In this paper, we propose PADriver, a novel closed-loop framework for personalized autonomous driving (PAD). Built upon Multi-modal Large Language Model (MLLM), PADriver takes streaming frames and personalized textual prompts as inputs. It autoaggressively performs scene understanding, danger level estimation and action decision. The predicted danger level reflects the risk of the potential action and provides an explicit reference for the final action, which corresponds to the preset personalized prompt. Moreover, we construct a closed-loop benchmark named PAD-Highway based on Highway-Env simulator to comprehensively evaluate the decision performance under traffic rules. The dataset contains 250 hours videos with high-quality annotation to facilitate the development of PAD behavior analysis. Experimental results on the constructed benchmark show that PADriver outperforms state-of-the-art approaches on different evaluation metrics, and enables various driving modes.
View on arXiv@article{kou2025_2505.05240,
title={ PADriver: Towards Personalized Autonomous Driving },
author={ Genghua Kou and Fan Jia and Weixin Mao and Yingfei Liu and Yucheng Zhao and Ziheng Zhang and Osamu Yoshie and Tiancai Wang and Ying Li and Xiangyu Zhang },
journal={arXiv preprint arXiv:2505.05240},
year={ 2025 }
}