Lecturer: Stefano Ramat  

Course name: Biomimetic systems
Course code: 503003
Degree course: Bioingegneria
Disciplinary field of science: ING-INF/06
L'insegnamento è caratterizzante per: Bioingegneria
University credits: ECTS 6
Course website: n.d.

Specific course objectives

The aim of the course is to provide students with some basic knowledge of tools and technology for the design and realization of artificial sensorimotor systems that can emulate the corresponding biological systems. The student will acquire knowledge of physiology and psychophysics related to perception and motor skills, along with technical and methodological skills for the realization of life-like robotic systems. To present these broad issues within a class we will focus on the study of the motor system and specifically on the ocular motor system and simple motor tasks related to tracking and / or grasping of an object in proximal space. At the end of the course the student will be able to use the methodological tools of machine learning, such as the various paradigms of neural learning and genetic algorithms, and have some technical knowledge about sensors, actuators and devices used in the field of anthropomorphic robotics.

Course programme

Perception and movement

• Processes of perception, discrimination, learning, memory, vigilance.
• Nervous System
• Biological neural networks.
• Neurophysiological and psychophysical tests.
• Evaluation of sensations.
• Models and extraction of descriptive parameters.
• The Motor Control: a computational approach.

Artificial neural networks.
* Neuron of McCulloch and Pitts. * Single layer perceptron. * Multilayer Perceptron. * Radial Basis Function Networks. * Self Organizing Maps * Discrete Hopfield Networks * Recurrent Layer Networks

Training Neural Networks

• Linear separability
• Principle of Hebb
• Delta rule
• Simulated annealing
• Back-propagation algorithm.
• Other neural algorithms.

Fuzzy Logic

• Introduction to fuzzy logic.
• The fuzzy reasoning and the FIS (Fuzzy Inference System).
• Sugeno ANFIS.

Evolutionary algorithms

• Introduction to evolutionary algorithms.
• Genetic Algorithms.
• Genetic Operators.
• Applications to optimization problems.
• Artificial Immune Systems: an evolutionary approach to classification.

Biomimetics in sensory and motor systems, and robotics applications

• Ocular motor and vestibular systems
• Biological sensors and information coding
• Sensory information processing models
• Objects identification
• Pattern recognition

Course entry requirements

Knowledge of calculus and physics (mechanics and electromagnetism), control systems, human physiology, basic digital signal processing, basic technologies of sensors. Programming in Matlab environment.

Course structure and teaching

Lectures (hours/year in lecture theatre): 45
Practical class (hours/year in lecture theatre): 0
Practicals / Workshops (hours/year in lecture theatre): 0

Suggested reading materials

S. Haykin. Neural Networks (2nd edition). Prentice Hall, 1999.

M. Mitchell. An introduction to genetic algorithms. MIT Press, 1996.

A. Berthoz. Il senso del movimento. Mc Graw Hill, 1998. Lettura.

Testing and exams

The final evaluation is made up of a written exam and a project consisting in a written report and a Matlab software. The final grade will be the mean of those obtained in the two parts of the examination.