Robots are example of controllable and autonomous physical devices. In real environments, robots are useful in providing some help on dangerous or daily tasks in a residential environment. There are two main issues concerning the development of a mobile robot to act as a domestic helper. The first one concerns the self-localization problem. The second one concerns decision making. In this paper we describe a domestic mobile robot which is capable of finding itself in a house environment, take decisions autonomously and communicate these decisions in a pervasive fashion. The localization algorithm is based on particle filter whereas a Bayesian Network has been designed to provide decision making.