Due to the proliferation of Internet of Things (IoT) devices and the emergence of the Ambient Intelligence (AmI) paradigm, the need to facilitate the interaction between the user and the services that are integrated in Intelligent Environments has surfaced. As a result, Conversational Agents are increasingly used in this context, in order to achieve a natural, intuitive and seamless interaction between the user and the system. However, in spite of the continuous progress and advancements in the area of Conversational Agents, there are still some considerable limitations in current approaches. The system proposed in this paper addresses some of the main drawbacks by: (a) automatically integrating new services based on their formal specification, (b) incorporating error handling via follow-up questions, and (c) processing multiple user intents through the segmentation of the input. The paper describes the main components of the system, as well as the technologies that they utilize. Additionally, it analyses the pipeline process of the user input, which results in the generation of a response and the invocation of the appropriate intelligent services.
The emergence of the Ambient Intelligence (AmI) paradigm and the proliferation of Internet of Things (IoT) devices and services unveiled new potentials for the domain of domestic living, where the line between “the computer” and the (intelligent) environment becomes altogether invisible. Particularly, the residents of a house can use the living room not only as a traditional social and individual space where many activities take place, but also as a smart ecosystem that (a) enhances leisure activities by providing a rich suite of entertainment applications, (b) implements a home control middleware, (c) acts as an intervention host that is able to display appropriate content when the users need help or support, (d) behaves as an intelligent agent that communicates with the users in a natural manner and assists them throughout their daily activities, (e) presents a notification hub that provides personalized alerts according to contextual information, and (f) becomes an intermediary communication center for the family. This paper (i) describes how the “Intelligent Living Room” realizes these newly emerged roles, (ii) presents the process that was followed in order to design the living room environment, (iii) introduces the hardware and software facilities that were developed in order to improve quality of life, and (iv) reports the findings of various evaluation experiments conducted to assess the overall User Experience (UX)
Intelligent Conversational Agents are already employed in different scenarios, both in commerce and in research. In particular, they can play an important role in defining a new natural interaction paradigm between them and humans. When these Intelligent Agents take a human-like form (embodied Virtual Agents) in the virtual world, we refer to them as Virtual Humans. In this context, they can communicate with humans through storytelling, where the Virtual Human plays the role of a narrator and/or demonstrator, and the user can listen, as well as interact with the story. We propose that the behavior and actions of multiple, concurrently active Virtual Humans, can be the result of communication between them, based on a dynamic script, which resembles in structure a screenplay. This paper presents CasandRA, a framework enabling real-time user interaction with Virtual Humans, whose actions are based on this kind of scripts. CasandRA can be integrated in any Ambient Intelligence setting, and the Virtual Humans provide contextual information, assistance, and narration, accessible through various mobile devices, in Augmented Reality. Finally, they allow users to manipulate smart objects in AmI Environments.
The proliferation of Internet of Things (IoT) devices and services and their integration in intelligent environments creates the need for a simple yet effective way of controlling and communicating with them. Towards such a direction, this work presents ParlAmI, a conversational framework featuring a multimodal chatbot that permits users to create simple “if-then” rules to define the behavior of an intelligent environment. ParlAmI delivers a disembodied conversational agent in the form of a messaging application named MAI, and an embodied conversational agent named nAoMI employing the programmable humanoid robot NAO. This paper describes the requirements and architecture of ParlAmI, the infrastructure of the “Intelligent Home” in which ParlAmI is deployed, the characteristics and functionality of both MAI and nAoMI, and finally presents the findings of a user experience evaluation that was conducted with the participation of sixteen users.
The proliferation of Ambient Intelligence (AmI) devices and services and their integration in smart environments creates the need for a simple yet effective way of controlling and communicating with them. Towards that direction, the application of the Trigger -- Action model has attracted a lot of research with many systems and applications having been developed following that approach. This work introduces ParlAmI, a multimodal conversational interface aiming to give its users the ability to determine the behavior of AmI environments, by creating rules using natural language as well as a GUI. The paper describes ParlAmI, its requirements and functionality, and presents the findings of a user-based evaluation which was conducted.
This paper leverages previous work on the concept of smart object federations and proposes a new dynamic programming language for implementing and simulating smart objects and their interactions. Following their description in the proposed programming language, smart objects can be fully simulated and used for describing ambient intelligence scenarios. In this context, the contributions of the paper are two-fold: (a) the introduction of a new programming language whose runtime semantics allows for a simple and effective description of smart objects, and (b) the description of meaningful interaction strategies, that are implemented in the proposed language, through which executable smart object federations can be used for simulating and implementing ambient intelligence scenarios.
In this paper we describe a novel methodology for performing real-time analysis of localization data streams produced by sensors embedded in ambient intelligence (AmI) environments. The methodology aims to handle different types of real-time events, detect interesting behavior in sequences of such events, and calculate statistical information using a scalable stream-processing engine (SPE) that executes continuous queries expressed in a stream-oriented query language. Key contributions of our approach are the integration of the Borealis SPE into a large-scale interactive museum exhibit system that tracks visitor positions through a number of cameras; the extension and customization of Borealis to support the types of real-time analysis useful in the context of the museum exhibit as well as in other AmI applications; and the integration with a visualization component responsible for rendering events received by the SPE in a variety of human readable forms.
Ambient Intelligence is an emerging research field that aims to make many of the everyday activities of people easier and more efficient. This new paradigm gives rise to opportunities for novel, more efficient interactions with computing systems. At a technical level, the vision of Ambient Intelligence is realized by the seamless confluence of diverse computing platforms. In this context, a software framework (middleware) is essential to enable heterogeneous computing systems to interoperate. In this paper we first consider the basic requirements of a middleware that can effectively support the construction of Ambient Intelligence environments. Subsequently, we present a brief survey of existing, general-purpose middleware systems and evaluate them in terms of their suitability for serving as the low-level communication platform of an Ambient Intelligence middleware. Finally, we argue that an Object-Oriented middleware such as the Common Request Broker Architecture (CORBA) is most suited for basing a middleware for Ambient Intelligence environments.