Utilising BCI Techniques for Controlling Robotic Arm Movements by Motor Imagery Signals

This research paper explores the use of brain-computer interface (BCI) techniques for controlling arm movements through motor imagery (MI) signals. BCI technology enables communication and control between the brain and computers, utilizing brain activity to operate virtual entities without physical movement. Our study compares classification approaches, such as K-nearest neighbor (KNN) and support vector machine (SVM), to distinguish between left and right-hand MI based on EEG trials. Both classifiers achieve satisfactory accuracy in distinguishing MI-induced EEGs. The study highlights the application of MI-based BCI systems as an alternative communication and control channel for individuals with motor impairments. By employing Mel Frequency Cepstral Coefficients (MFCC) for feature extraction, known for their effectiveness in speech recognition and music retrieval, we contribute to the development of BCI-controlled robotic arm systems for rehabilitation. Our findings offer insights into effective motor imagery signal classification, promoting the development of BCI-based arm movement control during rehabilitation.