--- license: apache-2.0 ---
# FlowDrive Model Card ## Model Details We propose FlowDrive, a novel framework that intriduces physically interpretable energy-based flow fields to encode semantic priors and safety cues into the BEV space. - **Model type:** An end-to-end autonomous driving model based on the energy-based flow fields. - **License:** Non-commercial license ### Model Sources - **Repository:** https://github.com/AstrixDrive/FlowDrive - **Paper:** https://arxiv.org/abs/2509.14303 ## Uses The primary use of FlowDrive is for the end-to-end autonomous driving. ## Citation Information @article{jiang2025flowdrive, title={FlowDrive: Energy Flow Field for End-to-End Autonomous Driving}, author={Jiang, Hao and Zhang, Zhipeng and Gao, Yu and Sun, Zhigang and Wang, Yiru and Heng, Yuwen and Wang, Shuo and Chai, Jinhao and Chen, Zhuo and Zhao, Hao and others}, journal={arXiv preprint arXiv:2509.14303}, year={2025} }