Building safer autonomous vehicles

Safety is important, yet still is a key challenge for autonomous driving systems, especially when extreme conditions are involved. In this study, you will design and implement a data-driven safe controller on a 1/10th scale F1 vehicle equipped with a depth camera, laser range scanner, inertial/wheel odometry, and a GPU enabled Nvidia Jetson Xavier NX. The research lies in the intersection of machine learning, control theory, and robotics.

Preferred skills: Numerical Python (numpy, scipy), experience with deep learning frameworks (e.g PyTorch), and Robot Operating System (good to have but not strictly necessary).