The University of Sheffield
Department of Computer Science

COM2005 Bio-Inspired Computing and Robotics

Summary

This module provides a foundation in biologically inspired computing, artificial intelligence and robotics. It takes inspiration from aspects of biological systems from intelligence, cognition and behaviour down to cellular inspired computing. It covers behaviour-based theory and the basics of learning techniques for robot control as well as biological collective behaviour and self-organisation. The module has a particular emphasis on robotics research and Human Robot Interaction (HRI), which is key to the development of theories and methods that create interactions between AI systems and the world.

Session Spring 2017/18
Credits 20
Assessment

There will be no written examination for this module. Various group and individual assignments will be set throughout the module.

Lecturer(s) Dr Dawn Walker & Prof. Roger Moore
Resources
Aims

The aims of this module are to:

  • provide an understanding of a range of features from the biological world that have influenced the world of computing.
  • foster a basic understanding of the nature biological inspiration for AI, Robotics and Computing - the goals and motivations
  • provide a broad overview of how robots are being used in the 21st century
  • develop an understanding of simple computer modelling of biological systems
  • provide a foundation for sensory-motor integration research
  • provide an historical context for biologically inspired systems
  • develop a foundation for biological learning models and self-organisation
  • provide experience of collaborative work that develops biologically inspired solutions to practical problems.
  • provide an understanding of the ethical issues involved in the application of modern robotics
  • to encourage curiosity and motivate further research in biologically inspired research
Objectives

At the end of the course we expect students to:

  • understand some of the essential features of biologically inspired systems
  • work effectively in a group environment to select and apply specific bio-inspired computing approaches to solve practical problems
  • be able to critically discuss ethical issues involved in the application of modern robotics
  • understand the interaction between cognition and control
  • be equipped with a knowledge of cellular computing and cellular automata.
Content

Bio-inspired Computing [10 lectures/practicals]

  • Introduction to Evolutionary Algorithms
  • Introduction to the characteristics of cellular systems
  • Fundamentals of Cellular Automata
  • Application of Cellular Automata in exploring Artificial Intelligence and understanding real-world dynamical systems
  • Introduction to Artificial Immune Systems
  • Application of Cellular Neural Computing
  • Computational models of biological populations

Cognition, Interaction and Robotics [10 lectures/practicals]

  • Traditional Artificial Intelligence versus modern robot control methods
  • Biological modelling and swarm robotics
  • Creating the illusion of intelligence in robots
  • Current and future robot applications
  • Robot ethics: concerns about control, deception, privacy and loss of human contact
  • The link between cognition and control - perceptual control theory
  • The role of memory in cognitive systems
  • The implications of mirror neurons - sensorimotor overlap
  • Human-robot interaction, theory of mind, communication and language
Teaching Method

Teaching will consist of two sessions per week and will be a mixture of lectures and practical laboratory classes.

Feedback
  • Informal feedback on the progress of the Lab group projects will be given during practical sessions. Feedback and marks for the formal assignments will be provided within 3 weeks of hand-in.
Recommended Reading
  • Stephen Wolfram: A New Kind of Science. Wolfram Media (2002) ISBN 1-57955-008-8. An extensive text relating to understanding complex systems and cellular automata: Available online at http://www.wolframscience.com/nksonline/toc.html (it's free, but registration is required)
  • Floreano and Matiussi: Bio-Inspired Artificial Intelligence: Theories, Methods, and Technologies (2008) ISBN: 0262062712
  • Relevant articles will be also be made available on the course website