The aim of the course is to make available to the students the knowledge of the following topics: - Methods and algorithms for building mathematical models of real-world dynamical systems from sets of noise-corrupted experimental data focusing in particular on system identification techniques devoted to the quantification of the model uncertainty (set-membership identification). - Methods and algorithms for the data-based feedback control system design. - Software and Hardware tools and methods for rapid prototyping of control systems. Students are expected to develop the following skills: - Ability to derive the mathematical model of a dynamical system from sets of noise-corrupted data collected by performing suitable experiments on a real plant, though the application of modeling and identification tools/algorithms. - Ability to design a feedback control system able to guarantee fulfillment of the prescribed performance requirements on the real plant, through the application of data-based control methods and algorithms. - Ability to practically implement and test on a real plant the designed robust control system by exploiting suitable software/hardware tools for rapid prototyping.

Students are assumed to own adequate knowledge of the following preliminary subjects: - Basic facts on mathematical analysis, linear algebra, differential/difference equations, Laplace/Fourier/Zeta transform - Basic facts on systems and control theory The course will also make explicit reference to the content of the lessons of the course “Modern Design of Control Systems” that will be taught to the students in parallel to the course of Laboratory of robust identification and control.

The course will focus on the following main subjects: - Fundamental principles of set-membership identification theory [about 20 hours] - Fundamental principles of direct data-based controller design [about 20 hours] - Fundamental notions about hardware/software tools for rapid prototyping of control systems (MATLAB real-time toolbox, LAbview, National Instrument PXI) [about 5 hours] - Application of set-membership identification and direct data-based control methods and algorithms to real-world case studies and/or laboratory processes [about 15 hours].

Theoretical and methodological lessons will be delivered, on a weekly scheduled basis, by face-to-face instruction in the classroom. Computer laboratory/experimental activities are scheduled in order to develop the student’s skill through proper training. Each student is supposed to practice individually with the aid of laboratory work stations. The primary purpose of the laboratory exercises is to apply the methodologies presented in class, through the use of MATLAB, Simulink, Control System Toolbox and rapid prototyping software/hardware platforms. Some exam simulations are presented in the last two weeks of the course.

The subject of the course is almost entirely based on the research activity of the teacher and collaborators. Therefore, reading materials is made by: - Selected chapters from the book "System Identification: Theory for the User (2nd Edition)" (author: Lennart Ljung; publisher: Prentice Hall, year: 1999)" - Lecture slides and notes provided by the teacher - Scientific publications by the teacher and co-workers - Laboratory practice handouts

Lecture slides; Exercises; Exercise with solutions ; Lab exercises; Lab exercises with solutions; Video lectures (previous years); Self-assessment tools;

Exam: Computer-based written test in class using POLITO platform;

Written examination in computer laboratory. The assessment will be performed through a written exam that will take place directly in the LADISPE lab. The aim is to check the ability of the students to perform system modeling, identification and feedback control design based on a set of input-output data collected on the plant. A detailed written report describing the proposed solution is required. More precisely, the student is required to solve two problems. Problem 1 is about set-membership identification. Problem 2 is about direct data-driven design. The two problem together allows the teacher to check all the expected learning outcomes reported in the list above. In Problem 1 the students are required to: (i) analyze the provided input-output data and information (4 points) (ii) formulate the set.membership identfication problem (4 points) (iii) solve the formulated problem by means of suitable sofware tools for mathematical optimization (4 points) (iv) answer additional questions designed by the teacher to check the ability of the student to apply to the tools provided in the course to solve problem characterized by a certain degree of novelty (4 points) In Problem 2 the students are required to: (i) analyze the provided input-output data and information (4 points) (ii) formulate the direct data-driven controller design problem (4 points) (ii) design a data-based controller able to satify a set of performance specifications (4points) (iii) answer additional questions designed by the teacher to check the ability of the student to apply to the tools provided in the course to solve problem characterized by a certain degree of novelty (4 points) The score 30/30 cum laude will be assigned only to students collecting the total amout of 36 points. The exam will last 2 hours. During the exam, the students are only allowed to use MATLAB software and related user manuals.

In addition to the message sent by the online system, students with disabilities or Specific Learning Disorders (SLD) are invited to directly inform the professor in charge of the course about the special arrangements for the exam that have been agreed with the Special Needs Unit. The professor has to be informed at least one week before the beginning of the examination session in order to provide students with the most suitable arrangements for each specific type of exam.