@Mail
Simple notions of systems theory and probability (the topics in use will be reviewed in the course).
The course aims to introduce the student to estimation problems for uncertain dynamic systems for monitoring and control purposes. The problem of state prediction will be addressed in particular. The course introduces the subject gradually. In order to maximize usability of the course, the preliminary knowledge required is minimal.
Description of uncertainty in systems and the problem of the study of uncertain dynamical systems. Introduction to the modeling of uncertainty. Estimation of state variables from output measurements (so-called "virtual sensors"). Techniques to minimize the estimation error: optimal estimators. Recursive methods. Application to dynamic systems: observers and the Kalman filter. Use of Kalman filtering in control problems (overview).
- description of uncertainty - uncertain dynamic systems - Bayesian estimate - state prediction in dynamic systems (virtual sensors) -one or more steps state and output prediction - the Kalman filter - use of state prediction in monitoring and control - applications in the automotive, GPS and water management sectors
Sergio Bittanti, "Teoria della predizione e del filtraggio". Pitagora Ed.
- frontal teaching - computer simulations - tutorial sessions
Written exam.