Estimating Emotion with Biological Information for Robot Interaction

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

To estimate emotion is difficult, not only for others, but even for people themselves. However, this is important for robots of the future, that are expected to operate in harmony with humans. This study proposes a method of estimating emotion using involuntary biological information. To be able to estimate emotion and feeling, a lot of work has been done in the field of artificial intelligence and robot engineering that focuses on human robot communications, especially where it applies to therapy. Generally, estimating emotions of people is based on expressed information such as facial expression, eye-gazing direction and behaviors that are observable by the robot. However, sometimes this information would not be suitable, as some people do not express themselves with observable information. In this case, it is difficult to estimate the emotion even if the analysis technologies are sophisticated. The main idea of our proposal is to use biological information, brain waves and heart rate for estimating the actual emotion of people that is the result or the nonconscious brain. The first experiment shows that our suggested method will outperform the traditional method, for the people who cannot express emotion directly. And, after changing the technique that measures the degree of joy for each scene and compared it with the subjective evaluation, the second experiment was performed. In the second experiment, accuracy did not change, but accuracy differed greatly for different people. In the analysis, we have found that there is a correlation between parts of the personality and the accuracy of results.

Original languageEnglish
Pages (from-to)1589-1600
Number of pages12
JournalProcedia Computer Science
Volume112
DOIs
Publication statusPublished - 2017 Jan 1

Fingerprint

Robots
Brain
Experiments
Artificial intelligence
Communication

Keywords

  • Biological Information
  • Concentration
  • Therapy

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Estimating Emotion with Biological Information for Robot Interaction. / Ikeda, Yuhei; Horie, Ryota; Sugaya, Midori.

In: Procedia Computer Science, Vol. 112, 01.01.2017, p. 1589-1600.

Research output: Contribution to journalArticle

@article{96f4af68127849a8a3dbe4062f90e8e5,
title = "Estimating Emotion with Biological Information for Robot Interaction",
abstract = "To estimate emotion is difficult, not only for others, but even for people themselves. However, this is important for robots of the future, that are expected to operate in harmony with humans. This study proposes a method of estimating emotion using involuntary biological information. To be able to estimate emotion and feeling, a lot of work has been done in the field of artificial intelligence and robot engineering that focuses on human robot communications, especially where it applies to therapy. Generally, estimating emotions of people is based on expressed information such as facial expression, eye-gazing direction and behaviors that are observable by the robot. However, sometimes this information would not be suitable, as some people do not express themselves with observable information. In this case, it is difficult to estimate the emotion even if the analysis technologies are sophisticated. The main idea of our proposal is to use biological information, brain waves and heart rate for estimating the actual emotion of people that is the result or the nonconscious brain. The first experiment shows that our suggested method will outperform the traditional method, for the people who cannot express emotion directly. And, after changing the technique that measures the degree of joy for each scene and compared it with the subjective evaluation, the second experiment was performed. In the second experiment, accuracy did not change, but accuracy differed greatly for different people. In the analysis, we have found that there is a correlation between parts of the personality and the accuracy of results.",
keywords = "Biological Information, Concentration, Therapy",
author = "Yuhei Ikeda and Ryota Horie and Midori Sugaya",
year = "2017",
month = "1",
day = "1",
doi = "10.1016/j.procs.2017.08.198",
language = "English",
volume = "112",
pages = "1589--1600",
journal = "Procedia Computer Science",
issn = "1877-0509",
publisher = "Elsevier BV",

}

TY - JOUR

T1 - Estimating Emotion with Biological Information for Robot Interaction

AU - Ikeda, Yuhei

AU - Horie, Ryota

AU - Sugaya, Midori

PY - 2017/1/1

Y1 - 2017/1/1

N2 - To estimate emotion is difficult, not only for others, but even for people themselves. However, this is important for robots of the future, that are expected to operate in harmony with humans. This study proposes a method of estimating emotion using involuntary biological information. To be able to estimate emotion and feeling, a lot of work has been done in the field of artificial intelligence and robot engineering that focuses on human robot communications, especially where it applies to therapy. Generally, estimating emotions of people is based on expressed information such as facial expression, eye-gazing direction and behaviors that are observable by the robot. However, sometimes this information would not be suitable, as some people do not express themselves with observable information. In this case, it is difficult to estimate the emotion even if the analysis technologies are sophisticated. The main idea of our proposal is to use biological information, brain waves and heart rate for estimating the actual emotion of people that is the result or the nonconscious brain. The first experiment shows that our suggested method will outperform the traditional method, for the people who cannot express emotion directly. And, after changing the technique that measures the degree of joy for each scene and compared it with the subjective evaluation, the second experiment was performed. In the second experiment, accuracy did not change, but accuracy differed greatly for different people. In the analysis, we have found that there is a correlation between parts of the personality and the accuracy of results.

AB - To estimate emotion is difficult, not only for others, but even for people themselves. However, this is important for robots of the future, that are expected to operate in harmony with humans. This study proposes a method of estimating emotion using involuntary biological information. To be able to estimate emotion and feeling, a lot of work has been done in the field of artificial intelligence and robot engineering that focuses on human robot communications, especially where it applies to therapy. Generally, estimating emotions of people is based on expressed information such as facial expression, eye-gazing direction and behaviors that are observable by the robot. However, sometimes this information would not be suitable, as some people do not express themselves with observable information. In this case, it is difficult to estimate the emotion even if the analysis technologies are sophisticated. The main idea of our proposal is to use biological information, brain waves and heart rate for estimating the actual emotion of people that is the result or the nonconscious brain. The first experiment shows that our suggested method will outperform the traditional method, for the people who cannot express emotion directly. And, after changing the technique that measures the degree of joy for each scene and compared it with the subjective evaluation, the second experiment was performed. In the second experiment, accuracy did not change, but accuracy differed greatly for different people. In the analysis, we have found that there is a correlation between parts of the personality and the accuracy of results.

KW - Biological Information

KW - Concentration

KW - Therapy

UR - http://www.scopus.com/inward/record.url?scp=85032384730&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85032384730&partnerID=8YFLogxK

U2 - 10.1016/j.procs.2017.08.198

DO - 10.1016/j.procs.2017.08.198

M3 - Article

VL - 112

SP - 1589

EP - 1600

JO - Procedia Computer Science

JF - Procedia Computer Science

SN - 1877-0509

ER -