Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This study addresses the challenge of 3D lane boundary detection in high-speed vehicle racing by introducing a novel dataset comprising over 250,000 images and inertial measurements from a Lexus LC 500. By implementing a multi-camera ensemble approach and leveraging odometry for pre-integration, the authors achieve processing rates of nearly 300Hz while significantly enhancing predictive performance. The results demonstrate a 3-point improvement in F1 scores and a reduction of over 30% in near-vehicle mean absolute errors, achieving F1 scores above 0.9 and lateral MAEs below 0.18m in real-world deployments.
Racing environments push lane detection to its limits, but with the right dataset and methods, we can achieve near-perfect accuracy at high speeds.
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.