3D Lane Detection with Odometry for High-Speed Vehicle Racing
This work addresses the underexplored problem of 3D lane detection for high-speed racing scenarios, providing a dataset and efficient method that enables real-time performance on hardware.
The paper introduces a new dataset for 3D lane detection in high-speed vehicle racing and proposes modifications to achieve processing rates of nearly 300Hz. By incorporating odometry and ensemble predictions, they improve F1 score by 3 points and reduce near-vehicle MAEs by over 30% compared to BevLaneDet, achieving F1 >0.9 and lateral MAEs <0.18m in deployment.
Lane boundary detection is a critical component in autonomous driving systems and has been rigorously studied in regular driving scenarios. However, it is less explored in vehicle racing, where the car moves at higher speeds across more extreme road geometries. To study this problem, we introduce a new dataset for 3D lane detection in racing, featuring >$250$k images from multiple camera feeds and inertial measurements taken with a Lexus LC 500 driving on a closed circuit. With this dataset, we compare various approaches to 3D lane detection and propose modifications that permit frames to be processed at rates of almost 300Hz while retaining high predictive performance in the racing application. This facilitates a multi-camera ensemble approach that is validated on hardware. We show that sensing modalities such as inertial measurements can be leveraged for pre-integration to regress road geometries over both cameras and time, yielding improvements in key metrics. Compared to methods such as BevLaneDet, adding odometry and ensemble predictions improves the F1 score by 3 points and reduces near-vehicle mean absolute errors (MAEs) by $>30 \%$. We show F1 scores $>$0.9 and lateral MAEs of $<$0.18m in vehicle deployments.