Brain-Driven Telepresence Robots: A Fusion of User’s Commands with Robot’s Intelligence

Abstract

This paper presents different methodologies to enhance the human-robot interaction during the control of brain-machine interface (BMI) driven telepresence robots. To overcome the limitations of BMIs, namely the low bit rate and the intrinsic uncertainty as a control channel, we hypothesize that the fusion of the user’s commands with the robot’s intelligence is essential to achieve robust and natural systems. Compared to most current neurorobotics works, we exploit the robot as an intelligent agent that contributes at different levels to choose the final action to perform. Furthermore, we present the first implementation of a BMI system inside the Robot Operating System (ROS) designed to facilitate the combination between BMI and robotics.