Investigation of impressions for approach motion of a mobile robot based on psychophysiological analysis

Abstract

This paper investigates impressions of approach motions of a mobile robot based on psychophysiological analysis. In one of our previous studies, we suggested that actuation noise caused by the robots tended to raise the sympathetic nervous system (SNS) response of heart rate variability. In another experiment it was observed that blocking out either the sound or the sight of the robot motion attenuated the electrodermal activity (EDA), which reflects the SNS. From these investigations, one candidate for motion rules for humanfriendly robots was deduced such that robots must reduce their motion speed in the immediate vicinity of human. To confirm validity of the motion rule, we constructed an experimental setup with a mobile robot approaching humans at several speeds, and investigated the human impressions by means of psychophysiological methods. The experimental results showed that robot motion adjacent to humans tended to increase EDA responses. We found especially that the approach motion tented to give stronger stimuli to humans than motions at a distance. The faster approach motion tended to increase EDA responses, but there was no significant statistical difference from the response to slower approach motion. From factor analysis of the subjective ratings two factors were extracted, which were interpreted as "relief of mind" and "observation on motion."

Human motion behavior while interacting with an industrial robot

“…The question whether the distance between human and robot, the velocity of the robot, or interference of the two variables, have the deepest impact on human perception, is not answered consistently. Tejima [13] and Hanajima [3] figured out, that the distance between the two entities has the biggest influence on human perception. On the other hand an interaction between the distance and the velocity was discovered as a result of the study of Yamada et al [15].…”

Navigation Among Humans

“…Other methods, like Hidden Markov Models, have been used to learn to identify the behaviour of humans Kelley et al (2008). Bayesian inference algorithms and Hidden Markov Models have also successfully been applied to modelling and for predicting spatial user information Govea (2007 Hanajima et al (2005). In Michalowski et al (2006) models are reviewed that describe social engagement based on the spatial relationships between a robot and a person, with emphasis on the movement of the person.…”

Interaction between human and robot

“…They focus on the two aspects of intent namely attention and approval, where attention was measured through gesture recognition and eye gaze tracking and approval was measured through facial expressions and physiological signals. Operator physiological response was also studied by Hanajima et al who investigated the impact of robot motion on operator's HRV and electrodermal activity (Hanajima, Ohta, Hikita & Yamashita 2005).…”