Neurobotics: blended brain-machine control for human assistance using hybrid smart systems

Connecting brains to machines will change the lives of thousands of patients suffering from brain or spinal cord disorders. However, driving the many inputs of an assistive articulated arm or wheelchair through brain signals only has proven impractical and conversely, artifical control systems have not yet reached a point where they can reliably cope with the complexity of the real world entirely on their own. If we want to help patients with brain or spinal cord injuries regain autonomy in their daily life by means of robotics proxies, it is crucial to integrate the subject’s brain signals with the artificial-intelligence layers of their robotic helpers. Therefore, we want to decode neural activity recorded in human and nonhuman primates by means of an invasive brain-machine interface and blend these signals with computer vision and robotic systems into ‘hybrid’ intelligent systems, with human-robot shared control as a guiding principle. Our multidisciplinary approach will pave the way for restoring independent living by means of smart robot assistance in patients with untreatable brain disorders.

Keywords:brain-machine interface, artificial intelligence, mirror system, motor control, robotics, vision

Disciplines:Neurophysiology, Adaptive agents and intelligent robotics, Knowledge representation and reasoning