Human-Robot Interaction
Module aims
Human-Robot Interaction is an exciting new field at the intersection of robotics and human-computer interaction. With the increasing presence of robots in our daily lives, it is important to understand both how robots impact humans, and how we can design robots to best suit people's needs. The module first motivates the importance of the field and presents relevant theoretical foundations. The module covers topics such as user-centric design, user study design, data analysis, and verbal and non-verbal robot behavior. Additionally, the module will cover several human-robot interaction applications such as healthcare, education, and in-home robots.
Learning outcomes
1. Rationalise about the different perceptions people have of humans, computers and robots.
2. Design a Human-Robot Interaction Study, control conditions and correct validation schemes.
3. Conduct Data analysis of interactions between people and robots.
4. Conduct end-user focused design.
5. Evaluate the impact of social robots on different sectors (such as education and healthcare), and consider their ethical implications.
6. Create action selection and planning algorithms, where a robot considers the best action given the current user.
7. Read a research paper, extract its core points, and clearly present it to others.
Module syllabus
Theoretical Aspects of HRI: how robots are similar and different to people and computers, trust in HRI, Theory of Mind, etc
Design in HRI: user-centric design, user-study design and analysis, robot behavior design, etc
Computational aspects of HRI: human-robot collaboration, planning under uncertainty, imitation learning, etc
Application in HRI: education, healthcare, home robots, assistive robotics, etc
Teaching methods
The module consists of 7 weeks of timetabled sessions. Most weeks consist of two 2-hour lectures. Each lecture involves a presentation of the day's topic followed by several student 10-15 minute paper presentations on the topic (which will be individual or in pairs depending on the number of students in the class). Some classes will involve formative in-class discussion on future directions in the area or on ethical implications. Additionally some sessions will have formative in-class practice exercises. The class is designed to be interactive with active student participation.
An online service will be used as a discussion forum for the module.
Assessments
70% of the grade will be from the final exam. The remaining 30% will come from (a) student presentations on the topic and (b) small-group work, which will involve programming a simulated robot to interact with users and analysing resulting dynamics.
There will be several in-class exercises during which students will be given feedback. I will also meet with each student group at least once to provide feedback on their project. Many aspects of the project will also be covered on the exam, therefore the student will have some clear feedback before the final exam. When marking student projects and the exam, a marking scheme will be used, and student's will be able to clearly see where they have lost marks.