Virginia Tech, Spring 2022
Tuesdays and Thursdays, 3:30-4:45pm
McBryde Hall, Room 318
ME4824 Class Average: 83.5
ME5824 + CS5844 Class Average: 88.3
As robots and artificial intelligence emerge, it is critical to ensure that these agents can co-exist and collaborate alongside humans. This course surveys recent literature to formalize interaction between robots and humans. We will cover diverse approaches to human-robot interaction, including learning from demonstration, Bayesian inference, intent detection, safe and optimal control, assistive autonomy, and user study design. Students will review and present existing literature, and conduct a research project.
ME 4524 (for ME 4824 students). Knowledge of probability and multivariable calculus is expected. Assignments will involve programming.
Learning ObjectivesHaving successfully completed this course, the student will be able to:
- Articulate the challenges of developing algorithms for human-robot interaction.
- Appraise and implement different methods to learn from human data.
- Use Bayesian inference to detect human intent and autonomously assist humans.
- Apply safe and optimal controllers so that robots can operate around humans.
- Combine human and robot control inputs for assistive autonomy tasks.
- Plan, conduct, and analyze a user study that involves human-robot interaction.
- Assess the scientific merits and weaknesses of research publications.
- Carry out a research project that involves human-robot interaction.
- Communicate scientific content and research to a peer audience.
Prof. Dylan Losey
Final Project: 40%
|Jan. 20||Decision Making||Artificial Intelligence 17.1 - 17.4||HW0|
|Jan. 25||Decision Making||Artificial Intelligence 17.1 - 17.4|
|Jan. 27||Probability||Artificial Intelligence 13.1 - 13.7||HW1|
|Feb. 1||Probability||Artificial Intelligence 13.1 - 13.7|
|Feb. 3||Reward Learning||Link. A unifying formalism for reward learning.||HW2|
|Feb. 8||Reward Learning||Link. Learning rewards from demonstrations and preferences.|
|Feb. 10||Inverse Reinforcement Learning||Link. Learning robot objectives online from human corrections.||HW3|
|Feb. 15||Inverse Reinforcement Learning||Link. Adversarial inverse reinforcement learning.|
|Feb. 17||Imitation Learning||
Keyframes for kinesthetic teaching.
Link. Dataset aggregation for interactive imitation learning.
|Feb. 22||Shared Autonomy||Link. Policy blending for shared autonomy.|
|Feb. 24||Shared Autonomy||Students Present,
|March 1||Physical Human-Robot Interaction||Link. Framework for physical interaction.|
|March 3||Physical Human-Robot Interaction||Students Present,
|March 15||Safety and Planning||Link. Movement primitives via optimization.|
|March 17||Safety and Planning||Students Present|
|March 22||User Study Design||Link. Effects of human-aware motion planning.|
|March 24||User Study Design||
Fluency in human-robot interaction.
Link. Likert scales in human-robot interaction studies.
|March 29||Human Models||Link. Theory of mind.|
|March 31||Human Models||Students Present|
|April 5||Legibility||Link. Legibility and predictability of robot motion.||Project Milestone|
|April 7||Legibility||Students Present|
|April 12||Trust and Influence||Link. Autonomous cars that leverage effects on human actions.|
|April 14||Trust and Influence||Students Present|
|April 19||Inclusive Learning||Link. Extrapolating beyond suboptimal demonstrations.|
|April 21||Inclusive Learning||Students Present|
|April 26||Multiagent Teams||Guest Lecture: Minae Kwon|
|April 28||Multiagent Teams||Link. Robots in Groups and Teams.|
|May 3||Final Project Presentations||Students Present|
|May 10||Final Report|
All students are invited to attend! You don't need to have a particular question — you’re welcome to just stop by for 5 minutes and introduce yourself, or talk about a recent lecture. I also encourage students to use the Discussion feature on Canvas, which I regularly monitor. I do not respond to emails requesting for help on an assignment.
Students will form teams to work on the homeworks, paper presentation, and final project. Students in ME4824 will form teams of four. Students in ME5824 or CS5844 will form teams of two. When possible, I encourage students to team up with peers that have similar areas of interest. You will form your team after the first lecture and maintain the same team throughout the semester.
To ensure that all students are receiving fair credit, and to enable students to provide feedback about their team, we will have team surveys throughout the semester. These surveys will occur on Canvas. If you are having an issue within your team, you should first meet with your team and attempt to resolve the problem. If the issue persists, please notify me and we can look for an appropriate solution.
There are five assignments to reinforce the fundamental concepts from lecture. You should complete each assignment with your team and submit a single copy for grading — clearly mark your name and your teammate name(s) on the assignment. I will drop your lowest homework grade at the end of the semester. Late assignments are not accepted, except when the team has an illness, emergency, or other pressing issue. If you need to ask for an extension due to one of these reasons do not hesitate to contact me; however, you must make your request before the homework deadline. All assignments submitted are considered graded work and are subject to the Honor Code.
Each team of students will select and present one research paper from the syllabus. Your presentation should be 25 minutes long, including time for questions and discussion. Your objective is to teach your classmates about the paper. You will be graded based on:
- Your demonstrated understanding of the technical content.
- Your ability to answer questions from the instructor and the class.
- How well you relate the paper to other papers and lecture material.
- The clarity, structure, and timing of your presentation.
- How well you lead a class discussion at the end of your presentation.
Each team of students will perform a research project. The objective for this project is to i) dive into a particular concept from lecture and ii) apply that concept to something that is of interest to you. The project is broken up into four stages.
Project Proposal (5%): 1 page + references. Your report should motivate the problem, briefly describe how the state-of-the-art tackles this problem, and clearly identify how your solution will advance the state-of-the-art. Include a timeline.
Project Milestone (5%): 2 pages. Describe your team's progress so far, and include preliminary results. Discuss any changes you have made to your project goals; provide an updated timeline. You should stop by office hours to go over your milestone report.
Project Presentation (8%): Present your findings in a 5 minute talk. Explain your problem motivation and solution. If possible, I encourage you to include a demo. Your presentation must articulate how your project applies and extends the concepts from lecture. All team members are expected to speak and equally participate in the presentation.
Final Report (22%): 6 pages + references. Divide your report into the following sections: Introduction, Related Work, Problem Statement, Methods, Experiments, Conclusion, and References. Include figures to visualize your method and results. You will be graded based on both the technical content of your report and the clarity of your writing.
This is a discussion-based class. All students are expected to ask and answer questions, particularly on days when we have paper presentations. I will not give participation points for just attending class regularly — if you do not join the discussion you will receive a zero. To earn full credit, I expect you to provide one insightful question or comment in at least 15 lectures throughout the semester.
Services for Students with Disabilities
Every student in this course should have an equal opportunity to succeed. If you anticipate or experience academic barriers that may be due to disability, including but not limited to ADHD, chronic or temporary medical conditions, deaf or hard of hearing, learning disability, mental health, or vision impairment, please contact the Services for Students with Disabilities (SSD) office (540-231-3788, email@example.com, or visit www.ssd.vt.edu). If you have an SSD accommodation letter, please email me as soon as possible so that I can accommodate your needs.
The Honor Code pledge that each member of the community agrees to abide by states:
“As a Hokie, I will conduct myself with honor and integrity at all times. I will not lie, cheat, or steal, nor will I accept the actions of those who do.”
Students enrolled in this course are responsible for abiding by the Honor Code. A student who has doubts about how the Honor Code applies to any assignment is responsible for obtaining specific guidance from the course instructor before submitting the assignment for evaluation. Ignorance of the rules does not exclude any member of the University community from the requirements and expectations of the Honor Code. Academic integrity expectations are the same for online classes as they are for in person classes. All university policies and procedures apply in any Virginia Tech academic environment. For additional information about the Honor Code, please visit: https://www.honorsystem.vt.edu/
Honor Code Pledge
The Virginia Tech honor code pledge for assignments is as follows:
``I have neither given nor received unauthorized assistance on this assignment.''
The pledge is to be written out on all graded assignments at the university and signed by the student. The honor pledge represents both an expression of the student's support of the honor code and a commitment to uphold the academic standards at Virginia Tech.
If you have questions or are unclear about what constitutes academic misconduct on an assignment or exam, please speak with me. The normal sanction I will recommend for a violation of the Honor Code is an F* sanction as your final course grade. The F represents failure in the course, and * identifies a student who has failed to uphold the values of academic integrity at Virginia Tech.