Planning safe and efficient motions for robots in complex environments, often shared with humans and other robots, is a difficult problem combining discrete and continuous mathematics, as well as probabilistic, game-theoretic, and learning aspects. This course will cover the algorithmic foundations of motion planning, with an eye to real-world implementation issues.
The students will learn how to design and implement state-of-the-art algorithms for planning the motion of robots executing challenging tasks in complex environments.
Discrete planning, shortest path problems. Planning under uncertainty. Game-theoretic planning. Geometric Representations. Configuration space. Grids, lattices, visibility graphs. Sampling-based methods. Potential and Navigation functions. Mathematical Programming. Local and global optimization, convex relaxations. Planning with limited information. Multi-agent Planning.
Course notes and other education material will be provided for free in an electronic form.
There is no required textbook, but an excellent reference is Steve Lavalle's book on "Planning Algorithms."
Prerequisites / Notice
Students should have taken basic courses in optimization, control systems, probability theory, and should be familiar with basic programming (e.g., Python, and/or C/C++). Previous exposure to robotic systems is a definite advantage.
Domain A - Subject-specific Competencies
Concepts and Theories
Techniques and Technologies
Performance assessment information (valid until the course unit is held again)
The performance assessment is offered every session. Repetition possible without re-enrolling for the course unit.
Mode of examination
written 150 minutes
Additional information on mode of examination
There is a written final exam during the examination session, which covers all material taught during the course, i.e. the material presented during the lectures and corresponding problem sets, programming exercises, and recitations. Additionally, there will be programming assignments, which are an optional learning task during the semester, requiring the students to understand and apply the lecture material. These contribute a maximum of 0.25 grade points to the final grade, but only if it helps to improve the final grade.
One sheet of A4 paper, front and back. Only handwritten material by the individual student is allowed --- no computer printouts or photocopies. (Preparing such a sheet would be an important part of the learning process.)
This information can be updated until the beginning of the semester; information on the examination timetable is binding.
No public learning materials available.
Only public learning materials are listed.
No information on groups available.
There are no additional restrictions for the registration.