Aerospace and Electronic Systems Magazine March 2017 - 58

Student Research Highlights:

DOI. No. 10.1109/MAES.2017.160038

Autonomous Quadrotor for Accurate Positioning
Lucas Moraes, Luiz C. Carmo Jr., Rafael F. Campos, Marco A. Jucá, Lucas S. Moreira, João
Pedro Carvalho, Alexandre M. Texeira, Daniel D. Silveira, Thiago V. N. Coelho, André Luis
M. Marcato, Alexandre B. dos Santos, Universidade Federal de Juiz de Fora, Brazil

INTRODUCTION
Surveillance missions in vast, difficult access environments are responsible for logistic difficulties in comparison to using an in loco
monitoring team. For this and many other reasons, solutions with
robotic platforms such as unmanned aerial vehicles (UAVs), present economic advantages [1].
Among the different UAV architectures, it can be noted that
fixed-wing aircraft allows higher autonomy, while multirotor aircraft allows vertical take-off and landing (VTOL), which reduces
the required free area for taking off and landing actions, in addition to allowing in-flight loitering. Currently, UAVs have different levels of autonomy; some UAVs are controlled by an operator
instead of a wireless connection to a remote control station. These
operators need to be duly trained for executing the mission safely,
regarding potential accidents involving risks to people nearby, as
well as economic losses and aircraft damage.
The combination of remote control operation with computerized automation was able to conduct semiautonomous missions
such as flying to preset targets. Even more sophisticated versions
have embedded control and/or trajectory systems to accomplish
missions that can be altered in real-time [2]. However, only a small
group of UAV systems are able to accomplish high complexity operations autonomously, such as taking off and landing the aircraft,
as well as being able to follow flight directives according to its
programmed mission. These systems are called autonomous unmanned aerial vehicles (AUAVs).
Due to the cited potentialities, the focus of this work was the
development of an AUAV with sufficient load capacity to embed
solutions for monitoring environmental areas, carrying out its missions with high precision in positioning based on the data processing from different sensors like optical flow (OF), camera, inertial
measurement unit (IMU) and GPS units, mostly in procedures of
loitering and landing.

Authors' address: Universidade Federal de Juiz de Fora, José
Lourenço Kelmer Street, Campus da UFJF4a Plataforma do
Setor de Tecnologia, 36036-330 - Juiz de Fora - MG, Brazil.
E-mail: (thiago.coelho@ufjf.edu.br).
Manuscript received February 12, 2016, revised June 8, 2016,
August 10, 2016, November 14, 2016, and ready for publication January 5, 2017.
Review handled by W. Dou.
0885/8985/17/$26.00 © 2017 IEEE
58

A high processing unit was embedded into the UAV architecture to perform autonomous tracking for a landing mark pattern,
based on computer vision algorithms and fuzzy logic rules developed in our laboratory, for precision landing procedures.

QUADROTOR FRAME TYPE
Figure 1 shows the orientation scheme of an X-type quadrotor,
where the arrow in the central body indicates the front of the aircraft. Its maneuver dynamics are based on the rotation about one of
three orthogonal orientation axes, whereas upward and downward
motion is controlled by the rotation of the four motors. The angles
of rotation about the axes are denoted as pitch (ϴ), roll (ϕ), and
yaw (ψ).
To ensure stability, motors 1 and 2 rotate in one direction and
motors 3 and 4 in the opposite direction. This occurs so that neighboring rotors cancel out their torque. To increase or decrease altitude, it suffices to increase or decrease, respectively, the rotation
in every motor.
To move forward, it is necessary to increase the rotation in
motors 2 and 4, and decrease the rotation in motors 1 and 3 by
the same rate. The backward maneuver is done by the same logic.
The rightward motion is accomplished by increasing the rotation
in motors 2 and 3, and decreasing the motors 1 and 4 by the same
proportion. The leftward maneuver follows the same logic. The
rotation about its own axis (z axis) is accomplished by accelerating
and decelerating neighboring motors, by the same proportion. In
order to rotate clockwise, the motors that rotate counterclockwise
must be accelerated.

Figure 1.

Quadrotor orientation scheme.

IEEE A&E SYSTEMS MAGAZINE

MARCH 2017

Table of Contents for the Digital Edition of Aerospace and Electronic Systems Magazine March 2017