During the last decade, efforts have been made to ensure a continuous development within smart home technologies in order to provide comfort, security and conviviality functionalities in home's managing. Among the researches aiming to improve the living conditions and the well-being of residents, the home support for people with loss of autonomy (PLA) living alone is of great importance. Indeed, because of their condition, their independence becomes restricted and their freedom jeopardized, since they are confronted to constantly rely on others for their needs. The idea of conceiving an integrative system came out as a mean to ensure life quality and independence to PLA. Our research project consists of designing and developing a new model of home assistance for PLA, based on smart home technologies such as wireless sensor networks, Data mining, and artificial intelligence. This paper presents an overview of smart home issues in order to introduce our solution of intelligent model support of PLA.

We present in this paper a method for distance computing in an encoded environment. The proposed method allows to find the closest point between a general polygonal shape obstacle (a robot) and some complex general shapes objects (the obstacles). Due to the discretized aspect of the model, the algorithm complexity does not depend on the obstacles number and shape. In order to have a preprocessed model, the multivalue number coding has been used. Such a distance computing can be used in mobile robotics field as a basic tool in local path planning algorithms or in more sophisticated general paths planners. All the proposed algorithms follow a general schema which makes their implementation easier in an oriented object language, using a series of class.

Proposes to exploit an uncertain man-made representation of an indoor environment. The problem of uncertainties valuation is solved by a human behavior modeling which consists in using proximity relations between the geometrical primitives of the objects. The multivalue numbers formalism provides an efficient structure in order to model the a priori representation. The dynamic aspect of such a model makes real-time alterations of the map components possible. A Kalman filter used in order to match ultrasonic data with the preprocessed map while navigating. This process allows the map to be incrementally improved while preserving the original shape of the obstacles.