rethinklab/Bench2Drive-Speed
收藏Hugging Face2026-03-27 更新2026-04-05 收录
下载链接：
https://hf-mirror.com/datasets/rethinklab/Bench2Drive-Speed
下载链接
链接失效反馈官方服务：
资源简介：
---
license: cc-by-nc-nd-4.0
task_categories:
- robotics
---
# Bench2Drive-Speed
[**Project Page**](https://thinklab-sjtu.github.io/Bench2Drive-Speed/) | [**Paper**](https://huggingface.co/papers/2603.25672) | [**GitHub**](https://github.com/Thinklab-SJTU/Bench2Drive-Speed)
**Bench2Drive-Speed** is a closed-loop benchmark for desired-speed conditioned autonomous driving, enabling explicit control over vehicle behavior through **target speed** and **overtake/follow commands**.
The **CustomizedSpeedDataset** contains **2,100 CARLA driving scenarios** with expert demonstrations and annotated **overtake/follow commands**. The released dataset includes expert target-speed signals only.
## Virtual Target Speed Annotation
You can use a script to generate **virtual target speed** from Bench2Drive-style driving dataset.
1. Find `tools/append_virtual_target_speed.py` in the GitHub repository.
2. Set the path to your dataset in the main function:
```python
# tools/append_virtual_target_speed.py
if __name__ == "__main__":
main(
["/path/to/dataset"] # << modify here
)
```
3. Run the script:
```bash
python tools/append_virtual_target_speed.py
```
## Citation
```bibtex
@article{Bench2DriveSpeed,
title={Can Users Specify Driving Speed? Bench2Drive-Speed: Benchmark and Baselines for Desired-Speed Conditioned Autonomous Driving},
author={Yuqian Shao and Xiaosong Jia and Langechuan Liu and Junchi Yan},
year={2026},
eprint={2603.25672},
archivePrefix={arXiv},
primaryClass={cs.RO},
}
```
提供机构：
rethinklab
