Instructor	Justin W. Hart
Class Times	Tuesday & Thursday	3:30-5:00pm
Classroom	Robert Lee Moore (RLM) 5.122
Laboratory	Gates-Dell Complex (GDC) 3.414
Course Syllabus	[PDF]

	Name	Email	Office Hours
Instructor	Justin W. Hart	hart@cs.utexas.edu	Monday	4:00pm-5:00pm
GDC 3.402 or BWI lab			Thursday	5:15pm-6:15pm
Mentors
	David Chen	xiangweichen99@gmail.com	TBD	- - -
	Blake Holman	blake.holman@utexas.edu	TBD	- - -
	Jamin Goo	jgoo@utexas.edu	TBD	- - -
	Jeffrey Huang	jeffreyhuang23@gmail.com	TBD	- - -
	Lucinda Nguyen	lucinda.onguyen@gmail.com	TBD	- - -
	Sydney Owen	seowen@nctv.com	TBD	- - -
	Mayuri Raja	mraja7@utexas.edu	TBD	- - -
	Connor Sheehan	c-she@utexas.edu	TBD	- - -
	Stone Tejeda	stonetejeda@utexas.edu	TBD	- - -

This course focuses on expanding on what was learned in the first semester of FRI by diving more deeply into a directed project chosen jointly by student project teams and the instructor. Students participate as researchers in a real laboratory. As such, the challenges presented differ greatly from those of a typical undergraduate university course. Successful projects will make significant contributions to the laboratory; either by completing novel research or laying the groundwork for longer-term projects, or by developing important infrastructure components which support the laboratory's research efforts.

There are two significant research efforts upon which students are encouraged to focus their contributions. The first, the Building-Wide Intelligence (BWI) Project, aims to develop a fleet of service robots with which people in the Gates-Dell Complex may interact. These robots are intended to carry out useful and fun tasks and to become an integral part of the building's environment. The second is the development of software and studies to support the UT Austin Villa @ Home RoboCup@Home team. RoboCup@Home is a competition in which domestic service robots are tested on a series of challenge tasks. Robots perform a variety of tasks; such as helping people carry groceries into their apartment from a car, or acting on verbal instructions. More details about BWI can be found at http://www.cs.utexas.edu/~larg/bwi_web/. More details about RoboCup@Home can be found at http://www.robocupathome.org/.

Some of the assignments you undertake this semester will require you to use the BWI Lab, in GDC 3.414.

In order to maintain productivity in the lab, please observe the following. Failure to do so will harm the lab's productivity, and repeat offenders will be banned from the lab and receive failing grades on assignments taking place in the lab.

• If you break something, immediately report it to a mentor, the TA, or the instructor. (In that order.)
  • Breakages happen, but we have to handle them quickly as downtime can harm research progress.
  • Students and faculty performing research for scientific publications work on strict deadlines, and problems must be quickly addressed.
• Always leave the robots on chargers when not in use, regardless of whether you were the last to use them.
• Do NOT over-tighten the screws on the robot charging cables. The screws will break, and this will lead to downtime.
  • A red light means that the charger is not connected. Report issues to a mentor, the TA, or the instructor. (In that order.)
  • A flashing green light means that the charger in a special charging cycle. It's fine.
• Do NOT leave computers locked. Some computers may be locked by researchers in the lab. Students should not lock them.
• The computers with the LED cases or connected to robots are for student use (except the Alienware laptop, or Alienware gaming PC).
  • On the first day in lab, a few computers with special purposes will be identified. Do not log into these machines either locally or remotely.
• Do not hold group meetings in the hallways or meeting areas on the AI corridor with no door. Those areas are for the use of AI lab graduate students, faculty, and staff only.
• Do not meet with groups other than your robotics final project group in the lab. That space is reserved for class and research purposes.

The goal of this class is to expose students to real experience researching as a robotics researcher. The challenges in this class are real. Unlike other classes in the undergraduate curriculum, the criteria for success are dictated by the needs of the laboratory and the state of the art of the field. This is a novel challenge for undergraduates, who must put themselves into the mindset of a doctoral student. The instructor and mentors will not know all of the answers to your scientific questions or how to implement all of your ideas. We will not know how easy or hard it is to accomplish what you are trying to do. We will provide support and advice on how to find the answers that you seek, and the result will be actual advancement of scientific knowledge in the fields of robotics, artificial intelligence, and human-robot interaction or the development of necessary infrastructure to support our research projects.
Students will:

• Perform literature surveys, and read about and critically assess current research.
• Perform research talks in a standard conference format.
• Write research reports a standard conference format.
• Develop real research infrastructure and/or perform real research.

There is no textbook for this course. Instead, students will perform a literature survey; finding papers relevant to their research project. Grading of these surveys will follow the criteria that peer reviewers use when performing the scientific peer review process. These papers should explore prior work on the problem that the student team is investigating, demonstrate an understanding of the problem or phenomenon, and support the investigation that the students perform. Much like the literature survey in an archived scientific paper, a successful literature survey both demonstrates that the students understand the current state of the art and supports the hypothesis tested or approach taken in their research. Students may also be assigned readings as part of class exercises, and will be expected to read the paper and possibly compose a written response to be emailed to the instructor prior to in-class discussion.

This class will incorporate a mixture of classroom instruction and laboratory practice. Class sessions will be held in RLM 5.122. The laboratory can be found at GDC 3.414. The first class will be held in RLM 5.122. On days when class is not held in RLM 5.122, students are expected to come to the laboratory in GDC 3.414. An up-to-date schedule of class locations and dates can be found on the course website.

During laboratory days, student teams will have scheduled meeting slots with the instructor in which progress will be evaluated and general guidance will be provided. These meetings will take a format similar to the regular weekly meetings that a PhD student would have with their doctoral advisor.

Participation in either FRI I or the Summer ARI program is a prerequisite for this course. Students are also expected to be able to work independently and in teams on projects utilizing ROS in either the C++ or Python programming languages. When choosing course projects, students should carefully consider the prerequisite or corequisite knowledge necessary for their success, as this will vary from project to project.

Early in the course, project ideas will be offered by the instructor. These project ideas are intended to seed discussion and may either become the project that a student team pursues or the inspiration for their project. Student teams may also come up with ideas all their own. Successful projects will balance ambition with pragmatism and make a real contribution to the lab. Projects will be mutually agreed upon by student teams and the instructor, and the instructor will provide guidance in the development of these ideas.

Successful projects will meet a handful of criteria:

• The research is of interest to the lab, and uses methods common to other research in the group.
• The topic is something that the instructor understands well enough to supervise you through.
• The project has clear objectives, and not just a rough idea of the area.
• The project can be well-situated in the literature, and the unique contribution to science is well-understood.

Grades will be based on:
Class participation and attendance 15% Final Project 85%
The final project will comprise the following components, taken as a whole:

• Difficulty & Effort
• Research Value
• Successful Completion
• Workshop Presentation
• Workshop Paper
• Conference Presentation
• Conference Paper

Plus and minus grades will be used in final grading of the course.

Final project reports will be due on Monday, December 9 by 11:59pm.
Final project presentations will be during the final exam slot on Tuesday, December 17 from 2:00pm-5:00pm.

Schedule subject to change due to pace of class, see website for updates.

Date	Title	Slides
08/29/19	First Class! Kickoff with Peter Stone. Discussion with Peer Mentors	[PDF]
09/03/19	Team Formation & Project Brainstorming	[PDF]
09/05/19	Debug Project Proposals	[PDF]
	What does a good research presentation look like?	[PDF]
	What does a good research report look like?
09/09/19	Project Proposals by 11:59pm
09/10/19	Project Proposal Presentations	Your Slides
10/29/19	FAIR SW WIP Papers by 11:59pm	No Slides
11/04/19	FAIR SW Peer-reviews by 11:59pm	No Slides
11/05/19	FAIR SW PC Meeting - In Class	No Slides
11/07/19	FAIR SW - In Class	Your Slides
12/09/19	FAIR Paper Submission by 11:59pm	No Slides
12/17/19	Final project presentations 2:00pm - 5:00pm	Your Slides

As this is a research course, it is important to use the many tools at your disposal to achieve your research goals. Students will work in groups in this course, and are expected to collaborate with their teams and outside of their immediate teams in order to achieve the best results possible. When you leverage someone else's work, cite them. Citations are the currency of the scientific community. Use third-party software, but make sure to honor licenses and cite the authors. In this course, you will be graded on what you accomplish above and beyond what is already freely available. If this means implementing an algorithm, state which parts were your original work or implementation in your progress reports, and which parts were downloaded or were someone else's ideas. In this class, leveraging such resources is encouraged. It makes code easier to maintain and update, and encourages potential collaborations with other institutions. Invest your efforts in making novel discoveries or implementing functionality beyond what is freely available. Do, however, abide by Computer Science Department's Academic Honesty Policy, which can be found at http://www.cs.utexas.edu/users/ear/CodeOfConduct.html

The University of Texas at Austin provides upon request appropriate academic accommodations for qualified students with disabilities. To determine if you qualify, please contact the Dean of Students at 471-6529; 471-4641 TTY. If they certify your needs, I will work with you to make appropriate arrangements. Further information can be found at http://www.utexas.edu/diversity/ddce/ssd/.

A student who misses an examination, work assignment, or other project due to the observance of a religious holy day will be given an opportunity to complete the work missed within a reasonable time after the absence, provided that he or she has properly notified the instructor. It is the policy of the University of Texas at Austin that the student must notify the instructor at least fourteen days prior to the classes scheduled on dates he or she will be absent to observe a religious holy day. For religious holy days that fall within the first two weeks of the semester, the notice should be given on the first day of the semester. The student will not be penalized for these excused absences, but the instructor may appropriately respond if the student fails to complete satisfactorily the missed assignment or examination within a reasonable time after the excused absence.

This course, as presented by Justin Hart in conjunction with Peter Stone. It is part of the University of Texas at Austin, Texas Institute for Discovery Education in Science, Freshman Resaerch Initiative Program. The course was originally conceived by Peter Stone, and this material is an evolution of material developed by Jivko Sinapov, who succeeded Matteo Leonetti. It is influenced by Brian Scassellati's CS 473b: Intelligent Robotics course at Yale University.