Assistive Technology Design And Preliminary Testing Of A Robot Platform Based On Movement Intention Using Low-Cost Brain Computer Interface

The process through which children learn about the world and develop perceptual, cognitive and motor skills relies heavily on object exploration in their physical world. New types of assistive technology that enable children with impairments to interact with their environment have emerged in recent years, and they could be beneficial for children's cognitive and perceptual skills development. Many studies have reported on brain computer interface (BCI) research. However, a conventional electroencephalography (EEG) system is generally bulky and expensive. It also requires special equipment and technical expertise to operate successfully. In this study, users EEG signals related to movement intention were measured and used as an input for a mobile robot control. EEG signals of three non-disabled adults were acquired by a compact low-cost BCI system and the movement intention was classified during physical movement and motor imagery. The average classification accuracies achieved during testing were 56.4% for the motor imagery and 72.7% for the physical movement. The results show moderate classification accuracy for the motor imagery; however, the classification accuracy for the physical movement showed high for all the subjects. Even though further improvement of the system is still needed, the experimental results demonstrated the feasibility of a BCI-based robotic system that is affordable and accessible for many people including children with disabilities.