Evolutionary robotics

MICRO-515 / 3 credits

Teacher: Floreano Dario

Language: English

Summary

The course gives an introduction to evolutionary computation, its major algorithms, applications to optimization problems (including evolution of neural networks), and application to design and control of robots. It includes software exercises and project to evolve, build, and test a robot.

Content

  • Natural and Artificial Evolution
  • Principles of Evolutionary Computation
  • Algorithms: Genetic Algorithms and Evolutionary Strategies
  • Algorithms: Multi-objective optimization (various algorithms, including NSGA-II)
  • Intorduction to neural netwok architectures and learning methods, including reinforcement learning
  • Evolution of Artificial Neural Networks and comparison to Reinforcement Leanring
  • Evolution of Neurocontrollers for mobile robots
  • Morphological Growth and Evolution
  • Evolution of Collective Systems: Competitive and Cooperative Evolution
  • Evolution of bio-hybrid robots

Learning Prerequisites

Important concepts to start the course

Programming skills (Phython, Java, C++)

Learning Outcomes

By the end of the course, the student must be able to:

  • Apply new tools for software and hardware engineering
  • Translate acquired theoretical knowledge in practical implementations during laboratory sessions

Teaching methods

Lectures, software exercises, and project involving evolution of morphology and neural control of robot, 3D printing of parts and assembly, test characterization of evolved robots.

Expected student activities

Attending lectures, asking critical questions, taking all exercises and completing assignments for the following week, forming groups and performing collaboratively project with physical robots, writing and presenting project results

Assessment methods

Mini-project report in powerpoint + presentation + written exam

Supervision

Office hours No
Assistants Yes
Forum Yes

Resources

Bibliography

Floreano, D. & Mattiussi, C. (2008) Bioinspired Artificial Intelligence. MIT Press (selected chapters)

Ressources en bibliothèque

Moodle Link

In the programs

  • Semester: Spring
  • Exam form: Written (summer session)
  • Subject examined: Evolutionary robotics
  • Courses: 2 Hour(s) per week x 14 weeks
  • Project: 1 Hour(s) per week x 14 weeks
  • TP: 1 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: Written (summer session)
  • Subject examined: Evolutionary robotics
  • Courses: 2 Hour(s) per week x 14 weeks
  • Project: 1 Hour(s) per week x 14 weeks
  • TP: 1 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: Written (summer session)
  • Subject examined: Evolutionary robotics
  • Courses: 2 Hour(s) per week x 14 weeks
  • Project: 1 Hour(s) per week x 14 weeks
  • TP: 1 Hour(s) per week x 14 weeks
  • Type: optional
  • Semester: Spring
  • Exam form: Written (summer session)
  • Subject examined: Evolutionary robotics
  • Courses: 2 Hour(s) per week x 14 weeks
  • Project: 1 Hour(s) per week x 14 weeks
  • TP: 1 Hour(s) per week x 14 weeks
  • Type: optional
  • Exam form: Written (summer session)
  • Subject examined: Evolutionary robotics
  • Courses: 2 Hour(s) per week x 14 weeks
  • Project: 1 Hour(s) per week x 14 weeks
  • Type: optional

Reference week

Related courses

Results from graphsearch.epfl.ch.