Real-time Human-like Robot Control based on Brain-Computer Interface

Abstract

Is the brain computer interface (BCI) ready to be used in daily-life contexts by patients with severe motor dysfunction? In this talk, possible engineering applications for the use of advanced BCI findings, by using the electroencephalogram and magnetoencephalogram will be described (e.g. increase the degree of freedom of BCI control-based applications). A brain computer interface provides the possibility of communicating with environment by controlling any external device (e.g., artificial robotic arm) for patients with severe motor dysfunction using their brainwaves only. However, BCI researchers have been using only unilateral movements for decoding motor imagery, which are not enough for daily life activities. To provide multidimensional control, Belkacem et al. have described the first reported decoding of bilateral hand movements by using single-trial magnetoencephalography (MEG) signals as a new approach to enhance a user's ability to interact with a complex environment through a multidimensional BMI. In addition, bilateral movement requires multitask processing whose neural correlates remains unclear. In order to study the multitasking neural process, Belkacem et al. have analyzed and decoded the brain activity for four types of bilateral hand movements using noninvasive MEG measurement. Furthermore, they have been working on extending the 2D control using bilateral arms to 3D control based on brainwaves using an additional third arm. The interest in advanced emerging technologies-based applications has guided BCI researchers to develop necessary and useful BCIs for rehabilitation, brain monitoring and neurological disorder diagnosis, and medical and non-medical game entertainment. New challenges are now coming from BCIs to be available in the markets that will get attention of healthcare companies to work together to facilitate the improvement of the quality of life of disabled users.