I was invited to give a talk on Human Interaction with Autonomous Agents at the ARTS 2nd Summer School called Autonomic Road Transport Support Systems: Behavioural Response and Impact.
I attended this venue on behalf of our research group of Human-Robot Interaction at the University of Twente lead by prof. dr. Vanessa Evers. So, I started with an introduction to our research and the current European projects we are involved in. And afterwards I presented some aspects of my own PhD research on Long-Term Acceptance of Social Robots in Home Environments.
Human-Robot Interaction at the University of Twente
Our research group focuses on human interaction with social technologies and our goal is to make these technologies more humanlike or socially intelligent in such a way that the interactions with these technologies are more intuitive. One of the current European projects we are working on is the Spencer project (http://www.spencer.eu/), which has the goal to employ a fully autonomous mobile robot for smart passengers flow management at Amsterdam Schiphol Airport in The Netherlands. Another EU-project we are involved in is the FROG project (https://www.frogrobot.eu/wordpress/), which aspires to turn autonomous outdoor robots into viable location-based service providers. And the last EU-project I would like to introduce is the Accompany project (http://accompanyproject.eu/), which will deliver a companion robot that provides services to elderly users in a motivating and socially acceptable way to facilitate independent living at home.
Key Findings of My PhD Research
The part about my own research started with some results I collected on people’s perceptions of robots. In an interview I asked people (N= 21) to tell me what they think of when hearing the word ‘robot’. Their answers mainly involved either descriptions of the appearance of robots or their utility. In terms of their appearance, most participants described and preferred robots with humanlike feature (e.g., having a body, a head, arms and legs). Only a few participants acknowledged that robot could also have an animal shape or that its embodiment should be purely functional. In terms of a robots utility, most people described robots as our helpers. Especially the idea of having a butler robot that could perform several household chores was admired, but also other types of supporting tasks were mentioned (e.g., fetching objects in and around the house, or help remembering things). Other purposes for a robots that people would prefer were playing music or videos, service as a walking encyclopedia, or as a companion. A separate online survey (N= 226) further investigated people’s preferences with robot embodiments. This study confirmed people’s preferences for humanoid robots, which they evaluated as more intelligent, more likeable, more safe, more animate, more socially present, and more sociable compared to either caricatured, zoomorphic or functional robots. Some of these results were presented as a LBR at the HRI conference 2014 (http://dl.acm.org/citation.cfm?id=2563683).
Next, I presented the results of my long-term home study. A summary of these findings can be found in my earlier post of March this year. Additionally, I presented my ideas of a (so far still) theoretical model of social robot acceptance. Some information on this can be found in my HRI Pioneers submission at the HRI conference 2014 (http://www.hripioneers.info/Proceedings/2014PioneersProceedings.pdf). At this very moment, I am working on the empirical validation of this model. I hope to be able to share my findings with you in the not so far future.
Last, I presented some initial results (N= approx. 400) on my evaluation of future roles that robots could have in society. Although people could see some positive outcomes of most scenarios, overall (and expectedly) they were quite negative about the more social roles for robots, such as teachers and nannies. When I have finalized the analysis of this data, I will share my findings with you.