See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/282870437

Dynamics Modeling and Control of Quadrotor Vehicle

Conference Paper · January 2012

CITATIONS READS

10 796

4 authors, including:

Ahmed Elruby Mohamed Elkhatib

Memorial University of Newfoundland 49 PUBLICATIONS   56 CITATIONS   
9 PUBLICATIONS   21 CITATIONS   
SEE PROFILE
SEE PROFILE

Noha Hany El-Amary

Arab Academy for Science, Technology & Maritime Transport, Cairo, Egypt
45 PUBLICATIONS   84 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Modeling View project

Intelligent laser tracking system View project

All content following this page was uploaded by Mohamed Elkhatib on 09 November 2015.

The user has requested enhancement of the downloaded file.

Proceedings of the 15th Int. AMME Conference, 29-31 May, 2012 RC 66

Military Technical College 15th International Conference

Kobry El-Kobba, on Applied Mechanics and
Cairo, Egypt. Mechanical Engineering.

DYNAMIC MODELING AND CONTROL OF QUADROTOR VEHICLE

A. Y. Elruby*, M. M. El-khatib**, N. H. El-Amary+ and A. I. Hashad++

ABSTRACT

The control of Unmanned Aerial Vehicles (UAVs) is a very challenging field of

research especially for Vertical Take-Off and Landing (VTOL) vehicles or aircrafts for
their numerous advantages over the traditional airplanes and due to the rapid
advances that were made in this field with the development of light weight Micro-
Electromechanical System (MEMS) sensors; it has become possible to build an
autonomous model for a light weight quadrotor and to develop various controls for it.
This paper focuses on the mathematical model of a quadrotor vehicle. A CAD model
has been built for estimating mass and inertial properties of the physical model.
Finally a PID controller for the proposed model is introduced then a Simulink model
has been implemented for estimating the response of flight dynamics.

KEY WORDS

Modelling, PID control, quadrotor.

---------------------------------------------------------------------------------------------------------
* Dept. of Mech. Mechatronics, AAST, Cairo, Egypt. Email: [email protected].
** Egyptian Armed Forces. Email: [email protected].
+ Dept. of Elec. Power, AAST, Cairo, Egypt. Email: [email protected].
++ Dept. of Comp. Eng., AAST, Cairo, Egypt. Email: [email protected].
Proceedings of the 15th Int. AMME Conference, 29-31 May, 2012 RC 67

INTRODUCTION

The quadrotor is a Vertical Take-Off and Landing (VTOL) aircraft [1] consists of 4
propellers arranged on x-shape or +-shape. Every arm holds a propeller on its end
as shown in Fig.1. The quadrotor has higher payload to total weight ratio compared
to other aircrafts like single rotor and tandem rotor helicopters , simplicity of control
and has a great manoeuvring attitude which can help in going into several areas
cannot be accessed by traditional airplanes nor helicopters [2]. VTOL craft also
offers direct access to buildings or areas making them a very fast form of transport
between areas, especially those are elevated or have limited access.

The symmetry of the quadrotor body gives simplicity to the controller design as it can
be controlled through varying the speed of the propellers [3]. Each two opposite
propellers rotate in the same direction as shown in figure 1. The direction of rotation
for each propeller reduces the mechanical complexity inherent in helicopters and
other VTOL aircraft. Directional control is produced by individually altering the speed
of the four motors. Collective pitch propellers are not required in a quadrotor design.
This greatly reduces the mechanical complexity [4]. A quadrotor consists of two fixed
pitch clockwise spinning propellers and two counter-clockwise spinning rotors which
diagonally oppose each other as shown in Fig.1. This result that the reactive force of
each propeller being effectively cancelled out by the diagonally opposite propeller’s
reactive component. This eliminates the need for a helicopter tail rotor.

Quadrotor craft has further efficiency and mechanical advantages. They have four
small propellers as opposed to a single large propeller. Having small propellers
reduces the torque on the system which means that the blades can be driven at
higher velocities without producing additional mechanical vibrations and also
increases motor efficiency. Less craft torque and vibration obviously put less stress
on the mechanical components of the craft.

In this paper, a mathematical model for an under actuated six-degrees of freedom (6

DoF) quadrotor aircraft is derived based on Newton-Euler formalism. In addition the
actuation forces are considered by modelling the aerodynamic forces and
coefficients.

The model consists of two main parts, the first one is the translational system of
equations and the second is the rotational system of equations. The quadrotor
aircraft is a highly non-linear, Multi-Input Multi-Output (MIMO), strongly coupled and
under actuated system with only four actuators [5]. This model is derived based on
some assumptions in order to simplify the dynamics of that complex system to be
suitable for simulation. These assumptions are as follows;

• The quadrotor structure is rigid and symmetric

• The propellers are rigid
• The ground effect is neglected
Proceedings of the 15th Int. AMME Conference, 29-31 May, 2012 RC 68

QUADROTOR SYSTEM MODEL

The development of a suitable attitude controller for the quadrotor prototype required
an accurate dynamic model to be developed. A Newtonian modelling method [2]was
chosen to define the quadrotor dynamics for control purposes. The Newtonian
method is the most popular choice for modelling rigid bodies in six degrees of
freedom and has been used extensively for the modelling of traditional helicopters [2,
6]. The derived model consists of 6 equations for the system dynamics and 4
equations describing the inputs of the system [1]. The dynamic equations are
summarized as follows:

(1)

(2)

(3)

(4)

(5)

(6)

The first 3 equations (1,2 and 3) describes the linear acceleration of the vehicle in
the direction of x,y, and z axes respectively while the last 3 equations (4,5 and 6) are
nominated for the angular accelerations of the vehicle about the same axes
respectively. ”ℓ” represents the length of the arm holding the propeller,
represent the Euler angles about the body axes (x, y, z) respectively.
are the inertial components about the x-axis, y-axis and z-axis respectively. Where
and are the translational drag coefficients, and are the propeller
disk areas while are the translational velocity components along the main
axes. Finally, ρ is the air density.

Considering to represent the input of the system, the following

equations can be given [1];

(7)

(8)

(9)
Proceedings of the 15th Int. AMME Conference, 29-31 May, 2012 RC 69

(10)

where represents the total thrust, represents the pitch moment, represents
the roll moment, while represents the yaw moment. Finally, “b” is the thrust factor
in hover and “d” is the drag factor in hover.

Increasing or decreasing of the speed of the four propellers together will be

responsible for the altitude change in position and velocity while varying the speed of
one pair of propellers (Ω3, and Ω1) will cause the aircraft to tilt about the y-axis
which is denoted as pitch angle theta “ ”. Similarly varying the speed of the
propellers pair (Ω4, and Ω2) will cause the aircraft to tilt about x-axis which is
denoted as roll angle phi “ ”.

Finally the vector summation of the reaction moment produced by the rotation of the
pair (Ω3, and Ω1) and the reaction moment produced by the rotation of the pair (Ω4,
and Ω2) will cause the quadrotor to spin about its axis (z-axis) which is denoted as
yaw angle epsi “ ”. These are the six degrees of freedom of the system consisting of
the position (x, y, z) and the orientation .

COMPUTER AIDED DESIGN MODEL

The CAD model was built up using SolidEdge (Synchronous Technology) ST in

order to estimate the mass of the vehicle and its inertial properties about the main
axes as to be used in the simulation as the system parameters. The assembly of the
model is shown in Fig.1 while the total estimated mass is equal to 688 grams and the
inertial properties about the main axes X,Y, and Z can be found in Table 1

MATLAB/SIMULINK MODELLING

A Matlab/simulink model was built using Matlab 2009 in order to simulate the
systems dynamics and response under various conditions Fig.3 shows the complete
simulink model of the system

The model was built such that it has 4 inputs which are the desired altitude (position
along the Z-axis) and the desired orientation of the vehicle consisting of the three
Euler angles about the axes X, Y and Z respectively.

CONTROLLER DESIGN AND TUNING

Given time restraints, a PID controller was selected to provide attitude control for the
quadrotor craft.
Proceedings of the 15th Int. AMME Conference, 29-31 May, 2012 RC 70

The Euler angles are used as a feedback to the proposed controller in order to
achieve the desired orientation. This may be used in the future with a path way
trajectory controller which is not our concern in this study.

The traditional implementation can be defined as follows according to [7]:

(11)

where K p is proportional gain, K i is integral gain, K d is derivative gain and e(t) is

defined as the difference between the actual controlled variable value and a desired
valued. The quadrotor controller will be targeted at forcing the attitude angles to
desired set points. As a result e(t) can be defined as:

e roll (t) = Set point − Measured Roll

e pitch (t) = Set point − Measured Pitch
e yaw (t) = Set point − Measured Yaw

These are the attitude and altitude controllers. First, the attitude controller, it is
shown in Fig.4. The controller has 3 inputs; the desired Euler angles and it takes the
actual Euler angles values as a feedback and it calculates the appropriate control
inputs (control command) which is responsible for the pitching moment, which
is responsible for rolling moment and finally which is responsible for yawing
moment.

The second is the altitude controller which is similar to the attitude controller but it
only controls the position of the craft along the z axis. It means that it only controls
the altitude of the craft. It has the desired altitude as an input in addition to the actual
altitude as a feedback signal while the output is the appropriate thrust force that
keeps the craft at the desired altitude. Both controllers were manually tuned.

STEP RESPONSE OF THE SYSTEM

The system was tested for a step input to the desired altitude and the output is
shown in Fig. 6. The system response is fast and has an acceptable overshoot about
12.3 % in addition to small steady state error of 0.02%. The system response for a
step input to the desired phi was not as better as the response of the system to the
desired altitude Fig.7 shows this response. The system reaches the desired phi after
more than 200 sec and with about 15% steady state error which is not acceptable.
Theta response is similar to the phi response and this is due to the similarity in
dynamic equations which came from the symmetry of the quadrotor structure.
Another method of control can be applied for improving the system output. This can
be done in future work.

PULSE RESPONSE OF THE SYSTEM

The system was tested for a pulse input of desired altitude equal to 5m for 4 sec and
Proceedings of the 15th Int. AMME Conference, 29-31 May, 2012 RC 71

then it should reach the zero altitude again and Fig.8 shows the response of the
system.

The response of the system for a desired phi pulse was not sufficient and it is not
acceptable Fig.9 shows the response of the system again it reaches the desired
input after very long time about 20 sec in addition to a very large over shoot about
more than 300% percent of the input which is not acceptable.

CONCLUSION

The quadrotor model mathematical model was different than the other built models
as the drag forces along the 3 axes is included, in addition to the coupling between
the translation dynamics and the angular dynamics. The proposed PID controller
used is sufficient and suitable for controlling the altitude (motion in Z-direction) of the
quadrotor craft but it is not sufficient for controlling the attitude or Euler angles of the
quadrotor craft. This suggests proposing an Artificial Intelligence (AI) controller in
future work. The system equations of the model shows that the Euler angles
dynamics is independent and it does not rely on the translational dynamics while the
translational dynamics is dependent on the angular dynamics, this is clear from the
equations of motion of translation as they are function of Phi, Theta, and Psi.
In future, an artificial intelligent controller may be introduced for the attitude controller
of the same model in order to achieve a better performance.

The major drawback in the model is that the motor dynamics is not included which
will be affecting the dynamics of system it can be added in the future work of this
research.

REFERENCES

[1] V. MARTÍNEZ, "Modelling of the Flight Dynamics of a Quadrotor Helicopter,"

in Aerospace Sciences. vol. Master of Science: Cranfield, 2007.
[2] K. P. Valavanis, Advances in Unmanned Aerial Vehicles vol. 33. Florida:
Springer, 2007.
[3] T. Bresciani, "Modelling, Identification and Control of a Quadrotor Helicopter,"
in Automatic Control. vol. Master of Science Sweden: Lund, 2008.
[4] G. M. H. Haomiao Huang, Steven L. Waslander and Claire J. Tomlin,
"Aerodynamics and Control of Autonomous Quadrotor Helicopters in
Aggressive Maneuvering," in IEEE International Conference on Robotics and
Automation Kobe, Japan, 2009.
[5] D.-b. W. Ashfaq Ahmad MIAN, "Dynamic modeling and nonlinear control
strategy for an underactuated quad rotor rotorcraft," Zhejiang University
SCIENCE, vol. 249, 2008.
[6] D. P. RAYMER, Aircraft Design: A Conceptual Approach. Washington: AIAA,
2004.
[7] M. M. Atheer L. Salih, Haider A. F. Mohamed and Khalaf Sallom Gaeid,
"Modelling and PID Controller Design for a Quadrotor Unmanned Air Vehicle,"
in IEEE International Conference on Automation Quality and Testing Robotics
(AQTR) Cluj-Napoca 2010.
Proceedings of the 15th Int. AMME Conference, 29-31 May, 2012 RC 72

Table and Figures:

Table 1. Inertial properties of the system.

Radii of
Mass Moment
Axis Gyration,
Inertia, [kg.m2]
[mm]

X 0.003325 88.19

Y 0.003263 87.36

Z 0.006430 122.64

Fig.1. Quadrotor operation.

Fig. 2. Axes notation.

Proceedings of the 15th Int. AMME Conference, 29-31 May, 2012 RC 73

Fig.3. Simulink model.

Fig.4. Attitude controller block.

Fig.5. Altitude controller block.

Proceedings of the 15th Int. AMME Conference, 29-31 May, 2012 RC 74

Fig.6. Step response of Z.

Fig.7. Step response of Phi.

Fig.8. Pulse response of Z.

 Proceedings of the 15th Int. AMME Conference, 29-31 May, 2012 RC 75

Fig.9. Pulse response of Phi.

View publication stats