This course rigorously describes state estimation, perception, and planning methods for mobile robot systems. Topics include Bayesian filtering, stochastic representations of the environment, motion and sensor models, algorithms for Simultaneous Localization and Mapping (SLAM). The course will illustrate these topics with applications to autonomous and human-on-the-loop robotic systems, including marine, ground, and air vehicles.
Upon completion of the course, successful students will have a principled and nuanced understanding of state-of-the-art methods. Students will develop new intuitions and a deeper mastery of probability and linear algebra.
Students will be evaluated on a combination of team and individual efforts.
- Team efforts include three problem sets, each with a substantial programming exercise. Team members will turn in peer evaluations.
- Students will complete a substantial final project in a team of their choice.
- Individual efforts include two quizzes and two midterm exams. The quizzes are intended, in part, to help study for the more heavily-weighted midterms.
|Problem Sets||30 %|
|Final Project||30 %|
|Submit end-of-term course evaluation||1 %|
Quizzes are given approximately one week before each midterm. Note that the quizzes are not weighted heavily: they are intended to help students "debug" their studying for the midterms.
The final project has a number of sub-components:
The showcase score will reflect the ability of your team to demonstrate and educate those who come by your booth. This is a combination of technical presentation skills, as well as the performance of a demo of your system.
The "artifact" is a lasting documentation of your project, however the medium of communication is not specified except that it must be self-contained and sharable via a web site. It could be a movie, a document, etc. This artifact should communicate what you learned (and to be informative to the consumer) in addition to describing your specific approach and results.
The problem sets (along with their programming exercises) are collaborative activities. Each team must turn in work that is wholly their own: teams are encouraged to discuss problems, strategies, ideas, algorithms, etc. with other teams, but their write-ups (including software) must be done independently.
Members of a team are required to work together on the problems; dividing up the problems and working separately is not acceptable. This will require at least one in-person meeting. On every problem set, each team will certify when their team met and who attended, noting any exceptions. With each problem's solution being the product of the group, all members will be held accountable for violations of the honor code. Students must sign the statement below, noting any qualifications necessary to render the statement true:
"I participated and contributed to team discussions on each problem, and I attest to the integrity of each solution. Our team met as a group on [DATE(S)]."
An example of a reasonable qualification might be: "Jimmy was out of town when we met, but Amy and Jimmy were able to meet later on. Jimmy's ideas were emailed to us and are reflected in our final solutions." We expect exceptions to be rare, but we understand that life can be complicated! The certification should be physically signed by each team member, and a photo attached to the submission.
Note that every student on a team will receive the same grade on that project. Students will evaluate their teammates; a student's evaluations may result in grade adjustments. As part of the college's honor policy, you have a responsibility to report students that have violated the honor code by (for example) using inappropriate outside resources, failing to contribute to a group assignment equitably and not documenting this on their certification, or cheating during an exam. Your peer evaluations provide a mechanism for this reporting (and a staff member will follow up if the situation warrants), but you are also welcome to contact a staff member directly.
It is not acceptable to use code or solutions from outside class (including those found online) unless the resources are specifically suggested by the problem set. This includes (but is not limited to) previous years' materials (regardless of whether it originated from staff, students, etc.), the textbook's solution manual, etc.
Assignment Turn-in and Lateness Policy
Problem sets are due at 11:59p on their corresponding due date and should be submitted electronically by email (including a photo of the signature page). Late problem sets will be penalized 10% for each day (or fraction of) that they are late; after three days, no credit will be given.
All team members are responsible for the assignment being turned on promptly (that's why team members get cc'd on the problem sets!) Play it safe--- guard against email typos, miscommunications, etc., and submit early.
Assignment/Exam Regrade Requests
Any requests to have an assignment or exam regraded should be done in writing. Attach a cover page describing what grading error you think was made and submit it to the course staff.
Regrade requests must be submitted no later than one week after the graded work is returned to the class.
Students who miss an exam for approved reasons will be given a make-up oral exam by the staff. Approval for anticipate-able absences should be sought at least two weeks before the scheduled exam. All other absences will result in a zero score.