An Indoor Navigation Control Strategy for a Brain-Actuated Mobile Robot

Abstract

In this paper, a brain-machine interface (BMI) with the capability of navigating a mobile robot in indoor environment, is proposed. The BMI, which is based on steady state visually evoked potentials (SSVEP), utilizes canonical correlation analysis (CCA) algorithm to classify electroencephalogram (EEG) signals and then translates the recognition results into motion commands for the trajectory planning based on artificial potential field (APF). In the navigation control strategy, the combination of particle filter-based simultaneous localization and mapping (S-LAM) and EEG-APF method is proposed to navigate the robot to a goal point with free obstacles, where SLAM algorithm would build a global map of the corridor and EEG-APF method would produce a trajectory free of obstacles. Extensive experiments have been conducted to verify the effectiveness of our system.