Drone Control Using Functional Near-Infrared Spectroscopy

Abstract

In this paper, we have used initial dip detection to generate early brain command for the drone control. The initial dips in functional near-infrared spectroscopy (fNIRS) signals are detected by using the vector phase analysis for the development of brain-computer interface (BCI). fNIRS signals are obtained from the prefrontal cortex of five healthy participants during the mental arithmetic task. In order to generate the early brain command, we have used vector phase analysis with a threshold circle based on the peak value from the resting state hemodynamics as a decision criterion. We are able to transmit a command to quadcopter in 0.1 sec after the initial dip is detected. The generated command is used to control the forward direction of the drone while keeping the forward velocity at 0.5 m/sec. In comparison to conventional classification method (support vector machine in this paper), our method was able to achieve 18% better control of the quadcopter. The results show the significance of the proposed method for BCI.