given above, ρspar is the density of the carbon material per unit
volume. To conclude, after summing skin and spar weights, the weight function for
the wing is derived as in Eq. 11 given below:
2 2
Wspar = · φouterspar − φinnerspar ·b ·ρspar + 2.06613·c¯wing ·b ·ρcomposite (11)
4
Stabilizer Weight Modeling The total weight of the stabilizers can be expressed as
the summation of horizontal and vertical tail weights. The manufacturing method,
used for the stabilizers, is slightly different from the one used for fuselage and wings.
In this manufacturing technique, stabilizers are going to be made of high density foam
covered with carbon fiber. Thus, wetted area and inner volume of the stabilizers are
used to derive the weight function as seen in Eq. 12:
Wstab ilizers = W HT + WVT (12)
J Intell Robot Syst (2010) 57:505–529 517
First of all, the calculation steps for horizontal tail are shown. After that, the
calculations of vertical tail weight determination is shown.
In Eq. 13, S HTwetted represents the wetted area of horizontal stabilizer, ρskin repre-
sents the density of stabilizers’ composite covering material, ν HT represents the inner
volume of horizontal stabilizer and ρ f oam represents the density of the foam that fills
the horizontal tail. Horizontal stabilizer area is written in terms of wing area, mean
aerodynamic chord, tail arm and horizontal tail volume coefficient and seen in Eq. 14:
ν HT ·c¯wing ·Swing
S HT = (14)
l HT
In the equation given above, ν HT is horizontal tail volume coefficient, c¯wing is
mean aerodynamic chord of the wing, Swing is wing area and l HT is horizontal tail
arm which is equal to the distance between quarter mean aerodynamic chords of
wing and horizontal stabilizer respectively. In ITU Tailsitter airplane, inverted V-tail
configuration is used with 25 degree of anhedral on both left and right horizontal
stabilizer parts. Therefore, the actual area of horizontal stabilizer in terms of the
anhedral angle, can be calculated by using Eq. 15.
1 ν HT ·c¯wing ·Swing
S HTactual = · (15)
cos 25 l HT
After the actual horizontal tail area is determined, wetted area can be found with
respect to the tail airfoil. As indicated before, NACA0014 is selected for horizontal
and vertical stabilizers. Therefore, upon inspecting the geometric properties of the
selected airfoil, it is found that wetted area is 2.133 m2 per unit planform area. Then,
wetted area can easily be found as given in Eq. 16:
1 ν HT ·c¯wing ·Swing
S HTwetted = 2.133· · (16)
cos 25 l HT
After determination of wetted area for horizontal stabilizer, the weight of the
composite skin covering material can be found as given in Eq. 17:
1 ν HT ·c¯wing ·Swing
W HTskin = 2.133· · ·ρskin (17)
cos 25 l HT
Next, the weight function for “foam core” must be found in order to derive weight
function for horizontal tail. According to the geometrical data, the airfoil side area
coefficient for one meter of chord is determined as 0.000951 fsm2 . Hence, the volume
of horizontal tail is expressed as given in Eq. 18:
ν HT ·Swing ·cwing 1
V HT = · ·9.51·10−4 (18)
l HT cos 25
In order to determine the weight of foam core, the volume is multiplied by the
foam density as given in Eq. 19:
ν HT ·Swing ·cwing 1
W f oamcore = · ·9.51·10−4 ·ρ f oam (19)
l HT cos 25
518 J Intell Robot Syst (2010) 57:505–529
As a result, the total weight of the horizontal stabilizer is derived by summing both
foam core and skin composite weights as seen in Eq. 20:
1 ν HT ·c¯wing ·Swing
WHT = 2.133· · ·ρskin
cos 25 l HT
ν HT ·Swing ·cwing 1
+ · ·9.51·10−4 ·ρ f oam (20)
l HT cos 25
In similar but slightly different manner, the weight function for vertical tail is
derived by using the same steps and obtained as Eq. 21:
νVT ·b ·Swing ν HT ·c¯wing ·Swing
WVT = − 2.133· tan 25· ·ρskin
l VT l HT
νVT ·b ·Swing ν HT ·Swing ·cwing −4
+ − · tan 25·9.51·10 ·ρ f oam (21)
l VT l HT
Finally, entire weight function for stabilizers is derived by summing horizontal tail
and vertical tail weight functions. By inserting Eqs. 20 and 21 in the equation given
below, the weight function for stabilizers is obtained as seen in Eq. 22:
1 ν HT ·c¯wing ·Swing
Wstabilizer = 2.133· · ·ρskin
cos 25 l HT
ν HT ·Swing ·cwing 1
+ · ·9.51·10−4 ·ρ f oam
l HT cos 25
νVT ·b ·Swing ν HT ·c¯wing ·Swing
+ − 2.133· tan 25· ·ρskin
l VT l HT
νVT ·b ·Swing ν HT ·Swing ·cwing −4
+ − · tan 25·9.51·10 ·ρ f oam (22)
l VT l HT
Structural Weight Modeling Structural weight consists of the weight of spar box,
bulkheads, longerons and glue. Spar box weight is assumed to be a constant, while
glue weight is approximated as 10% of bulkheads, longerons and spar box weights.
Moreover, for building fuselage frame, it is planned to locate four longerons along
the fuselage and one bulkhead per each 0.15 m of fuselage length. The total structural
weight is then expressed as given in Eq. 23:
l f uselage
Wstructure = 1.1· Wb ulkhead · + 4·l f uselage ·Wlongeron + Wsparb ox (23)
0.15
Where Wb ulkhead is the assigned constant average weight of a bulkhead, Wlongeron
is the weight of a carbon longeron per unit length and Wsparb ox is the preassigned
constant weight value of the spar box.
propeller consumes more energy than hovering mode during vertical climb; the
energy consumption is reduced to a level below that of hovering during the vertical
descent maneuver.
For hover mode, the thrust produced by the propeller is equal to the sum of weight
of the aircraft and the drag force created by the propeller’s induced velocity, Which
is calculated using helicopter theory [2]. The battery weight determination logic for
the propeller system is shown in Fig. 8 step by step.
The battery weight of the propeller propulsion system is a function of the VTOL
operation duration, propeller’s characteristics, drag force created by geometry of the
aircraft, and the takeoff weight of the aircraft. Note that, in order to calculate the
“worst case” and simplify the calculations, instead of the parasite drag coefficient
that is caused by the “induced velocity wetted” part of the aircraft,the parasite drag
coefficient for the whole aircraft is used.
Battery Weight Modeling of The EDF Propulsion System The selected EDF unit is
mainly dedicated to cruise flight. For this reason, thrust generated by the EDF unit is
determined as equal to the drag force on the aircraft. Therefore, the battery weight
calculation methodology is based on cruise flight and seen in Fig. 9.
Drag Force Modeling for Aerodynamic Surfaces The parasite drag coefficient for
the aerodynamic surfaces, C Daero , composed of three discrete parts; wing, horizontal
stabilizer and vertical stabilizer, and figured out in Eq. 25:
1
C Daero = · C fw ·F Fw ·Qw ·F Fw Swetw + C f HT ·F F HT ·Q HT ·F Fc Swet HT
Sref
+ C fVT ·F FVT ·QVT ·F FVT SwetVT (25)
Flat-plate skin friction varies depending on the type of flow, laminar or turbulent,
over the surfaces. For wing and stabilizers, turbulent flow assumption has been made
and flat-plate skin friction coefficient is written as Eq. 26:
0.455
Cf = 2.58 0.65 (26)
log10 Re + 1 + 0.144·M2
Where “Re” represents Reynolds number, which is taken 500.000 for wings and
300.000 for stabilizers; “M” indicates mach number, which is selected 0.1 as constant
for each design during optimization. To continue, form factor(F F) equation for wing
and tails is written in Eq. 27.
4
0.6 t t
FF = 1 + x · + 100· · 1.34M0.18 (cos m )0.28 (27)
c m
c c
In Eq. 27, the term “( xc )m ” is the chordwise location of the airfoil maximum
thickness point, which is 0.3 for the selected airfoils. m refers to the sweep of the
maximum thickness line, which are 0 and 23 degrees for wing and stabilizers used in
ITU Tailsitter respectively. The interference factor, “Q”, is chosen 1.1 for both wing
and stabilizers.
Drag Force Modeling for Fuselage To calculate the drag coefficient for fuselage,
the steps must be followed are similar to the steps followed in wing and stabilizer
calculations. However, there’s a change on form factor estimation. The form factor
for fuselage or smooth canopy is calculated using Eq. 28.
60 f
FF = 1 + 3 + (28)
f 400
522 J Intell Robot Syst (2010) 57:505–529
Where;
l l
f = = (29)
d 4
A
max
4.3 Optimization
Table 3 Primal variables and side constraints for the optimization problem
Primal variable Lower bound # upper bound Explanation (gr)
MTOW (W0 ) 30 N 100 N Lower bound is determined according
to the known weights where the upper
bound is set by the structural restrictions
of the propeller.
W0 N N
Wing loading S 100 m2 m2
The limits are based on the similar
types of UAVs
b 1m 2m The lower limit is set by wing efficiency
issues and the upper limit is restricted
by the VTOL operation area.
l HT 0.6 m 1.5 m The lower limit is determined by considering
the tail size and the upper limit is set
by the structural limits during vertical
landing phase.
W f anb attery 3N 30 N The boundaries are set by the previous
experiences
524 J Intell Robot Syst (2010) 57:505–529
Before going into the details of objective functions, some descriptions are neces-
sary to lay out the components of the functions. Aircraft’s maximum takeoff weight
(W0 ) can be expressed as in Eq. 30:
Where Winputs is described in Section 4.1.1 and can be reproduced here as Eq. 31:
In Eq. 31; Wservo , Wrec , Wrecb att , W ESC , Wregulator , Wcable , W propunit , W EDFunit de-
notes the weights of servos, receiver, electronic speed controller of the brushless mo-
tors, voltage regulator, cables, propeller propulsion unit and EDF unit respectively.
In the light of the given formulations, one of the objective functions can be expressed
in a compact form in Eq. 32. Note that, Winputs , W propb attery , W f anb attery andWempty
were described in the previous chapters as functions or invariant values.
The second objective function is the cruise time for EDF propulsion system. As
seen in Fig. 9, the general process to determine the battery weight for the EDF system
was summarized, but not formulated. Therefore, cruise time formulation, which is
seen in Eq. 33, is obtained with the help of the battery weight equation.
5 Results
As stated previously, two objective functions had been determined in order to see
the performance of the aircraft with different payload weights. With the help of the
pareto chart, the optimization results for different configurations are seen in Fig. 11.
The final specifications and CAD drawing screenshot of ITU Tailsitter UAV are
seen in Table 5 and Fig. 12, respectively.
According to the results, the tailsitter design with hybrid propulsion system has
both payload and cruise duration advantages over same-class rotary wing and fixed
wing alternatives. So, it shows that the accurate selection of propulsion system is one
of the most vital parameters in aircraft design.
6 Conclusions
In this work, the design optimization study of a tailsitter aircraft with a revolutionary
hybrid/dual propulsion system has been described. The results obtained in this paper
are based on analytical calculations on the propeller propulsion system, experimental
data on the EDF propulsion system and the design inputs which are in close
relation with both design constraints and design criteria. Initial system performance
analyses with candidate propulsion units indicate that up to 27.5 m/s cruise speed
and maximum 2 hours of flight endurance, including 3 min of vertical take-off and
landing duration, can be achieved while carrying a 1 kg payload and 90 km of range -
a markedly superior performance in comparison to the similar class rotary-wing and
OAV alternatives in the same class. As seen in Figs. 13 and 14 Prototyping of ITU
Tailsitter is almost completed and flight tests are going to be conducted so as to
compare real and the calculated performance.
References