Motor imagery, P300 and error-related EEG-based robot arm movement control for rehabilitation purpose

Abstract

The paper proposes a novel approach toward EEG-driven position control of a robot arm by utilizing motor imagery, P300 and error-related potentials (ErRP) to align the robot arm with desired target position. In the proposed scheme, the users generate motor imagery signals to control the motion of the robot arm. The P300 waveforms are detected when the user intends to stop the motion of the robot on reaching the goal position. The error potentials are employed as feedback response by the user. On detection of error the control system performs the necessary corrections on the robot arm. Here, an AdaBoost-Support Vector Machine (SVM) classifier is used to decode the 4-class motor imagery and an SVM is used to decode the presence of P300 and ErRP waveforms. The average steady-state error, peak overshoot and settling time obtained for our proposed approach is 0.045, 2.8% and 44 s, respectively, and the average rate of reaching the target is 95%. The results obtained for the proposed control scheme make it suitable for designs of prosthetics in rehabilitative applications.

EEG based gesture mimicking by an artificial limb using cascade-correlation learning architecture

“… Jaco robot arm [1]- [4], [9] (manufactured by Kinova) is manipulated based on the information obtained from EEG signal acquisition device and Kinect sensor.…”

“…Now the target subject is wearing an EEG sensor [2]- [4], [8]- [12] to record physiological signals from the brain. Also his/her body gestures are captured by the Microsoft's Kinect sensor [5]- [9], [11]- [14].…”

“…Maheu et al [1] states the advantages of Jaco arm to do tasks related to rehabilitation. Bhattacharyya et al proposed how the efficient movement of robot can be done from the acquired EEG signals [3], [4]. All of these literatures are based on directing the robot using EEG signals.…”

“…To implement our novel system of gesture mimicking based on EEG signals for artificial limb movement, we require three devices, namely EEG acquisition device [2]- [4], [8]- [12], Kinect sensor [5]- [9], [11]- [14] and Jaco robot arm [1]- [4], [9]. For healthy subjects, when a ball is thrown, then they are instructed to catch the ball.…”

“…1 depicts the schematic diagram of our proposed system. The system utilizes three devices, namely  21-channel EEG signal acquisition device (manufactured by Nihon Kohden) [2]- [4], [8]- [12] to record the physiological signals from the brain.…”