In a new study, a team of researchers described their design of a transradial prosthesis controlled by electroencephalographic (EEG) signals as an alternative to electromyographic (EMG) signals, which can be challenging for users to execute.
The researchers collected EEG signal data using an Emotiv Insight Headset, which was then processed to control the movement of the prosthesis, known as the Zero Arm. The Zero Arm was 3D printed with PLA filament and easily obtainable servomotors and controllers, chosen by the research team to make the prosthesis affordable and accessible. It featured four degrees of freedom and a haptic feedback system based on skin stretching to simulate the function of mechanoreceptors in the skin, providing the user with a sense of touch when using the prosthesis.
Photograph courtesy of HardwareX and the study’s authors.
The control of the arm was carried out with a Raspberry Pi Pico card. The mechanical design included 29 modeled parts, including finger phalanges, palm, palm cap, garter cap, forearm, forearm cover, and finger thread guides. Machine learning algorithms were used to classify different types of objects and shapes.
Performance tests of the prosthesis demonstrated an average success rate of 86.67 percent across various tasks, indicating its reliability and effectiveness, the study’s authors concluded. Additionally, the prosthesis had an average recognition rate of 70 percent for different types of objects.
The open-access study, “A low-cost robotic hand prosthesis with apparent haptic sense controlled by electroencephalographic signals,” was published in HardwareX.
