Visual Navigation for Autonomous Vehicles (VNAV) covers theoretical foundations of vision-based navigation as well as implementation and testing of advanced algorithms in a photo-realistic Unity-based simulator. Lectures will explore fundamental tools and results from a wide spectrum of disciplines (optimization, estimation, geometry, probabilistic inference) that underlie modern techniques for real-time robot perception and 3D computer vision (including visual-inertial navigation and SLAM), control and trajectory optimization, and machine learning. Implementation and testing will be based on C++ and ROS, the Robot Operating System. Students will be given access to an advanced drone simulator and will be able to implement and test state-of-the-art algorithms and learn about the bleeding edge of autonomous navigation. The final portion of the class includes an individual or team-based project that has the goal of advancing the state of the art in vision-based navigation, according to students’ interests.