Correcting Robot Mistakes in Real Time Using EEG Signals
Lab member Andrés F. Salazar-Gómez worked on the paper titled, Error-related potential for human-robot interaction: Correcting Robot Mistakes in Real Time Using EEG Signals.
In this project we are using EEG signals to automatically detect errors made by a robot. Our work focuses on identifying Error-related Potentials (ErrP), EEG signals generated in response to an observed or executed error, to correct in real time a robot’s selection. Our results demonstrate the potential of ErrPs for seamless robotic control, and moves closer towards the goal of real time intuitive human-robot interaction.
This video is part of results submitted to the IEEE International Conference on Robotics and Automation (ICRA 2017) in the paper “Correcting Robot Mistakes in Real Time Using EEG Signals”.