Abstract
In the IT world, the development of systems based on artificial intelligence or the integration of autonomous components into existing applications is becoming increasingly important for the resolution of problems that until now have remained unanswered. The goal of this approach is trying to automate the learning process of the machines to make them independent and able to perform certain operations freely and without the need for the human component. The work carried out in this thesis focuses on the concept of "sense of agency", that is, the awareness of being able to perform and control one's own activities and to understand what impact they will have on the world around us. The basic idea is to investigate how a machine can become aware of what it does without a specific task assigned a priori and what influence its actions can have on the environment in which it is located. In particular, the scenario analysis is carried out by implementing a smart home prototype and using innovative and still little used tools in this sector such as Bayesian inference networks.
|