Final Activity Report Summary - BRAIN2ROBOT (A Robotic-Arm Orthosis Controlled by Electroencephalography and Gaze for Locked-In Paralytics)
In the MC-EXT project BRAIN2ROBOT a significant step forward was taken in adapting non-invasive brain computer interfaces for neuroprosthetic applications such that they may eventually enter the homes of severely disabled persons with needs for such assistive technology. An assistive robot arm was guided by a combination of eye and head tracking and electroencephalography (EEG) based brain computer interface (BCI) in a demonstration application, whereby subjects (both healthy and disabled) were able to grab objects from a table without moving, a level of functionality previously achieved only with invasive brain computer interfaces in non-human subjects.
For this purpose, a 3-dimensional gaze tracking system had to be developed and implemented, combining head and eye tracking, and the Berlin Brain Computer Interface developed at the host institute had to be adapted for robotic control. The resulting combination of eye-tracking and BCI was shown to be complementary and beneficiary for human-machine interface use. Importantly, a cooperative investigation was undertaken in which important lessons were learned from experiments involving 9 tetraplegic volunteers, about the proficiency of BCI users in the targeted population, the challenges of BCI training and use for these users, and the feasibility and rehabilitative benefit of robot control by gaze and noninvasive BCI. The gaze-BCI-robot control application was also shown in various public demonstrations - this receiving significant public and media exposure.
While the core technology of Brain2Robot promises eventual benefits to disabled persons, at little or no risk due to its non-invasive nature, the basic components of the user interface must be made more practical, at lower cost, such that they may enter everyday use. To this purpose, patented innovation in sensor technology was brought forth: dry contact electrodes were developed which can be applied within minutes and used repeatedly, unlike the previous standard of EEG recording equipment which about a half-hour of expertly applied liquid gel.
This technology was also publicly demonstrated using volunteers, another first in BCI development. Other advances led by the Team and its members include the development of a method of BCI use which does not require training - also needed to enhance practicability of BCI use, the use of EEG BCI in motor adaptation to continuous position control towards future wearable neuroprosthetics and the testing of video- and virtual reality- inspired feedback and training cues for improving BCI performance in novel subjects.
For this purpose, a 3-dimensional gaze tracking system had to be developed and implemented, combining head and eye tracking, and the Berlin Brain Computer Interface developed at the host institute had to be adapted for robotic control. The resulting combination of eye-tracking and BCI was shown to be complementary and beneficiary for human-machine interface use. Importantly, a cooperative investigation was undertaken in which important lessons were learned from experiments involving 9 tetraplegic volunteers, about the proficiency of BCI users in the targeted population, the challenges of BCI training and use for these users, and the feasibility and rehabilitative benefit of robot control by gaze and noninvasive BCI. The gaze-BCI-robot control application was also shown in various public demonstrations - this receiving significant public and media exposure.
While the core technology of Brain2Robot promises eventual benefits to disabled persons, at little or no risk due to its non-invasive nature, the basic components of the user interface must be made more practical, at lower cost, such that they may enter everyday use. To this purpose, patented innovation in sensor technology was brought forth: dry contact electrodes were developed which can be applied within minutes and used repeatedly, unlike the previous standard of EEG recording equipment which about a half-hour of expertly applied liquid gel.
This technology was also publicly demonstrated using volunteers, another first in BCI development. Other advances led by the Team and its members include the development of a method of BCI use which does not require training - also needed to enhance practicability of BCI use, the use of EEG BCI in motor adaptation to continuous position control towards future wearable neuroprosthetics and the testing of video- and virtual reality- inspired feedback and training cues for improving BCI performance in novel subjects.