Saved in:
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2602.21904 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915816091418624 |
|---|---|
| author | Baidachna, Mariia Carty, James Ferguson, Aidan Agrane, Joseph Kulkarni, Varad Agub, Aubrey Baxendale, Michael David, Aaron Horton, Rachel Atkinson, Elliott |
| author_facet | Baidachna, Mariia Carty, James Ferguson, Aidan Agrane, Joseph Kulkarni, Varad Agub, Aubrey Baxendale, Michael David, Aaron Horton, Rachel Atkinson, Elliott |
| contents | Accurate cone localization in 3D space is essential in autonomous racing for precise navigation around the track. Approaches that rely on traditional computer vision algorithms are sensitive to environmental variations, and neural networks are often trained on limited data and are infeasible to run in real time. We present a UNet-based neural network for keypoint detection on cones, leveraging the largest custom-labeled dataset we have assembled. Our approach enables accurate cone position estimation and the potential for color prediction. Our model achieves substantial improvements in keypoint accuracy over conventional methods. Furthermore, we leverage our predicted keypoints in the perception pipeline and evaluate the end-to-end autonomous system. Our results show high-quality performance across all metrics, highlighting the effectiveness of this approach and its potential for adoption in competitive autonomous racing systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_21904 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | UNet-Based Keypoint Regression for 3D Cone Localization in Autonomous Racing Baidachna, Mariia Carty, James Ferguson, Aidan Agrane, Joseph Kulkarni, Varad Agub, Aubrey Baxendale, Michael David, Aaron Horton, Rachel Atkinson, Elliott Computer Vision and Pattern Recognition Robotics Accurate cone localization in 3D space is essential in autonomous racing for precise navigation around the track. Approaches that rely on traditional computer vision algorithms are sensitive to environmental variations, and neural networks are often trained on limited data and are infeasible to run in real time. We present a UNet-based neural network for keypoint detection on cones, leveraging the largest custom-labeled dataset we have assembled. Our approach enables accurate cone position estimation and the potential for color prediction. Our model achieves substantial improvements in keypoint accuracy over conventional methods. Furthermore, we leverage our predicted keypoints in the perception pipeline and evaluate the end-to-end autonomous system. Our results show high-quality performance across all metrics, highlighting the effectiveness of this approach and its potential for adoption in competitive autonomous racing systems. |
| title | UNet-Based Keypoint Regression for 3D Cone Localization in Autonomous Racing |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2602.21904 |