Cognitive load estimation for adaptive human-machine system automation

Document Type

Article

Publication Date

1-1-2020

Publication Title

Learning Control: Applications in Robotics and Complex Dynamical Systems

First Page

35

Keywords

Cognitive load detection, Detection, Eye-tracking, Human-computer interface, Machine learning, Psychophysiological signals, Pupil dilation, Signal processing

Last Page

58

Abstract

An important goal in automation is to create machines that are able to better understand human cognitive states so that the overall system efficiency can be enhanced. For example, an advanced driver assistance system (ADAS) equipped with the ability to understand the cognitive state of human will enhance the overall safety of the roads. Similar scenarios can be found in numerous emerging industries, such as smart-manufacturing, robotics, virtual reality, video games, media, online learning, and entertainment. In order to achieve such an intelligent automation system involving humans we need to develop approaches that can estimate cognitive load through non-invasive means and to develop control strategies for real-time system adaptation with humans. The focus of this chapter is on presenting some recent advances in cognitive load estimation based on non-invasive measures, such as pupil diameter, eye-gaze patterns, eye-blink patterns, heart rate, and heart-rate variability. Finally, we present some results from an experiment where the pupil diameter data, among other measures, was collected for varying cognitive difficulty levels.

DOI

10.1016/B978-0-12-822314-7.00007-9

ISBN

9780128223147

Link to Full Text

Share

COinS