A New Paradigm for Collaborative Smart Objects to Enable the Development of User-centric Services in Pervasive Environments

Abstract

In the near future, people will not be the only consumers of Web content, but also an increasing number of machines will be able to independently search and interpret data received from web servers in order to perform tasks for users. If a machine is an Internet-connected everyday object and its functionalities can be remotely invoked through REST API, then such machine will be part of the Web of Things. In addition to core features, objects will be augmented with sensing and adaptive capabilities, reasoning and decision-making abilities, and, as consequence, intelligence will be transferred to the environment. The new properties of these spaces will change the way in which people interact with objects, as well as services, which users will access to, will become absolutely innovative. From one hand, in fact, you want to reduce or facilitate human-machine interaction. From the other hand, you want to provide context-aware services that are consistent with the context where users are located, personalized services that take into account the preferences and habits of users, and complex services that are based on the aggregation of basic services ( mashups ). The respective evolutions, that machines and people, inhabitants of Smart Spaces, are going through, are closely connected: if machines become smart, the role and attitude of users change, and vice versa to improve and simplify people s lives, it is necessary to design advanced capabilities for machines. In this thesis we analyze in parallel both the aspects in the context of the Web of Things: we want to make everyday objects intelligent and cooperative in order to introduce innovative forms of interaction between users and machines, satisfy people s expectations, and increase users eco-awareness to induce them to change their wrong behaviors that generate energy waste. Underlying the process of collaboration among objects, there is the issue to find a machine-understandable format to describe the effects produced by invoking services exposed by a device, namely REST APIs, and a semantic language that allows to universally interpret exchanged data. Furthermore, to make machines proactive (i.e. a goal-driven attitude), it is necessary to adopt a strategy to determine all the possible plans , in the form of communication flows involving real objects or Web services (i.e. physical mashups ), that satisfy a specific objective. In this thesis we propose to use standard semantic reasoners and Web technologies to overcome these problems. Considering that pervasive environments are populated by people with different needs and abilities, this thesis presents a platform in which users express goals through their voice or via a web app, and Smart Objects cooperate with each other in order to execute tasks for users. The platform monitors three types of contextual data: the user s indoor and outdoor position, the elapsed time, and the state of objects. Moreover, the plan, that is selected to be executed, is personalized on the base of user s preferences and feedback. Exploiting the method to describe REST APIs in machine-understandable format, this thesis proposes new user-object interactions. Using the Augmented Reality and the user-experience of web applications, we demonstrate how to overcome the heterogeneity in the interfaces to control objects. To motivate people in to put more attention to energy consumption, in this thesis we describe a method in which everyday objects provide eco-feedback to users giving them advice about the more convenient working-mode (between on/off and standby) to set in order to save energy. These appliances are able to apply predictive algorithms to determine their next-week usage forecast and, thus, the working-mode to use per hours.