- UCAS course code
- GG41
- UCAS institution code
- M20
BSc Computer Science and Mathematics with Industrial Experience
Year of entry: 2024
- View tabs
- View full page
Course unit details:
Cognitive Robotics
Unit code | COMP34212 |
---|---|
Credit rating | 10 |
Unit level | Level 3 |
Teaching period(s) | Semester 2 |
Available as a free choice unit? | No |
Overview
The course will provide an introduction to the methods and software/hardware technologies for intelligent robotics. It will analyse the selection and application of AI and machine learning methods, such as deep learning, for designing intelligent behaviour and cognitive skills (e.g. vision, motor control, language, social skills).
Aims
This unit provides an in-depth understanding of the field of cognitive robotics. This will analyse the selection, use and combination of methods and approaches in robotics, in artificial intelligence and in psychology and neuroscience to design intelligent behaviour and cognitive skills in interactive robots.
Learning outcomes
At the end of this course a student will be able to:
1. analyse the methods and software/hardware technologies for robotics research and applications
2. understand how our psychology and neuroscience understanding of behaviour and intelligence informs the design of robotics models and applications
3. compare, select and apply different machine learning methods for intelligent behaviour in robots
4. Discuss the state of the art in cognitive and intelligent robotics models, and how this informs the design of future robot applications
5. Discuss the role of ethics and responsible research and innovation in robotics
Syllabus
Lecture topics:
- Introduction to Cognitive Robotics
- Overview of robot technologies, sensors and actuators
- Robot platforms
- Machine learning for robotics
- Developmental Robotics
- Neuro-robotics
- Evolutionary and swarm robotics
- Social robotics and human-robot interaction
- Language learning and speech interfaces
- Robot tutors for children
- Ethics for robotics and AI
Practical Labs:
The practical lab sessions will focus on the use of machine learning methods, such as deep learning, for robot vision and language and on the software tools for robotics.
Teaching and learning methods
Lectures
24 in total, 2 per week
Labs with TA support plus coursework and exam preparation and independent study
Practical skills
Aim and Deliverable
The aim of this coursework is to develop skills on the design, execution and evaluation of deep neural networks experiments for robotics. It also aims at discussing the role of the deep learning approach within the context of the state of the art in robotics. The assignment will in particular address the learning outcome LO1 on the analysis of the methods and software technologies for robotics, and LO3 on applying different machine learning methods for intelligent behaviour.
Your task is to extend the deep learning laboratory exercises (e.g. Multi-Layer Perceptron (MLP) and/or Convolutional Neural Network (CNN) exercises for image datasets) and carry out and analyse new training simulations. This will allow you to evaluate the role of different hyperparameter values and explain and interpret the general pattern of results to optimise the training for robotics (vision) applications. You should also contextualised yoru work within the state of the art, with a discussion of the role of deep learning and its pros and cons for robotics research and applications.
You can use the standard object recognition datasets (e.g. CIFAR, COCO) or robotics vision datasets (e.g. iCub World[1], RGB-D Object Dataset[2])
The deliverable to submit is a report (5 pages including figures/tables) to describe and discuss the training simulations done and their context within robotics research and applications.
Marking Criteria (out of 30)
A clear introductory to the problem and the methodology to be used, with justification of the network topology and hyperparameters chosen [3]
- Contextualisation of deep learning methodologies within the state of the art of deep learning for robotics [8]
- Complexity of the network(s), hyperparameters and dataset [7]
- Description, interpretation and assessment of the results on the hyperparameter testing simulations, including appropriate figures and tables to support the results. [12]
Due Date: 184.00 on 20 April 2020, uploaded to BlackBoard as a PDF. Use standard file name: 34212-Lab-S-Report
Assessment methods
Method | Weight |
---|---|
Written exam | 70% |
Report | 30% |
Feedback methods
Feedback on report and additional oral feedback during office/surgery hours and during labs.
Recommended reading
Cangelosi & Schlesinger (2015). Developmental Robotics: From Babies to Robots. MIT Press
Matari¿, M. J. (2007). The Robotics Primer. MIT Press.
Study hours
Scheduled activity hours | |
---|---|
Assessment written exam | 2 |
Lectures | 24 |
Practical classes & workshops | 8 |
Independent study hours | |
---|---|
Independent study | 66 |
Teaching staff
Staff member | Role |
---|---|
Angelo Cangelosi | Unit coordinator |