You Drive Me Crazy! Interactive QoE Assessment for Telepresence Robot Control
This addresses the neglected aspect of remote operation QoE for telepresence robot users, but it is incremental as it builds on existing QoE assessment methods.
The study tackled the problem of assessing remote navigation Quality of Experience (QoE) for telepresence robots by conducting a subjective study where participants performed tasks under impaired network conditions, finding that users differentiate between visual and control aspects and sensitivity varies by task.
Telepresence robots (TPRs) are versatile, remotely controlled vehicles that enable physical presence and human-to-human interaction over a distance. Thanks to improving hardware and dropping price points, TPRs enjoy the growing interest in various industries and application domains. Still, a satisfying experience remains key for their acceptance and successful adoption, not only in terms of enabling remote communication with others, but also in terms of managing robot mobility by means of remote navigation. This paper focuses on the latter aspect of remote operation which has been hitherto neglected. We present the results of an extensive subjective study designed to systematically assess remote navigation Quality of Experience (QoE) in the context of using a TPR live over the Internet. Participants were 'beamed' into a remote office space and asked to perform characteristic TPR remote operation tasks (driving, turning, parking). Visual and control dimensions of their experience were systematically impaired by altering network characteristics (bandwidth, delay and packet loss rate) in a controlled fashion. Our results show that users can differentiate well between visual and navigation/control aspects of their experience. Furthermore, QoE impairment sensitivity varies with the actual task at hand.