International Master Program in

Mechatronic Systems for Rehabilitation

Booklet of teaching units

Third semester (M2S1)

Master Sciences de lIngnieur Universit Pierre et Marie Curie Paris 6 Boite 164, 4 place Jussieu, 75252 Paris cedex 05 FRANCE

Course title

HUMAN VERBAL COMMUNICATION ANALYSIS

Code

Course description Course components: CM (lectures), TD (tutorials), TP 18 h CM + 12 h TP (practical work), Projects European Credits 3 Master specialization Mechatronic Systems for Rehabilitation (MSR) Semester S3 Language English Maximum number of students 15 a) Objective The verbal communication channel conveys many information such as linguistic, speaker identity but also emotional, intentional or pathological states. Focus on understanding on key algorithms including speech analysis, prosody modeling, sound classification and segmentation. Theory will be presented and practice will be provided in analyzing typical situations such as speech and language disorders, patient and speech therapist interaction. This course teaches the principle behind current speech processing techniques for a human behavior analysis purpose. b) Content

1- Introduction to interaction 2- Introduction to speech processing (signal and phonetic) 3- Speech representations (Acoustical, Phonetic, Prosodic, Semantic) 4- Psycho-acoustic 5- Algorithmic aspects for speaker states classification 6- Application to speech assessment and Human-Robot Interaction c) Pre-requisite Basic knowledge in signal processing d) Evaluation methods / Knowledge acquisition control Written examinations and practical work evaluation

Teaching organization University work Lectures (CM) Tutorials (TD) Practical work (TP) Projects Other Total time volume 18 h 12 h Weekly number of hours 3h 4h Number of students per group 15 15

Course title

HAPTICS AND APPLICATIONS IN TELEOPERATION AND VIRTUAL REALITY

Code

Course description Course components: CM (lectures), TD (tutorials), TP 20 h CM + 10 h TP (practical work), Projects European Credits 3 Master specialization Mechatronic Systems for Rehabilitation (MSR) Semester S3 Language English Maximum number of students 15 a) Objective This course is designed to give an overview of the role of haptics (tactile and force) feedback in medical applications. The course presents an overview of haptics in a teleoperated or virtual world, and will discuss the history, technology, methodologies, and applications. The course is primarily lecture and lab based. The lab component of the course will include a simple haptic rendering and a teleoperation exercise. There will also be worksheets completed during class time. b) Content

Fundamentals of human haptic perception, such as the sensory elements involved, their characteristics, the cognitive process of recognizing haptic properties, and cross-modal interaction. haptics perception and acquisition of the environment. Modeling and rendering of contact, manipulation, and interaction among objects. Discrimination of roughness, shape, and compliance in static or dynamic contact settings, and how this knowledge can be exploited in rendering algorithms. Method of synthesis and object mechanical properties and their applications to virtual reality systems. Force feedback control in teleoperation. Master-slave control. Methods and performances. Force feedback in haptic interfaces and the transparency challenge. c) Pre-requisite None d) Evaluation methods / Knowledge acquisition control Written examinations and practical work evaluation

Teaching organization University work Lectures (CM) Tutorials (TD) Practical work (TP) Projects Other Total time volume 20 h 10 h Weekly number of hours 4h 3h Number of students per group 15 15

Course title

MODELS OF SENSORI-MOTOR FUNCTIONS AND LEARNING

Code

Course description Course components: CM (lectures), TD (tutorials), TP 16h CM + 14h Projects (practical work), Projects European Credits 3 Master specialization Mechatronic Systems for Rehabilitation (MSR) Semester S3 Language English Maximum number of students 15 a) Objective The goal of the course is to present the modelling activities dedicated to motor control and motor learning at the level of computational principles. We present the basics of the theories underlying state-of-the-art computational tools and approaches and we give the students a general background in sensori-motor functions and sensori-motor learning, pointing to the main scientific issues of the domain. Finally, we give the students an opportunity to test state-of-the-art methods on a concrete experimental setting. After attending this course, a student should be able to take part to a research project at the interface between robotics and modelling activities in life sciences. b) Content

The course presents general computational principles of motor control and motor learning. It consists of eight lessons: The organization of movement: issues, observations, concepts and models Mathematics of control Motor control I: anatomo-physiological organization of motor control (muscles, reflexes and circuits) Motor control II: computational motor control (dynamical systems, internal models, optimality) Motor control III: models of motor control (muscle models, equilibrium point and impedance control, optimal control) Motor Learning I: introduction (motor adaptation, skill acquisition, action selection, reinforcement learning) Motor Learning II: reinforcement learning tools for motor learning Motor Learning III: supervised learning tools for motor learning Furthermore, the students realize a project consisting of the simulation of a relevant motor system borrowed from the literature, with time dedicated to work on the project in the presence of the tutors. c) Pre-requisite Some Matlab programming ability will be required to realize the project. d) Evaluation methods / Knowledge acquisition control Project evaluation e) References Shadmehr R, Wise SP (2005) Computational Neurobiology of Reaching and Pointing: A Foundation for Motor Learning. Cambridge, MA: MIT Press. Arbib MA (2002) The Handbook of Brain Theory and Neural Networks, 2nd ed. Cambridge, MA: MIT Press. Bryson AE (1999) Dynamic Optimization. Englewood Cliffs, NJ: Prentice-Hall. Sutton RS, Barto AG (1998) Reinforcement Learning: An Introduction. Cambridge, MA: MIT Press.

Teaching organization University work Lectures (CM) Tutorials (TD) Practical work (TP) Projects Other Total time volume 16 h Weekly number of hours 4h Number of students per group 15

14 h

4h

15

Course title

MULTIMODAL PERCEPTION FOR HUMAN POSE ESTIMATION

Code

Course description Course components: CM (lectures), TD (tutorials), TP 18 h CM + 12 h TP (practical work), Projects European Credits 3 Master specialization Mechatronic Systems for Rehabilitation 5MSR) Semester S3 Language English Maximum number of students 15 a) Objective The objective of this course is to present to students the sensors and tools useful for human pose estimation. Various type of sensors like force sensors, accelerometers or motion capture systems will be introduced. Their use in human pose (or gesture) tracking system will be too presented. As generally, the human pose is not directly measured by these sensors, this pose has to be estimated from the measures. This course will focus on state estimators like the Kalman filter and its extensions or the particle filter. b) Content Sensors for human pose estimation: force sensors, accelerometers, EMG, motion capture Estimation theory : bias, covariance, efficiency, asymptotic properties, classical estimators (least square, maximum likelihood,.) and state estimators for stochastic systems : Kalman filter, extended Kalman filter, unscented Kalman filter, particle filter,. Sensors fusion

c) Pre-requisite None d) Evaluation methods / Knowledge acquisition control Written examinations and practical work evaluation

Teaching organization University work Lectures (CM) Tutorials (TD) Practical work (TP) Projects Other Total time volume 18 h 12 h Weekly number of hours 4h 4h Number of students per group 15 15

Course title

PROGRAMMING REAL-TIME SYSTEMS

Code

Course description Course components: CM (lectures), TD (tutorials), TP 12h CM + 2 h TD + 16h TP (practical work), Projects European Credits 3 Master specialization Mechatronic Systems for Rehabilitation (MSR) Semester S3 Language English Maximum number of students 15 a) Objective This course will provide an understanding of real-time embedded systems with key modeling concepts and fundamental programming aspects of multithreaded applications running with real-time operating systems currently used in rehabilitation robotics. b) Content

Introduction to real-time operating systems, tasks and kernel objects (semaphores, message queues, etc), scheduling, synchronization and communication. Timer services, I/O subsystem, exceptions and interrupts. c) Pre-requisite C Langage d) Evaluation methods / Knowledge acquisition control Written examinations and practical work evaluation

Teaching organization University work Lectures (CM) Tutorials (TD) Practical work (TP) Projects Other Total time volume 12 h 2h 16 h Weekly number of hours 2h 2h 4h Number of students per group 15 15 15

Course title

PHYSICALLY INTERACTING ROBOTS AND THEIR APPLICATIONS TO REHABILITATION

Code

Course description Course components: CM (lectures), TD (tutorials), TP 22 h CM + 8 h TP (practical work), Projects European Credits 3 Master specialization Mechatronic Systems for Rehabilitation (MSR) Semester S3 Language English Maximum number of students 15 a) Objective Being able of properly handling the specific constraints pertaining to the design and control of robotic devices aimed at physically interacting with a human being. Understanding how they can be used in a motor rehabilitation protocol. b) Content

The course focuses on those robots that are more and more used for rehabilitation purpose, i.e. that can physically assist the motion of a human being and eventually apply corrective forces. It will be divided in three main parts. Firstly, a 10 hour series of lectures will focus on design. It will start by an exhaustive overview of the state of the art in rehab robotics, which will also cover the medical aspects such as protocol definition, pathology description, motor learning fundaments, etc. This series will then cover the kinematics, statics and dynamics aspects of the interactive robot design, all based on elementary biomechanics, as well as the most popular actuation solutions. Secondly, an 6 hour series of lectures will introduce the problems and solutions encountered in force control of robots, in a conventional way. Attached to this part will be a 4 hour lab. Finally, a 6 hour series of lectures will focus on current tendencies of anthropocentric robot control, i.e. robot control methods explicitly accounting for human being models in their controller, with, again, a 4 hour lab. c) Pre-requisite Bases in robot design, modeling (kinematics and dynamics) and control. Programming with Matlab d) Evaluation methods / Knowledge acquisition control Written examinations and practical work evaluation

Teaching organization University work Lectures (CM) Tutorials (TD) Practical work (TP) Projects Other Total time volume 22 h 8h Weekly number of hours 4h 4h Number of students per group 15 15

Course title

ADVANCED ROBOTIC SYSTEMS

Code

Course description Course components: CM (lectures), TD (tutorials), TP 16 h CM + 16 h TP (practical work), Projects European Credits 3 Master specialization Mechatronic Systems for Rehabilitation (MSR) Semester S3 Language English Maximum number of students 15 a) Objective This course introduces the methods for mechanical modelling, analysis and control of robotics systems. It aims at giving to the students the possibility of beginning in complete autonomy the resolution of certain number of elementary problems of robotics as the configuration description, the generation of trajectories, the dynamic control as well as of being capable of analyzing the behavior of biomechanical systems or simply mechanical from the point of view of their properties of kinematostatic transmission, their static and dynamic balance. b) Content

Kinematic description and parametric representation of systems. Holonome and non-holonome constraints Homogeneous transformations. Direct and inverse geometrical model by analytical, numeric methods and by methods of homotopie. Laws of transmission of movements in the simple and complex systems. Symbolic methods of elimination for the obtaining of the input-output relationships. Singularities. Resolution of the kinematic problems. Resolution of the inverse kinematic problems - forced systems and redundant systems. Dynamic model by the general theorems and the equations of the movement by the equations of Lagrange. Generation of articular and Cartesian trajectories. c) Pre-requisite Rational mechanics d) Evaluation methods / Knowledge acquisition control Written examinations and practical work evaluation

Teaching organization University work Lectures (CM) Tutorials (TD) Practical work (TP) Projects Other Total time volume
16 h 16 h

Weekly number of hours

Number of students per group

15 15

Course title

SAFETY, EVALUATION AND ACCEPTABILITY ASPECTS IN REHABILITATION AND ASSISTIVE TECHNOLOGIES

Code

Course description Course components: CM (lectures), TD (tutorials), TP 30 h CM (practical work), Projects European Credits 3 Master specialization Mechatronic Systems for Rehabilitation (MSR) Semester S3 Language English Maximum number of students 15 a) Objective This course examines several central aspects with respect to the safety and the acceptability of assistive and rehabilitation technologies. Learn risk management concepts and techniques for the development of medical robots and more particularly rehabilitation robots. Developing equipment and devices that are well accepted by potential users is a central question in rehabilitation. The engineering aspects here are only a part of the issues to be considered at the design stage, the human factors and the impact of the potential robotic aids in the users life as to be carefully analyzed. b) Content

Definition and application of basic concepts for risk management: risk, harm, safety, hazard, hazardous situation, safety integrity, safety integrity level, etc. Risk analysis methods (Fault Tree Analysis, Failure Mode Effects and Criticality Analysis, HAZOP,etc.) based on system modeling Risk reduction techniques for medical robots (Hardware and software fault avoidance techniques, fault tolerance mechanisms, etc.) Certification principles and regulations Acceptability and ergonomy of the assistive technologies. Relationship between Assistive Technology and Inclusive Design. Clinical trials methodology and protocol design. Engineers, designers and technicians collaboration methodology. c) Pre-requisite None d) Evaluation methods / Knowledge acquisition control Written examinations and practical work evaluation Teaching organization University work Lectures (CM) Tutorials (TD) Practical work (TP) Projects Other Total time volume 30 h Weekly number of hours Number of students per group 15

Course title

IMAGE PROCESSING FOR HUMAN ANALYSIS

Code

Course description Course components: CM (lectures), TD (tutorials), TP 30 h CM (practical work), Projects European Credits 3 Master specialization Mechatronic Systems for Rehabilitation (MSR) Semester S3 Language English Maximum number of students 15 a) Objective The objective of this course is to provide to students image processing tools required for any real application based on vision and, more particularly, applications centred on the human. This course addresses several issues to present all the steps of image processing system. Several applications such as human detection or motion estimation will be proposed during tutorials. b) Content

Introduction Presentation of images Images filtering Mathematical Morphology Edge detection Human detection with background subtraction Motion estimation Basis of tracking c) Pre-requisite None d) Evaluation methods / Knowledge acquisition control Written examinations and practical work evaluation

Teaching organization University work Lectures (CM) Tutorials (TD) Practical work (TP) Projects Other Total time volume 14 h 16 h Weekly number of hours 2h 4h Number of students per group 15 15