2. Professors
  3. CHIACCHIO Pasquale
  4. Teaching

ADVANCED CONTROL

Pasquale CHIACCHIO ADVANCED CONTROL

0622700101
DIPARTIMENTO DI INGEGNERIA DELL'INFORMAZIONE ED ELETTRICA E MATEMATICA APPLICATA
EQF7
COMPUTER ENGINEERING
2022/2023

OBBLIGATORIO
YEAR OF COURSE 2
YEAR OF DIDACTIC SYSTEM 2017
AUTUMN SEMESTER
CFUHOURSACTIVITY
324LESSONS
324EXERCISES
324LAB
GIOVANNI RUSSO T Curriculum
PASQUALE CHIACCHIO Curriculum
Objectives
THIS COURSE WILL PROVIDE THE STUDENT WITH A COMPREHENSIVE UNDERSTANDING OF ADVANCED CONTROL METHODS FOR DYNAMIC SYSTEMS AND ROBOTS. THE FOCUS OF THE COURSE WILL BE MAINLY ON THE FUNDAMENTAL THEORY, WITH THE USE OF SIMULATION ENVIRONMENTS TO BETTER UNDERSTAND THE POTENTIAL OF THE PRESENTED METHODS. HANDS-ON EXPERIENCE ON THE ROBOTS AVAILABLE AT THE ROBOTICS LAB WILL BE PROVIDED.

KNOWLEDGE AND UNDERSTANDING
STRUCTURAL PROPERTIES OF DYNAMIC SYSTEMS; STATE-FEEDBACK CONTROL; OPTIMAL CONTROL; ROBUST CONTROL; NONLINEAR CONTROL; ADVANCED ROBOT CONTROL

APPLYING KNOWLEDGE AND UNDERSTANDING
DESIGN OF ADVANCED CONTROL ALGORITHMS, USING SOFTWARE TOOLS; DESIGN OF ADVANCE CONTROL ALGORITHMS FOR ROBOTS; USE OF SIMULATION AND DESIGN SOFTWARE TOOLS FOR ROBOTICS APPLICATIONS
Prerequisites
FOR THE SUCCESSFUL ACHIEVEMENT OF THE COURSE GOALS, KNOWLEDGE ON CLOSED-LOOP CONTROL SYSTEMS IS REQUIRED. THIS KNOWLEDGE CAN BE ACQUIRED IN THE COURSES: AUTOMAZIONE
Contents
UNIT 1 - STATE SPACE CONTROL
(LECTURE/PRACTICE/LABORATORY HOURS 6/4/2)
-1 (2 hours lecture/1 hour practice). INTRODUCTION TO THE MODULE AND MOTIVATIONS: BASIC CONCEPTS OF NONLINEAR CONTROL/INTERACTIVE EXAMPLES ON THE ANALYSIS OF NONLINEAR SYSTEMS
-2 (2 hours practice). FIRST EXAMPLES OF CONTROL ALGORITHMS FOR NONLINEAR SYSTEMS AND PHASE PLANE ANALYSIS
-3 (2 hours lecture/1 hour practice). REACHABILITY AND INTRODUCTION TO POLE PLACEMENT/EXERCISES ON THE PHASE PLANE AND CONTROL VIA POLE PLACING
-4 (2 hours lecture). POLE PLACEMENT: FULFILLMENT OF TRANSIENT AND STEADY-STATE REQUIREMENTS FOR SECOND ORDER SYSTEMS AND SYSTEMS IN REACHABILITY CANONICAL FORM
-5 (2 hours lab). SIMULATION OF NONLINEAR SYSTEMS AND PHASE PLANE.
KNOWLEDGE AND UNDERSTANDING. Topological properties of nonlinear systems, phase-plane analysis, reachability and pole placement
APPLYING KNOWLEDGE AND UNDERSTANDING. Via off-the-shelf tools: analysis of the properties of nonlinear systems for control design purposes, design of controllers via pole placement, simulation of nonlinear systems and phase plane.





UNIT 2 - OBSERVERS, PRINCIPLES OF ADVANCED STABILITY NOTIONS AND OPTIMAL CONTROL
(LECTURE/PRACTICE/LABORATORY HOURS 10/2/2)
-1 (3 hours lecture). POLE PLACEMENT ENGINEERING: LQR CONTROL AND CONTINUOUS/DISCRETE TIME EXAMPLES, INTEGRAL ACTIONS, EXERCISES
-2 (2 hours lecture). OBSERVABILITY AND EXAMPLES
-3 (1 hour lecture/2 hours practice). OBSERVER SYNTHESIS AND CONTROL WITH OUTPUT FEEDBACK/EXERCISES ON THE SYNTHESIS OF OBSERVERS AND CONTROLLERS WITH OBSERVED STATE
-4 (2 hours lecture/1 practice). KALMAN FILTER AND STABILITY OF NONLINEAR SYSTEMS/LYAPUNOV FUNCTIONS VIA EXAMPLES
-5 (2 hours lecture). DIRECT LYAPUNOV METHOD FOR GLOBAL ASYMPTOTIC STABILITY OF NONLINEAR SYSTEMS, LASALLE’S THEOREM, EXERCISES
-6 (2 hours lab). POLE PLACEMENT, LQR AND STATE OBSERVERS
KNOWLEDGE AND UNDERSTANDING. Design of closed-loop systems with state feedback, Lyapunov stability, observability, observer design, basic concepts of optimal control and separation principle
APPLYING KNOWLEDGE AND UNDERSTANDING. Via off-the-shelf tools: control synthesis via Lyapunov functions, observers implementation, design of optimal control algorithms and controllers with output feedback, application of the techniques to use-cases of interest for engineering applications.







UNIT 3 - GEOMETRIC CONTROL: FEEDBACK LINEARIZATION
(LECTURE/PRACTICE/LABORATORY HOURS 3/2/2)
-1 (2 hours lecture/1 hour practice). FEEDBACK LINEARIZATION: INTUITIVE CONCEPTS, LIE DERIVATIVE AND EXAMPLES/DESIGN OF FEEDBACK LINEARIZATION CONTROLLERS FOR NONLINEAR SYSTEMS, CASE STUDIES
-2 (1 hours lecture/1 hour practice). I/S FEEDBACK LINEARIZATION AND HIDDEN DYNAMICS, EXAMPLES ON THE DESIGN OF CONTROLLERS VIA I/O AND I/S FBL
-3 (2 hours lab). IMPLEMENTATION OF FEEDBACK LINEARIZATION CONTROL ALGORITHM
KNOWLEDGE AND UNDERSTANDING. Basic concepts of geometric control, input-state (I/S) and input-output (I/O) feedback linearization (FBL), FBL control synthesis
APPLYING KNOWLEDGE AND UNDERSTANDING. Via off-the-shelf tools: guidelines for the design of FBL controllers, implementation of FBL controllers on a case study






UNIT 4 - VARIABLE STRUCTURE CONTROLLER
(LECTURE/PRACTICE/LABORATORY HOURS 2/3/2)
-1 (1 hours lecture/2 hour practice). GAIN SCHEDULING CONTROL - BASIC CONCEPTS/SYNTHESIS OF A GAIN SCHEDULING CONTROLLER ON AN INDUSTRIAL PLANT
-2 (1 hours lecture/1 hour practice). ALGORITHM FOR THE DESIGN OF GAIN SCHEDULER/PRACTICE ON THE DESIGN PROCEDURE
-3 (2 hours lab). IMPLEMENTATION OF GAIN SCHEDULING CONTROL ALGORITHM
KNOWLEDGE AND UNDERSTANDING. Basic concepts of variable structure control, gain scheduling control synthesis
APPLYING KNOWLEDGE AND UNDERSTANDING. Via off-the-shelf tools: guidelines for the design of gain scheduling controllers, controller evaluation via simulations, implementation of gain scheduling controllers on a case study




UNIT 5 - DESIGN OF AN ADVANCED CONTROL
(LECTURE/PRACTICE/LABORATORY HOURS 3/1/4)
-1 (2 hours lab). IDENTIFICATION OF THE USE CASE AND ITS CHARACTERISTICS, CHOICE OF THE CONTROL TECHNIQUE
-2 (1 hour lab).ANALYSIS OF THE CHOSEN CONTROL TECHNIQUE AND DESIGN CONSIDERATIONS
-3 (1 hour lab). RESULTS DISCUSSION WITH TRADE-OFFS ANALYSIS
-4 (3 hours lecture/1 hour practice). PRESENTATION ON SOME OF THE FRONTIER CONTROL APPLICATIONS.
KNOWLEDGE AND UNDERSTANDING. Systematic analysis of the control requirements, critical assessment of design choices, trade-offs analysis.
APPLYING KNOWLEDGE AND UNDERSTANDING. Via off-the-shelf tools: presentations of the technical solution, discussions of requirements, design choices and results





Didactic Unit 6 – ADVANCED MOTION CONTROL OF ROBOTS
(LECTURE/PRACTICE/LABORATORY HOURS 4/2/6)
- 1 (2 hours Lecture): Feedforward compensation in independent joint control. Need for centralized control to improve performance. Gravity compensation PD control. Inverse dynamics control.
- 2 (2 hours Lecture): Task space control: motivations and concept schemes. Gravity compensation PD control. Inverse dynamics control.
- 3 (2 hours Practice): Design of robot controllers.
- 4 (3 hours Laboratory): References to ros-control. Dynamic solvers. Design of a custom ROS controller with ros-control.
- 5 (3 hours Laboratory): Implementation of a ROS motion controller with gravity compensation PD control and/or inverse dynamics control. Validation in the Gazebo simulated environment.
KNOWLEDGE AND UNDERSTANDING: Joint space motion control techniques. Task space motion control techniques. Dynamic solvers.
APPLYING KNOWLEDGE AND UNDERSTANDING: Selection of control techniques. Design of advanced motion controllers. Development of advanced motion controllers with ros-control. Computer-aided controller validation in simulated 3D environments.






Didactic Unit 7 – ROBOT-ENVIRONMENT INTERACTION CONTROL
(LECTURE/PRACTICE/LABORATORY HOURS 6/0/6)
- 1 (2 hours Lecture): Interaction control: motivations, environment modeling, compliance control.
- 2 (2 hours Lecture): Impedance control, impedance control with force measurements, impedance control with force set-point, implicit impedance (or admittance) control.
- 3 (2 hours Lecture): Direct force control with or without position feedback. Hybrid force/position control. Natural and artificial constraints.
- 4 (2 hours Laboratory): Task space control with ros-control. Gazebo simulation of force/torque sensors.
- 5 (2 hours Laboratory): Assisted interaction controller design with ros-control.
- 6 (2 hours Laboratory): Demonstrations of motion and interaction control on real robots.
KNOWLEDGE AND UNDERSTANDING: Interaction with the environment and methods to control the exchanged forces.
APPLYING KNOWLEDGE AND UNDERSTANDING: Selection of interaction control techniques. Design of interaction controllers. Development of interaction controllers with ros-control. Computer-aided validation of simulated and real robots.




TOTAL LECTURE/PRACTICE/LABORATORY HOURS 34/14/24
Teaching Methods
THE MODULE IS ORGANIZED IN LECTURES (34 HOURS), PRACTICES (14 HOURS) AND LABS (24 HOURS).

THE LECTURES WILL ALLOW STUDENTS TO ACQUIRE KNOWLEDGE RELATED TO THE ANALYSIS OF DYNAMICAL SYSTEMS AND THE SUBSEQUENT DESIGN OF ADVANCED CONTROLLERS, ALSO FOR ROBOTS.

PRACTICES AND LABS WILL ALLOW STUDENTS TO ACQUIRE THE ABILITY TO APPLY THE ABOVE KNOWLEDGE.


THE MODULE FORESEES THE DEVELOPMENT OF A PROJECT GROUP WORK AIMED AT GATHERING ALL THE ABOVE SKILLS. THE PROJECT WORK GROUP WILL BE DISCUSSED WITHIN CLASS HOURS (4 HOURS) AND INDIVIDUALLY DURING THE PROFESSOR OFFICE HOURS.



LECTURES ATTENDANCE IS MANDATORY AND THE EXAM CAN BE HELD BY A STUDENT IF 70% OF LECTURES AND PRACTICES HOURS ARE ATTENDED.

Verification of learning
THE EXAM CONSISTS OF THE DISCUSSION OF THE PROJECT AND OF AN INTERVIEW. THE INTERVIEW IS AIMED AT VERIFYING THE METHODOLOGICAL KNOWLEDGE, ALSO WITH RESPECT TO THE PROJECT TOPICS.

THE MAXIMUM GRADE IS 30 AND LAUDE WILL BE GIVEN TO OUTSTANDING STUDENTS. THE PROJECT WEIGHT IS 60% ON THE FINAL GRADE AND THE INTERVIEW WEIGHTS 40%.

THE MINIMUM GRADE TO PASS THE EXAM IS (18/30) AND IS GRANTED WHEN THE STUDENT, WHILE SHOWING EFFORT, ALSO SHOWS: UNDERSTANDING GAPS IN THE METHODS AND THEIR APPLICATION WITH A FAIR DISCUSSION ABILITY.


THE MAXIMUM GRADE (30/30) IS GRANTED WHEN THE STUDENT DEMONSTRATES A DEEP KNOWLEDGE OF THE TOPICS, IS ABLE TO SOLVE ALL THE DESIGN PROBLEMS WITH THE ABILITY TO IDENTIFY THAT ARE MOST APPROPRIATE.

THE LAUDE IS GRANTED WHEN THE STUDENT DEMONSTRATES AN EXCEPTIONAL FAMILIARITY WITH THE METHODS AND THEIR APPLICATIONS. ALSO, THE STUDENT MUST DEMONSTRATE AN EXCEPTIONAL ABILITY TO WORK INDEPENDENTLY, ALSO IN CONTEXTS NOT EXPLICITLY ADDRESSED WITHIN THE MODULE/PROJECT, AND TO PROPERLY COMMUNICATE THE FINDINGS.
Texts
[T1] K.J. ASTROM, R.M. MURRAY, FEEDBACK SYSTEMS: AN INTRODUCTION TO SCIENTISTS AND ENGINEERS, 2020 DISPONIBILE SUL WEB

[T2] SLOTINE, LI, APPLIED NONLINEAR CONTROL, PEARSON EDUCATION, 1991

[T3] F. BULLO, LECTURES ON NETWORK SYSTEMS, 2022, DISPONIBILE SUL WEB

[T4] S. STROGATZ, NONLINEAR DYNAMICS AND CHAOS, 2ND EDITION, CRC PRESS

[T5] B. SICILIANO, L. SCIAVICCO, L. VILLANI, G. ORIOLO, ROBOTICS - MODELLING, PLANNING AND CONTROL, SPRINGER, LONDON, 2009, ISBN 978-1-84628-642-1, ENGLISH LANGUAGE. (IN ITALIAN: B. SICILIANO, L. SCIAVICCO, L. VILLANI, G. ORIOLO, “ROBOTICA - MODELLISTICA, PIANIFICAZIONE E CONTROLLO”, TERZA EDIZIONE, MCGRAW-HILL, 2008)

[T6] ADDITIONAL MATERIAL WILL BE MADE AVAILABLE ON THE ELEARNING PLATFORM FOR THE MODULE (HTTP://ELEARNING.UNISA.IT). THE MATERIAL WILL BE MADE AVAILABLE TO THE STUDENTS REGISTERED FOR THE MODULE
More Information
THE COURSE IS HELD IN ITALIAN.
  BETA VERSION Data source ESSE3 [Ultima Sincronizzazione: 2024-08-21]