Work

A Brain-Machine Interface for a Sequence Movement Control of a Robotic Arm

Year: 2017
Type: conference-paper
Abstract: Brain Machine Interfaces (BMI) have become of interest during the past years. The brain activities are recorded by invasive or noninvasive approaches and translated into command signals to control external prosthetic devices such as computer cursor, wheelchair, and robotic arm. Although many studies confirmed the capability of BMI systems in controlling multi-degrees of freedom (DOF) prosthetic devices using invasive approaches, this area of research in noninvasive approaches is at the beginning stage. In this work, a new BMI robotic platform has been developed using noninvasive Electroencephalography (EEG) technology. EEG signals were acquired using 14 channels and through BCI2000 software (with high pass filter at 0.1Hz and low pass filter at 30Hz). A low-cost 6-DOF robotic arm was used for real-time object manipulation between two fixed points in the workspace. A successful manipulation task consisted of 6 sequential movements of the robotic arm to pick up an object from a point on the left side of the workspace and drop it at a fixed point on the right side of the workspace. The programmed movements were controlled by a developed 4-target cursor control task. The subject was instructed to use imagined body kinematics to control the cursor and the robotic arm movements by hitting the targets. After initial calibration and testing, the subject attempted to control the robotic arm. The most efficient target sequence consisted of hitting 6 targets in sequential order. A successful manipulation run was reported when the object was placed at the final point regardless of any mistakes. Among 10 runs of manipulation runs, a 70% success rate was reported for the overall object manipulation task (Table 1). The experiments here serve as a concept proof for the feasibility of developing a noninvasive BMI robotic platform for manipulation task with minimum training. In future work, we will test the platform on a greater subject population with more complicated tasks such as manipulating an object between two arbitrary locations and in three dimensional space.
Source: HAL
Language: English
Open Access status: green
OpenAlex ID: W2906003822
FWCI: 1.24
Cites:
Cited by: 3
Related to: 10