Beskrivelse
Recommended prerequisitesIt is recommended that the student has followed the 1st semester course in Classical Autonomous Systems and has good knowledge of programming, control theory, robotics, and mathematics.
Learning objectives - Knowledge
Having completed the course, the successful student will have knowledge about:
The broad, general principles of the field of Embodied AI
Fundamental mechanisms of bio-inspired perception, actuation, control and navigation in the context of autonomous systems
The process through which coordination and other emergent behaviours appear in bio-inspired robot swarms
Learning objectives - Skills
Having completed the course, the successful student will be able to:
Choose appropriate bio-inspired learning techniques to solve learning problems in autonomous robotic systems
Identify relevant bio-inspired perception, control, navigation, and learning techniques for the design of autonomous robotic systems
Develop and implement bio-inspired methods for perception, control, navigation, and learning in simulated environments
Evaluate and improve bio-inspired techniques applied to autonomous systems
Learning objectives - Competences
Having completed the course, the successful student will be able to:
Propose solutions to problems and scenarios within the context of autonomous systems for which bio-inspired approaches are appropriate
Design autonomous robotics systems integrating principles of embodied AI and bio-inspiration
Content
Introduction to bioinspiration and Embodied AI
Models of animal uni- and multi-sensorial perception
Biological sensing and perception
Motion perception and optical flow
Active perception
Non-visual perception
Multisensory integration
Bio-inspired actuation and locomotion
Central pattern generators
Legged locomotion and passive-dynamic walking
Snake-robot locomotion
Flapping-wing robots
Soft robotic actuators
Biological strategies and principles of control and navigation for autonomous systems
Braitenberg vehicles
Path integration models for bio-inspired navigation
Bio-inspired SLAM
Fundamentals of bio-inspired learning
Biological basis of Reinforcement Learning
Evolutionary robotics
Input Correlation Learning
Bio-inspired coordination of robot swarms
Self-organization
Collective decision-making
Boids