Logo

Publikacije (45118)

Nazad
A. Chella, H. Dindo, Ignazio Infantino

Marek P. Michalowski, S. Šabanović, Reid G. Simmons

Even if a socially interactive robot has perfect information about the location, pose, and movement of humans in the environment, it is unclear how this information should be used to enable the initiation, maintenance, and termination of social interactions. We review models that have been developed to describe social engagement based on spatial relationships and describe a system developed for use on a robotic receptionist. The system uses spatial information from a laser tracker and head pose information from a camera to classify people in a categorical model of engagement. The robot's behaviors are determined by the presence of people in these different levels. We evaluate the system using observational behavioral analysis of recorded interactions between the robot and humans. This analysis suggests improvements to the current system: namely, to put a stronger emphasis on movement in the estimation of social engagement and to vary the timing of interactive behaviors

S. Šabanović, Marek P. Michalowski, Reid G. Simmons

This paper discusses the use of observational studies of human-robot social interaction in open human-inhabited environments as a method for improving on the design and evaluating the interactive capabilities of social robots. First, we discuss issues that have surfaced in attempts to evaluate social interactions between humans and robots. Next, we review two observational studies involving robots interacting socially with humans and discuss how the results can be applied to improving robot design. The first is an analysis of a mobile conference-attending robot that performed a search task by augmenting its perception through social interaction with human attendees. The second is an analysis of a stationary robotic receptionist that provides information to visitors and enhances interaction through story-telling. Through these examples, we show how observational studies can be applied to human-robot social interactions in varying contexts and with differing tasks to quantitatively and qualitatively evaluate (and discover unanticipated aspects of) the social interaction. Finally, we discuss design recommendations suggested by insights gained through these analyses

Nema pronađenih rezultata, molimo da izmjenite uslove pretrage i pokušate ponovo!

Pretplatite se na novosti o BH Akademskom Imeniku

Ova stranica koristi kolačiće da bi vam pružila najbolje iskustvo

Saznaj više