Statistical Filtering and Control for AI and Robotics


PhD Course

Lesson Schedule

  1. lesson 1 (23/11/17; Sala Riunioni Piano Terra; 13:30--15:30); Introduction to Probabilistic Robotics; Basics of Probability; Bayes filter (last changed: 22/11/2017) [R.M.]
  2. lesson 2 (27/11/17; Sala Riunioni II Piano; 10:00--12:00); Basics of Linear methods for Regression; Kalman filter and applications [R.M.]
  3. lesson 3 (29/11/17; Sala Riunioni II Piano; 10:00--12:00); Nonparametric filters; Particle filter [R.M.]
  4. lesson 4 (30/11/17; Sala Riunioni Piano Terra; 10:00--12:00); Planning and Control: Markov Decision Processes [A.F.]
  5. lesson 5 (01/12/17; Sala Riunioni II Piano; 10:00--12:00); Exploration and information gathering [A.F.]
  6. lesson 6 (04/12/17; Sala Riunioni II Piano; 10:00--12:00); Plan monitoring for robotics; Applications for mobile robots [A.F.]

Course Descripton

Statistical filtering and control are both well studied topics in computer science and engineering. In particular, filtering and control are crucial components for two important and highly related fields: Robotics and Artificial Intelligence. This course will provide an introduction to the most widely used models and approaches to perform filtering and control in robotics and AI. In more detail, the course will first provide the basics of probability and statistical filtering and then present algorithms and approaches to planning, control and exploration for robotics platforms.

Class Activity

In the first part of the course the instructors will offer a block of lectures to provide students with basic background knowledge on relevant issues, methodologies and techniques regarding the above topics. The evaluation will be based on class participation and on a final exam. The exam will take the form of a seminar on a set of selected papers or a project on a selected topic to be discussed with the teachers.

Docenti/Teachers: Alessandro Farinelli and Riccardo Muradore