NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913568766558208 |
|---|---|
| author | Dauner, Daniel Hallgarten, Marcel Li, Tianyu Weng, Xinshuo Huang, Zhiyu Yang, Zetong Li, Hongyang Gilitschenski, Igor Ivanovic, Boris Pavone, Marco Geiger, Andreas Chitta, Kashyap |
| author_facet | Dauner, Daniel Hallgarten, Marcel Li, Tianyu Weng, Xinshuo Huang, Zhiyu Yang, Zetong Li, Hongyang Gilitschenski, Igor Ivanovic, Boris Pavone, Marco Geiger, Andreas Chitta, Kashyap |
| contents | Benchmarking vision-based driving policies is challenging. On one hand, open-loop evaluation with real data is easy, but these results do not reflect closed-loop performance. On the other, closed-loop evaluation is possible in simulation, but is hard to scale due to its significant computational demands. Further, the simulators available today exhibit a large domain gap to real data. This has resulted in an inability to draw clear conclusions from the rapidly growing body of research on end-to-end autonomous driving. In this paper, we present NAVSIM, a middle ground between these evaluation paradigms, where we use large datasets in combination with a non-reactive simulator to enable large-scale real-world benchmarking. Specifically, we gather simulation-based metrics, such as progress and time to collision, by unrolling bird's eye view abstractions of the test scenes for a short simulation horizon. Our simulation is non-reactive, i.e., the evaluated policy and environment do not influence each other. As we demonstrate empirically, this decoupling allows open-loop metric computation while being better aligned with closed-loop evaluations than traditional displacement errors. NAVSIM enabled a new competition held at CVPR 2024, where 143 teams submitted 463 entries, resulting in several new insights. On a large set of challenging scenarios, we observe that simple methods with moderate compute requirements such as TransFuser can match recent large-scale end-to-end driving architectures such as UniAD. Our modular framework can potentially be extended with new datasets, data curation strategies, and metrics, and will be continually maintained to host future challenges. Our code is available at https://github.com/autonomousvision/navsim. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_15349 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking Dauner, Daniel Hallgarten, Marcel Li, Tianyu Weng, Xinshuo Huang, Zhiyu Yang, Zetong Li, Hongyang Gilitschenski, Igor Ivanovic, Boris Pavone, Marco Geiger, Andreas Chitta, Kashyap Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Robotics Benchmarking vision-based driving policies is challenging. On one hand, open-loop evaluation with real data is easy, but these results do not reflect closed-loop performance. On the other, closed-loop evaluation is possible in simulation, but is hard to scale due to its significant computational demands. Further, the simulators available today exhibit a large domain gap to real data. This has resulted in an inability to draw clear conclusions from the rapidly growing body of research on end-to-end autonomous driving. In this paper, we present NAVSIM, a middle ground between these evaluation paradigms, where we use large datasets in combination with a non-reactive simulator to enable large-scale real-world benchmarking. Specifically, we gather simulation-based metrics, such as progress and time to collision, by unrolling bird's eye view abstractions of the test scenes for a short simulation horizon. Our simulation is non-reactive, i.e., the evaluated policy and environment do not influence each other. As we demonstrate empirically, this decoupling allows open-loop metric computation while being better aligned with closed-loop evaluations than traditional displacement errors. NAVSIM enabled a new competition held at CVPR 2024, where 143 teams submitted 463 entries, resulting in several new insights. On a large set of challenging scenarios, we observe that simple methods with moderate compute requirements such as TransFuser can match recent large-scale end-to-end driving architectures such as UniAD. Our modular framework can potentially be extended with new datasets, data curation strategies, and metrics, and will be continually maintained to host future challenges. Our code is available at https://github.com/autonomousvision/navsim. |
| title | NAVSIM: Data-Driven Non-Reactive Autonomous Vehicle Simulation and Benchmarking |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Robotics |
| url | https://arxiv.org/abs/2406.15349 |