A Review on the Platform Design, Dynamic

Modeling and Control of Hybrid UAVs

Adnan S. Saeed, Guowei Cai, Ahmad Bani Younes, Shafiqul Islam, Jorge Dias, Lakmal Seneviratne
Khalifa University, Abu Dhabi, UAE
{adnan.saeed, guowei.cai, ahmad.younes, shafiqul.islam, jorge.dias, lakmal.seneviratne} @kustar.ac.ae

Abstract— Unmanned aerial vehicles (UAVs) are environments and rugged surfaces. Moreover, their
experiencing a tremendous development as they are being maneuvering and hovering capabilities enable better
utilized in a wide range of reconnaissance and surveillance performance in observation, reconnaissance, monitoring and
missions. For now, mini UAV platforms are dominated by two other missions. On the other hand, they cannot accomplish
main types, i.e., fixed-wing conventional aircrafts and vertical missions requiring high speed, long flight range or payload
take-off and landing (VTOL) aircrafts and each type has its own capacity. In contrast, fixed-wing UAVs require runways or
inherent limitations. The aim of this paper is to present a launching and recovery equipment for take-off and landing but
technical overview of Hybrid UAVs which integrate the beneficial they have higher cruising speed, better payload capacity and
features of both. The platform design types and technical details
longer flight range. Hybrid UAVs combine the advantages of
are introduced first. The modeling is then explained in terms of
kinematics, dynamics, complexity and fidelity. Several flight
both where they have the ability of vertical take-off and
control strategies implemented in Hybrid UAVs are then landing as well as high cruising speed and enhanced endurance.
discussed and compared in terms of theory, linearity and This enables the possibility of performing wider range of
implementation. missions or same missions with better performance.
Combing the advantages of fixed-wing and VTOL aircrafts
Keywords— Hybrid UAVs, Fixed-wing VTOL, Platform design, has long been a concern for many aerospace and aviation
Dynamic modeling, Control, Automation
industries. Over the years, there have been several attempts to
I. INTRODUCTION build manned Hybrid aircrafts such as Bell Boeing V-22
Osprey, Vertol VZ-2, Sikorsky X-wing, Convair XFY Pogo
During the last two to three decades, Unmanned Aerial and Harrier GR7 as shown in Fig 1a-1e respectively [8-12].
Vehicles (UAVs) have experienced a tremendous development. Some of the attempts did succeed and the aircrafts are still
They are the fastest-growing sector in the aerospace industry as operating up to the moment such as V22-Osprey and Harrier
the UAV market is expected to encounter a robust growth of GR7. Nevertheless, within the last four years, the concept
7.73% by 2020 [1] [2]. Also, they have been extensively used invaded the UAV field as a number of research groups
for military applications such as surveillance, tracking, security documented their pioneer work in literature and a couple of
and data acquisition. Moreover, the total sector sales of the companies even commercialized the idea. It is believed that
global military UAV market is expected to increase by more Hybrid UAVs will be having a bright future and will promptly
than 60% between 2011 and 2020 with the lion’s share coming dominate the miniature UAV market. Still in its infancy, there
from the U.S. [3] [4] [5]. Nevertheless, UAVs’ applications is a huge space for the miniature Hybrid UAVs to become
were not limited to that, instead, the market of civilian UAVs is more mature, in terms of design philosophy, dynamics
growing quickly and steadily and it is possible that it will dwarf modeling, control, guidance, navigation, robustness, etc.
the military demand in the future [3] [6] [7]. These civilian
UAVs can utilize variety of new areas such as monitoring of The central objective of this paper is to present an overview
wildlife, surveillance of road traffic, disaster and crisis on the platform design, dynamic modeling, control methods
management, gas leakage detection, infrastructure inspection, and implementation of mini electric powered Hybrid UAVs. It
law enforcement, agriculture etc. Furthermore, they also gain is the authors’ desire that this work provides a fairly complete
increasing popularity in academia as the potential of miniature picture of the mini Hybrid UAVs and serves as a base line for
(or mini) UAVs, those ranging from micro aerial vehicles further developments and modifications on that field. The
(MAVs) to vehicles with few tens of kilograms, is researched paper is organized as follows: Section II discusses the platform
actively. design types of all the hybird UAVs, Section III gives an
overview of the modeling work done on hybrid aerial vehicles
For now, mini UAV platforms are dominated by two main and section IV includes the different control strategies
types, i.e., fixed-wing conventional aircrafts and vertical take- implemented in Hybrid UAVs.
off and landing (VTOL) aircrafts, also known as rotorcrafts or
multi-copters, and each type has its own inherent limitations. A II. PLATFORM DESIGN
new and promising trend is to develop a fixed-wing VTOL
Generally, hybrid UAVs can be categorized into two main
UAV or the so-called Hybrid UAV, which effectively
types: Convertiplanes and Tail-sitters. A Tail-sitter is an
integrates the features of fixed-wing and VTOL aircrafts and
aircraft that takes off and lands vertically on its tail and the
thus inherits the advantages of both. VTOL UAVs have the
spot take-off and landing functionality even at hazardous
entire aircraft tilts to achieve horizontal flight. On the other A Tilt-wing has a similar concept to Tilt-Rotor except that
hand, a the assembly of the wing tilts instead of the rotors only. The
wings will be directed upwards which makes the aircraft more
vulnerable to cross winds during takeoff, landing and hovering.

Figure 1 Examples of manned hybrid aerial vehicles

Convertiplane maintains its airframe orientation in both flights

and certain transition mechanisms such as tilting rotors or Figure 2 Examples of Tilt Rotor Hybrid UAVs
wings apply to make the conversion. Each of them can be
further categorized into a few sub-types, depending on the Because of that, they require complicated control mechanisms
transition mechanism and airframe configuration. In what and higher use of available power to maintain stability during
follows of Section II, explanation of the subtypes, analysis on vertical flight. Also, this makes landing on moving deck
their design features, pros and cons, and brief introduction of environments very difficult. However, since rotors are fixed to
representative examples will be addressed. wings, this allows various design options for the wing
geometry and therefore enhance aerodynamic performance of
A. Convertiplane the aircraft. Vertol VZ-2 (Fig. 1b) was a manned Tilt-Wing
As mentioned earlier, a Convertiplane is a type of Hybrid aircraft built in 1957 by Boeing Vertol, completed several
UAV that takes off, cruises, hovers and lands with the aircraft’s successful flights and was retired in 1965 [9]. Like Tilt-
reference line remaining horizontal (i.e. the main body Rotors, Tilt-Wings are also researched actively and there are
configuration does not change during flight). Different many examples applying the idea such as AVIGLE (Fig. 3a)
transition mechanisms are applied to convert from vertical [22-24], Sabanci University UAV (SUAVI) (Fig. 3b) [25] [26],
flight to horizontal flight or vice versa. Based on that, AT-10 Responder (Fig. 3c) [13], Quad Tilt Wing VTOL UAV
Convertiplanes can be further categorized into Tilt-Rotors, Tilt- (QTW) [27], and HARVee [28].
Wings, Rotor-Wings and Dual-Systems.
A Rotor-Wing is another type of convertiplane aircrafts
A Tilt-Rotor is an aircraft where multiple rotors are where rotary wings spin to provide lift during vertical flight
mounted on rotating shafts or nacelles. During transition, the and hovering and stop to act like a fixed wing in horizontal
rotors tilt gradually towards flight direction providing the flight. Sikorsky X-wing (Fig. 1c) is a manned aircraft
aircraft forward speed until level flight mode is achieved. It is implementing the Rotor-Wing concept but the program was
important to note that some Tilt-Rotors might have fixed rotors cancelled in 1988 [10] [29]. Moreover, another Rotor-Wing
always directed upwards and operate only during vertical flight UAV are the Boeing X-50 Dragon Fly developed by DARPA
to provide extra lift for takeoff, landing and hovering. An (See Fig. 3d) and the one presented in [30] developed by
example of manned Tilt-rotor is Bell Boeing V-22 Osprey (Fig. Arizona state university and was also withdrawn after several
1a) where it included two tilting jet engines to perform the crashes because of some aerodynamic problems [31]. It is true
transition [8]. As in V-22 Osprey (Fig. 1a), Tilt-Rotors usually that Rotor Wings might be light weight and easy for takeoff
have their engines mounted at the wing tips forcing shorter and landing, but many stability issues arise.
wing span and thicker airfoil. This results in lower aspect ratio
and increased drag respectively causing poor aerodynamic Another type of Convertiplanes could be referred to as
performance when compared to fixed-wing aircrafts or other Dual-System where it implements multiple rotors directed
Hybrid UAVs. On the other hand, the shafts and nacelles are upwards for vertical flight and another separate tractor or
only required to rotate rotors instead of wings or other heavy pusher for level flight. However, during horizontal flight, the
structures which in turn saves power and weight. Moreover, multiple lifting rotors used for vertical flight are useless and
due to their controllability and stability in vertical flight when add extra weight to the aircraft which results in requiring more
compared to other Hybrid UAVs, Tilt-Rotors are actively power from the tractor or pusher. That is why this method
researched in academia and there exist several vehicles might not be as efficient as tilting rotors or tilting wings. Apart
implementing the idea such as IAI Panther [13][14], TURAC from the propulsion system, this concept is very simple to
[15-18], Orange Hawk [19], FireFLY6 [20] and apply in terms of design, control system and modeling because
AgustaWestland Project Zero [21] shown in Fig. 2a-2e the two flight modes could be analyzed separately. The idea
respectively. was employed in Arcturus JUMP (Fig. 3e) developed by
Arcturus UAV [32-34] and in Airbus’ Quancruiser, shown in
Fig. 3f, which is a joint effort of Airbus Group Innovations, reduced efficiency in horizontal flight. However, in terms of
Airbus Defense, and Steinbeis Flugzeug-und Leichtbau GmbH stability, DTTTs are far more stable in takeoff, hovering, and
[35`]. landing and require simpler control strategies than CSTTs.
ITU Tail-sitter [39], T-Wing (Fig. 3b) [40] [41] and [42] are
examples of CSTT UAVs. On the other hand, there are three
main examples for DTTTs, namely, VertiKUL (Fig. 3c) [43]
ATMOS (Fig. 3d) [44] and Quadshot [45] [46]. All three have
four rotors to provide differential thrust to make the transition
to horizontal flight and back. However, ATMOS and Quadshot
have control surfaces or tilting rotors for control during
horizontal flight unlike VertiKUL which depends on
differential thrust for to control the aircraft in all modes as well
as performing the transition.
Reconfigurable wings are another type of Tail-sitters where
Figure 3 Examples of Tilt Wing, Rotor Wing and Dual the wings extend during horizontal flight and retract during
System UAVs vertical flight. This concept is based on the idea that more lift
at lower speed is desired in cruising and therefore the wings
B. Tail-sitter
extend to provide larger wing span. Moreover, the wings
A Tail-sitter is an aircraft that takes off and lands vertically retract during vertical flight to minimize the effect of wind
on its tail and the whole aircraft tilts forward using differential disturbance and ease maneuverability. The idea is applied in U-
thrust or control surfaces to achieve horizontal flight. This Lion (Fig. 3e and 3f), an aircraft developed by the National
concept could also be denoted as Tilt-Plane since the whole University of Singapore in 2014 in which a four bar linkage
plane tilts to achieve level flight. Due to its ability to make the was designed to make the reconfiguration which in turn
transition without the need of extra actuators, this concept is minimizes platform weight and keeps the structure simple [47].
mechanically simple and saves a huge amount of weight when
compared to Convertiplanes. Moreover, since tail-sitters land
on their tails, they require relatively stronger tails to be able to
withstand landing impacts. In the last 50 years, Tail-sitters
have been analyzed extensively and there were several trials to
build manned ones such as Convair XFY Pogo (Fig. 1e) and
Lockheed XF-V1. However, due to the difficulty of control in
transition and vertical flight, none of the projects was
successful [11] [36]. Tail-sitters can be classified into several
types as follows.
Duct-Fan VTOL UAV is a type of tail-sitter UAVs where a
large duct fan usually coaxial forms the main body of the
aircraft and several control surfaces are installed for stability,
control and transition. It is true that this method is
mechanically simple but the control and stability of such Figure 4 Examples of Tail-sitter UAVs
designs require complex strategies. Payload capacity and
cruising speed are other drawbacks for this design. RMIT III. DYNAMICS MODEL
University ducted fan aimed to aid law enforcement activities Understanding the flight dynamics of the aircraft is an
[37], Vertical Bat developed by Brigham Young University essential part for developing an autonomous flight control
partnering with MLB Company [38], Bertin Hovereye [13], system. Therefore, a reliable mathematical model that captures
Selex Galileo Asio (Fig. 3a) [13] are several examples of duct- the full flight envelope should be established to describe the
fan tail-sitters. vehicle’s orientation and motion in the entire mission profile.
For Hybrid aerial vehicles, the flight envelope consists of three
Apart from Ducted-fans, other tail-sitters can be classified into modes, namely, vertical flight mode, transition mode, and level
those performing the transition using large control surfaces and flight mode. As a result, developing a reliable flight dynamics
those performing that using differential thrust from their model becomes more challenging. Over the past decades, a
multiple rotors. The former will be referred to as Control number of modeling works, highlighted in Table 3, has been
Surface Transitioning Tail-sitters (CSTT) and the latter will be documented in literature. This section shows a brief overview
referred to as Differential Thrust Transitioning Tail-sitters of the listed dynamics modeling works via three aspects: 1)
(DTTT). Both require complicated control mechanisms and model structure of the hybrid UAV dynamics, 2) representative
higher use of available power to maintain stability during modeling work and 3) the associated key observations.
vertical flight because they are more susceptible to cross winds
which also makes landing on moving decks difficult. CSTTs Table 1
usually require only a single or sometimes twin rotors. In Aircraft Type Representative Modeling Work
contrast, DTTTs require multiple rotors to enable producing
Tilt-Rotor [17, 59, 61, 70-72, 77, 89, 101-103]
sufficient differential thrust to make transitions resulting in
Tilt-Wing [22, 26, 104-107] will address, is derived from Newton’s Laws and Euler’s
CSTT [38, 53, 56, 62, 64, 66-67, 74, 90-91, 93, 97-100] Laws which are expressed in vector equations as:
d
DTTT [43]
 F  dt (mv)
A. Model Structure
d
Based on a comprehensive study on the work listed in
Table 3, it was concluded that the flight dynamics of the M 
dt
(H)
hybrid aerial vehicles can be genericly depicted in Fig. 5, in
which six key components are contained. More specific It should be noted that, corresponding to the different
explanations of these components are provided as follows. orientation expressions (i.e., Euler angle or quaternion), the
First, as hybrid aerial vehicles are only capable of expressions of the combined forces F and moments M differ
operating in a relatively confined area, two main Cartesian accordingly. On the other hand, Euler-Lagrange approach does
coordinate systems, i.e., local coordinate (mainly north-east- not require a particular identification of the coordinate system
down) and body-frame coordinate, are generally sufficient for and makes use of the conservation of energy to derive the
describing the kinematic and dynamic motions of hybrid aerial equations of motion instead. More specifically, the Lagrangian,
vehicles. For the level flight mode, the stability coordinate is L , which sums the total kinetic and potential energy is
additionally taking into account the wind effect. The adopted and expressed as follows
definitions of all these coordinate systems can be easily found
in variety of textbooks (e.g., [51, 108]). L  Ttrans T rot V
where T trans , T rot and V and are the kinetic energy due to
translation, kinetic energy due to rotation and potential energy
of the system respectively. Then, the translational and
rotational equations of motions could be derived from Euler-
Lagrange equation as follows
d L L
 F
dt  q  q
Figure 5 Block diagram of the dynamics of the hybrid where q is the generalized coordinates of the system and F
aerial vehicles represents the combined forces and moments w.r.t. the C.G of
the hybrid aircraft. According to the survey conducted, only
With the coordinate systems defined, the kinematics, i.e., few research works [74, 90, 91, 93] adopt this formulation
orientation and translation relations w.r.t. the operational without highlighting the particular reason for their selection.
environment, of hybrid aerial vehicles can be determined. The Aerodynamic forces (lift and/or drag) and the associated
former can be expressed via two formulations, namely, Euler moments are constantly generated by the various control
angles and quaternions, depending on the specific type of the surfaces and the fuselage of hybrid aerial vehicles in
hybrid vehicles as well as the specified mission. Again, operation. Compared with the conventional fixed-wing or
detailed explanations of both expressions can be found in rotary aircrafts, hybrid vehicles dynamic modeling is more
textbooks such as [51]. Generally, Euler angle representation complex as the aerodynamic features of the aforementioned
dominates the convertiplanes and single mode (e.g., hover) of components in different flight conditions or envelopes should
partial tail-sitters because the fuselage does not change with be taken into account comprehensively. A prevalent method
large amplitude. On the other hand, the tail-sitters operating dominantly adopted by the documented works is to express the
over full flight envelope commonly adopt quaternion aerodynamic forces and moments using dimensionless
representations because the transition between hover and level coefficients which are determined via practical flight
flight modes leads to approximately 90-deg pitch angle change experimental data as in [97], wind tunnel experimental data as
which has a high chance to raise the singularity of Euler-angle in [26, 103-104], or CFD results as in [17, 53, 102].
representation. Representative implementation of quaternion Regarding the dynamics of the propulsion system, two
coupled sub components which are the propeller aerodynamics
expression on hybrid aerial vehicles will be addressed later in
and the motor dynamics are involved. For the former, the
Section III.B.
majority of the modeling works on hybrid aerial vehicles
The second component to be addressed is the rigid body
adopt highly simplified dynamics model which involves very
dynamics which concerns the translational and rotational fundamental aerodynamic analysis. For instance, in [97-98]
equations of the hybrid vehicles. Newton-Euler and Euler- only quasi-steady equations are utilized to model the
Lagrange formulations are generally employed to achieve this aerodynamic forces and moments. For the latter, according to
aim. The former, which has been widely utilized in the the survey conducted, no research work has particularly paid
majority of the hybrid aircraft modeling works as section III.B attention to the motor dynamics. Thus, the response of
propulsion systems to the actuator input is assumed dynamics and tilting mechanisms as mentioned in Section
instantaneous. III.A. One distinguished feature shared by all the documented
Finally, the titling mechanism uniquely belongs to work on tilt-wing hybrid aircraft listed in Table 3 is that wind
convertiplane aircraft. Thus, in Fig. 5 it is displayed in a tunnels are uniformly used to determine the aerodynamic
dashed block. Similar to the propeller aerodynamics part, the coefficients involved in the developed models.
current hybrid aerial vehicle modeling works adopt highly A number of research works on CSTTs, which is either
simplified models to account for the titling motion of the single-rotor- or bi-rotor-based, have been carried out and
propulsion systems. For instance, a common method has been documented in literature. Part of them only focuses on vertical
documented in [59, 70, 77] in which two instantanous shaft flight mode and attitude stabilization. For instance, in [74, 90-
tilting angles, and , are defined for the rotation of the left 91, 93], two types of bi-rotor CSTTs have been developed and
and right front propulsion systems and the tilting motion is Euler-Lagrange formulation is adopted in modeling their
reflected by a rotation matrix based on them. dynamics. As attitude stabilization in hover model is the focus,
Euler angle instead of quaternion is used for more
B. Representative Modeling Work
straightforward attitude representation. The rest of the
Table 3 provides a complete list of the documentations modeling works for CSTTs address the dynamics model
related to the dynamics modeling of hybrid aerial vehicles covering the full envelope, that is, hover, transition, and level
following the categorization method introduced in Section II. flight. More specifically, [38, 53, 56 62, 66, 98, 100]
According to review conducted, no systematic modeling work concentrate on the modeling of a single-rotor hybrid aerial
on 1) rotor/wing, 2) dual system, and 3) reconfigurable hybrid vehicles. All these works adopt 1) quaternion formulation for
aircrafts have been documented in the literature, thus in Table avoiding the singularity in pitch angle expression and 2)
3 only the remaining four sub-categories are listed. The focus simple expression for propulsion systems. In order to enhance
of this section is to analyze some unique features of the the accuracy of the proposed model, additional effort has been
modeling work given in Table 3. made in some documented works, mainly on motor dynamics
Starting with the tilt-rotor hybrid aerial vehicles, most of and aerodynamic coefficients determination. For instance, in
the works (e.g., [59, 61, 70, 77] for bi-rotor convertiplane, [71, [53], ducted-fan design code is employed to account for the
103] for tri-rotor convertiplane, and [89] for quad-rotor unique duct fan feature of the custom-made CSTT developed
convertiplane) employ highly simplified motor dynamics and at the KAIST and Navier-Stokes solver integrated in FLUENT
titling mechanisms to minimize the complexity of the overall toolkit is used to determine aerodynamic control coefficients.
model. An exception that can be treated as a benchmark is the In another two documentations [62, 100] based on a miniature
modeling work documented in [103], in which a fairly single-rotor hybrid UAV developed at BYU, aerodynamic
complete flight dynamics model for a bi-rotor convertiplane coefficients are determined by maximally matching the flight
has been proposed. The propulsion system is modelled in test data collected in experiments. Furthermore, [62] also
depth by introducing additional coordinate systems (such as addresses a technique of modeling the angular dynamics as a
Nacelle axis system, hub-axis system, and blade axis system) combination of one bias acceleration term and one actuator-
and including the flapping motion of the propellers. based input term which aims at reducing the computational
Furthermore, the aerodynamics of the control surfaces and load of physical parameter estimation. However, except the
fuselage are carefully determined via variety of wind-tunnel work documented in [56] which provides identification results
experiments. Model validation in both time- and frequency- for the longitudinal motions, none of the aforementioned
domains is presented and the results indicate the relative high works have included results on model fidelity validation. On
fidelity of the proposed model. In another work documented in the other hand, research works documented in [64, 67, 97, 99]
[101], the essential role of the wind tunnel usage in focus on the modeling of bi-rotor hybrid aerial vehicles.
determining various aerodynamic coefficients is clearly Quaternion expression is dominantly adopted and certain
demonstrated via both large amount of data and model unique features such as variable pitch propeller [50] and motor
validation results. Instead of using the experimental results dynamics [67, 99] are additionally considered aiming at
collected in the wind-tunnel, the authors of [17, 102] have covering the key dynamic features. Validation results and
explored the possibility of using CFD to determine the analysis are again rarely addressed with the exception of [97],
aerodynamic coefficients for a 0.15-scale MV-22 bi-rotor in which a comparison between the model responses and
convertiplane and a custom-built tri-rotor convertiplane actual experimental data is conducted and non-ignorable
TURAC respectively. Partial validation results have also been deviations have been observed for all channels which indicates
presented in [102] to prove the efficiency of the CFD-based that the model accuracy can be further enhanced.
estimation. For DTTT, very rare work on dynamical modeling has
Compared with tilt-rotor hybrid aircraft, less interest in been documented in the literature, as DTTT-based UAV is
modeling tilt-wing hybrid aircraft has been observed. still a relatively new topic to the academia and very less
Furthermore, [26, 104-106] are based on an identical custom- systematic research has yet been conducted. One
built miniature quad tilt-wing hybrid UAV and only one representative work on DTTT dynamics modeling is presented
modeling work on twin tilt-wing hybrid aircraft [22] has been in [43], in which a quaternion-based Newton-Euler
found. All the proposed models adopt Euler-angle expression, formulation model is proposed for the custom-made
Newton-Euler formulation, and highly simplified motor
quadcopter tail-sitter named VertiKUL. However, no detailed the low-level controllers (innermost) should have a quicker
description of the aerodynamic coefficients determination and response than the higher ones. Fig Y shows the block diagram
propeller dynamics is contained, which makes the validation of successive loop closure. For a particular case of Hybrid
of model fidelity difficult. UAVs, the decomposition of the states could result in a low-
level attitude controller directly linked to the control surfaces
IV. HYBRID UAV FLIGHT CONTROL TECHNIQUES of the UAV. It gets the reference commands from a mid-level
Flight control systems are required to regulate the UAVs’ velocity controller which gets the reference commands from
motion to complete their flight missions successfully. This is the high-level position controller as shown in Fig. 7 [48]. This
done in terms of delivering the desired performance and control theory was implemented in Ducted-Fan Tail-sitters as
following the desired path with sufficient accuracy in the detailed in [73], [38], [53] and [54], Tilt-Rotors and Tilt-Wings
presence of wind and other external disturbances. In general, as detailed in [25-26], [57-58] and [86] and other tail-sitters,
the goal of the control system is to stabilize the UAV as well as mainly for hovering, as detailed in [43], [55] and [56].
to minimize the tracking error between the desired reference
command and the measured response of the UAV. Therefore,
flight control and stability is a very critical part of the UAV
because the failure of the system due to components
malfunctioning, lack of robustness or even improper design
will lead to deprived mission performance or high risk of
damage. Figure 7 Successive loop closure [48]
The core of the control system depends on the derived For the classical control theory, the reference commands
dynamics model. As seen in section III of the paper, the are saturated through the outer loops prior to reaching the
equations of motion are highly complicated and nonlinear. innermost loop which mean that this approach is it results in
Particularly speaking, the dynamics of the Hybrid UAVs can good handling of the flight variables and actuator inputs.
be inherently unstable because it inherits the operation of a Moreover, every single control loop design is simple as it
fixed-wing and VTOL UAVs. Even if horizontal and vertical involves few variables and sometimes even single variable.
modes were analyzed separately, the transition phase remains a However, the difficulty arises when decomposing and
critical part of the control system due to the multiple separating the variables for the controllers and also when
nonlinearities in the model. That is why feedback control is linking the controllers together especially in terms of
essential which ensures more accurate and quicker response to determining whether the inner loops are faster than the outer
meet the desired reference command. A typical feedback ones. The design algorithms of this classical approach depend
control loop is given in Fig. 6 [48]. on one loop at a time and are very effective for Single-Input
Single-Output (SISO) systems. However, complex systems
such as Hybrid UAVs are Multi-Input Multi-Output (MIMO)
systems. Therefore, the success of the classical control system
when applied to Hybrid UAVs is not guaranteed because it
comprises multiple actuators such as elevators, rudders,
ailerons, motors and tilting rotors and multiple sensors for
velocity, position and attitude [48][50-52].
Figure 6 Feedback Control Loop [48] The second approach, known as modern control, is to
design a control system that handles the full dynamics of the
Before classifying the different approaches of flight UAV. The stability and control specifications can be expressed
control systems in Hybrid UAVs, it is important to note that in terms of a system of first-order differential equations which
there are several specifications, requirements and constraints results in matrix equations that can be solved using
for each control system depending on the specified mission commercially available computer software to compute the
and the UAV’s model. Therefore, evaluating control systems control gains simultaneously [50-52]. This means that all the
depends on the state variables, the UAV actuators and sensors, feedback loops are closed at the same time. Therefore, a better
the desired response and the likely disturbances. Based on performance for MIMO systems is achieved compared to the
that, some control systems might seem poor for certain UAVs classical approach in which the control gains are selected
but perfect for others. individually. This quick and direct modern control approach
can be utilized for time varying and time invariant systems,
A. Flight Control System Theory whereas, the classical approach is mainly for time invariant
This deals with the synthesis and analysis of the logic systems [79]. Also, there are many optimal control techniques
behind which the flight control system is designed. There are that could be applied in the modern control theory to improve
two strategies for that, namely, the classical control theory and the controllability and stability of the UAV [73]. However, not
the modern control theory. The former, also known as all states correspond directly to a single actuator as was the
successive loop closure, considers the decomposition of the case for the first approach. Therefore, it is difficult to handle
states derived from the model to form successive control loops actuator saturation [48]. However, modern control theory is
such that the output of the innermost loop (low-level) is linked also popular in Hybrid UAVs as it was employed in Tilt-rotors
to the actuators of the UAV [48-50]. It is important to note that as described in [59], [61], [60] and [77], Tail-sitters as detailed
in [67], [55], [62] and [65] and Ducted-Fans as detailed in [53- (PI) and Proportional (P). The controller gain values are
54] and [57]. determined by empirical tuning until some preconceived ideal
response of the system is achieved. Those gains are sometimes
B. Control Laws Classification estimated theoretically using Ziegler and Nichols method to
Apart from that, flight control systems can be classified reduce the amount of tuning [50]. It is important to note that
into linear and nonlinear based on the dynamics of the Hybrid the PID controllers can be implemented in Hybrid UAVs for
UAV model. As previously mentioned, Hybrid UAVs’ models altitude control, attitude angles control and velocity control by
are nonlinear. However, those models are commonly just changing the control gains accordingly. Since PID control
linearized using relative equilibrium conditions. Although strategy only requires appropriate adjustment of the control
linear controllers are simple, easy to implement, reduce the gains, it serves as a concrete starting design point for many
computational effort and minimize the design time but their Hybrid UAVs as it does not require extensive knowledge of the
performance degrade when operating away from the local model. However, PID controller is applicable only for SISO
equilibrium point or while performing agile maneuvers. This systems, therefore it does not account for the cross coupling
is very critical during the transition flight for the case of effects present in UAVs. For such cases, multiple independent
Hybrid UAVs because changing from vertical flight mode to PID controllers are usually utilized in Hybrid UAVs such as in
[43], [53-55], [64-65], [68], [70], [73] and [75].
horizontal flight mode and vice versa results in operation far
away from the relative equilibrium condition. That is the 2) Linear Quadratic Regulator (LQR) Controller: LQR
reason behind which some current Hybrid UAVs implement controller’s goal is to find a control input of the form, that
nonlinear controllers such as [53-55], [59], [60-63], [67], minimzes the performance index,  , which is given by
[75],[77] and [86] or three separate linear controllers, one for tf
1
2 t 
the horizontal mode, one for the vertical mode and one for the   x T (t )Qx (t )  uT (t ) Ru(t )  dt
transition such as [25-26], [38], [55], [75-76].
0
Nonlinear controllers are extensively studied and
investigated theoretically for application in Hybrid UAVs but subject to
in terms of implementation, linear controllers are far more x(t )  Ax(t )  Bu(t )
popular. However, nonlinear controllers operate in a much where x (t )  n
is the (n×1) state vector, u(t )  m
is the
wider profile than linear ones which are restricted within a
specific operating region. Also, they consider the full and true (m×1) control input vector, A is the system matrix, B is the
dynamics of the UAV and account for the nonlinear control influence matrix, R and Q are real positive
aerodynamic and kinematic effects, actuator saturations and weighting matrices. The feedback control input is of the form,
rate limitations [109]. u (t )  G (t )x (t ) .
The classical PID controller and the Linear Quadratic By applying the necessery and sufficient conditions of the
Regulator (LQR) are the most common linear control laws optimility, the control signal is given by
applied in Hybrid UAVs while the backstepping, gain-
scheduling and dynamic inversion are the common nonlinear u(t )   R1 (t ) BT (t ) P(t ) x(t )
laws. where P (t ) is known as the riccati matrix that is found by
solving the general matrix differential riccati equation given
1) Proportional-Integral-Derivative (PID) Controller: by
PID controllers are very common in the UAV field. [25-26],
[38], [43], [53], [55-56], [58], [63-64], [68] and [70-76] have
P  A T P  PA  PBR 1B T P Q
applied PID controllers in their Hybrid UAV. The PID control [87][96]. The only design freedom in this approach is
law consists of a proportional, integral and derivative choosing the weighting matrices. Bryson’s Rule is usually
elements. The general mathematical description of the PID used to select those weighting matrices based on normalizing
controller is the signals [85] [48]. Therefore, LQR easily handles complex
t dynamic systems and multiple actuators [48]. It is robust with
de
u (t )  k pe (t )  k i  e ( )d   k

d
dx
(t ) respect to process uncertainty, asymptotically stable given that
the system is at least controllable and has very large stability
margins to errors in the loop (gain margin of infinity for gain
where u (t ) is the control signal, e (t ) is the tracking error
increase and -6 dB for gain decrease and phase margin of 60°
for each control signal) [48] [80]. On the other hand, LQR
between the reference command and output from the sensors
requires access to the full state which is not always possible
k k k [48]. Due to their robustness, LQR controllers are extremely
and p , i and d are the proportional, integral and
derivative gains respectively. When utilizing the PID control suitable for Hybrid UAV flight control systems and they are
law algorithm, it is essential to decide which of these three implemented in [55] [57] [66-67] [69].
elements are to be used since each has particular effect on the
control signal as given in [48], [50] and [73]. Generally, for 3) Backstepping: In Hybrid UAVs, there are several
Hybrid UAVs, three main controllers are implemented: parameters that irregularly change especially during the
Proportional-Integral-Derivative (PID), Proportional-Integral transition between horizontal and vertical flight modes.
Therefore, a controller that handles these parameters changes which is a linear relation and therefore can be used to control
might be necessary. One example is the backstepping the system easily. Here, a SISO system is considered for
controller which is based on Lyapunov stability and provides a simplicity, however, the concept could be generalized for
powerful recursive approach for nonlinear systems that can be MIMO systems as detailed in [83]. An example of this
transformed into triangular form. The key idea is to let certain approach is given in fig 8 which illustrates how the NDI
states act as virtual controls of others. For a system linearizes the inner loop making the dashed box a linear
system [82].
x  f 0 (x )  g 0 (x )z 1
z i  f i (z 1 , z 2 ,..., z n )  g i (z 1 , z 2 ,..., z n ) z i 1
where z represent the general coordinates of the aircraft,
i  1,..., n and z n 1  u . The goal is to desgin a feedback
control law to stabilize z . The detailed backstepping
procedure is described in [81-82] [109]. This method is
beneficial for Hybrid UAVs since it takes into consideration
Figure 8 NDI [82]
all the states of the system and accounts for the nonlinearities
Unlike Gain-scheduling, a single controller is required for
present in the model. From the literature review conducted, it
the full flight envelope. Another advantage of this method is
was observed that backstepping method was mostly coupled
that closed loops can be easily tuned as in PID controllers.
with using Euler-Lagrange approach for the dynamic
However, to apply this method a precise knowledge of the
modeling as in [60-62] [77][88].
aerodynamic coefficients is necessary [82]. From the review,
it was noted that NDI was as common and effective as gain
4) Gain-Scheduling: As previously mentioned, the UAV scheduling. It was implemented and tested in many Hybrid
dynamics could be linearized by applying relative equilinrium UAVs such as in [53-54] [59] [63].
conditions around a steady state operating point and then 6) Other control strategies: There are several other
applying linear controllers such as PID or LQR. However, the controlling methods implemented in Hybrid UAVs. [62-63]
controller performance degrades effectively when deviating and [71] apply adaptive control techniques which accout for
away from that point. Therefore, a prevailing control design the nonlinearities and uncertainities present in the model. J.A.
approach, known as gain-scheduling, is to divide the flight Guerrero et al. [95] presents a robust control desingn based in
envelope into a finite number of small partitions. For each sliding mode of a mini birotor tail-sitter for the hovering
small region, the UAV dynamics is linearized around a mode. The work in [94] shows the control of hovering flight
corresponding steady state operating point. Then, linear and vertical landing using optical flow. Fault tolerant flight
controllers such as PID or LQR each having different control system for a tilt-rotor UAV was discussed by S. Park
consistent control gain values could be applied for each small et al. in [69]. Moreover, other control strategies based on
region effectively. Therefore, this allows easy understanding Lyapunov stability concepts can be found in [88-93].
and simple implementation of the control laws over the full
flight envelope. However, since for each small region, a linear V. CONCLUSION
controller has to be designed, this method might be tedious A novel technical overview of Hybrid UAVs platforms is
and time conusming [48][82]. For Hybrid UAVs, Gain- presented. The common platform design types and technical
scheduling was mainly utilized to enhance the control during details are introduced first with the advantages and
transition as in [51], [53-55], [67] and [75]. disadvantages of most common conceptual platforms designs;
5) Nonlinear Dynamic Inversion (NDI): The dynamic convertiplanes and tail-sitters. The modeling is then explained
model of a SISO system can be written in companion form as in terms of kinematics, dynamics and simulation. Similar to
 x1   x 2   0  other aerial vehicles, the dynamic nature of hybrid UAVs
      results in nonlinear model. Several flight control strategies
   u implemented in Hybrid UAVs are then discussed and
 x n 1   x n   0  compared in terms of theory, linearity and implementation.
      The application of control theory on the hybrid UAV is highly
 x n  b (x )  a (x )  linked with the specified mission and the UAVs model. The
review presented in this paper is expected to be broadly useful
In this form, b (x ) and a (x ) are functions of the state
for UAVs applications.
vector x which linearize the nth state which is the only state
affected by the input u . All other elements of the state vector REFERENCES
derivative are linear. In a similar manner to the backstepping [1] P. Hindle, “UAVs unleashed,” Microwave Journal, Military Micorwave
approach, a virtual control input v could be defined as Supplement, pp. 8-22, August 2013.
v  b (x )  a(x )u [2] “UAV market by class, subsystem, application, funding & by aayload -
forecast & analysis to 2014 – 2020,” MarketsandMarkets, October 2014.
[3] J. R. Wilson, “UAV roundup 2013,” Aerospace America, vol. 51, no. 7, [30] A. V. Clara and S. Redkar, “Dynamics of a vertical takeoff and landing
pp. 26-36, July-August 2013. (VTOL) unmanned aerial vehicle (UAV),” in International Journal of
[4] C. Drubin, “UAV market worth $8.3 B by 2018,” Microwave Journal, Engneering and Innovation, vol. 3, no. 1, pp. 52-58, 2011.
Defense News, p. 37, August 2013. [31] J. T. McKenna, “One step beyond,” Rotor & Wing,
[5] “Eyes of the army: U.S. Army roadmap for umanned aircraft systems,” http://www.aviationtoday.com/rw/military/heavylift/One-Step-
U.S. Army UAS Center of Excellence. Beyond_8439.html#.VKrXGCuUeSo, Feb 1, 2007.
[6] J. T. Ping et al.,“Generic unmanned aerial vehicle (UAV) for civilian [32] “Introducing the new arcturus JUMP™ fixed wing VTOL UAV,”
application,” in 2012 IEEE Conf. Sustainable Utilization and Acrturus UAV, http://www.arcturus-uav.com/aircraft_jump.html.
Development in Engineering and Technology, pp. 289-294, Oct. 6-9, [33] R. Park, “Arcturus UAV upgrades the JUMP15 VTOL UAV,”
2012. http://gisuser.com/2014/12/arcturus-uav-upgrades-the-jump15-vtol-
[7] “Unmanned aircarft systems for civilian missions,” Brandenburg uav/?print=pdf, Dec 4, 2014.
Institute for Society and Security, February 2012. [34] Arcturus UAV, “The Arcturus UAV JUMP™15 electric,”
[8] “V-22 Osprey,” Boeing Defense, Space and Security. http://www.youtube.com/watch?v=omkay1cJmXI
[9] D. C. Dugan, “Thrust control of vtol aircraft – part deux,” in the 5th [35] R. Tomkins, “Airbus' VTOL Quancruiser UAV successfully transitions
Decennial AHS Aeromechanics Specialists’ Conf. , p. 4, Jan. 22-24, to fixed-wing flight,” Space Daily,
2014. http://www.spacedaily.com/reports/Airbus_VTOL_Quancruiser_UAV_s
uccessfully_transitions_to_fixed-wing_flight_999.html, May 23, 2014.
[10] J. Richmond, “It’s a helicopter! It’s a plane,” Military Aerospace
Technolgy, High Technology, Nov. 1985. [36] R. H. Stone, “The T-Wing: a VTOL UAV for defense and civilian
applications,”.
[11] “Rollout week,” United States Navy Centinnel Newsletter, p.12, 2009.
[37] Z. Omar, C. Bill, and R. Hill, “The development of a new VTOL UAV
[12] B. Handy, “Harrier GR7,” Royal Air Force Aircraft and Weapons, pp. 8- configuration for law enforcement,” in the Proc. of Int. Conf. on
9.
Mechanical & Manufacturing Engineering (ICME 2008), Johor Bahru,
[13] M. Streetly, IHS Jane’s all the world aricraft: Unmanned 2013-2014, Malaysia, May 21-23, 2008.
IHS, 2013.
[38] M. E. Argyle, R. W. Beard, and S. Morris, “The Vertical Bat Tail-sitter:
[14] “Mini panther fixed wing VTOL mini UAS,” dynamic model and control architecture,” in 2013 American Control
http://www.iai.co.il/2013/35673-41637-en/IAI.aspx Conf. (ACC), Washington, DC, USA, June 17-19, 2013.
[15] U. Ozdemir et al., “Design of a commercial hybrid VTOL UAV [39] M. Aksugur and G. Inhalan, “Design methodology of a hybrid
system,” in Journal of Intelligent & Robotic Systems, vol. 74, pp. 371- propulsion driven electric powered miniature unmanned aerial vehicle,”
393, 2014. in Journal of Intelligent & Robotic Systems, vol. 57, pp. 502-529, 2010.
[16] Y. Aktas et al., “A low cost prototyping approach for design analysis [40] R. H. Stone and K. C. Wong, “Preliminary design of a Tandem-Wing
and flight testing of the TURAC VTOL UAV,” in the International tail-Sitter UAV using multi-disciplinary design optimisation,” 2006.
Conf. on Unmanned Aircraft Systems (ICUAS), Orlando, FL, USA, May [41] R. H. Stone and G. Clarke, “The T-Wing: a VTOL UAV for defense and
27-30, 2014. civilian applications”.
[17] A. Vuruskan et al, “Dynamic modeling of a VTOL fixed-wing UAV,” in [42] A. Oosedo, “Design and simulation of a quad rotor tail-sitter unmanned
Int. Conf. on Unmanned Aircraft Systems (ICUAS), Orlando, FL, USA, aerial vehicle”.
May 27-30, 2014.
[43] M. Hochstenbach and C. Notteboom, “Design and control of an
[18] U. Ozdemir et.al, “Design of a commercial hybrid VTOL UAV unmanned aerial vehicle for autonomous parcel delivery with transition
System,” in the Int. Conf. on Unmanned Aircraft Systems (ICUAS), from vertical take-off to forward flight”.
Atlanta, GA, USA, May 28-31, 2013.
[44] S. Hulsman, J. de Groot, and D. Dokter, “ATMOS UAV,” in Leonardo
[19] S. Carlson, “A hybrid tricopter/flying wing VTOL UAV,” in 52nd Times, March 2013.
Aerospace Sciences Meeting, AIAA SciTech, Jan. 13-17, 2014.
[45] Tara, “Interview with the inventors of the Quadshot a remote controlled
[20] “FireFLY6,” http://www.birdseyeview.aero/products/firefly6
aircraft,” http://ideasuploaded.com/interview-with-the-inventors-of-the-
[21] M. Hirschberg, “Project zero: The exclusive story of AgustaWestland’s quadshot-a-remote-controlled-aircraft/
all electric technology incubator,”
[46] B. Coxworth, “Quadshot RC aircraft combines quadricopter hovering
[22] J. Holsen, T. Ostermann, and D. Moormann, “Design and wind tunnel with airplane flight,” http://www.gizmag.com/quadshot-hovers-and-
tests of a tilt-wing UAV,” in CEAS Aeronaut J, vol. 2, pp. 69-79, 2011. flies/19449/, August 8, 2011.
[23] T. Ostermann et al., “Control concept of a tiltwing UAV during low [47] K. Z. Y. Ang et al., “Development of an unmanned tail-sitter with
speed manoeuvring,” in the 28th Int. Congress of the Aeronautical reconfigurable wings: U-Lion”, in 2014 11th IEEE Int. Conf. on Control
Sciences, 2012. & Automation (ICCA), Taichung, Taiwan, June 18-29
[24] T. Ostermann et al., “Model validation of a tiltwing UAV in transition [48] J. P. How, E. Frazzoli, and G. Chowdhary, “Linear flight contol
phase applying windtunnel investigations,” in the 28th Int. Congress of techniques for unmanned aerial vehicles,” in Handbook of Unmanned
the Aeronautical Sciences, 2012. Aerial Vehicles, Springer, 2012.
[25] E. Cetinsoy et al., “Design and development of a tilt-wing UAV,” in [49] R. W. Beard, and T. W. McLain, “Autopilot design using successive
Turk J Eng & Comp, vol. 19, no. 5, pp. 733-741, 2011. loop closure,” in Small Unmanned Aircraft: Theory and Practice,
[26] E. Cetinsoy et al., “Design and construction of a novel quad tilt-wing Princeton University Press, 2012, ch. 6, pp. 95-119.
UAV,” in Mechatronics, vol.22, pp. 723-745, 2012. [50] R. C. Nelson, Flight Stability and Automatic Control, 2nd ed., McGraw
[27] K. Muraoka, N. Okada and D. Kubo, “Quad tilt wing VTOL UAV: Hill, ch. 7-10, pp. 236-394.
aerodynamic characteristics and prototype flight test,” in AIAA [51] B. L. Stevens and F. L. Lewis, Aircraft Control and Simulation, Wiley,
Aerospace Conf., Seattle, WA, USA, April 6-9, 2009. 2003.
[28] J. J. Dickeson, “Robust LPV H∞ gain-scheduled hover-to-cruise [52] W. S. Levine, “Optimal control,” in The Control Handbook, CRC Press,
conversion for a tilt-wing rotorcraft in the presence of CG variations,” in 1996, ch. 48, pp. 759-777.
the Proc. of the 2007 American Control Conference, New York City, [53] Y. Jung, D. H. Shim, and N. Ananthkrishnan,” Controller synthesis and
USA, July 11-13, 2007.
application to hover-to-cruise transition flight of a Tail Sitter UAV,” in
[29] “High speed tilt-rotoer ousts X-Wing project,” Flight Internation AIAA Atmospheric Flight Mechanics Conf., Toronto, Canada, Aug. 2-
Technology, Jan 23, 1988. 8,2010.
[54] N. Ananthkrishnan and H.C. Shim,”Controlled near-hover to cruise
transition using a dynamic inversion law,” AIAA.
[55] R. H. Stone, “Control architecture for a Tail-Sitter Unmanned Air [79] K. Ogata, “PID controllers and modified PID controllers,” in Modern
Vehicle” Control Engineering, 5th ed., ch. 8, pp. 567-647, Prentice Hall, 1970.
[56] P. R. Bilodeau and F. Wong, “Modeling and control of a hovering mini [80] R. Vepa, “Aircraft flight control,” in Flight Dynamics, Simulation, and
Tail-sitter,”in Int. Journal of Micro Air Vehicles, vol. 2, no. 4, pp-211- Control for Rigid and Flexible Aircraft, CRC Press, ch. 8, pp. 333-486,
220, 2010. 2014.
[57] O. Tekinalp, T. Unlu, and I. Yavrucuk, “Simulation and flight control of [81] A. J. Koshkouei and A. S. Zinober, “Adaptive backstepping control of
a tilt duct UAV,” in AIAA Modeling and Simulation Conference, nonlinear systems with unmatched uncertainity”
Chicago, Illinois, Aug. 10-13, 2009. [82] O. Harkegard, “Flight control design using backstepping”, Dept. of Elec.
[58] S. Yanguo and W. Huanjin, “Design of flight control system for a small Eng, Linkopings universitet, 2011.
unmanned tilt rotor aircraft,” in Chinese Journal of Aeronautics, vol. 22, [83] “Nonlinear dynamic inversion”
pp. 250-256, 2009.
[84] S. F. Campell and J. T. Kaneshige, “A nonlinear dynamic inversion L1
[59] X. Fang et al., “Control strategy design for the transitional mode of adaptive controller for a generic transport model”
tiltrotor UAV,” IEEE, 2012.
[85] A. E. Bryson and Y. C. Ho, Applied Optimal Control, Blaisdell
[60] A. B. Chowdhury, A. Kulhare, and G. Raina, “A generalized control Publishing Company, Waltham, 1969.
method for a tilt-rotor UAV stabilization,” in the Proc. Of the 2012
IEEE Int. Conf. on Cyber Technology in Automation, Control and [86] K. Muraoka et al., “Transition flight of quad tilt wing VTOL UAV,” in
Intelligent Systems, Bangkok, Thailand, May 27-31, 2012. the 28th Int. Congress of the Aeronautical Sciences, 2012.
[61] F. Kendoul, I. Fantoni, and R. Lozano,”Modeling and control of a small [87] J. L. Junkins and Y. Kim,”Design of linear state feedback control
autonomous aircraft having tow tilting rotors,” in the Proc. Of the 44th systems,” in Introduction to Dynamics and Control of Flexible
IEEE Conf. on Decision and Control, Seville, Spain, Dec. 12-15, 2005. Structures, AIAA, Washington, ch. 5, pp. 235-296, 1993.
[62] N. B. Knoebel and T. W. McLain, “Adaptive quaternion control of a [88] G. R. Flores and R. Lozano,”A nonlinear control law for hover to level
miniature tailsitter UAV,” in American Control Conference (ACC), flight for the quad tilit-rotor UAV,” in the Preprints of the 19th World
Seattle, WA, USA, June 11-13, 2008. Congress, The international Federation of Automatic Control, Cape
Town, South Africa, Aug. 24-29, 2014.
[63] Y. Jung and D. H. Shim, “Development and application of controller for
[89] G. R. Flores and R. Lozano,”Transition flight control of the quad-tilting
transition flight of tail-sitter UAV,” in Journal of Intelligent & Robotic
Systems, vol. 65, pp. 137-152, 2012. rotor convertible MAV,” in 2013 International Conference of
Unmanned Aircraft Systems, Atlanta, GA, USA, May 28-31, 2013.
[64] K. C. Wong et al., “Attitude stabilization in hover flight of a mini tail-
[90] A. Sanchez et al.,”Autonomous hovering of a noncyclic tilt-rotor UAV:
sitter UAV with variable pitch propeller,” in 2007 IEEE/RSJ Int. Conf.
on Intelligent Robots and Systems, San Diego, CA, USA, October, 2007. modeling, control and implementation,” in the Proc. off the 17th World
Congress, The international Federation of Automatic Control, Seoul,
[65] A. Oosedo et al., “Design and simulation of a quad rotor tail-sitter Korea, July. 2, 6-11, 2008.
unmanned aerial vehicle,” IEEE, 2010.
[91] J. Escareno, S. Salazar and R. Lozano,”Modeling and control of a
[66] A. Frank et al.,” Hover, transition, and level flight control design for a convertible VTOL aircraft,” in the Proc. of the 45th IEEE Conference on
single-propeller indoor airplane,”in AIAA Guidance and Control Conf. Decision & Control, San Diego, CA, USA, December 13-15, 2008.
and Exhibit, Hilton Head, SC, USA, Aug. 20-23, 2007.
[92] J. Escareno, S. Salazar and R. Lozano,”Modeling and control of a
[67] R. Casau,” Autonomous transition flight for a vertical take-off and convertible VTOL aircraft,” in the Proc. of the 2006 IEEE Int.
landing aircraft” Conference on Control Applications, Munich, Germany, October 4-6,
[68] T. Matsumoto et al, “A hovering control strategy for a tail-sitter VTOL 2006.
UAV that increases stability against large disturbance,” in 2010 Int. [93] J. Escareno et al.,”Modeling and global control of the longitudinal
Conf. on Robotics and Automation, Anchorage, AK, USA, May 3-8, dynamics of a coaxial convertible mini-UAV in hover mode,” in Journal
2010. of Intelligent & Robotic Systems, vol. 54, pp. 261-273, 2009.
[69] S. Park et al, “Fault tolerant flight control system for the tilt-rotor [94] B. Herisse,”Hovering flight and vertical landing control of a VTOL
UAV,” Journal of the Franklin Institute, vol. 350, pp. 2535-2559, 2013. Unmanned Aerial Vehicle using Optical Flow”
[70] C. Papachristos, K. Alexis and A. Tzes, “Design and experimental [95] J. Guerrero et al.,”Robust control design based on sliding mode control
attitude control of an unmanned tilt-rotor aerial vehicle,” in 15th Int. for hover flight of a mini tail-sitter Unmanned Aerial Vehicle”
Conf. on Advanced Robotics, Tallinn, Estonia, June 20-23, 2011.
[96] F. L. Lewis and V. Syrmos,”Optimal control of continuous –time
[71] D. A. Ta, I. Fantoni, R. Lozano, “Modeling and control of a tilt tri-rotor systems,” in Optimal Control, 2nd ed., ch. 3, pp. 129-214, John Wiley
airplane,” in 2012 American Control Conf., Montreal, Cananda, June 27- and sons, Canada, 1995.
29, 2012.
[97] J. T. VanderMey, “A Tilt Rotor UAV for Long Endurance Operations in
[72] C. Papachristos and A. Tzes, “Modeling and control simulation of an Remote Environments,” M.Sc thesis, Massachusetts Institute of
unmanned tilt tri-rotor aerial vehicle,” IEEE, 2012. Technology, Department of Aeronautics and Astronautics, 2011.
[73] A. A. Manouchehri, H. Hajkarami and M. S. Ahmadi,”Hovering control [98] S. Kohno and K. Uchiyama, “Design of robust controller of Fixed-Wing
of a ducted fan VTOL UAV based on PID control,” IEEE, 2011. UAV for transition flight,” in the 2014 Int. Conf. on Unmanned Aircraft
[74] J. Escareno, S. Salazar and R. Lozano, “Attitude stabilization of a Systems, Orlando, FL, USA, May 27-30.
convertible mini birotor,” in the Proc. of the 2006 IEEE Int. Conf. on [99] P. Casau, D. Cabecinhas and C.Silvestre, “Hybrid control strategy for
Control Applications, Munich, Germany, Oct. 4-6, 2008. the autonomous transition flight of a Fixed-Wing aircraft,” in IEEE
[75] T. Ostermann, J. Holsten, Y. Dobrev, D. Moormann, “Control concept transactions on control systems technology, vol. 21, no. 6, November
of a tiltwing UAV during low speed manoeuvring,” in 28th Int. Congress 2013.
of the Aeronautical Sciences (ICAS). [100] S. R. Osborne, “Transitions between hover and level flight for a tailsitter
[76] J. L. Forshaw, V. J. Lappas and P. Briggs, “Transitional control UAV,” M.Sc thesis, Brigham Young University, Department of
architecture and methodology for a twin rotor tailsitter,” in Journal of Electrical and Computer Engineering, 2007.
Guidance, Control and Dynamics, vol. 37, no. 4, pp. 1289-1298, July- [101] C. Abdollahi, “Aerodynamic analysis and simulation of a twin-tail tilt-
Aug. 2014. duct unmanned aerial vehicle,” M.Sc thesis, University of Maryland,
[77] A. B. Chowdhury, A. Kulhare and G. Raina, “Back-stepping control Department of Aerospace Engineering, 2010.
strategy for stabilization of a tilt-rotor UAV,” IEEE, 2012 [102] J. Abras and R. Narducci, “Analysis of CFD modeling techniques over
[78] D. Xue, Y. Chen, D. Atherton, “PID controller design,” in Linear the MV-22 tiltrotor,” presented at the American Helicopter Society 66th
Feedback Control: Analysis and Design with MATLAB, SIAM, ch. 6, pp. Annual Forum, Phoenix, AZ, May 11-13, 2010.
183-235, 2007.
[103] K. M. Kleinhesselink, “Stability and control modeling of tiltrotor [106] K. T. Oner et al. “Mathematical modeling and vertical flight control of a
aircraft,” M.Sc thesis, University of Maryland, Department of Aerospace tilt-wing UAV,” in Turk J Eng & Comp, vol. 20, no. 1, pp. 149-157,
Engineering Department, 2007. 2012.
[104] K. T. Oner et al., “Dynamic model and control of a new quadrotor [107] K. Muroaka, N. Okada and D. Kubo, “Quad tilt wing VTOL UAV:
unmanned aerial vehicle with tilt-wing mechanism,” in Int. Journal of Aerodynamic characteristics and prototype flight test,” in AIAA
Mechanical, Aerospace, Industrial and Mechatronics Engineering, vol. Aerospace Conf., Seattle, WA, USA, April 6-9, 2009.
2, no. 9, pp. 12-17, 2008. [108] G. Cai, B. M. Chen and T. H. Lee, Unmanned Rotorcraft Systems,
[105] K. T. Oner et al., “LQR and SMC stabilization of a new unmanned aerial Advances in Industrial Control Series, Springer, New York, 2011.
vehicle,” in World Academy of Science, Engineering and Technology, [109] G,. Chowdhary et al., “Nonlinear flight contol techniques for unmanned
vol. 3, no. 10, pp. 367-372, 2009. aerial vehicles,” in Handbook of Unmanned Aerial Vehicles, Springer,
2012.