KAIST Launches Platform to Control Robots Using Brain Signals, 'New Rehabilitation Paradigm'

KAIST (President Ik Kwang-hyung) announced that a research team led by Professors Gongsung Cheol and Jeong Kim has initiated development of the world's first bidirectional brain-to-robot (Brain-to-Robot) system as part of a cross-ministerial advanced medical device R&D flagship project in collaboration with Angel Robotics. Current brain-computer interface technology enabling cursor movement or smartphone control via brain signals has entered clinical trials, with global companies like Neuralink and Synapse accelerating development. However, connecting actual movement and sensation remains challenging. The research team aims to create a complete bidirectional UI by reading users' movement intentions through brain signals to operate robots while relaying sensory feedback like ground reaction force, joint torque, and tactile information back to the brain. The target is an exoskeleton robot, with no prior global implementations of such a fully bidirectional system. KAIST's team leads core technology development, including wearable robot control and AI-based intent interpretation, alongside designing a somatosensory interface system to accurately transmit sensory data. The project also involves AI encoding-decoding algorithms to process large-scale cortical neural signals in real-time with minimal latency.