Start Date

9-5-2019 1:15 PM

End Date

9-5-2019 2:45 PM

Document Type

Event

Keywords

Request, Help, Autonomous mobile robots, Human-robot interaction, Source of control, Situational awareness

Description

Most attention in the development of robots has been focused on the robots’ hardware and software to improve functionality. However, the interaction between humans and robots can be key to enhancing robots’ performance. Autonomous mobile robots can more often complete their tasks if they are enabled to effectively solicit help from humans when robots face limitations. Hence, in this paper, we examine three different manners of addressing humans: friendly, polite, and indirect. These approaches combine different interaction modalities, like speech and gestures. We conducted experiments where the NAO robot addresses bystanders in one of the aforementioned three manners to get help with opening a public door. The aim of the experiment was to identify the best ways for robots to successfully draw the attention, interact and request help from humans. We investigated if passers-by would respond differently to each manner. Each passer-by was then asked to fill a questionnaire to better understand and classify his actions. In addition, we measure the effects of bystanders’ source of control assumptions and the situational awareness desire from the response of the bystanders, with respect to the robot request.

DOI

https://doi.org/10.5038/CETI2288

Share

COinS
 
May 9th, 1:15 PM May 9th, 2:45 PM

Robots That Successfully Elicit Help from Humans

Most attention in the development of robots has been focused on the robots’ hardware and software to improve functionality. However, the interaction between humans and robots can be key to enhancing robots’ performance. Autonomous mobile robots can more often complete their tasks if they are enabled to effectively solicit help from humans when robots face limitations. Hence, in this paper, we examine three different manners of addressing humans: friendly, polite, and indirect. These approaches combine different interaction modalities, like speech and gestures. We conducted experiments where the NAO robot addresses bystanders in one of the aforementioned three manners to get help with opening a public door. The aim of the experiment was to identify the best ways for robots to successfully draw the attention, interact and request help from humans. We investigated if passers-by would respond differently to each manner. Each passer-by was then asked to fill a questionnaire to better understand and classify his actions. In addition, we measure the effects of bystanders’ source of control assumptions and the situational awareness desire from the response of the bystanders, with respect to the robot request.