We present insights gleaned from a series of child-robot interaction experiments carried out in a hospital paediatric department. Our aim here is to share good practice in experimental design and lessons learned about the implementation of systems for social HRI with child users towards application in "the wild", rather than in tightly controlled and constrained laboratory environments: a trade-off between the structures imposed by experimental design and the desire for removal of such constraints that inhibit interaction depth, and hence engagement, requires a careful balance.
Categories and Subject Descriptors
General TermsDesign, Experimentation, Human Factors, Theory.
KeywordsChild-robot interaction, experimental practice, human-robot interaction, interaction design, social robotics.
THE CHALLENGES OF LONG-TERM HUMAN-ROBOT INTERACTIONOver the last few decades electromechanical progress in robotics has outpaced developments in artificial cognition. This has been noticeable across the various sub-fields of . robotics, but never more so than in social robotics. Robots embedded in a social environment not only need to deal with the challenges encountered by field robots, but in addition need capabilities for coping with highly dynamic and stochastic elements inherent in the social milieu. As such, social robots have to face a number of challenges which are to a large extent still beyond the current state of the art in science and technology. The three main challenges yet to be surmounted by social robotics are:1. Contingent responding in dynamic social environments:In field robotics, an action by the robot does not necessarily induce a significant reaction from the environment: an unmanned vehicle driving around a rock, does not precipitate a response from the rock. In a social environment however, an action by a robot will almost always prompts some reaction from the user or users. Endowing robots with capabilities for contingent and appropriate handling of user responses, plus some capacity for the prediction of responses to their subsequent actions, is one of the main challenges in HRI.2. Interpretation of social sensory information: in contrast to robots which do not engage in interaction over natural communication channels, social robots are required to make sense of social signals. For interactions that go beyond proto-social responses, this has proven to be a tremendous challenge. The multi-modal aspect of social interaction thus continues to elude us, but even the interpretation of single modalities, such as gesture or language, has proven to be extremely challenging. The early promise of multi-modal communication channels constraining the interpretation of social information has delivered limited progress.3. Designing sociable robots: while progress in the electromechanical design of robots has been impressive and we seem to have a fairly clear understanding of which elements of the physical embodiment of a robot engage the user, the biggest challenge remains the creation believable, contingent...