2020.5-2021.6 Autonomous System of the Formula Student Driverless Car

Introduction

During my undergraduate study, I joined in BITFSD and spent a lot of time working on formula student driverless car. Thanks to our teammates’ hard work, we won the championship in Formula Student Autonomous China for several consecutive years.

Main Work

1. Proposed a fast, accurate and large-scale perception system of a formula student driverless car, includes objects detection, point clouds segmentation, and point cloud cluster.
2. Applied PP-YOLO to make cones detection, GPF algorithm to segment the ground, and Euclidean cluster to extract the cones.
3. Conducted the data fusion of binocular cameras and LIDAR to effectively reduce the noise, and eliminated the influence of environmental factors including light weather and obtained stable sensory results.

Awarded Excellent Oral Presentation in ICRCA2022

Nice Formula Student Driverless

Tech Stack

• Theory
  • Computer Vision(Object Detection, YOLO)
  • Artificial Intelligence (Cluster)
  • Machine Vision(Calibration, Perception System Design)
• Programming
  • C++
  • Python
  • ROS