Neurorobotics promotes the fusion of neuroscience and robotics.
Image credit: IEEE Robotics and Automation Society Technical Committee for Neuro-Robotics Systems .
Course Description | Learning Objectives | Outcomes | Lecture Notes | Course Textbook | Recommended Reading | Resources
Course Description
Neurorobots are robots whose control is based on a model of some aspect of the operation of the brain. This course explores neurorobotics as a possible means to develop autonomous systems that have some level of biological intelligence. It will give students an understanding of intelligence in biological systems from the perspective of cognitive neuroscience and insights into how this understanding can be used to augment alternative knowledge-based approaches to the development of cognitive robots. The course is delivered through a mix of teaching, reading, and in-class discussion. Students will conduct simple neurorobotics experiments using a robot simulator to illustrate and consolidate what they have learned in class. Student progress is assessed by a series of multiple choice tests and written individual & group assignments. There are no prerequisites for taking this course, although it would be an advantage to have taken 18-799-K Artificial Cognitive Systems.
Learning Objectives
Students will be introduced to roots of neurorobotics in cybernetics and cognitive neuroscience and some simple principles that underly emergent behaviour. They will learn about the different forms of control in robotics and neuroscience. They will learn about natural and artifical neurons and networks of the neurons, how they are modelled, and how they operate. Students will be introduced to the structure of the brain, the principles of neuromorphic computing, and the principles of neurorobot design. They will lean how navigation is accomplished using neural models and they will learn how neurophysiology and psychology impact of cognitive development and social cognition. They will also learn about the importance of emotion in social interaction. They will learn what the future of neurorobotics might hold and what applications might benefit from the use of neurorobotics.
Outcomes
After completing this course, students will be able to do the following.
Lecture Notes
Module 1: Background and Foundations
Module 2: Neurorobot Design Principles
Hwu, T. and Krichmar, J. (2022). Neurorobotics: Connecting the Brain, Body and Environment , MIT Press.
Recommended Reading
Mataric, M. The Robotics Primer , MIT Press, 2007.
Lecture 1. Neurorobotics. Origins and background.
Lecture 2. Neuroscience. Neurons and synapses; systems neuroscience.
Lecture 3. Neuroscience. Case study: visual navigation in insects and mammals.
Lecture 4. Learning and memory. Types of learning; neural networks basics.
Lecture 5. Learning and memory. Weight stabilization; classical conditioning; spiking neural networks.
Lecture 6. Reinforcement learning and prediction. Braitenberg Vehicle 4; Markov decision processes; prediction.
Lecture 7. Reinforcement learning and prediction. Case study: Darwin VII - perceptual categorization and conditioning.
Lecture 1. Design Principle 1: Every Action has a Reaction.
Lecture 2. Design Principle 1: Case study.
Lecture 3. Design Principle 2: Adaptive behaviour I.
Lecture 4. Design Principle 2: Adaptive behaviour II.
Lecture 5. Design Principle 3: Behavioral Tradeoffs.
Lecture 6. Design Principle 3: Case study.