Miroslav Sili

A modular platform for event recognition in smart homes

Ambient Assisted Living technologies try to integrate intelligent assistance-systems in elder peoples homes to maintain a high degree of independence, autonomy and dignity. To speed up the development process and to make the applications more adaptive and flexible to special user needs as well as to ensure compatibility throughout systems a common middleware with standardized interfaces is desirable. The integration of off-the-shelf sensor hardware is an important aspect to assure longterm applicability and interoperability. AIT Austrian Institute of Technology has developed a modular software platform providing services to enable independent living for elder people at home. The platform is based on state-of-the-art software development techniques like OSGi and Spring, which enable modularity and flexibility. The aspect of interoperability at the hardware level is considered by integrating standards from the two areas of medical informatics and home automation. A hardware abstraction module harmonizes data from different sensor networks in terms of a common data format. Based on sensor data, functionalities for the detection of specific events and situations in the AAL domain have been implemented using finite state machines and simple statistical methods.

User interfaces for older adults

Needs and wishes regarding the interaction with ICT solutions change over time and vary between older adults. They depend on the users physical and mental capabilities and preferences. Thus the user interface, which is considered critical to the success or failure of an ICT product or service, should be adaptable. AALuis provides an open middleware layer to guarantee accessible and usable user interfaces for Ambient Assisted Living services. The general idea is to foster a detachment of the user interface from the service and its functionality, respectively. Furthermore an input fusion and output fission regarding I/O modalities based on the users preferences is striven for. At the heart of AALuis lays a dynamically adapted, personalized interaction between an older adult and the service, with various I/O devices. The first results of the project look promising to achieve flexibility in the creation and usage of interfaces. The chosen approach allows further developments expanding the functionalities and improving the generated user interfaces.

Interaction Model-Based User Interfaces: Two Approaches, One Goal - Comparison of Two User Interface Generation Approaches Applying Interaction Models

Nowadays, we use a variety of devices to interact with local and cloud-based systems and services. We are accustomed to utilize an esthetic, functional and well-structured user interfaces. Most of these user interfaces are designed for the broad mass following the “one fits all” strategy. Undoubtedly, such a universal design approach has some benefits, but it has also some essential limitations. Inflexibility and the lack of possibilities for personalization are just two examples for these limitations. General designed user interfaces are not able to take specific user needs and requirements into account. Model-based user interface design tackles this challenge by creating abstract models which are used as starting point for the generation of tailored user interfaces. This work targets the comparison of two different interaction modeling techniques (CTT vs. SCXML) used for the design of multimodal user interfaces. Next to the general concepts, the corresponding execution frameworks and the practical exploration results are presented. The summary of elaborated advantages and disadvantages of each approach clarifies that the CTT approach is more applicable for large and complex user interaction scenarios, whereas the SCXML approach is more suitable for lightweight and structurally simpler user interaction scenarios.

CogniWin – A Virtual Assistance System for Older Adults at Work

This paper presents an innovative virtual assistant system, which aims to address older adults’ needs in a professional environment by proposing promising and innovative virtual assistance mechanisms. The system, named CogniWin, is expected to alleviate eventual age related memory degradation and gradual decrease of other cognitive capabilities (i.e. speed of processing new information, concentration level) and at the same time assist older adults to increase their learning abilities through personalized learning assistance and well-being guidance. In this paper we describe the overall system concept, the technological approach, the methodology used in the elicitation of user needs, and describe the first pre-trials’ evaluation.

Talking Faces in Lab and Field Trials

In recent years, there has been an increasing interest in Ambient Assisted Living technology to support older adults. Research and industry are working jointly on reliable and suitable solutions to help older adults to remain healthy and safe while living independently. Appropriate interaction methods play an important role for the acceptance of such supporting systems. Today, solutions mainly rely on common and well-evaluated interaction techniques such as TV remotes or touch screens to enhance the usability. Projects presented in this work are based on the same interaction techniques, but additionally enrich the interaction experience with a real-time, empathic virtual assistance avatar. In this paper, we present evaluation settings and user involvement results acquired from three different Ambient Assisted Living projects focusing on avatar-based user interaction. Our results show that avatar-based interaction in the Ambient Assisted Living context is very well applicable, especially when combined with speech recognition.

Personalization in the User Interaction Design

User diversity plays an essential role in the design of modern Human Computer Interaction HCI systems. Users differ among their perception and utilization of technology. Thus, designers of modern Information and Communication Technologies ICT systems are asked to consider these aspects and to build interaction systems which are able to support and react to changing user wishes and needs. This work focuses on the identification and elaboration of this additional user-related and contextual information and facilitates to structure the design process of new user-adaptive systems. Based on a comprehensive literature review this work presents methods, tools and systems used to pre and post process user- and context-related information as well as different approaches for the adaption decision process. Additionally, based on a set of selected systems the work illustrates the adaption process.

Automatic IO Device Selection for Ambient Environments

Active and Assisted Living (AAL) aims at providing services for elderly or disabled people in their homes using modern smart home technology and AAL software. The accessibility of user interfaces for such systems is of particular interest. This article proposes a model-based solution for selecting the best device and modality for user interactions of AAL services using the Ambient Assisted Living user interfaces (AALuis) framework. The best device and modality for a given situation depends on context information provided by the AAL system. An exemplary household was modeled as a Bayesian Network, incorporating a selection of devices and their modalities, together with relevant context information regarding the user and the environment. Each entity of the network is assigned with a probability. For devices and modalities these probabilities represent a measure of their suitability for output for the user, given the context. This model was then used to simulate different scenarios, in order to review the results of this selection mechanism.

Interaction Modeling in PRACTICE - CTT Vs. SCXML - A Comparison of Two Practical Solutions Applying Interaction Modeling Techniques for Multimodal User-System Interaction

Nowadays, we are surrounded by various devices to interact with digital media and services. Each device and its inand output modalities can support users’ abilities differently. Thus, it is important to cover a wide range of interaction devices. Modeling user interaction instead of modeling single user interfaces customized to the device is a starting point to do so. This work targets the comparison of two different user interaction modeling techniques used for the design of multimodal user interfaces. Next to the general concepts of the two interaction modeling techniques, the corresponding execution frameworks and the practical exploration results are presented. This paper summarizes advantages and disadvantages of each approach and the comparison clarifies that the CTT approach applied in AALuis is more applicable for large and complex user interaction scenarios. The SCXML approach applied in the ibi project is more suitable for lightweight and structurally simpler user interaction scenarios.

A Practical Solution for the Automatic Generation of User Interfaces --- What Are the Benefits of a Practical Solution for the Automatic Generation of User Interfaces?

Older adults benefit from information and communication technology solutions in the Ambient Assisted Living AAL domain. The offered user interfaces for these ICT solutions often do not take the special needs, preferences and the physical and mental capabilities of older adults into account. The project AALuis focuses on solutions to increase accessibility, adaptability and usability of user interfaces in the AAL domain. The paper describes the functionality of the AALuis layer and the different steps involved stakeholders have to cover to benefit from the user interface generation framework. A detailed comparison between the traditional user interface design and the AALuis approach lists similarities and identifies differences in the user interface generation process.