Towards Hybrid Multimodal Brain Computer Interface for Robotic Arm Command

Abstract

A hybrid brain-computer interface (BCI) is a system that combines multiple biopotentials or different types of devices with a typical BCI system to enhance the interaction paradigms and various functional parameters. In this paper we present the initial development of a hybrid BCI system based on steady state evoked potentials (SSVEP), eye tracking and hand gestures, used to command a robotic arm for manipulation tasks. The research aims to develop a robust system that will allow users to manipulate objects by means of natural gestures and biopotentials. Two flickering boxes with different frequencies (7.5 Hz and 10 Hz) were used to induce the SSVEP for the selection of target objects, while eight channels were used to record the electroencephalographic (EEG) signals from user’s scalp. Following the selection, the users were able to manipulate the objects from the workspace by using the Leap Motion controller to send commands to a Jaco robotic arm.