MIT 16.S398 and 16.485: Visual Navigation for Autonomous Vehicles (VNAV)

Class materials will be made available on MIT OpenCourseWare.

I co-created and taught MIT 16.485 (VNAV: Visual Navigation for Autonomous Vehicles) in 2018 and 2019. VNAV is a new advanced subject (for both graduate and senior undergraduate students) on vision-based navigation, perception, and state estimation. It rigorously covers the theoretical foundations of vision-based navigation (e.g., geometric control, 3D vision, visual-inertial navigation, place recognition, SLAM, optimization on matrix Lie groups), while also providing students with hands-on experience using real aerial and ground robots. The course culminates in a critical review of recent advances in the field and a team project aimed at advancing the state-of-the-art.

VNAV 2018 (last day)

VNAV 2019 (last day)

Subject Evaluation (2018 and 2019):

  • Subject rating (median): 7/7
  • Rating as lecturer (median): 7/7

“This is how a robotics course should be structured. Math fundamentals in lecture, interesting lab to demand understanding and show utility. Well done, especially considering it was the first time through.”

“Kasra makes lectures fun and interesting, even when the content is quite dense mathematics. Takes the effort to ensure lectures are engaging by welcoming and effectively answering questions and giving fun examples. Very approachable for questions regarding the course and projects.”

More reviews …

UTS 49274 Advanced Robotics and 49329 Control of Mechatronic Systems

I was the tutor (teaching assistant) for postgraduate subjects Advanced Robotics and Control Mechatronic Systems from 2012 to 2014. I led the recitation lectures (covering topics from linear algebra, estimation theory, optimal control, and probabilistic robotics) and lab sessions. I also played an active role in redesigning new problem sets and projects and updating the curriculum.