cookieImage
2026-2027 / GNEU0002-1

Brain Inspired Computing

Duration

25h Th, 20h Pr, 20h Proj.

Number of credits

 Master MSc. in Biomedical Engineering, professional focus5 crédits 
 Master Msc. in Electrical Engineering, professional focus in Neuromorphic Engineering5 crédits 

Lecturer

Alessio Franci

Language(s) of instruction

English language

Organisation and examination

Teaching in the second semester

Schedule

Schedule online

Units courses prerequisite and corequisite

Prerequisite or corequisite units are presented within each program

Learning unit contents

NEW COURSE TITLE: Neuromorphic Systems and Control

Despite huge advances in data-driven control, reinforcement learning, physical systems modeling and physics-informed machine learning, as well as in smart and distributed sensing, big data assimilation and analysis, artificial systems remain orders of magnitude less efficient, performant, and resilient than their biological counterparts.

Neuromorphic engineering aims at understanding and bridging the gap between biology and machines by focusing on the most notable difference between them: the spiky, event-based, on-demand, and intrinsically adaptive nature of biological systems. This course presents the foundations of neuromorphic systems and control theory, the branch of neuromorphic engineering aiming to embed neuromorphic intelligence into real-time control systems, from sensors to controllers and actuators.

Part I of the course starts by revisiting the classical problem of static output feedback stabilization of LTI systems from a neuromorphic perspective. The intrinsic multiscale nature of the neuromorphic approach is also presented. Dual to stabilization, the course then revisits state observation from the neuromorphic perspective and presents an overview of ongoing research on neuromorphic dynamic output feedback stabilization.

Part II presents a summary of established and ongoing extensions of the material of Part I to non-stationary dynamical behaviors. This includes perceptual and motor decision-making, the generation of stable and tunable rhythmic behaviors, adaptive sensorimotor control, and the connections between adaptive neuromorphic control and on-line learning.

The course will present rigorous theory in an accessible way and thoroughly illustrate the theory on extensive numerical examples and real-time hardware control. The course will mostly be taught in Scilab, a free software which provides Xcos, a block diagram-based graphical user interface similar to Matlab Simulink.

Introduction

  • The biology of neuromorphic systems: neural sensing, neural deciding, neural acting
  • The biology of neuromorphic control: multiscale sensorimotor loops
  • Neuromorphic sensors, decision-makers, and actuators
  • A tentative definition of neuromorphic systems and control
  • Linear control systems preliminaries.
Part I: Neuromorphic stabilization

  • Neuromorphic static output feedback stabilization of one-dimensional linear systems
  • Neuromorphic static output feedback stabilization of SISO LTI systems
  • Muscle-inspired multi-scale neuromorphic stabilization
  • Neuromorphic sensors and neuromorphic PID control of MIMO systems.
Part II: Neuromorphic control of non-stationary dynamical behaviors

  • Decision-making and deadlock breaking through bifurcations of neuromorphic control systems
  • Neuromorphic control of oscillatory behaviors
  • (Optional) Flexible sensorimotor control from controlling low-dimensional bodies through high-dimensional neuromorphic controllers
  • (Optional) Neuromorphic control and learning
Part III: Group Project

  • Hardware and neuromorphic control loop design of a simple robotic system based on the cart-pendulum system
  • The class will work in cooperative teams, focused on different sub-aspects of the project, depending on the students' preferred topics and previous knowledge and background
  • Willing students will be able to present the project's results

Learning outcomes of the learning unit

Through theoretical classes and computational exercises developed in the Julia environment, at the end of the course students will:

- Understand the fundamental biological and mathematical principles of how brain control the body and how this can inspire new control-theoretical and robotics methods.

- Be able to design spiking control systems.

- Spiking control of real-time (hardware-in-the-loop) systems.

Prerequisite knowledge and skills

Good knowledge of linear systems and control.

Elements of nonlinear dynamical systems.

Interest in neuroscience and systems biology.

Good programming skill (the course will mostly be taught in Scilab)

Planned learning activities and teaching methods

The course includes theoretical lectures, practical sessions (mostly in Scilab), and two guided group project sessions.

For the practical sessions and the project, simple mechanical systems with real-time control loops (through Scilab+Arduino) will be used and available to the students.

Mode of delivery (face to face, distance learning, hybrid learning)

Face-to-face course

Course materials and recommended or required readings

Platform(s) used for course materials:
- eCampus


Further information:

Course slides, lecture notes, reference books and papers, and Scilab models will be distributed during the course.

Exam(s) in session

Any session

- In-person

written exam AND oral exam

Written work / report


Further information:

Two in-person exams (80%)

Final group project (20%)

Work placement(s)

Organisational remarks and main changes to the course

Contacts

Alessio Franci.
https://sites.google.com/site/francialessioac/

Association of one or more MOOCs