Prosthetic AI Enabled Arm for Rehabilitation and Advanced Dynamics

Abstract

Brain Computer Interface (BCI) is wide range of system were signal generated by the human brain is transformed into commands/messages that are communicated via computer or robotic limb to the outside world. In the presented research here, Motor Imagery based Brain Computer Interface (MI-BCI) to control the prosthetic hand is proposed. The hand features an electric motor and an angle mechanism to deliver haptic feedback and enable local machine control. With the utilization of this system, participants demonstrated the capacity to regulate the grasp of the prosthesis with an accuracy close to that of the control scheme. The SVM classification algorithm is employed to interpret and transmit commands for operating the prostheses. Utilizing model predictions as commands for device control and other Brain-Computer Interface (BCI) applications, real-time brain signal input has been incorporated into the user interface. Based on the conducted pragmatic study, Random Forest delivers better efficiency in terms of accuracy in comparison to other machine learning classifiers.