A brain-machine interface to navigate a mobile robot in a planar workspace: Enabling humans to fly simulated aircraft with EEG

Abdullah Akce, Miles Johnson, Or Dantsker, Timothy Wolfe Bretl

Research output: Contribution to journalArticle

Abstract

This paper presents an interface for navigating a mobile robot that moves at a fixed speed in a planar workspace, with noisy binary inputs that are obtained asynchronously at low bit-rates from a human user through an electroencephalograph (EEG). The approach is to construct an ordered symbolic language for smooth planar curves and to use these curves as desired paths for a mobile robot. The underlying problem is then to design a communication protocol by which the user can, with vanishing error probability, specify a string in this language using a sequence of inputs. Such a protocol, provided by tools from information theory, relies on a human user's ability to compare smooth curves, just like they can compare strings of text. We demonstrate our interface by performing experiments in which twenty subjects fly a simulated aircraft at a fixed speed and altitude with input only from EEG. Experimental results show that the majority of subjects are able to specify desired paths despite a wide range of errors made in decoding EEG signals.

Original languageEnglish (US)
Article number6381524
Pages (from-to)306-318
Number of pages13
JournalIEEE Transactions on Neural Systems and Rehabilitation Engineering
Volume21
Issue number2
DOIs
StatePublished - Mar 19 2013

Keywords

  • Brain-machine interface (BMI)
  • information theory
  • robotics
  • semi-autonomous navigation

ASJC Scopus subject areas

  • Neuroscience(all)
  • Biomedical Engineering
  • Computer Science Applications

Fingerprint Dive into the research topics of 'A brain-machine interface to navigate a mobile robot in a planar workspace: Enabling humans to fly simulated aircraft with EEG'. Together they form a unique fingerprint.

  • Cite this