Catálogo de publicaciones - libros

Smart Homes have become firmly established as an active research area with potential for huge social and economic benefits. The concept of a Smart Home refers to the enrichment of a living environment with technology in order to offer improved habitual support to its inhabitants and therefore an improved quality of life for them.

In this article we purport how advances from the device and technological side have not necessarily been matched with a similar level of development in processing of the information recorded within the living environment from an algorithmic or ‘intelligent’ perspective. We surmise how traditional areas in Artificial Intelligence can bridge this gap and improve the experience for the user within a Smart Home.

An important aspect in smart homes is the ability to reason about space and time. Certain things have to be done at certain times or at certain places, or they have to be done in relation with other things. For example, it might be necessary to switch on the lights in a room during the night and while a person is present in that room, but not if the room is the bedroom and the person is asleep. In this chapter, we discuss several AI-techniques for dealing with temporal and spatial knowledge in smart homes, mainly focussing on qualitative approaches to spatiotemporal reasoning.

In this chapter, we propose a logic-based approach to describe temporal constraints with multiple time granularities related to events occurring in Smart Homes. We identify a time granularity as a (possibly) infinite sequence of time points properly labeled with propositional symbols marking the starting and the ending points of each granule. In particular, describe time intervals during which Smart Home sensors are in the state “ON”. Both time and sensor granularities and temporal constraints are expressed by means of PPLTL formulae. Temporal constraints for Smart Home are satisfied when the specific relationships between time/sensor granularities, involved in the described constraints, hold.

In this chapter we discuss some of the features of causal reasoning, and present a simple formalism for handling a restricted form of such reasoning applicable to the Smart Homes domain. The formalism handles the manner in which certain conjunctions of (i.e., readings from the sensors) can be used to trigger a variety of , notably those which give alert regarding potential emergency situations.

In this chapter, we review the use (and uses) of plans and planning in Smart Homes. Plans have several applications within Smart Homes, including: sharing task execution with the home’s inhabitants, providing task guidance to inhabitants, and to identifying emergencies. These plans are not necessarily generated automatically, nor are they always represented in a human-readable form. The chapter ends with a discussion of the research issues surrounding the integration of plans and planning into Smart Homes.

Temporal data mining is a relatively new area of research in computer science. It can provide a large variety of different methods and techniques for handling and analyzing temporal data generated by smart-home environments. Temporal data mining in general fits into a two level architecture, where initially a transformation technique reduces data dimensionality in the first level and indexing techniques provide efficient access to the data in the second level. This infrastructure of temporal data mining provides the basis for high-level data mining operations such as clustering, classification, rule discovery and prediction. These operations can form the basis for developing different smart-home applications, capable of addressing a number of situations occurring within this environment. This paper outlines the main temporal data mining techniques available and provides examples of where they can be applied within a smart home environment.

Artificial intelligence (AI) methods have the potential for broad impact in smart homes. Different AI methods offer different contributions for this domain, with different design goals, tasks, and circumstances dictating where each type of method best applies. In this chapter, we describe motivations and opportunities for applying (CBR) to a human-centered approach to the capture, sharing, and revision of knowledge for smart homes. Starting from the CBR cognitive model of reasoning and learning, we illustrate how CBR could provide useful capabilities for problem detection and response, provide a basis for personalization and learning, and provide a paradigm for home-human communication to cooperatively guide performance improvement. After sketching how these capabilities could be served by case-based reasoning, we discuss some design issues for applying CBR within smart homes and case-based reasoning research challenges for realizing the vision.

This chapter aims to illustrate a possible way of using decision trees to make Smart Homes smarter. Decision trees are popular modelling technique, and the corresponding models are both predictive and descriptive. We formulate the modelling problem by defining the generic question “Is the undergoing activity or event in the Smart Home usual?” Then we explain how it is possible to gather appropriate data from the sensors and pre-process these data to form appropriate input for a decision tree algorithm. We further explain the mainstream approaches in decision trees algorithms rather then analysing them in detail, and we give short overview of available software. Finally, we explain some measures for quantitative and qualitative evaluation of the induced decision tree models (e.g. expert opinion, cross-validation, statistical tests etc.).

Many wonderful technological developments in recent years have opened up the possibility of using smart or intelligent homes for a number of important applications. Typical applications range from overall lifestyle improvement to helping people with special needs such as the elderly and the disabled to improve their independence, safety and security at home. Research in the area has looked into ways of making the home environment automatic and automated devices have been designed to help the disabled people. Also, possibilities of automated health monitoring systems and usage of automatic controlled devices to replace caregiver and housekeeper have received significant attention. Most of the models require acquisition of useful information from the environment, identification of the significant features and finally usage of some sort of machine learning techniques for decision making and planning for the next action to be undertaken. This chapter specifically focuses on neural networks applications in building a smart home environment.

The goal of the MavHome (anaging n Intelligent ersa- tile ) project is to create a home that acts as a rational agent. The agent seeks to maximize inhabitant comfort and minimize operation cost. In order to achieve these goals, the agent must be able to predict the mobility patterns and device usages of the inhabitants. Because of the size of the problem, controlling a smart environment can be effectively approached as a multi-agent task. Individual agents can address a portion of the problem but must coordinate their actions to accomplish the overall goals of the system. In this chapter, we discuss the application of multi-agent systems to the challenge of controlling a smart environment and describe its implementation in the MavHome project.