Control of Soap Bubble Ejection Robot Using Facial Expressions

Control of Soap Bubble Ejection Robot Using Facial Expressions

Naoya Hasegawa, Yoshihiko Takahashi
DOI: 10.4018/IJMMME.2021040101
OnDemand:
(Individual Articles)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

This research has developed a soap bubble ejection robot as an amusement system that reads emotions from human facial expressions and controls the ejection of soap bubbles to improve human-robot interaction. A subject's response to soap bubble ejection is read by a built-in face recognition sensor which sends data to a control system which in turn controls the next ejection. Soap bubbles are often used to research children's emotions/emotional responses. First, evaluation experiments of the control system were performed using face photographs that show human emotions. The experimental results revealed that soap bubbles were ejected in the case of indifference, and the ejection stopped in the case of joy. Through the experimental results, it was confirmed that the control system worked properly when face photographs were used and also verified the effectiveness of the facial recognition sensor. Secondly, evaluation experiments were conducted with an actual human, and it was confirmed from the results that the control system operates as designed.
Article Preview
Top

1. Introduction

It is expected that mechanical systems that take our emotions into consideration, in order to improve the human experience, will increase. In order to realize control of such mechanical systems, advanced technology that fuses mechanical and ergonomic engineering is required. However, it cannot be said that there has been sufficient research in this field so far. This is because human emotional expressions are extremely complex and difficult to gauge and evaluate. Furthermore, it is very difficult to control any mechanical system using such complex data. Therefore, in this research, we used a soap bubble machine because humans usually react to such machines in a positive, and more easily gaugeable way. The fusion of mechanical engineering and ergonomics could be the best way to enrich humans emotionally. We hope that many such studies will be published.

At various exhibitions, amusement robots interact with humans. Questionnaires and interviews are often used to evaluate how humans feel when a robot moves (Odashima et al., 2014). It takes a long time for the data from these questionnaires and interviews to be collected, analyzed and incorporated in robot design. However, short-term or real-time feedback is desired. There are examples of research on human facial expressions (Kitamura et al., 2014) and on human responses to robot facial expressions (Takahashi et al., 2006), but not real-time robot control.

GSR (Galvanic Skin Reflex) sensors are often used as a method for measuring human psychological stress. There have been many basic research cases over the years (Yokota et al., 1959; Fujimori et al., 1962; Darrow et al., 1970; Panigrahy et al., 2017). In addition, GSR has been applied in many fields: food support robots (Takahashi et al., 2001 and 2002), stress measurement for workers (Sriramprakash et al., 2017; Jeroh et al., 2018), driver's stress measurement (Dogan et al., 2019), stress measurement for wheelchair users (Shirai et al., 2018 and 2019), measurement of effects of high frequency sound (Fletcher et al., 2018), and so on. Measurement systems using GSR have also been studied: a ZigBee system (Villarejo et al., 2012), a wristband sensor (Kikhia et al., 2016).

However, because GSR sensors can only detect emotional change and cannot identify emotions such as Joy or Sadness, it is limited in its scope. In order to incorporate such human emotions in robot control, it is necessary to identify them.

Many children seem to like soap bubbles. Therefore, there have been many studies which evaluate child emotion/reaction to soap bubbles (Yamada, 2017; Kato, 2013; Matsumura et al., 2016; Yabe, 2012). Also, events using soap bubbles and bubble robots have been studied (Nakamura et al., 2006; Hasegawa et al., 2017, 2019a, and 2019b).

Complete Article List

Search this Journal:
Reset
Volume 13: 1 Issue (2024)
Volume 12: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 11: 4 Issues (2021)
Volume 10: 4 Issues (2020)
Volume 9: 4 Issues (2019)
Volume 8: 4 Issues (2018)
Volume 7: 4 Issues (2017)
Volume 6: 4 Issues (2016)
Volume 5: 4 Issues (2015)
Volume 4: 4 Issues (2014)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing