ROBOTICS ENGINEERING

The goal of the module is to give the students the fundamentals of POSIX programming, concurrent programming, and inter-process communication (i.e., interrupts, signals, pipes, threads, semaphores, shared memory, sockets, publish/subscribe methods). The objective involves both theoretical knowledge and practical work (coding for multiprocess / distributed systems). State-of-the-art programming languages are used in coding, in particular C and rust.

Robots are programmed machines based on multi-task, distributed and concurrent programming. The ARP course expands the so called system programming level. The student will learn how to use the operating system's services to design such concurrent sitributed applications. In addition, robotic programs must be reliable and portable through differenet kernels, different versions and different architectures/electronics. In general, programming in Robotics encompasses different topics: 1) writing and debugging a soft/hard real time programme, 2) the operating system structure, and its relevant features, 3) concurrent / distributed programming, 4) designing complex, distributed, modular software applications. ARP covers mostly the mid topics of the list above. ARP, RTOS and SOFAR cooperate in addressing all topics. State-of-the-art languages are used, in particular C, C++ and Rust, together with ISO/IEC 9945 POSIX framework

The final exam is only one othe component of the assessment process. The exam mark will contribute approximatively with 22% - 33% of the full assessment.
The exam consists in an on line test on the AulaWeb portal, with questions related to the learning outcomes. The test is done in presence in the classroom.

The course implements a continuous assessment process. All the assignments proposed to students every week may contribute to it. Intermediate and final, more complex, assignments will be also proposed. Assignments are evaluated if delivered in a regular basis, respecting specific deadlines.
The final mark will take into account both the exam and the continuous evaluation of assessments, giving specific values to the different components.
All the assignment deadlines will be closed before the last exam session of the semester (approximatively end of February for semester 1 and mid September for semester 2).

The whole study material (reference textbooks, slides, examples) is available in the AulaWeb portal and / or in the Teams main page of the course.

The goal of the course is to enable students to understand the Ambient Intelligence computing paradigm, which envisions a world where people (and possibly robots) are surrounded by intelligent sensors/actuators and interfaces embedded in the everyday objects around them.

Ambient Intelligence assumes the presence of a number of devices (sensors and/or actuators) that are embedded in the environment and able to communicate with each other, and can support humans to perform their daily activities. The class explores how Ambient Intelligence applications can be designed, by presenting methodological solutions and technological tools.

The exam is written and requires the student to solve problems related to the design of an Ambient Intelligence application using the theoretical and practical tools seen during the year.
Assignments will be given during the year, the evaluation of which will contribute to the final grade.

The exam requires the student to address the design of an Ambient Intelligence application with given characteristics, using the theoretical foundations and programming tools learned during the lessons and exercises.
The final grade results from the composition of the exam grade (70%) and the grade obtained from the exercises carried out during the year (30%).

Relevant material (technical reports, slides, papers) will be distributed by the teachers during the lessons

The teaching unit consists of two modules, which together provide a comprehensive introduction to the principles and applications of AI techniques in autonomous robotic systems. Overall, the course aims to equip students with the theoretical understanding and practical tools needed to design and implement intelligent, autonomous robotic systems capable of operating in complex environments.

The course introduces to the main themes of deductive-based Artificial Intelligence with a focus on practical applications on Knowledge Representation and Planning.

The class consists of two modules, and the final exam mark is obtained from the marks of intermediate tests and projects from both modules. Please refer to each module for a detailed description of its respective intermediate tests and projects.

The class is divided into two modules, with specific assessment methods for each module’s learning outcomes detailed within the corresponding module descriptions.

The purpose of the module is to introduce the students to the foundations of logic and logic-based intelligent agents. The logics to be covered include propositional logic, first order logic and decidable sublasses of first order logic useful in robotic applications (modal and temporal logics, satisfiability modulo theory). For each logic, the formal syntax and semantics will be presented, as well as the related decision problems and reasoning algorithms.

The course introduces to the main themes of deductive-based Artificial Intelligence.

Test with open and closed questions and individual project on ontology design.
Students with certification of Specific Learning Disabilities (SLD), disabilities, or other special educational needs must contact the instructor at the beginning of the course to agree on teaching and examination methods that, while respecting the course objectives, take into account individual learning styles and provide appropriate compensatory tools. It is reminded that the request for compensatory/dispensatory measures for exams must be sent to the course instructor, the School representative, and the “Settore servizi per l'inclusione degli studenti con disabilità e con DSA” office ([email protected]) at least 10 working days before the test, as per the guidelines available at the link: https://unige.it/disabilita-dsa

Ability to formalize domain of practical interest and solve them using the techniques shown in the course.

Stuart Russell, Peter Norvig - Artificial Intelligence, a Modern Approach (third edition) - Prentice Hall

The purpose of the module is to introduce concepts related to knowledge representation and reasoning, including ontologies, description logics, OWL, subsumption and instance checking, planning for hybrid domains with a particular focus on discrete and continuous domains, and AI-based robot motion algorithms such as RRTs, probabilistic roadmaps, and belief-space planning, as a logical continuation of Artificial Intelligence for Robotics 1.

Artificial Intelligence for Robotics II (AI4RO2) describes advanced concepts, methods and techniques at the intersection between Artificial Intelligence (AI) and Robotics. The aim of the subject is to provide future scholars and professionals with conceptual tools and practical knowledge about how to integrate cutting-edge AI techniques in robot software architectures, and to guide future engineers about how to do so. The subject presents real-world, advanced scenarios, e.g., autonomous vehicles, collaborative robots, assistive robots.

The AI4RO2 final mark is based on assignments. Assignments work as follows:
./ at the beginning of the semester, a number of assignments are proposed; ./ students, self-organized in groups of up to 5 people, bid on assignments; each group is required to express 3 ordered preferences; ./ we'll try to satisfy the preferences at best and allocate assignments to groups accordingly; ./ at that point, work on assignments can start.
Please note that:
./ EMARO-wannabe and JEMARO students have a strict deadline to complete their assignment; ./ Ph.D. students attending the course can propose a topic on their own, agreed with us, as an assignment, and do not have any specific deadline.
Properly carrying out an assignment means providing:
1/ an appropriate sketch of the solution the group will aim at designing and developing; 2/ a (possibly working, maybe with limitations) solution to the given problem; 3/ properly commented code (in a specific format) and, where appropriate, a tutorial; 4/ a video showing how the developed solution works.

The overall grade will be determined on the basis of:
./ the exhibited group's capability in applying the notions and insights discussed during the classes in the assignments; ./ the "quality" of the provided documentation about the work done in the assignments.

Relevant material will be given by the teachers and the instructors during the lectures.

The purpose of this course is to provide a perspective on robotic technologies applied to (and inspired by) themes of biomedical research and practice. The first part of the course is intended to offer a background on biological signals and their applications in human-machine interfaces. The second part is devoted to in-depth analysis of specific applications. These include basic research in sensory-motor systems, advanced surgical and diagnostic techniques, body and brain machine interfaces, robots for assistance and rehabilitation, prosthetics, biomimetic robotics

This course provides a perspective on robotic technologies applied to—and inspired by—biomedical research and clinical practice. Applications include sensory–motor system research, advanced surgical and diagnostic techniques, brain–computer and body–machine interfaces, prosthetics, and rehabilitation robotics. The course aims to strike an appropriate balance between foundational concepts and practical applications. Students will engage in a range of activities, including scientific presentations and the development of research projects, both individually and in teams.

Assessment Methods Students will be evaluated based on the following components:
a) Assignments These will be related to the robotics topics presented during the course and may include simulations or software development.
b) Active participation During the second part of the course, students will be asked to present research articles on specific robotic applications and/or take part in discussions with invited experts from the field.
c) Final exam The final written or oral exam will assess the overall understanding of the course contents, including the ability to apply and explain key concepts.

Students will be evaluated based on the following components:
a) Assignments Students will develop and comment on a computer simulation and/or software project related to the robotics topics presented in the first part of the course. This component assesses the ability to understand, explain, and solve problems using basic robotics concepts.
b) Active participation and/or knowledge of application topics In the second part of the course, students are asked to present research articles on specific robotic applications and/or engage in technical discussions with the instructor and invited experts. This part evaluates the ability to communicate scientific content, deepen understanding, and apply knowledge in context.
c) Final exam The final written or oral exam will assess the student’s overall understanding of course content, including the ability to critically explain and apply theoretical and practical knowledge.
Components a) and b) represent ongoing assessments (continuous evaluation), while c) is conducted at the end of the course. Students who do not attend classes are still required to complete both a) and b) prior to the final exam; part b)may be integrated into the exam itself.
In addition to disciplinary knowledge, the assessment will also consider the quality of oral or written presentations, the use of technical language, critical reasoning, and the ability to collaborate effectively and communicate clearly—transferable skills that are relevant across domains.

Readings and software materials will be made available through the course website (AulaWeb and/or teams).

The module aims to provide students with a general understanding of robots as multi-purpose, multi-form, and multi-function machines, and of how next-generation robots are expected to exhibit new and unique behaviors compared to current machines, in terms of their objectives, structure, and capabilities. The course further aims to introduce the design of cognitive architectures for robots capable of coping with this diversity in form and function, enabling robots to perceive their environment, interpret it through knowledge representation and reasoning models, and act effectively upon it. All these topics are addressed with a research-oriented approach.

Cognitive Architectures for Robotics (COGAR) is a novel and highly experimental field involving research and development activities aimed at providing scholars and future researchers in robotics with conceptual tools and practical knowledge about how to design and implement complex architectures for robots involving advanced capabilities in perception, reasoning, and action. The subject starts by covers a few basics in software design, and goes on by investigating how to design and integrate perception, reasoning, and motion modules into a coherent workflow. Then, it showcases a number of proposed architectural designs in the literature, and features examples borrowed research-related use cases.

The COGAR final mark is based on assignments. Assignments work as follows:
./ assignments are foreseen to evaluate in itinere how students, individually, understood the concepts and the methodologies after TOPIC 2, 3, and 4; ./ for TOPIC 2, the assignment consists of a multiple-choice exam on the matters discussed during classes; ./ for TOPIC 3, it consists of a programming-related assignment, in which students will have to design and develop a specific part of a robot architecture; ./ for TOPIC 4, it consists of a programming-related assignment, in which students will have to implement relevant traits of a cognitive architecture.
Please note that:
./ EMARO-wannabe and JEMARO students have a strict deadline to complete their assignment; ./ Ph.D. students attending the course can propose a topic on their own, agreed with us, as an assignment, and do not have any specific deadline; ./ the final grade will be a weighted mean of the grades of single assignments.

The overall grade will be determined on the basis of the exhibited individual student's capability in applying the notions and insights discussed during the classes in the assignments.

Relevant material will be given by the teacher and the instructors during the lectures.

The teaching unit aims at providing knowledge on theory and tools on the basics of Computer Vision, for the extraction of semantic and geometric information about a scene from an image or a sequence of images. Topics of interest include: camera models and image formation; camera calibration; connection between 2D images and 3D scene structures; image processing basics as image filtering, local features extraction (edge, corner, blob), including the use of multi-scale image representations; image matching, with reference to classification and retrieval problems; stereo vision and scene depth estimation; motion detection in image sequences, including change detection and optical flow estimation.

The course intoduces image processing and computer vision techniques for 3D static and dynamic scene interpretation and discusses applications to object tracking, depth perception, object recognition and automatic guidance

50% from a continuous assessment through practical laboratory exercises done throughout the semester.
A multiple-choice quiz (~5%) that is a threshold for attending the oral exam;
An oral exam (~45%).
It is not allowed to consult books, notes, or other written material.

The course is organized in theory classes and practical (hands-on) classes.
Practical activities cover about 1/3 of the course. The goal of such activities is presented in class by the instructors, and should be completed by the students as a homework. They can be carried out individually or in groups; some of them are associated with an assignment and constitute a continuous assessment of the student's work.
The final assessment of the theory part is carried out through a final exam.

Recommended texts:
R.C. Gonzalez and R.E. Woods, Digital image processing, Prentice-Hall, 2008.
E. Trucco and A. Verri, Introductory Techniques for 3-D Computer Vision, Prentice Hall, 1998.
Further readings: Material distributed by lecturers through the Aulaweb portal

The objective of the module is that of presenting the basic methodologies for the analysis and control of multivariable systems with particular emphasis on linear time invariant ones. The module will start with a review of the basic concepts relevant to (linear) systems in state space form. Stability and structural properties of state space multivariable dynamic systems will be addressed with particular emphasis on the linear time invariant case. Matrix pseudo inversion methods will be discussed with reference to robot inverse kinematics and control allocation problems. The module will end with the treatment of some specific topics concerning multivariable control methods as applied in robotics scenarios (eg. linear state space observers used for robot navigation or control allocation used for robot control).

The aim of the course is to present the basic concepts and methods for the analysis and design of controllers for multivariable linear time invariant systems (MIMO LTI systems) of interest within robotics applications. The models used in the course are mainly those based on state-space equations. Basic robot control allocation problems will be illustrated as well as motion control ones.

Oral exam: students will be asked describe, explain and provide examples of multivariable systems properties illustrated during the course. Simple exercises will be discussed to highlight multivariable systems properties and to assess the students ability to analyses and solve practical cases. The lab exercises reports and codes developed by the students during the course will also be discussed during the exam. Students with learning disorders ("disturbi specifici di apprendimento", DSA) will be allowed to use specific modalities and supports that will be determined on a case-by-case basis in agreement with the delegate of the Engineering courses in the Committee for the Inclusion of Students with Disabilities. In particular, students who have valid certification of physical or learning disabilities on file with the University and who wish to discuss possible accommodations or other circumstances regarding lectures, coursework and exams, should speak both with the professor of the course and with Professor Federico Scarpa ([email protected] ), the Polytechnic School - Engineering's disability liaison.

The oral exam will test the effective acquisition of basic knowledge of the multivariable systems properties addressed during the course. Open questions will allow the assessment of the ability to apply the knowledge on practical robotic examples using a clear and correct terminology. The student should be able to link and integrate knowledge learned during laboratory activities with that provided during the face-to-face lectures. Overall, the oral examination will aim to assess not only whether the student has achieved an adequate level of knowledge, but whether he or she has acquired the ability to critically analyze multivariable systems problems in robotics that will be posed during the exam.

Papers and notes distributed by the lecturer; H. Kwakernaak, R. Sivan - Linear Optimal Control Systems - Wiley 1972

The goal of the course is to first introduce a modern task-priority based control of robotic systems such as dual arm robots, mobile manipulators, floating underwater vehicle-manipulator systems, which are all characterized by a high number of degrees of freedom. The framework is extended to the case where multiple robots need to work together, for example to manipulate and transport objects cooperatively.

Different robotic agents can be employed for achieving a set of objectives via cooperative activities. From example, multiple sensoride vehicles can be employed for distributed area exploration, monitoring or surveying, while cooperating multi-mobile manipulators can be employed for manipulating, transporting and assembling objects. This course will present an efficient task-priority based control framework and its extension for the use in a cooperative context.

Development of a project assigned during the course. The project is carried out in the MATLAB simulation environment used during the laboratory lessons. The project requires the development of a task priority controller for a robotic system employed in a specific case study. The exam will be based on the discussion of the project and of the content covered during the lessons.

At the end of the course, the student must be able to design a task priority controller for a given robotic system. This skill is evaluated through the discussion of the developed project (70%) and the continuous assessment during the lessons (30%).
The following items will be part of the evaluation:
the correctness of the employed Jacobian relationships
the student's ability in discussing the developed project
which and how many tasks were successfully implemented by the robotic system

The notes of the course will be available on Aulaweb and cover all the contents of the course.
For further readings, students can read the following books and papers:
Siciliano, B., Sciavicco, L., Villani, L., Oriolo, G. (2009) Robotics: Modelling, Planning and Control. ISSN: 1439-2232
Antonelli, G. (2018) Underwater robots. ISSN: 1610-7438
Simetti, E., & Casalino, G. (2017). Manipulation and transportation with cooperative underwater vehicle manipulator systems. IEEE Journal of Oceanic Engineering, 42(4), 782-799.
Simetti, E., & Casalino, G. (2016). A novel practical technique to integrate inequality control objectives and task transitions in priority based control. Journal of Intelligent & Robotic Systems, 84(1-4), 877-902.

By the end of the course, students will be able of selecting appropriate motors and power electronic topology and applying advanced control methods to achieve high-precision motion and energy-efficient performance. Moreover, they will be prepared to tackle real-world challenges in robotics by leveraging integrated sensing, actuation and energy storage solutions.

Electrical drive systems are essential in modern robotics, providing precise motion control, high efficiency, and reliable performance. They combine power electronics, electrical machines, control strategies, sensing technologies, and energy storage systems.
The module introduces DC-DC and DC-AC converters, modulation techniques, and the operating principles of DC motors, AC synchronous motors, and stepper motors. Particular attention is given to Field-Oriented Control (FOC), which enables accurate and efficient torque and speed regulation in AC machines.
Students will also study current, voltage, and position sensors, including encoders and resolvers, which provide the feedback required for safe and precise motion control. Battery technologies, battery management systems, and state-of-charge estimation methods are also covered.
The module provides the knowledge required to design and implement efficient electrical drive systems for industrial, mobile, and autonomous robotic applications.

Development of a project assigned at the end of the course. Discussion of the project and of the content covered during the lessons.

At the end of the course, the student must be able to design in MATLAB/Simulink and simple drive system for a specific robotic application.
The final oral examination will focus on the project report and on the topics covered during lectures.

The purpose of the teaching unit is to introduce the fundamentals of embedded systems, including a brief review of the most relevant architectures and a focus on microcontroller programming for control applications, with particular attention to peripheral configuration and real-time and event-based programming techniques.

An embedded system is a computer-equipped electronic circuit designed to perform a specific, embedded function on a platform that manages control with different design constraints such as cost of production, size, and energy consumption. The course will provide the skills needed to design these systems, demonstrating how the same technologies can be used for the realization of modern electronic devices.

Development of a project assigned at the end of the course. Discussion of the project and of the content covered during the lessons.
Students with certification of Specific Learning Disabilities (SLD), disabilities, or other special educational needs must contact the instructor at the beginning of the course to agree on teaching and examination methods that, while respecting the course objectives, take into account individual learning styles and provide appropriate compensatory tools. It is reminded that the request for compensatory/dispensatory measures for exams must be sent to the course instructor, the School representative, and the “Settore servizi per l'inclusione degli studenti con disabilità e con DSA” office ([email protected]) at least 10 working days before the test, as per the guidelines available at the link: https://unige.it/disabilita-dsa

At the end of the course, the student must be able to design and write the firmware of an embedded system in compliance with the technical specifications. The evaluation will be 15% on the continuous assessment during the lessons and 85% on the final project.
The following items will be part of the evaluation:
whether the firmware respects the given technical specifications
the student's ability to discuss the proposed software solution
the knowledge of the different peripherals
the possible problems that might be encountered in extending the project to implementation on a real system

Slides will be available through aulaweb. In general, notes taken during the module and the slides available on aulaweb will be sufficient to prepare the exam.
The following books can be used for further reading on embedded systems:
Q. Li, C. Yao, Real-Time Concepts for Embedded Systems, CMP Books, 2003. (ISBN:1578201241).
D. E. Simon, An Embedded Software Primer, Addison-Wesley Professional, 1999. (ISBN: 020161569X).
T Noergaard. 2013. A Comprehensive Guide for Engineers and Programmers, Embedded Systems Architecture (2 ed.). Newnes, Newton, MA, USA.
Peter Hintenaus. 2014. Engineering Embedded Systems: Physics, Programs, Circuits. Springer Publishing Company, Incorporated.

The purpose of the teaching unit is for participants to acquire knowledge and understand technical terminology, technical documentation, and oral and written communication skills in professional contexts, to enhance intercultural skills for collaboration in global teams, and to develop the linguistic and technical proficiency needed for effective communication and collaboration in complex international settings within the robotics industry.

This course is designed to develop students’ English proficiency in the field of robotics engineering. Using a communicative approach, it emphasizes practical language skills through interactive discussions and technical speaking activities related to robotics. Students will engage with authentic materials, to expand their vocabulary and enhance their ability to communicate effectively in academic and professional settings. The course aims at strengthening students' skills in all key language areas — writing, speaking, listening, and reading — to ensure well-rounded communication skills.

Written exam (Reading and Vocabulary + Writing).

Final exam.

All materials will be available on Aulaweb.

The course aims to provide students with practical and theoretical skills for developing advanced robotic applications. It covers robot modeling for simulation, the integration of computer vision and planning libraries with ROS and ROS2, the use of the ROS and ROS2 navigation stack, and the implementation of REST APIs for robotic systems.

Experimental work is a fundamental aspect of robotics, where theoretical knowledge from automatic control, computer vision, software architectures, and real-time programming is integrated into real robotic systems and mechatronic platforms. The course aims to provide students with a methodology for bridging the gap between theory and practical implementation through lectures, laboratory activities, and projects. In particular, the course introduces methods and tools for robotic simulation, software architecture design, computer vision, planning, and mapping.

During the semester, students will develop two projects, either individually or in groups, based on the topics covered in the course.
The final examination consists of a discussion about the assignment.
The final grade is determined by the evaluation of the course projects (35%) and the final examination (65%).

The final exam and the projects aim to ascertain the following aspects of the students' preparation:

• Knowledge of methodologies and tools for robotic simulation.
• Ability to apply appropriate methodologies to the practical implementation of theoretical concepts.
• Ability to adopt software architectures suitable for solving robotic problems.

Lecture slides, laboratory material, and additional resources will be made available through AulaWeb. The provided teaching material is sufficient for exam preparation and project development.
Students with valid certifications for Specific Learning Disorders (SLDs), disabilities or other educational needs are invited to contact the teacher and the School's contact person for disability at the beginning of teaching to agree on possible teaching arrangements that, while respecting the teaching objectives, take into account individual learning patterns. Contacts of the teacher and the School's disability contact person can be found at the following link Comitato di Ateneo per l’inclusione delle studentesse e degli studenti con disabilità o con DSA | UniGe | Università di Genova

This course aims to introduce the key concepts and methods of Quantum Information and Computation. The first part will provide the operational elements of quantum mechanics and quantum information: superposition principle, quantum entanglement, the quantum bit (qubit) and quantum logical gates. The second part will introduce the basic quantum algorithms and applications to informatics such as quantum database search algorithm, quantum teleportation and superdense coding.The final part will deal with some possible applications to robotics. It will be shown as the above ideas and concepts can be introduced in software architecture for robots that exploit quantum-inspired perception, reasoning and action techniques.

The course is an introduction to quantum information and computation. The theoretical part is followed by an applicative one aimed to show the more recent developments in the field.

The final examination will consist of a discussion of the subjects presented during the lessons.

At the end of the course, the student must be able to handle the basic concepts and techniques of quantum mechanics and quantum information. To this aim the possibility to develop a project as a home assignment during the course will be considered.
The following items will be part of the evaluation:
knowledge of the basic concepts in quantum information, e.g., measurement process entanglement and so on.
knowledge of the proposed quantum algorithms and protocols

M. A. Nielsen e I. L. Chuang "Quantum Computation and Quantum Information", Cambridge University Press (2011) N. S. Yanofsky e M. A. Mannucci "Quantum Computing for Computer Scientists", Cambridge University Press (2008)
E. G. Rieffel and W. H. Polak "Quantum Computing: A Gentle Introduction (Scientific and Engineering Computation)" The MIT Press (2011)

The teaching unit aims to provide students with a general understanding of oral and written comprehension of the local language, as well as an introduction to the country’s culture.

The course aims to provide foreign students with an Italian language competence at A1 level (contact) according to the levels of the Common European Framework of Reference for Languages (CEFR https://rm.coe.int/common-european-framework-of-reference-for-languages-learning-teaching/16809ea0d4).

Final written test and oral test
The written test is based on the lexical, grammatical and communicative topics covered during the course.
In the oral test candidates must be able to introduce themselves and communicate with the professor through simple exchanges of information.

During the course this manual will be used:
Fumagalli, L., Rota, D., Bonzano Costa, R. (2024) Italiano da zero, Loescher Editore
and reference will be made to:
Morano, L. (2024), Alfa Beta Gramma, Nina Edizioni

The teaching unit is designed to equip students with advanced knowledge and skills in the fields of machine learning and data analysis. Building upon foundational concepts, students delve into cutting-edge techniques and methodologies essential for tackling real-world problems in diverse domains. The teaching unit addresses a comprehensive review of fundamental machine learning algorithms, including supervised and unsupervised learning, and deep learning architectures. Through hands-on exercises and projects, students gain proficiency in implementing these algorithms using popular libraries.

In the information age any system or device generates some form of data for diagnostic purposes or analysis. he course details the techniques for analyzing data in order to extract useful information and knowledge for decision making.

Oral by appointment.

The student will solve a real problem at will by applying the techniques learned during the course.

T. Hastie, R.Tibshirani, J.Friedman "The Elements of Statistical Learning: Data Mining, Inference, and Prediction" 2009. S. Shalev-Shwartz, S. Ben-David "Understanding machine learning: From theory to algorithms" 2014 C. M. Bishop, H. Bishop. Deep learning: Foundations and concepts. Springer Nature, 2023. L. Oneto "Model Selection and Error Estimation in a Nutshell" 2020

The teaching unit introduces the basics of Machine Learning and Artificial Neural Networks, as well as other well-known techniques for solving supervised and unsupervised learning problems, with a specific emphasis on Robotics applications. Such learning systems can be applied to pattern recognition, function approximation, time-series prediction and clustering problems. Some mention will be made to the use of ANNs as static systems for information coding, and dynamical systems for optimization and identification.

The goal of the Machine Learning module is both to provide the basics of machine learning and pattern recognition theory and to expose the student to machine learning methods, workflows, and best practices, with emphasis on applications in Robotics and a focus on artificial neural networks as well as several other techniques.

The exam consists of (1) evaluation of assignments and a summary report, and (2) a written quiz.
Assignments must have been uploaded within the required deadlines during the first semester. The course cannot be attended during the second semester.

Evaluation of homeworks
Homeworks address the following learning outcomes:
apply selected techniques to relevant problems
code simple and medium-complexity machine learning methods using standard programming tools, without being limited to using software libraries
tackle the workflow of a machine learning assignment from data wrangling to result presentation
Quiz grading
Quizzes address the following learning outcomes:
demonstrate knowledge of a range of techniques and problems in machine learning and pattern recognition, including the underlying scientific and technical rationale
use critical thinking to analyse a problem and select the appropriate machine learning method to apply

Course slides and assignments are available on the official study portal.
A selection of suggested readings (journal articles and textbooks) will be provided during lectures.

Students who hold valid certificates relating to Specific Learning Difficulties (SLD), disabilities or other educational needs are invited to contact the lecturer and the school’s disability liaison officer at the start of the course to agree on any teaching arrangements which, whilst respecting the course objectives, take into account individual learning styles.
The contact details for the university’s disability liaison officer are available at the following link: https://unige.it/commissioni/comitatoperlinclusionedeglistudenticondisabilita.

The purpose of the teaching unit is to introduce the design process, including specification, conceptual design, and product design, and to provide students with the basic principles of industrial robot design.

Recall of the basics of mechanical engineering design. Introduction to methods and tools used in the design of robotic systems.The course will focus on the learning of a 3D modelling and computer aided engineering sw tool, CREO of PTC. Examples. Development of a design case during the course.

In two parts:
+ written: modelling of an item using the SW for 3D modelling thaught in the course: this is then discussed and evaluated as first part of the exam
+ oral: presentation of the classwork developed during the semester: discussion and questions. The classwork is common to more students in a group: this is taken into account and the focus is on questions answered by the student

Continuous assessment during the semester: 30%
Assessent of performance at the interview: 70%

+ Springer Handbook of Engineering Systems Design. Anja Maier, Josef Oehmen, Pieter E. Vermaas.
+ Springer Handbook of Mechanical Engineering. Karl-Heinrich Grote, Hamid Hefazi.
+ Shigley's Mechanical Engineering Design. Richard Budynas, Keith Nisbett. McGraw-Hill Higher Education

The module aims to provide students with the basic knowledge of the fundamentals of the theory of mechanisms and machines, including synthesis, analysis, modeling, singularities, kinematics and elements of dynamics, serial and parallel architectures, compliant mechanisms, architectures for robotics, the Lie group of rigid body displacement, and screw theory.

The course will introduce students to modern mathematical methods of modelling rigid-body motion as applied to the study, design, and control of robotic mechanisms. The focus will be on the geometry, kinematics, and statics of articulated multi-body systems, with targeted applications in mechanism analysis and synthesis, as well as robot dynamics, flexibility, and control.

Written exam or alternatively an oral interview based on the number of students per session.

final exam

Lecture notes and slides.
Hunt, K., 1978, Kinematic geometry of mechanisms, Clarendon Press.

The technig unit is composed of two modules, each focusing on key aspects of robot dynamics, control, and mobile robot design. Overall, the course aims to provide students with a solid understanding of the dynamics, control, and planning techniques required to design and implement robotic systems, including mobile robots, in real-world applications.

This course will cover two fundamental topics in robotics: the dynamics and control of holonomic robots, and the modelling and control of mobile robots (non-holonomic vehicles).

The class consists of two modules, and the final exam mark is the average of the intermediate tests from both modules. Please refer to each module for a detailed description of its respective intermediate test.

The class is divided into two modules, with specific assessment methods for each module’s learning outcomes detailed within the corresponding module descriptions.

The module aims to provide students with a general understanding of the kinematic modeling and motorization of mobile robots, illustrated through the detailed study of the differential drive robot, as well as of localization techniques based on the Extended Kalman Filter, including laboratory activities using real data and the presentation of tuning methodologies. The course further aims to introduce observability issues through practical examples, planning methods applicable to mobile robots such as potential field methods and Rapidly-exploring Random Trees, and control techniques with direct application to mobile robots, including static and dynamic feedback control and Lyapunov-based control, illustrated on the case of the differential drive robot.

Modeling of wheeled Robots: Constraint equations, Classification of robots using degrees of mobility and steerability, Posture kinematic model, Configuration kinematic model, Motorisation of wheels.
Localization: Relative localization using odometry, Absolute localization, Localization sensors, Localization using extended Kalman filtering.
Control: Controllability and stabilization, static and dynamic feedback linearization, nonlinear control based on Lyapunov functions.
Practical Work: The students will study various control laws in simulation. They will also implement a Kalman filter-based localization algorithm using data recorded with a real robot.

The exam is based on exercises that remain fairly close to class examples and homework examples. The exam typically is a mix of questions that must be solved by calculation, and questions that must not. The exam may also contain questions that are meant to evaluate the knowledge gained from the labs.

Final exam 100%, but as stated above, the exam may contain questions to evaluate the understanding of the labs.

“Theory of robot control”, Carlos Canudas de Wit, Bruno Siciliano, Georges Bastin, Springer Science & Business Media, 2012 - 392 pages.
PDF documents provided by the teacher.

The purpose of the module is to introduce the fundamentals of kinematic modeling and control techniques of serial manipulators, including geometric modeling, task Jacobian matrices, inverse kinematics, and closed-loop kinematic control.

The course presents the fundamentals of the modeling and control techniques for serial manipulators. Main topics are the geometric and kinematic modeling of serial manipulators, and the kinematic based control in the operational space.

The exam is done as it follows.
During the course, typically 3 assignments will be given. The students will complete them at home in groups of maximum 3/4 people.
Then, a written exam will be conducted. The written exam will require solving an exercise (possibily using MATLAB), and then asnwering a quiz whose questions will regard the exercise and general knowledge too.
A non-mandatory oral exam will allow students to gain or lose up to 20% of the final score.
The final mark will be the sum of the on-going assessment, of the written exam and, eventually, of the oral exam. The student must retake the written exam if they want to take the oral exam again.

At the end of the course, the student must be able to model a robotic arm, derive its kinematic model and be able to control it.
During the on-going assessment, the following items will be part of the evaluation:
The correctness of the numerical answers
The explanations provided
The quality of the figures (captions, units of measure, legends, etc.)
During the written exam, the following items will be part of the evaluation:
The capability of adapting the MATLAB code to new models
The capability of creating geometric models of a robot
The kinematic modeling of a robot
The inverse kinematic control of a robot
General knowledge about the course contents
In the oral exam, the following items will be part of the evaluation:
The understanding of how the equations were derived from
The properties of the mathematical tools
Demonstration of the equations

• B. Siciliano, L. Scxiavicco, L. Villani, L. Oriolo: "Robotics: Modelling, Planning and Control"; Mc Graw-Hill, 2009
• W. Khalil, and E. Dombre, "Modeling, identification and control of robots", Hermes Penton, London, 2002.
  Further readings - J. Angeles, Fundamentals of Robotic Mechanical Systems, Springer-Verlag, New York, 2002.

For a robot, perception and actions are fundamental, defining features of stereotyped or purposive behaviour. Especially when interacting with humans, robots must be capable of employing mental models of the human they are interacting with, perceiving the environment and their actions using common, shared categories, and act in a credible manner. This subject will provide advanced knowledge and theoretical insights about these matters.

For a robot and for natural intelligent system, perception and actions are fundamental capabilities necessary during the interaction with the environment and with other agents in the environment. When interacting with humans, humans are capable of employing mental models of the human counterpart they are interacting with in order to anticipate and predict. Such capabilities in cognitive science are based on how the other`s actions using common, shared categories, and perceiving the environment changes as function of the others.

The exam will be held in presence and the main modality is through written form. The questions will be provide the students either in the form of essay or as questions with multiple choices. The questions will cover equally all the topics of the course with an equal numer of question per topic.

The assessment will based on an accurate analysis of the answers and consequently the intention is to assess how much details of the lectures and knowledge about the topics has been acquired by the students.

The purpose of the teaching unit is to introduce issues concerning real-time applications and real-time operating systems, as well as real-time design and programming and embedded systems.

A real-time system is characterized by having a predictable behaviour concerning execution times. Its main characteristic is not speed, but rather guaranteeing that timing constraints are met: in particular, guaranteeing a maximum time within which computing processes will terminate. The class is aimed at presenting basic principles, methodologies, and the main tools to understand real-time operating system and the design of real-time software.

The exam is written and requires the student to solve problems related to the design of a real-time system using the theoretical and practical tools seen during the year.

The exam requires that the student is able to design, using theoretical bases and practical tools presented during lectures and during exercises, a real-time application with given characteristics.

Slides will be made available on aulaweb. For a deeper investigation of the topics presented in the class, please refer to the following books or websites.
Buttazzo, Giorgio C. Hard Real-time Computing Systems, Kluwer Academic publishers, 1997
Alessandro Rubini and Jonathan Corbet, Linux Device Drivers, Third Edition, O'Reilly and Associates, June, 2001 (available online at http://oreilly.com/openbook/linuxdrive3/book/)
Tom Wagner and Don Towsley, Getting Started With POSIX Threads (available online)
https://www.rtai.org/

This course is intended to provide the student with the necessary skills and tools to carry out and present a research topic. It presents the profession of university staff, researchers in research institutions, and in R&D departments in enterprises and how to apply for them. This course includes also the beginning of the bibliographical study and collect information part for the MSc thesis topic.

The course will test and develop certain basic skills necessary for conducting academic research in science and engineering. The focus is on the writing, reading, and research presentation of papers and reports.

There are no written assignments.
There is no written examination.
The students are invited to discuss about some course topics during the oral examination.

The oral examination assesses how well the student has understood and made the method presented during the course his or her own.

This is marketing - Seth Godin
Shape Up - Basecamp
The Machine That Changed the World - Daniel Roos, Daniel T. Jones, and James P. Womack
Gemba Kaizen - Masaaki Imai
What every new manager needs to know - Gerard H. Gaynor
Project based enterprise - Alfredo Biffi
Empowered - Marty Cagan, Chris Jones
Inspired - Marty Cagan
Zero to One - Peter Thiel
Business Model generation - Alexander Osterwalder, Yves Pigneur

The teaching unit aims to equip students with the foundational skills for software programming in robotics. It covers essential development tools (Docker, GitHub, Linux), basic programming in Python and C++, and introduces the Robot Operating System (ROS and ROS2) for building and managing robotic applications.

The Research Track I Course aims at providing students with advanced concepts needed for developing software for robotics, performing their research and presenting their results. The course will help the students in learning and analyzing the most relevant instruments in these contexts, such as data visualization libraries, software documentation tools, notebooks for robotics, statistical tests.
Furthermore, students will work on a specific line of research, elaborating an analysis of the state of the art on a topic chosen from among those proposed. A part of the course will therefore be dedicated to some aspects related to the research methodology.

During the final exam the student will discuss his/her Research Line (i.e., a "fake" scientific paper) carried out during the course.
The final grade will be composed of the evaluation of the assignment (50%).and the related discussion (50%)

With the assignment and the final exam, students will demonstrate that they are able to:

• use tools for data visualization
• write a correct software documentation
• design a notebook for a robotic simulation
• define research hypotheses and perform statistical analyses
• perform a literature analysis
• present their work
• discuss and motivate the work carried out and the choices made

All slides shown during the lessons and other teaching materials will be available on the Aulaweb platform. Generally speaking, notes taken during lectures and the teaching materials uploaded to AulaWeb will be sufficient.
Students with valid certifications for Specific Learning Disorders (SLDs), disabilities or other educational needs are invited to contact the teacher and the School's contact person for disability at the beginning of teaching to agree on possible teaching arrangements that, while respecting the teaching objectives, take into account individual learning patterns. Contacts of the teacher and the School's disability contact person can be found at the following link Comitato di Ateneo per l’inclusione delle studentesse e degli studenti con disabilità o con DSA | UniGe | Università di Genova

The teaching unit aims to provide students with key skills related to the presentation of results and research methodology in the context of robotics. Specifically, it covers software documentation, data visualization techniques, the use of interactive notebooks, basic statistical analysis, and the fundamentals of research design.

The Research Track II Course aims at providing students with concepts needed for developing software for robotics. Frontal lessons and class exercises will help the students in learning and analyzing the most relevant software instruments for robotic programming (Linux, Git, C++, Python, Docker) and the robotic framework ROS2.

During the course, students will have to complete two assignments, based on the contents of the course.
The final exam consists of an oral discussion regarding the assignments, aimed primarily at assessing the candidate's ability to manage and modify the implemented code in response to changes in project requirements.

The final exam and the two projects aim to ascertain the following aspects of the students' preparation:

• Acquired knowledge about the principles of robotic programming;
• Ability to apply the functionalities provided by the ROS framework.

All slides shown during lectures and other teaching materials will be available on the Aulaweb platform. Generally speaking, notes taken during lectures and teaching materials uploaded to Aulaweb will be sufficient for the course.
Students with valid certifications for Specific Learning Disorders (SLDs), disabilities or other educational needs are invited to contact the teacher and the School's contact person for disability at the beginning of teaching to agree on possible teaching arrangements that, while respecting the teaching objectives, take into account individual learning patterns. Contacts of the teacher and the School's disability contact person can be found at the following link Comitato di Ateneo per l’inclusione delle studentesse e degli studenti con disabilità o con DSA | UniGe | Università di Genova

The module aims to provide students with a general understanding of the dynamic modelling of robot manipulators and the fundamentals of dynamic control of robots, as key elements for the design of robot controllers and for the implementation of robot-controlled operations involving interaction with objects (e.g. for their manipulation), the environment (e.g. force control), and humans (e.g. human–robot collaborative tasks).

Robots are programmable machines capable of executing a wide range of operations according to their mechanical design.
In order to execute proper tasks alone, in cooperation with other robots, or in collaboration with humans, robots' motion and interaction with objects and people must be properly controlled.
In order to control the robot actions software must be designed and implemented ensuring the accomplishment of the programmed operations according to the planned task specifications (e.g. positioning accuracy, speed etc.) and safety conditions (e.g. maximum interaction forces etc.).
Therefore, the robot control software must implement algorithms capable to compute the correct control commands for the robot actuators (joint motors) using the feedback provided by robot sensors.
The course will provide the fundamentals for the design and implementation of these algorithms.

Final grades will be based on assignments.
Written+oral exams are also possible, but not encouraged since the outcome of the course is to give operational capabilities for the design of new robot systems.

Correction of the assignments or of the written exams

Lecture notes
Text books: none suggested (since there are many very good around but using different notations) Suggested reading:
Fundamentals of Robotic Mechanical Systems. J. Angeles
Robotics. Modelling, planning, and control. B. Siciliano et al.

This teaching unit is composed of two modules, each contributing to the development of core competencies in the modeling, analysis, and control of complex mechanical systems, with a particular focus on robotics. Overall, the course aims to provide students with a solid theoretical and practical foundation for the design, control, and analysis of manipulators and complex mechanisms in robotics contexts.

This class covers the fundamentals of robotics, structured into two modules. The first module focuses on the basics of robot manipulator modeling and control. The second module explores the fundamentals of mechanisms and machines from a mechanical perspective.

The class consists of two modules, and the final exam mark is the average of the intermediate tests from both modules. Please refer to each module for a detailed description of its respective intermediate test.

The class is divided into two modules, with specific assessment methods for each module’s learning outcomes detailed within the corresponding module descriptions.

TThe goal of the teaching unit is to provide the basic notions for the design of analog (both passive and active) and digital filters tailored for processing sensor measurements in robotic applications. The topics are proposed to students through both theoretical lessons and practical activities such as the simulation and the hardware realization of filters.

This subject aims to provide basic elements for the design and synthesis of analog and digital filters. One half of the lectures is given in laboratory, where the students are driven to design, simulate, implement and test different kinds of filters, by using tools such as PSPICE, MATLAB and laboratory equipment.

Oral: two discussions starting from two questions (one chosen by the student and one asked by the examiner) concerned with the topics treated during the lessons and to the design techniques applied during lab activities.

The learning results are assessed through the lab activities and the oral exam.

• Notes provided by the lecturer (main reference)
• C. Bowik, "RF circuits design," Newnes, 1997.
• J.G. Proakis, D.G. Manolakis, "Digital signal processing: principles, algorithms, and applications," Prentice Hall, 1996.
• M.E. Van Valkenburg, "Analog Filter Design," Oxford University Press, 1995.
• A. Liberatore, S. Manetti, "La progettazione dei filtri elettronici," Edizione Medicea, 1985.
• L.B. Jackson, "Digital filters and signal processing," Kluwer Academic Publishers, 1996.

Students will learn how to model the transducers based on smart materials (e.g., shape memory alloys, piezoelectric, electromagnetic) and their interaction with the hosting structure, as well as how to test and characterize the systems experimentally. They will be able to model and test coupled systems. They will be acquainted with the material peculiarities, their main sensing/actuation features as well as their use in practical applications. Moreover, they will be aware of their possible advanced use in mechanical systems in which the material properties are exploited in the context of multi-domain interaction with the hosting structure (e.g., vibration attenuation, monitoring, energy harvesting, adaptability).

I materiali intelligenti sono materiali che possono cambiare alcune delle loro caratteristiche se sottoposti ad uno stimolo esterno. Per questo motivo sono largamente utilizzati per lo sviluppo sia di sensori sia di attuatori.
Tuttavia, quando questi materiali sono integrati in un sistema, è possibile realizzare un dispositivo accoppiato (struttura+materiale), detto smart, dotato di caratteristiche e funzioni avanzate. Controllando le proprietà di sensing e attuazione dei materiali intelligenti è possibile, infatti, progettare sistemi in grado di percepire eventuali cambiamenti (del sistema stesso, dell’ambiente in cui si trova o delle condizioni operative) e di applicare contromisure per garantire il proprio corretto funzionamento.
Il corso si propone di rendere lo studente capace di concepire sistemi e strutture intelligenti dotate di proprietà avanzate, come ad esempio adattabilità, efficienza energetica, auto-monitoraggio ed energy harvesting, attraverso lo studio delle caratteristiche di questi materiali e del comportamento del sistema accoppiato struttura-materiale intelligente.

L’esame consta di una prova scritta e di una prova orale in cui viene discusso un progetto di gruppo sviluppato durante il corso.
Si consiglia agli studenti con certificazione di DSA, di disabilità o di altri bisogni educativi speciali di contattare il docente all’inizio delle lezioni per concordare modalità didattiche e d’esame che, nel rispetto degli obiettivi dell’insegnamento, tengano conto delle modalità di apprendimento individuali.

La prova scritta ha lo scopo di verificare che lo studente abbia raggiunto un’adeguata conoscenza delle peculiarità dei materiali intelligenti trattati durante il corso e delle loro possibili applicazioni in ambito ingegneristico. Il progetto, discusso in sede di colloquio, permette di accertare la capacità dello studente di applicare le conoscenze acquisite durante il corso, la sua capacità di giudizio critico ed indipendente dei dati e dei risultati ottenuti durante lo sviluppo del progetto ed infine l’abilità di presentare in maniera adeguata l’attività svolta.

A. Preumont, “Vibration Control of Active Structures: an introduction”, Springer Dordrecht 2011
Dimitris C. Lagoudas, “Shape Memory Alloys: Modeling and Engineering Applications”, Springer New York, NY, 2008
Anders Brandt, “Noise and Vibration Analysis: Signal Analysis and Experimental Procedures”, John Wiley & Sons, 2011

The objective of the course is to make students aware about the most relevant issues in the fields of social robotics, including: verbal and nonverbal human-robot interaction; cultural factors in the design of social robots; anthropomorphic and zoomorphic robots and robot behaviours; sensors for human-robot interaction; methodology and constraints in making experiments with robots and human participants; application scenarios. The student will face these problems both from a theoretical perspective and through practical assignments, by exploring in depth one of the topics above on real robots for social interaction.

Social Robotics (SoRo) is a research field that provides robots with a brand new set of skills, specifically related to social behaviour and natural interaction with humans. Different aspects must be considered when seeking to the design of robots capable of social interaction: how robots can enable an intuitive interaction through proxemics, gestures, gaze or head movements, as well as speech; how timely, purposive, and appropriate communication is established; which environmental (including human) cues must be detected, represented and considered to ground such interaction.

The SoRo final mark is based on a written exam at the end of the course plus assignments during the course
Please note that:
EMARO+, JEMARO, and Robotics students have a strict deadline to complete their assignments;
Ph.D. students attending the course do not have any specific deadline.
Properly carrying out an assignment means providing:
a (possibly working, maybe with assumptions) solution to the given problem;
properly commented code, and where appropriate a tutorial;
a video showing how the developed solution work.

The written exam and the assignments will be evaluated based on:
the capability in applying the notions and insights discussed during classes;
the quality of the provided documentation (for assignments);

Relevant material (tech reports, slides, papers) will be distributed by the teachers during the lectures.

The teaching unit consists of two modules, each focusing on essential aspects of programming and cognitive systems for robotic applications. Overall, the course aims to equip students with both the software development skills and the understanding of cognitive architecture necessary for designing robots capable of complex interactions with their environments.

The exam covers two essential modules in robotics, focusing on different aspects of programming and architecture design. Here follows a brief overview of the two modules, but further information is available on the individual module pages.
Advanced Robotics Programming (ARP) This module delves into system programming for robotics, emphasizing multi-tasking, distributed, and concurrent programming. Students will learn to utilize operating system services for designing reliable and portable robotic programs.
Cognitive Architectures for Robotics (COGAR) This module introduces the highly experimental field of cognitive architectures in robotics. It aims to equip scholars and future researchers with the conceptual tools and practical knowledge needed to design and implement complex robotic architectures with advanced perception, reasoning, and action capabilities.

The class consists of two modules, and the final exam mark is the average of the intermediate tests from both modules. Please refer to each module for a detailed description of its respective intermediate test.

The class is divided into two modules, with specific assessment methods for each module’s learning outcomes detailed within the corresponding module descriptions.

The goal of the module is to provide methodologies and tools for designing systems’ models to be used for control, estimation, diagnosis, prediction, etc. Different identification methods are considered, both in a “black box” context (where the structure of the system is unknown), as well as in a “grey box” (uncertainty on parameters) one. Methods are provided for choosing the complexity of the models, for determining the values of their parameters, and to validate them. Moreover, state estimation problems are addressed and their connections with control and identification are considered.

• Different models for dynamic systems and their applications. - Parametric and non-parametric models - Identification techniques for linear models. - Nonlinear models. Examples and identification methods. - Validation procedures. - Introduction to state estimation. - State estimation in the presence of disturbances. - Kalman filter and its extension to the nonlinear case. - Techniques for parameter identification of linear systems in the presence of disturbances.

Oral exam with discussion of system identification and state estimation methods and possible applications.
Students with learning disorders ("disturbi specifici di apprendimento", DSA) will be allowed to use specific modalities and supports that will be determined on a case-by-case basis in agreement with the delegate of the Engineering courses in the Committee for the Inclusion of Students with Disabilities.

The students will be evaluated on the basis of their capability to describe system identification and state estimation algorithms, to choose suitable models depending on the application context and to appropriately use a data set to optimize the complexity and the parameters of models.

L. Ljung, “System Identification: Theory for the User”, Prentice Hall Y. Bar-Shalom, X. R. Li, T. Kirubarajan, “Estimation with Applications to Tracking and Navigation”, John Wiley & Sons Further readings will be given by lecturer.

The teaching unit is composed of two modules, each focusing on critical aspects of system analysis, identification, and control in the context of multi-variable systems. Overall, the teaching unit provides students with the theoretical and practical skills necessary for the analysis, identification, and control of multi-variable systems, particularly in robotics and dynamic systems applications.

The course consists of two modules, Control of Multi-Variable Systems (80181, 5 ECTS) and System Identification (111106, 5 ECTS), both during the first semester. Central to the course is the concept of dynamical model. While the System Identification module focuses on methodologies and tools for designing systems’ models, the Control of Multi-Variable Systems module focuses on the structural properties of multivariable (linear) time invariant dynamical models and their use within robotics applications. Please refer to each module for a detailed description of their contents, learning outcomes, prerequists, teaching methods and bibliography.

The final course mark is the average of the intermediate tests from both modules. Please refer to each module for a detailed description of their contents, learning outcomes, prerequists, teaching methods and bibliography.

The class is divided into two modules, with specific assessment methods for each module’s learning outcomes detailed within the corresponding module descriptions.

Specific use cases on the evaluation of the security of the object recognition system of the iCub robot will be addressed. Students will also get ability to answer open-ended questions with closed books, solve numerical exercises, use open-source libraries for the security evaluation of machine learning algorithms used by modern robots.

Introduction to machine learning security: introduction by practical examples from computer vision, biometrics, spam filtering, malware detection.
Threat models and attacks to machine learning. Modelling adversarial tasks. The two-player model (the attacker and the classifier). Levels of reciprocal knowledge of the two players (perfect knowledge, limited knowledge, knowledge by queries and feedback). Attack models against machine learning: evasion, poisoning, backdoor and privacy attacks. The concepts of security by design and security by obscurity.
Evaluation of machine learning algorithms in adversarial environments. Vulnerability assessment via formal methods and empirical evaluations. Adaptive and non-adaptive evaluations.
Design of machine learning algorithms in adversarial environments: Taxonomy of possible defense strategies. Evasion attacks and countermeasures. Defenses against evasion, poisoning, backdoor and privacy attacks. Poisoning attacks and countermeasures. Backdoor poisoning, privacy-related threats and defenses.
Hands-on classes on attacks against machine learning and defences of machine-learning algorithms in robotics

Intermediate in class assignments

Solutions of numerical/coding exercises and open-ended, closed questions.

A. Joseph, B. Nelson, B. Rubinstein, D. Tygar, Adversarial machine learning, Cambridge University Press, 2018
B., Battista, F. Roli. Wild patterns: Ten years after the rise of adversarial machine learning. Pattern Recognition 84 (2018): 317-331.
B. Biggio, F.Roli, Wild Patterns, Half-day Tutorial on Adversarial Machine Learning: https://www.pluribus-one.it/research/sec-ml/wild-patterns
Biggio, B., Corona, I., Maiorca, D., Nelson, B., Srndic, N., Laskov, P., Giacinto, G., Roli, F. Evasion attacks against machine learning at test time. ECML-PKDD, 2013.
Biggio, B., Fumera, G., Roli, F. Security evaluation of pattern classifiers under attack. IEEE Trans. Knowl. Data Eng., 26 (4):984–996, 2014.

The aim of this teaching unit is to provide students with fundamental and advanced concepts on the security of machine learning and trustworthy artificial intelligence.

Today machine-learning algorithms and AI-based systems are used for many real-world applications, including image recognition, spam filtering, malware detection, biometric recognition. In these applications, the learning algorithm may have to face intelligent and adaptive attackers who can carefully manipulate data to purposely subvert both the learning and the operational phases. Part 1 of the course aims to introduce the fundamentals of the security of machine learning and the related field of adversarial machine learning. Part 2 introduces the international regulations behind the so called “trustworthy AI. The course uses application examples including object recognition in images, biometric recognition, spam filtering, and malware detection.

Intermediate in class assignments or home assignment+oral exam.

Intermediate in class assignments (closed-book solutions of numerical/coding exercises and open-ended/closed questions), or home assignment+oral exam. Grading policy = open/closed questions (15/30) + numerical/coding exercises (15/30)
Students with certification of Specific Learning Disabilities (SLD), disabilities, or other special educational needs must contact the instructor at the beginning of the course to agree on teaching and examination methods that, while respecting the course objectives, take into account individual learning styles and provide appropriate compensatory tools. It is reminded that the request for compensatory/dispensatory measures for exams must be sent to the course instructor, the School representative, and the “Settore servizi per l'inclusione degli studenti con disabilità e con DSA” office ([email protected]) at least 10 working days before the test, as per the guidelines available at the link: https://unige.it/disabilita-dsa

A. Joseph, B. Nelson, B. Rubinstein, D. Tygar, Adversarial machine learning, Cambridge University Press, 2018
B., Battista, F. Roli. Wild patterns: Ten years after the rise of adversarial machine learning. Pattern Recognition 84 (2018): 317-331.
Biggio, B., Corona, I., Maiorca, D., Nelson, B., Srndic, N., Laskov, P., Giacinto, G., Roli, F. Evasion attacks against machine learning at test time. ECML-PKDD, 2013.
Biggio, B., Fumera, G., Roli, F. Security evaluation of pattern classifiers under attack. IEEE Trans. Knowl. Data Eng., 26 (4):984–996, 2014.
All slides used during the lessons and other teaching materials will be available on Aulaweb. In general, the notes taken during the lessons and the material on Aulaweb are sufficient for preparing for the exam.

The course is composed of two modules, each focusing on the security and reliability of AI systems in robotics. Overall, the course aims to provide students with both theoretical and practical expertise in ensuring the security, safety, and trustworthiness of AI systems used in robotics.

Today machine-learning algorithms and AI-based systems are used for many real-world applications, including image recognition, spam filtering, malware detection, biometric recognition. In these applications, the learning algorithm may have to face intelligent and adaptive attackers who can carefully manipulate data to purposely subvert both the learning and the operational phases. Part 1 of the course aims to introduce the fundamentals of the security of machine learning and the related field of adversarial machine learning. Part 2 introduces the international regulations behind the so called “trustworthy AI. The course uses application examples including object recognition in images, biometric recognition, robotics.

Intermediate in class assignments, or home assignmemt and oral exam.

Solutions of open and closed questions, numerical excercies (15/309 + coding exercises (15/30).

Starting from the knowledge on the fundamentals of graphics, modeling and animation of 3D digital objects, the aim of the course is to get to the programming skills necessary to build applications and systems based on simulation in virtual / mixed / augmented / extended reality (VR / AR / MR / XR). The fundamental objectives of this course are to make students aware of the necessary interdisciplinarity of VR for Robotics: from mobile programming to biomechanics, sensory perception, humanoid robotics and video games, in order to manage complex interactions between simulated and / or physical objects and actors (both FPV first-person view and TPV third-person view).

The aim of the course is to start with the knowledge on the fundamentals of graphics, modeling and animation of 3D digital objects, to arrive at the programming skills necessary to build applications and systems based on simulation in virtual / mixed / augmented / extended reality (VR / AR / MR/ XR).
This objective will serve to make students aware of the necessary interdisciplinarity of contributions from mobile programming, biomechanics, sensory perception, robotics and video games to manage complex interactions between simulated and / or physical objects and actors (both in first person view FPV and third person view TPV).

The exam modality is based on a team-based or individual project, followed by an oral discussion.
Requested submissions:
Pitch Document before the end of the lessons
Report of the Project (or GDD Game Design Document), 7 working days before the chosen exam date.
Final Demo of the project, 2 working days before the chosen exam date.
Presentation of the project and oral discussion, on the exam date

The assessment method is based on the continuous assessment of learning and the final group or individual project, followed by an oral discussion.
The distribution of grades will be as follows:
Active partecipation and submission on time of the Pitch Document, 10%
Submission on time of the Report of the Project (or Game Design Document), 25%
Submission on time of the Final Demo of the project, 40%
Presentation of the project and oral discussion, 25%

Steven M. La Valle, Virtual Reality, 2023, Cambridge Univ. Press, freely available at http://vr.cs.uiuc.edu/
S.lacono - G.Vercelli, Dalla fotogramnmetria con i droni alla Virtual Reality (cod.113560) (in Italian), https://raiseliguria.aulaweb.unige.it/course/view.php?id=33
Epic Games, Tutorial su Unreal Engine, available at: https://dev.epicgames.com/community/unreal­ engine/learning