Content uploaded by Süleyman Canan
Author content
All content in this area was uploaded by Süleyman Canan on Aug 13, 2018
Content may be subject to copyright.
International Conference on Advanced Technologies, Computer Engineering and Science (ICATCES’18),
May 11-13, 2018 Safranbolu, Turkey
Abstract - Nowadays everyone is trying to record the moment
everywhere and wants it to be perfect. Beyond resolution, there
is a desire to get steady shots regardless of the environmental
conditions. The gimbal stabilization system ensures a stable
image by blocking motion-related vibrations before they are
transferred to the camera lens axes. Thanks to the Three Axis
Gimbal, perfect images can be achieved by minimizing the
vibrations while jogging, climbing or coming down stairs,
cycling, or using any kind of vehicle. In short, a three-axis
gimbal can be integrated everywhere a fixed image is needed. It
is envisaged that gimbal stabilization system will be needed in
many scientific studies in the following periods.
The aim of this study is to present the Three Axis Gimbal
mechanism. Three separate brushless servo motors are installed
on each axis for absorbing unwanted movements. The gimbal is
also equipped with an inertial measurement unit consisting of a
gyroscope and accelerometer close to the camera mount point.
The general control system and PID controller are simulated by
using MATLAB and the results are shown graphically.
Keywords - Inertial Measurement Unit, Brushless Servo
Motor, Gimbal System, PID Controller.
I. INTRODUCTION
imbal is the system used to prevent the shaking, one of
the biggest problems in video recordings. Figure 1
illustrates a simple block diagram of the gimbal assembly.
There are two or three engines on the systems called as
gimbal and they aim to prevent or eliminate vibration [1].
The basic logic of this system which can minimize the
vibration in video recording devices is to create a reverse
motion in the opposite direction of the vibration. This reverse
motion is provided by the Inertial Measuring Unit (IMU)
Sensor which is placed on the camera. The IMU Sensor
detects the camera movements and reports motion to three
brushless servo motors positioned in line with the camera
lens. The sensor detects the relative position of the camera
according to the ground. Based on the predetermined
optimum position, it is detected how much the optimum
position defined in each movement of the camera deteriorates.
The main aim is to protect this optimum position. The
information received from the sensor is processed on the
electronic board and transmitted as a command to the
brushless servo motors, which provide smooth motion. Thus,
the brushless servo motor that produces the opposite
movement of the camera allows to obtain a smooth image.
G
Thanks to the Three Axis Gimbal; a cameraman shooting
on the baseline of the field in a football match can record
smooth images while running in order not to miss the event;
in unmanned armed vehicles used in the defense industry,
cameras mounted at the barrel level can produce smooth
images which helps to achieve accurate targeting, and perfect
images can be obtained in many land and air shots as well
[2].
In this study, a three-axis gimbal system which uses PID
controller as the general control system is presented [3,4]. In
the second chapter, the components of the moving platform
system are detailed. In the third chapter, the filtering of the
values read from the sensor, and the control system are
referred. Finally, in the fourth chapter, proposals and
conclusions are submitted.
Figure 1: Gimbal block diagram.
II.MECHANICAL DESIGN
The frame carrying the system should be enduring enough to
carry the camera and light enough to provide ease of use.
Carbon fiber pipe is the material qualified for the conditions
that we are looking for.
A six-axis IMU sensor card (which is often used in
multicopter and robotic projects) which has a three-axis
gyroscope and a three-axis angular accelerometer is often
needed to detect camera movements. Thus, we can obtain the
information such as orientation, speed, and position from a
single unit.
The three axes mentioned in the Three-Axis Gimbal are
shown in Figure 2. These three axes are called pitch, yaw, and
roll, which carry the same name as the axes of the movement
of a plane. In order to absorb the unwanted movements of
these three axes, three separate brushless servo motors are
mounted on these axes corresponding to the camera lens. The
brushless servo motor mounted on the pitch axis absorbs the
unwanted up-down movement of the camera lens, undesired
right-left motion is absorbed by the brushless servo motor
camera lens mounted on the yaw axis, and undesired rolling
motion from one edge to the other is absorbed by the
1
Three Axis Gimbal Design And Its Application
E.DERE1, M.OZCAN2 and S.CANAN 1
1 Elfatek Elektronik Ltd. Şti., Konya/Turkey, eminecnby@gmail.com
2Necmettin Erbakan University, Konya/Turkey, mozcan@konya.edu.tr
1 Elfatek Elektronik Ltd. Şti., Konya/Turkey, suleyman.canan@gmail.com
brushless servo motor mounted on the roll axis. The biaxial
gimbal does not have yaw axis, so that the recorded videos are
shakier than the three-axis gimbal. Because, there is no
absorption in the sudden and involuntary turns towards right
or left.
Figure 2: Gimbal axes.
Motors mounted in the line with the camera lens receive
feedback from the sensor and are used to provide a rotational
motion in the opposite direction of the movement to keep the
camera lens steady. In this study, a brushless servo motor is
preferred because of its ability to tolerate the fault quickly and
smoothly.
III. SOFTWARE DESIGN
A. IMU Data Fusing
Gyros filter out accelerometer outputs to make a more
accurate measurement. There are various algorithms for
filtering. One of the most commonly used is the Kalman filter
[5]. But it has a complicated algorithm. It makes a calculation
by variable weighted average ratio and can use many different
methods to calculate this ratio, but it is difficult to
understand. The system works to predict new outputs by using
previous outputs and measurements. In sum, the Kalman
filter predicts the best value of the next output by monitoring
the constantly changing inputs of the system. This filter is
used in many areas such as image processing, orientation,
motion tracking.
Figure 3: Complementary filter block diagram.
Block diagram of the Complementary Filter is illustrated in
Figure 3. This filter is a method of taking averages of the
reference data with a constant weighted average ratio [6]. It is
the simplest algorithm. It is generally used for hobby
purposes and in the applications to understand the operating
logic. It simultaneously manages both high-pass and low-pass
filters. Low pass filter filters high frequency signals (such as
accelerometer in the case of vibration) and low frequency
signals (such as gyroscope drift). By combining these filters
with a complementary filter, a good signal can be obtained
without the complications of the Kalman Filter.
Consequently, the Complementary Filter can be used instead
of the Kalman Filter. Smoothing is better, and its algorithm is
much simpler than Kalman. Therefore, The Complementary
Filter is preferred in this study. An example of how the system
is used with the Complementary Filter is shown in Figure 4.
A graphical comparison between the signal filtered by the
Complementary Filter and the true angle is shown in Figure
5.
Figure 4: Complemenraty filter with IMU.
Figure 5: Complemenraty filter output and comparison of actual
value.
2
B. Controller Design
Although there are a variety of modern techniques for
controlling gimbal systems, PID controllers are preferred
because of their low cost, ease of implementation and high
performance.
The block diagram in Figure 6 is a feedback mechanism
controller commonly used in PID (Proportional-Integral-
Derivative) industrial control systems. In 1942, Ziegler and
Nichols presented two classical methods for determining the
parameters of the PID controller [7]. These methods are still
widely used in the original form or in some modifications.
The methods are based on determining some properties of the
process dynamics.
Figure 6: PID block schema.
The PID control system continuously calculates the error
value as the difference between the desired value and the
measured process variable. The controller tries to minimize
the error over time. In doing so, sets a new value (determined
by a weighted sum) for a control variable, such as power fed
to a control valve, a dashpot, or a heating element.
Transfer function of PID controller:
(1)
KP: proportional gain,
Ki: integral gain,
Kd: derivative gain,
As shown in Table 1, the proportional gain (Kp) reduces the
rise time and the steady-state error (but never removes it).
The integral gain (Ki) removes the steady-state error but can
worsen the transient response. Derivative gain (Kd) increases
the stability of the system, reduces overshoot and improves
the transient response.
Table 1 : Effect of PID gains on system
Controller
Response
Rise
Time Overshoot Settlement
Time
S-S
Error
KpDecreases Increases Changes
Slightly Decreases
KiDecreases Increases Increases Removes
Kd
Changes
Slightly Decreases Decreases Changes
Slightly
Figure 7 shows the PID controller designed with MATLAB
Simulink, where all controller coefficients are set. The
program is started and the necessary parameters are obtained
by PID tuning [8]. Once Simulink is running, we can see the
signal shown in Figure 8.
Figure 7: Design of PID control system in MATLAB Simulink.
Figure 8: Output signal in case of PID
IV. CONCLUSION AND RECOMMENDATION
In this study, three-axis gimbal system is introduced. Under
the title of mechanical design, the basic elements of the
gimbal system are introduced. Under the software design title,
gimbal control system, sensors and sensor information
filtering are mentioned. PID control is used to stabilize three
axes of this platform model. The PID control system is
simulated in the MATLAB / Simulink environment and the
results are shown in the diagram. PID control is often
preferred because of its high performance, but control
methods such as PI, PI2, optimal control, and compensator
can also be used in two-axis and three-axis gimbal systems.
Brushless servo motors are used in the system as motion
providers. Step motor or servo motor may be preferred instead
of brushless servo motor. The error signal is detected by using
the IMU sensor while the Complementary Filter is preferred
for sensor filtering. Besides Complementary and Kalman
filters; the Mahony Madwick Filter, which is used to calculate
3
the quaternion with the information received from the sensor,
may also be preferred. This system is being implemented and
it is targeted to be used in military applications carried out
within our company.
REFERENCES
[1] MD. Shusterand SD. Oh, Three-axis attitude determination from vector
observations, J Guidance Control 1981; 4: 70-77
[2] J.J. Xiong, E.H. Zheng, "Position and Attitude Tracking Control for a
Quadrotor UAV", ISA Trans., vol. 53, no. 3, pp. 725-731, 2014
[3] D. Stojcsics, A. Molnar, "Fixed-Wing Small-Size UAV Navigation
Methods with HIL Simulation for AERObot Autopilot", IEEE 9th Int.
Symp. on Intel. Sys. and Inf. (SISY), pp. 241-245, 2011
[4] J.M. Sullivan, "Evolution or Revolution? The Rise of UAVs", IEEE Tech.
and Soc. Mag., vol. 25, no. 3, pp. 43-49, 2006
[5] J.L. Marins, X. Yun, E.R. Bachmann, R.B. McGhee, M.J. Zyda An
Extended Kalman Filter for Quaternion-Based Orientation Estimation
Using MARG Sensors. International Conference on Intelligent Robots and
Systems, Maui, Hawaii, USA, pages 2003{2011, 2001
[6] G. Baldwin, R. Mahony, J. Trumpf, T. Hamel, T. Cheviron Complementary
Filter Design on the Special Euclidean Group SE. 9th European Control
Confrerence on behalf of the EUCA, Kos, Greece, 2007.
[7] J. G. Ziegler and N. B. Nichols: Optimum Settings for Automatic
Controllers, Trans. ASME, Vol. 64, 1942, s. 759-768.
[8] P.M. Meshram,, and R.G. Kanojiya, Tuning of PID controller using
ZieglerNichols method for speed control of DC motor, Advances in
Engineering, Science and Management (ICAESM), 2012 International
Conference on (pp. 117-122). IEEE.
4
