The Future of Work: Towards Service Robot Control through Brain-Computer Interface

Methods for brain-computer interfaces (BCI) have undergone major advancements in the past decade, providing mobility for disabled individuals through the control of wheelchairs and prosthetic devices. However, not many advancements have been made in allowing people with disabilities to control a robot in a remote location. Currently, people with disabilities are heavily underrepresented in the labor force. This paper presents the idea of using telepresence robots controlled by BCI in the workforce, providing the ability for a disabled individual to work in a wide variety of scenarios. This can be achieved by developing a system with 4 major components: Data Acquisition, Digital Signal Processing, Feature Extraction, and Classification. This work presents the preliminary results toward using traditional data acquisition, signal processing, feature extraction, and classification methods in order to control the primitive movements of a wheeled telepresence robot. The results show a promising classification accuracy of navigational direction for the telepresence robot. Future work will explore using state-of-the-art methods in data acquisition, signal processing, feature extraction, and classification to control a telepresence robot using a BCI headset, and other future work will also demonstrate the near real-time classification of BCI signals for robot control.