Companion-Technology: Towards User- and Situation-Adaptive Functionality of Technical Systems

Abstract

Abstract-The properties of multimodality, individuality, adaptability, availability, cooperativeness and trustworthiness are at the focus of the investigation of Companion Systems. In this article, we describe the involved key components of such a system and the way they interact with each other. Along with the article comes a video, in which we demonstrate a fully functional prototypical implementation and explain the involved scientific contributions in a simplified manner. The realized technology considers the entire situation of the user and the environment in current and past states. The gained knowledge reflects the context of use and serves as basis for decision-making in the presented adaptive system.

Automated Planning for Ubiquitous Computing

“…Preferences. A user may select and customise recipes [Kotsovinos and Vukovic 2005], indicate personalised choices on various devices [Ding et al 2006;Fraile et al 2013], and provide preferences on services [Liang et al 2010;Song and Lee 2013;Chen et al 2014], daily activities [Mastrogiovanni et al 2010;Vaquero et al 2015;Köckemann et al 2014], domains Yau and Buduru 2014], treatments [S ánchez-Garzón et al 2012], timetables ], companion systems [Honold et al 2014], and products ]. In Ranganathan and Campbell [2004], user preferences are in the form of utility u for each predicate in different contexts, where u ∈ [−10, 10].…”

“…Human Actions. Human actions are primary entities in the domain of assisted living for implementing applications for care giving [Courtemanche et al 2008;Hidalgo et al 2011;S ánchez-Garzón et al 2012], cooking [Kotsovinos and Vukovic 2005;Sando and Hishiyama 2011;Ortiz et al 2013], shopping , training [Bacon et al 2013], and system assembling [Honold et al 2014]. It is practically difficult to extract a generalisation of human actions because of the diversity of adopted models and the lack of representation details (for examples of human actions encoded in PDDL, we refer to Appendix B.2).…”

“…The majority of primary studies provide support for simple failures, and only few additionally for permanent failures and/or timeouts [Vukovic et al 2007;Kaldeli et al 2012;Madkour et al 2013]. The provided support is based on domain representation, such as nondeterministic effects [De Giacomo et al 2012;Vaquero et al 2015], conditional effects [De Giacomo et al 2012;Marquardt et al 2008;Eppe and Bhatt 2015], and other domain-specific knowledge [Ranganathan and Campbell 2004]; and conditions incorporated during plan execution and recovery [Vukovic et al 2007;Kaldeli et al 2012;Madkour et al 2013;Honold et al 2014]. For example, Ranganathan and Campbell [2004] encode additional domain knowledge in specific actions to inform the planner that action execution can be retried, if failed.…”

What is the Value of Embedding Artificial Emotional Prosody in Human–Computer Interactions? Implications for Theory and Design in Psychological Science

“…If it were possible to further improve the quality of artificial speech with emotional prosody, it would have significant consequences for those involved in creating HCI-AI ( Brenton et al, 2005 ). As computerized technology becomes an ever greater fixture at home and at work, our future interactions with it will need to become even more sophisticated ( Wendemuth and Biundo, 2011 ; Honold et al, 2014 ). Some time ago, it was recommended that artificial speech synthesis technology should not only have the ability to control prosody based on meaning, but also the capability to control individual speaking style (another form of prosody), choosing application-oriented speaking styles, and be able to add emotion ( Furui, 1995 ).…”

SPAM – A Process Model for Developing Smart Personal Assistants

“…This new way of support enables the user to get to know a new system in a more efficient way. For example, Honold et al [5] developed a home theater companion system that considers the entire situation of users and their environment in current and past states (e.g., the system recognizes which step the user is currently trying to conduct) in order to help users to understand and set up the home theater system in a more efficient way.…”