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

net/publication/313803439

Design & implementation of an UAV (Drone) with flight data record

Conference Paper · October 2016

DOI: 10.1109/ICISET.2016.7856519

CITATIONS READS

5 1,243

3 authors, including:

Tuton Mallick Mohammed Saifuddin Munna

Premier University Premier University
15 PUBLICATIONS   40 CITATIONS    16 PUBLICATIONS   25 CITATIONS   

SEE PROFILE SEE PROFILE

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

IOT based system View project

All content following this page was uploaded by Tuton Mallick on 01 October 2018.

The user has requested enhancement of the downloaded file.

 Design & Implementation of an UAV (Drone) with
Flight Data Record
Tuton Chandra Mallick , Mohammad Ariful Islam Bhuyan, Mohammed Saifuddin Munna
Department of Electrical and Electronic Engineering
Premier University, Chittagong
Chittagong- 4203, Bangladesh.
[email protected] , [email protected] , [email protected]

Abstract— This paper proposed the development of an world and also developing world use it for their own purposes.
autonomous unmanned aerial vehicle (UAV) which is controlled Uses of drone are rapidly increased for both public & Private
by wireless technology through graphical user interface (GUI). sector. Peoples of Canada & North America now use drone as
This proposed design capable to fly autonomously and also their assistant of housework & office work. At present not only
capable to track pre loaded mission automatically. Proposed in Canada but also in others countries drone technology
mathematical model and artificial algorithm control technique by increasing day by day.
which quad rotor can capable to fly autonomously, trajectory
tracking, graceful motion and accurate altitude hold For the domestic user they have to pay attention on government
performance. In this system author used IMU 9DOF (3-axis rules regarding use of drone. For this propose small & cost
accelerometer, 3-axis gyroscope & 3-axis magnetometer) which effectives drones are available in markets. Its popularity
ensure it smooth movement, graceful motion and trajectory increased day by day.
tracing. GPS system and barometric sensor make it more
efficient in autonomous mode. Several PID loops designed to get A. Applications of UAV( Drone)
better stability and performance in different mode. All signals
are processed by a powerful high speed controller board which There are some many important activates of drones. Which
makes it more efficient and effective. This work aimed to design a are:
quad copter that will try stable its position according to preferred
altitude. Also here stability check has been done with pitch and • National security
roll. All data and result discussed at the end of paper. • As a long range weapon of military war
• Public safety
Keywords—autonomous UAV system, inertial
• Environmental research
measurement unit, graphical user interface, PID controls
• Scientific research
loops, stability and performance.
• Small household work.
I. INTRODUCTION
At present world drone technology is very familiar & B. Related Works
versatile technology. Drone can drift in air .We can also use
wireless camera with it. So that we can use it to do different
In recent years many scientist works into this area of UAV.
types of tasks. Nowadays Drones are used in long range wars
As in reference paper [5], [6], [7], Samir Bouabdallah et al.
as a weapon and also as a helper of fighter in the war. Scientist
make a good number of research on smart controlling if UAV.
use drone as a part of their research assistant. Drones not only
This author also used back stepping method [7]. For a better
help us in society but also a threat for us, as because many of
solution in stability & smooth controlling many scientists used
developed countries use it as their weapon of destruction. So
PID and root locus method [8]. PD controller is also a very
drones have their ability to predetermined work so that it
popular method for control drone & robotic arm [9]. Many
becomes important in today’s world. [1]
scientist use PD control as their experimental platform [1], [10]
Drone has many importances but it also creates some
questions about privacy. So for this issue many governments II. PROPOSED DESIGN
declared some rules and regulation to fly drone in different
purposes. A mature quad rotor system can use for educational A. System Overview
and experimental porpoise [2], [3]. Photo shoot for films &
To desing a stable multicopter we need maintain some
drama are also use drone.
physics, mathmatics and aeirodynamic term. Aerodynamic help
At past drones was used only by military in their war. But to define its movement and inertial motion. In the other hand
day by day it is now used in various household works as it mathematical calculation helps to manipulate required lift
operation and control become easy day by day. Developed

978-1-5090-6122-8/16/$31.00 ©2016 IEEE

force, angular position, graceful motion and trajectory Figure 2. Air part hardware system in alive. Right figure shows Processing
definition. unit of an autonomous UAV.

We designed drone’s body according with dynamics and

also designed artificial algorithm to make it autonomous and B. System Block Diagram
well behaved. Hardwire system consists of different sensors,
powerful controller unit and electronic equipments so on. For a
desire movement controller takes data from different sensors.
3-axis accelerometer and 3-axis gyroscope provide data of its
orientation, acceleration and angular rate. Then these data
processed and compare with reference and desire value. This
operation performs with the help of PID loop. Several PID
loops used in these case like pitch control, roll control, yaw
control, hover, altitude holding and orientation control. IMU
(inertial measurement unit) provides real altitude, angular
movement and orientation. After that required pulse sends to
ESC (Electronic Speed Controller) for desire speed of rotation.
Magnetometer provides real time direction with the global
magnetic field reference. Barometric pressure sensor also
provides real time altitude. GPS (global position system)
module helps to make system autonomy. It helps find out any
coordinate and reach to this coordinate. Telemetry kit helps to
observe flight data wirelessly from ground station. It also send
mission file and communicate with air part like USB serial
mode (TTL mode).
Figure 3. Proposed block diagram of UAV’s air part.
In ground part consists of powerful ground station.
PC/Laptop used for sending data through telemetry and coding
or data logging from air part. Another radio transmitter used to
switch different mode and operate in manual mode.

Figure 4. Proposed block diagram of UAV’s ground station.

III. SYSTEM MODELING

To design a stable multi-rotor copter we have to
concentrate its structure and dynamics. We have to develop a
Figure 1. System Overview of Autonomous Unmanned Aerial Vehicle firmware in which contains different control strategy, mode of
operation, data evaluation and different PID loops for stability:

A. Body Dynamics
Body dynamic of multi-rotor copter governs the response of
attitude control. Let consider a multi-rotor copter frame. We
can derive expression in two coordinate system i.e. one is
inertial coordinates and another one is body fixed coordinates.
 U 1 = Th1 + Th 2 + Th3 + Th 4
 U 2 = (Th3 − Th1) L
 (2)
U =
 U 3 = (Th 2 − Th 4) L
U 4 = M 1 + M 3 − M 2 − M 4

Here Thi is thrust generated by four motor, M i are

momentums, and L is lever length.
By applying the force and moment balance laws, motion
formulations are given as
x = {U 1( SinψSinφ + CosψSinθCosφ ) − K1x} / m
y = {U 1( SinψSinθCosφ − CosψSinφ ) − K 2 y} / m
z = (U 1CosφCosθ − K 3z ) / m − g
Where k i is drag coefficient (assume zero since drag is
Figure 5. Force diagram of quad copter. negligible at low speed).

U1 is summation of the thrust of every individual motor. The angle φd and ψd can determine as follows from
Th1, Th2, Th3 and Th4 are thrust generated by front, rear, left Fig.7,
and right motor respectively. m is Quad-copter mass, g is
marked as the gravity acceleration and L is the liver distance of
Quad- copter. x, y and z are the three axis position. φ, θ, ψ are
three Euler angles representing pitch, roll and yaw.

Figure 6. Attitude control definition.

From the Fig. 6 we get transformation matrix with defined Figure 7. Angle movement of quad copter.
attitude angles:
yd − y
cψcθ cψsθsψ −sψcθ cψsθcφ + sψsφ φd = tan −1
xd − x
R(θ,φ,ψ) = sψcθ sψsθsφ +cψcφ sψsθcφ − sφcψ (1) zd − z
ψd = tan −1
 − sθ cθsφ cθcφ  ( x d − x) 2 + ( y d − y ) 2
Where s represents sin, c represents cosine, and φ, θ, ψ
stands for attitude angles of roll, pitch, and yaw respectively.
The second derivatives of each angle given as
Direct inputs are RPM commands for every motor in body
fixed coordinates. So the resultant outputs are Z directional φ = U 2 / I xx
thrusts in body fixed coordinates. Attitude & position is only
concern of our outputs. U1, U2, U3 and U4 are four control
θ = U 3 / I yy
variables used to eliminate this gap. Each of the affects the ψ = U 4 / I zz
attitude, rotation in roll angle, rotation in pitch angle and yaw
angle respectively.
 Where I xx , I yy I zz is rotary inertia around X, Y, Z axis τ : torque of motor
respectively τ d : load
B. Aerodynamics Effect
D. System PID Control
The thrust T produced by each motor is calculated as
PID (proportional-integral-derivative) is a closed-loop
T = ρCT Awm R 2
2 control system. It helps to get c our results as must as close to
the actual result by responding to our inputs. Scientist uses it
Where while controlling drone or robot for achieves stability.

CT : thrust coefficient
ρ : Air density
A: rotor disk area
R: blade radius
Propeller diameter & pitch-
D ~τ , p ~τ
τ ~E
Where, d: diameter of propeller, τ : torque, E: energy
Frame parameters-
Figure 8. PID loop for system.
Blade tip speed, v ~ R
PID basically works with three algorithms.
3
Lift, F ~ R • P depends on the present result
3 5
Inertia, m ~ R , I ~ R • I on the accumulation of past errors.

1 • D is prediction of future errors based on current

Acceleration, linear a ~ 1 , angular a~ data.
R
Different coding systems are available based on these
Where, R: frame center to motor distance algorithms.

C. Dynamics of Rotor Per axis PID structure shown in Fig.9. For controlling quad
copter or any types of multi copters, output of sensors (like the
The dynamics of DC motor is generally described as pitch angel) is very much needed. From the sensor data we can
di easily estimate the error (how far we are from the desired pitch
Li + Ri + k e wm = u angle, e.g. horizontal, 0 degree). Then we can use PID
dt algorithms for eliminating errors
dwm
J = τ −τ d
dt
Where
Li : Coefficient of inductance
i: armature current
R: armature resistance
k e : back emf constant
Figure 9. Per axis PID structure.
wm : speed of motor
u: armature voltage
J: inertia of motor
 IV. RESULTS AND IMPLEMENTED DESIGN 60
DesPitch
40 Pitch
A. On flight Simulation Data
All the data collected after 18-20 experimental fight of 20

Output
implemented UAV. 0
Fig. 10, indicates the altitude hold performance. This is one
of the important curves which indicate the stability and -20

behavior of our drone. Here Holt for desire altitude and Alt for -40
altitude at which drone travels. This carve contains several 0 200 400 600 800 1000 1200 1400 1600
Time (0.1s Scale)
mode of operations characteristics. When we shift from
stabilize (manual) mode to hover mode or GPS lock mode then Figure 12. Response of pitch in different mode.
DAlt curve generate and then altitude calculated from IMU 20 DesRoll
unit and also generate the barometer altitude curve shown in Roll
Fig. 11, as blue line. Alt and DAlt line lies close together that 0
means altitude hold performance is good in different modes.

Output
DAlt -20
30 Alt
-40
20
Output

-60
0 200 400 600 800 1000 1200 1400 1600
10 Time (0.1s Scale)

Figure 13. Response of roll in different mode.

0

0 200 400 600 800 1000 1200 1400 1600 Fig. 14, shows the throttle response in different modes. We see
Time (0.1s Scale)
that throttle fluctuate in manual mode due to manual control of
Figure 10. Altitude hold performance data in different mode. throttle fluctuation. In autonomous mode throttle
automatically control by processor unit. So throttle response
40 DAlt does not fluctuate significantly.
BarAlt 600
ThrRes
500
20
O u tp u t

400
Output

300
0
200

100
-20
0 200 400 600 800 1000 1200 1400 1600 0
0 200 400 600 800 1000 1200 1400 1600
Time (0.1s Scale) Time (0.1s Scale)

Figure 11. Altitude hold performance with barometer. Figure 14. Throttle response of proposed design in different mode.

Fig. 12, shows the pitch response and Fig. 13, shows the
roll response of UAV in different modes. Here, DesPitch and B. Implemented Design
DesRoll for desire pitch and roll respectively. And only pitch Total hardware design of proposed UAV is given below.
and roll indicate the pitch roll response. If we observe the
curves than we see that pitch and roll curve is similar to desire
pitch and roll curve. We see in stabilize (manual) mode and
altitude hold mode we get some spike which indicate pitch &
roll action. Though stabilize mode is manual mode so we
change pitch manually. In altitude hold mode- altitude maintain
automatically but pitch & roll is also manual control. In loiter
mode that means in GPS lock mode- all parameters controlled
automatically without human interface. So in loiter mode spike
is less which indicate it holds a constant coordinate with
stability.

Figure 15. Real view of implemented design.

 V. CONCLUSION [3] D. Mellinger, M. Shomin, and V. Kumar, “Control of quadrotors for
robust perching and landing,” in International Powered Lift
From all of curve analysis we can conclude that overall Conference, October 5-7, 2010, 2010.
performance of our system is stable. This system is capable to [4] The standing Senate Committee on National Security and Defense -
fly in different mode without complexity. It’s performance, Evidence, April 23, 2012.
movement, orientation, motion, stability also good. This drone [5] S. Bouabdallah, R. Siegwart, S. Bouabdallah, and R. Siegwart,
is capable to fly in several modes. The main modes are manual “Backstepping and sliding-mode techniques applied to an indoor
mode, hover mode, auto mode and return to lunch mode. In micro quadrotor,” in Robotics and Automation, 2005. ICRA 2005.
Proceedings of the 2005 IEEE International Conference on. Ieee,
manual mode drone is controlled by remote device and in 2005, pp. 2247–2252.
others mode drone flies autonomously. This system have [6] S. Bouabdallah and R. Siegwart, “Full control of a quadrotor,” in
facility to see flight data by using powerful ground station and Intelligent Robots and Systems, 2007. IROS 2007. IEEE/RSJ
user can upload or override a mission in real time flight International Conference on. Ieee, 2007, pp. 153–158
condition when a mission running. The distance between motor [7] S. Bouabdallah, “Design and control of quadrotors with application to
to motor is 0.61m. The overall total weight of implemented autonomous flying,” Lausanne Polytechnic University, 2007.
design is 1.46kg and its carrying capacity is 0.5kg.
[8] P. Pounds, R. Mahony, and P. Corke, “Modelling and control of a
quadrotor robot,” in Proceedings Australasian Conference on
REFERENCES Robotics and Automation 2006. Australian Robotics and Automation
[1] N. Michael, D. Mellinger, Q. Lindsey, and V. Kumar, “The grasp Association Inc., 2006.
multiple micro-uav testbed,” Robotics & Automation Magazine, IEEE, [9] Q. Lindsey, D. Mellinger, and V. Kumar, “Construction of cubic
vol. 17, no. 3, pp. 56–65, 2010. structures with quadrotor teams,” Proc. Robotics: Science & Systems
[2] V. Kumar and N. Michael, “Opportunities and challenges with VII, 2011.
autonomous micro aerial vehicles,” in Int. Symp. on Robotics [10] R. Mahony, V. Kumar, and P. Corke, “Multirotor aerial vehicles:
Research, 2011. Modeling, estimation, and control of quadrotor,” Robotics &
Automation Magazine, IEEE, vol. 19, pp. 20 – 32, 2012.

View publication stats