Autonomous motorsports aim to replicate the human racecar driver with software and sensors. As in traditional motorsports, Autonomous Racing Ve- hicles (ARVs) are pushed to their handling limits in multi-agent scenarios at extremely high (≥ 150mph) speeds. This Operational Design Domain (ODD) presents unique challenges across the autonomy stack. Standard autonomous vehicles (AVs) rely on High Definition (HD) Maps, that provide information such as lanes, right of way information, and more, to simplify many problems. For ARVs, an HD map includes information about the race track, including its boundaries and banking, which the autonomy software can exploit to improve performance. In this project, we created a pipeline to take in noisy sensor LiDAR data and GPS information in order to create an HD map for an ARV, automatically building a globally registered pointcloud and extracting the relevant information.
Poster and paper shown below: