BIOMEDICAL ENGINEERING VOLUME: 20 | NUMBER: 2 | 2022 | JUNE
Drone Movement Control by
Electroencephalography Signals Based on BCI
System
Ali Hussein ABDULWAHHAB
1
, Indrit MYDERRIZI
2
, Musaria Karim MAHMOOD
3
1
Electrical and Electronics Engineering Program, Institute of Graduate Studies, Istanbul Gelisim University,
Duygu Sokak No: 2, 34310 Istanbul, Turkey
2
Department of Electrical and Electronics Engineering, Faculty of Engineering and Architecture,
Istanbul Gelisim University, Petrol Ofisi Caddesi No: 5, 34310 Istanbul, Turkey
3
Department of Energy Systems Engineering, Faculty of Engineering, Ankara Yildirim Beyazit University,
Gazze Caddesi No: 7, 06010 Ankara, Turkey
ahabdulwahhab@gmail.com, imyderrizi@gelisim.edu.tr, mkmahmood@aybu.edu.tr
DOI: 10.15598/aeee.v20i2.4413
Article history: Received Nov 14, 2021; Revised Mar 20, 2022; Accepted Apr 13, 2022; Published Jun 30, 2022.
This is an open access article under the BY-CC license.
Abstract. Brain Computer Interface enables indi-
viduals to communicate with devices through Elec-
troEncephaloGraphy (EEG) signals in many applica-
tions that use brainwave-controlled units. This paper
presents a new algorithm using EEG waves for control-
ling the movements of a drone by eye-blinking and at-
tention level signals. Optimization of the signal recog-
nition obtained is carried out by classifying the eye-
blinking with a Support Vector Machine algorithm and
converting it into 4-bit codes via an artificial neural
network. Linear Regression Method is used to cate-
gorize the attention to either low or high level with
a dynamic threshold, yielding a 1-bit code. The con-
trol of the motions in the algorithm is structured with
two control layers. The first layer provides control with
eye-blink signals, the second layer with both eye-blink
and sensed attention levels. EEG signals are extracted
and processed using a single channel NeuroSky Mind-
Wave 2 device. The proposed algorithm has been vali-
dated by experimental testing of five individuals of dif-
ferent ages. The results show its high performance
compared to existing algorithms with an accuracy of
91.85 % for 9 control commands. With a capability of
up to 16 commands and its high accuracy, the algorithm
can be suitable for many applications.
Keywords
Attention level, Brain Computer Interface
(BCI), ElectroEncephaloGraphy (EEG),
eye-blink, NeuroSky MindWave 2.
1. Introduction
Nowadays there is a huge demand for Brain Computer
Interface (BCI) that can be used in situations where
typical control interfaces are not an option. The con-
cept of BCI based system has been developed to pro-
vide alternate control methods for handicap people,
gamming and for special purpose applications [1]. BCI
is an interfacing technology between the Human Mind
(HM) and a processor by sensing ElectroEncephaloG-
raphy (EEG) signal and employing it to perform dif-
ferent tasks. There are two types of mind-sensing tech-
niques for the BCI system, which are invasive and the
non-invasive [2] and [3]. The invasive and/or partially
invasive sensing technique requires surgical interven-
tion for implanting the electrodes under the scalp to
communicate with the human brain. Although this in-
vasive sensing technique provides high sensing accuracy
and good signal-to-noise ratio, some scar tissues can be
formed after surgery causing weakness in the acquisi-
tion of the brain signal and a severe medical state [4].
The non-invasive sensing technique works by installing
the electrodes in external headset placed on scalp to
© 2022 ADVANCES IN ELECTRICAL AND ELECTRONIC ENGINEERING 216
