EMG estimation from EEGs for constructing a power assist system

Brain-Machine Interface (BMI), has been becoming gradually an effective way to help and support the disabled person by using his/her brain activity information. It is also expected to be used for both of the disabled and the healthy people. In this paper, aiming to estimate the force/torque information from brain activity to help and support human's daily life, we estimate the humans muscular activity from elec-troencephalography (EEG) by Principal Components Analysis (PCA) and Recursive Least Squares (RLS). The concept of the proposed approach using PCA and RLS is explained, and then the proposed approach is verified by experiments. The results show that the estimation of electromyography (EMG) from EEG is possible and this implies a great potential to use EEG signals for supporting human activities.