Matthew S. Johannes
Semi-autonomous brain-machine interface that uses an intelligent control scheme to aid users in neutrally controlling a robotic limb.
We have developed a semi-autonomous brain-machine interface that uses an intelligent control scheme to aid users in neurally controlling a robotic limb. We have enabled two users to use a hybrid of their eye-tracking signals and their electrocorticography (ECoG) cortical signals to control the modular prosthetic limb developed by the JHU Applied Physics Laboratory. Users were able to reliably manipulate objects presented in a workspace in front of them without the large cognitive burden typically associated with using neural signals to directly each degree of freedom of a robotic apparatus. This marks the first brain-machine interface to use invasive cortical signals to deliver shared control of a robotic limb to human users.