Wireless brain-machine interface using EEG and EOG: brain wave classification and robot control

Abstract

A brain-machine interface (BMI) links a user's brain activity directly to an external device. It enables a person to control devices using only thought. Hence, it has gained significant interest in the design of assistive devices and systems for people with disabilities. In addition, BMI has also been proposed to replace humans with robots in the performance of dangerous tasks like explosives handling/diffusing, hazardous materials handling, fire fighting etc. There are mainly two types of BMI based on the measurement method of brain activity; invasive and non-invasive. Invasive BMI can provide pristine signals but it is expensive and surgery may lead to undesirable side effects. Recent advances in non-invasive BMI have opened the possibility of generating robust control signals from noisy brain activity signals like EEG and EOG. A practical implementation of a non-invasive BMI such as robot control requires: acquisition of brain signals with a robust wearable unit, noise filtering and signal processing, identification and extraction of relevant brain wave features and finally, an algorithm to determine control signals based on the wave features. In this work, we developed a wireless brain-machine interface with a small platform and established a BMI that can be used to control the movement of a robot by using the extracted features of the EEG and EOG signals. The system records and classifies EEG as alpha, beta, delta, and theta waves. The classified brain waves are then used to define the level of attention. The acceleration and deceleration or stopping of the robot is controlled based on the attention level of the wearer. In addition, the left and right movements of eye ball control the direction of the robot.

Eye blink characterization from frontal EEG electrodes using source separation and pattern recognition algorithms

“…Here, the blinks are just detected, but not characterized. Along the same lines, [16] detect horizontal eye movements from electrodes placed on the forehead in order to pilot a robot. Ref.…”

Recent Progress of Tactile and Force Sensors for Human–Machine Interaction

“…However, these sensors are attached to one arm and palm, requiring two hands to achieve dexterous 3D control, which is not easy for people with disabled arms. The combination of EOG and the noisy brain activity signals such as electroencephalography (EEG) to generate robust control signals for HMI can completely free hands to achieve wireless interaction [ 133 , 134 , 135 ]. However, due to the low signal-to-noise ratio of EEG signals and the complexity of signal acquisition, further research and application are needed.…”

Electrooculograms for Human–Computer Interaction: A Review

“…Some atypical electrode placements are shown in Figure 3c–f. The placement in Figure 3c involves embedding electrodes in a headband [27,42], which may increase the usability of the measuring device. Under this placement, the vertical eye movements are ignored.…”

“…In a comparable manner, some researchers suggested the use of EOG communication for robot control. Rusydi et al and Oh et al independently suggested moving a robot according to the moving direction and angle of the user’s eyes [41], and to control a robot’s moving direction and speed by using eye movements in four directions [42]. The studies of both Rusydi et al and Oh et al suggested the idea of the application without implementation.…”

“…Figure 12 shows eye- and eyelid-related signals observed at EEG channels. Similarly, Oh et al proposed a hybrid system utilizing both EEGs and EOGs, which classified eye closing/opening, and the movement of eyes in the left/right direction [42]. It was different from Ramli’s system in terms of the placement of electrodes: The EEGs and EOGs were measured from the prefrontal region in the proposed system.…”