Fast and Accurate: The Perception System of a Formula Student Driverless Car

Published in ICRCA, 2022

Proposed a perception system that fused point clouds and two monocular cameras frames to detect the cones of FSAC tracks quickly and accurately.
Developed a vision-based boundary regression and drivable lane segmentation algorithm.
Build and realesed FSACOCO dataset which is the first open-sourse dataset for FSAC event in China.