Australian Journal of Science and Technology

ISSN Number (2208-6404)

Volume 7; Issue 1; March 2023

Original Article

UAV gimbal position control using PID compensator

Odera Ohazurike, Sunday Anwansedo

Department of Electrical Engineering, Southern University and A&M College, Baton Rouge, Louisiana, USA

ABSTRACT
In this paper, the authors developed a proportional-integral-derivative (PID) controller for unmanned aerial vehicle gimbal stability. During
aerial flights, the gimbal instability as a result of actuator fluctuation leads to the capture of blurred images and joggled videos. This poses a
post-flight data analysis problem. Viewing the gimbal as a robotic manipulator, independent joint control method is employed in obtaining
the gimbal transfer function. Employing Matlab’s control systems analysis toolbox, the results obtained showed a stable convergence as seen
from the bode plot. The PID controller parameters showed that the percentage required for the system value was 8.63%. The time taken for
the system response to reach the target value from an initial state of zero was 0.496 s. The time taken for the system to reach steady state after
the initial rise was 1.54 s.
Keywords: Gimbal, modulation, proportional-integral-derivative controller, unmanned aerial vehicle
Submitted: 02-01-2023, Accepted: 03-02-2023, Published: 30-03-2023

INTRODUCTION the system and also to have a very good disturbance rejection
capability.[6]
The unmanned aerial vehicle (UAV)[1,2] is an aerial robotic
system that serves a variety of purpose. Such purposes could be The gimbal is a typical example of a stabilized platform.
civilian or military, of which operations, surveillance, disaster It is a robotic manipulator that positions the visual sensor
management, and aerial photography may be mentioned as during aerial flights. It is a mechanical system consisting of
examples.[1] Such applications require live video recording a servo, position encoder, and an inertia measurement unit.
and transmission. Thus, image stabilization is a primary However, the gimbal being a mechanical system is susceptible
requirement for the success of such applications. How best to vibrations which lead to blurred images. This mechanical
can a stable image be achieved? Image stabilization can be instability can be compensated for using control algorithms
achieved when the visual sensor payload is compensated to effectively control the camera position.[7,8] This control
for vibration, wind gust, and vehicle attitude transients.[3] mechanism is known as gimbal control.
An object developed for maintaining and keeping the visual
sensor in a steady and desired orientation despite the UAV’s In control theory, various control algorithms have been
motion is needed. Such an object is termed a stabilized proposed. With specific application to control theory,
platform.[4] algorithms developed for gimbal control include; coulomb
Friction control algorithms,[9] command filtered back stepping
A stabilized platform is an object used to isolate payloads algorithms,[10] PD controllers,[11] and genetic algorithms[7].
such as sensors, cameras, telescopes, and antennas from the
motion of the carrier (in this instance, the UAV), on which the In this paper, the design scheme of a proportional-integral-
payload is mounted.[5] It enables the camera mounted on the derivative (PID) controller is presented. The gimbal system is
platform to aim and track objects rapidly and exactly. Thus, modeled using Simulink. The proposed controller is built around
the basic requirements of the platform are to maintain stability the mathematical model and the controller behavior is analyzed
during operation even when changes occur in the dynamics of and tested for stability by employing bode plots techniques.

Address for correspondence:

Ohazurike and Anwansedo, Southern University and A&M college, Baton Rouge, USA. E-mail: [email protected]

Available at www.aujst.com 36
 Ohazurike and Anwansedo: PID as Compensator

Figure 1: Simulink model of PID gimbal control system[9]

METHODOLOGY

In developing the controller algorithm, the model of the gimbal

device and its actuator system is presented.

Gimbal Modeling
In modeling the gimbal device, it is viewed as a robotic
manipulator.[12] Thus, using the Euler –Lagrange equation, the
dynamic equation of the gimbal is developed. The LaGrange
equation of motion provides a systematic approach to obtain
the dynamic equation of a robot manipulator.[13] Defining the
LaGrangean (L), it is defined as the difference between the
kinetic and potential energies. That is;
Figure 2: Step response Plot of Gimbal system
o
1 T o 1
Ki  vi mi vi  Ti i I io i (1)
2 2o


1 T T 1 To
 qi ( ( J Di mi J Di  J Ri I i J Ri ))qi (2)
2 i 1
2

1 T
K= qi Dqi  (3)
2

Where D is a n*n matrix known as the manipulator inertia

matrix. Thus

n Figure 3: Reference tracking Plot of PID Controlled

 T 1 To Gimbal system
D J Di mi J Di  J Ri I i J Ri (4)
i 1
2
and pneumatic piezoelectric means. Electrically actuated
manipulators are powered by Dc motors due to their clean,
The equation of motion for a robot manipulator can be shown quiet, and precision qualities.
in the concise form
di 1
 (u  Rm i  ke aw )  (6)
D  q  q  H  q, q   G  q   Q (5) dt L

Actuator Dynamics Controller Design

Actuators are devices which translate power into motion. A PID controller is a control feedback loop mechanism
Manipulators are actuated using electrical, hydraulic, commonly employed in control systems. In describing the

Available at www.aujst.com 37
 Ohazurike and Anwansedo: PID as Compensator

Figure 4: Bode Plot response of PID Controlled Gimbal system

PID controller, the existing relationship between the controller

output (u(t)) and the error e(t) is considered. The relationship
is expressed in the continuous form as

de  t 
t


u  t   K p e  t   Ki e T  dT  K d
0
dt
(7)

Where the terms kp, ki, and kd are the gains of the proportional
integral and derivative parameters of the controller. E is the
error signal, and it is represented as

e  SP  PV  (8)
Figure 5: Joint Angular velocity output plot of PID gimbal
control system
Developing the controller within the Simulink design
environment, we obtain our controller parameters to be used Due to the non-linear nature of the system, it was necessary to
in controlling the gimbal model. PID gain values obtained add a compensator. Figure 3 shows the response of the system
that gave a stable operation to the system are; Kp = 3.9072, to a unit step input when a PID controller is tuned alongside
Ki = 0.090406, Kd = 2.5714, and a filter coefficient the plant. It is observed that stability was achieved as a result
(N) = 306.9477. The Simulink model of the closed loop of the auto tuning technique employed.
feedback pid gimbal control system is shown in Figure 1.[9]
The control performance and robustness parameters obtained
The Pid controller model is thus given as are a rise time of 0.496 s, settling time of 1.54 s, a percentage
793.1s 2 + 1119 s + 27.75 overshoot of 8.63%. Figure 4 shows the Bode plot response
(9) of the PID controlled system.
s 2 + 306.9 s
From the plot in Figure 4, a phase margin of 60° with a
RESULTS AND DISCUSSION gain margin of 39.8 decibels was obtained. The overall
performance of the system showed a system with a stable
In analyzing the controller results, the initial system response is performance. The system output showing the angular
considered. Figure 2 shows the response of the gimbal system velocity and position of the gimbal is shown in Figures 5,
to a unit step input. respectively. From the plots, it is seen that the response

Available at www.aujst.com 38
 Ohazurike and Anwansedo: PID as Compensator

of the velocity plot exhibited a drop in overshoot before Bengaluru, India: IEEE International Conference on Computer,
settling. Communication and Control; 2015.
4. Sudhanwa K, Akash M. Analysis of the Gimbaled Platform for
the Three Degrees of Freedom Using Differential Equations.
CONCLUSION AND FUTURE WORK Hyderabad, India: International Conference on Advanced
Research in Mechanical Engineering; 2015. p. 31-4.
In trying to develop an adaptive controller for the gimbal 5. Leghmizi S, Sheng L, Naeim F, Latifa B. Modelling, design and
model, which is a non-linear system, an auto tuned PID control of a ship carried 3DOF stabilized platform. Res J Appl
controller is developed and implemented on a Simulink model. Sci Eng Technol 2012;4:3843-51.
The results obtained show that while stability was achieved, 6. Hilkert JM. Inertially stabilized platform technology, concepts
and principles. IEEE Contr Syst Mag 2008;28:26-46.
the disturbance rejection capability of the developed controller
7. Rajesh RJ, Kavitha P. Camera Gimbal Stabilization Using
was not ideal. Evolutionary Algorithms. Hoogly, India: IEEE International
Conference on Computer, Communication and Control; 2015.
Hence, it is recommended that other control system 8. Spong WM, Hutchinson S, Vidyasagar M. Robot Modelling and
development techniques be looked into so as to achieve a more Control. New York: John Wiley and Sons; 2004.
robust and performance enhancing controller for stabilizing 9. Hwan SP, Jea-Ho P, Hong SK. Control system design for a UAV
and control of gimbal systems. mounted gimbal subject to coulomb friction. J Inst Contr Robot
Syst 2012;18:680-7.
10. Triputra FR, Trilaksono BR, Adiono T, Sasongko RA,
REFERENCES Dahsyat M. A nonlinear camera gimbal visual servoing control
using command filtered backstepping. J Unmann Syst Technol
1. Ole CJ, Eric NJ. Control Architecture for a UAV Mounted Camera 2015;3:49-60.
Gimbal. Arlington, Virginia: America Institute of Aeronautics and 11. Shiino T, Kawada K, Yamamoto T, Komichi M, Nishioka T.
Astronautics (AIAA) Conference on Information Technology Gimbals Control with the Camera for Aerial Photography in RC
and Aerospace; 2005. p. 1-10. Helicopter. Seoul, Korea: International Conference on Control,
2. Altan A, Hacıoğlu R. Model predictive control of three-axis Automation and System; 2008. p. 1135-9.
gimbal system mounted on UAV for real-time target tracking 12. Johansson J. Modelling and Control of an Advance Camera
under external disturbances. Mech Syst Signal Process Gimbal. M.S Thesis. Linkoping, Sweden: Department of
2020;138:150. Electrical Engineering, Linkoping University; 2012.
3. Rajesh RJ, Ananda CM. PSO Tuned PID Controller for 13. Jazar RN. Theory of Applied Robotics: Kinematics, Dynamics
Controlling Camera Position in UAV Using 2 Axes Gimbal. and Control. New York: Springer; 2007.

This work is licensed under a Creative Commons Attribution Non-Commercial 4.0

International License.

Available at www.aujst.com 39