| Category | Difficulty |
|---|---|
| HW | 4 |
| Project | 4 |
This course introduces one of the core problem in robotics, simultaneous localization and mapping (SLAM). Throughout the course, you will learn various techniques to solve the localization and mapping problem. Kalman Filter, Extended Kalman Filter, Particle Filter, Occupancy Maps, Sparse and Non-linear Least Squares techniques, PTAM, DTAM, Bayes Tree, ORB SLAM, etc. are the major topics covered in this course. Having a good understanding of probabilities and statistics related concepts would be helpful for this course. The course is not difficult and is a good course to get familiar with the state-of-the-art SLAM based research.
- HW: Programming assignments. Except for HW1 other HW are easy. If you get stuck you can ask during OH or on Piazza.
- Project: A group based project which needs to build on the topics learned in the course.
- Attend and understand lectures
- Start HW early
- Get the concepts and doubts clarified by the professor and course TAs
- Make use of Office Hours to help with the HW
- Follow discussions on Piazza
- Assignment 1 and 4 are time consuming, especially the 1st assignment will require you to fine-tune the hyperparameters which is not trivial.
- Start the project early as you may encounter unexpected issues. Run through your project ideas with the professor and course TAs.
- Probabilistic Robotics Text Book by Dieter Fox, Sebastian Thrun, and Wolfram Burgard
- Factor Graphs for Robot Perception