Towards Movement Intention.pdf (1.82 MB)

Towards symbiotic human-robot collaboration: human movement intention recognition with an EEG

Download (1.82 MB)
conference contribution
posted on 02.03.2021, 11:07 by Achim Buerkle, Niels Lohse, Pedro Ferreira
In order to meet the trend of customers demanding more customised and complex products, human workers and robots need to collaborate in closer proximity. Working in shared environments raises safety concerns of humans getting injured by robots. Current safety systems are mostly vision based and detect movement after it has started. This work proposes the use of an electroencephalography (EEG) which measures the brainwaves in order to detect a worker’s intention to move. This is expected to provide 0.5 seconds gain for the system to react in advance of the actual movement. In this paper the details on how EEG sensors can be integrated to detect intentions and how these can be extrapolated using machine learning techniques, are presented. The ultimate vision is to deliver an early warning system to enhance existing safety systems.

History

School

  • Mechanical, Electrical and Manufacturing Engineering

Published in

UK-RAS19 Conference: “Embedded Intelligence: Enabling & Supporting RAS Technologies” Proceedings

Pages

52 - 55

Source

2nd UK-RAS Robotics and Autonomous Conference, Loughborough, 2019

Publisher

UK-RAS Network

Version

VoR (Version of Record)

Rights holder

© UK-RAS Network

Publication date

2019-01-24

Language

en

Location

Loughborough, UK

Event dates

24th January 2019 - 24th January 2019

Depositor

Mr Achim Buerkle. Deposit date: 1 March 2021