DriveRX: A Vision-Language Reasoning Model for Cross-Task Autonomous Driving
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911370626203648 |
|---|---|
| author | Diao, Muxi Yang, Lele Yin, Hongbo Wang, Zhexu Wang, Yejie Tian, Daxin Liang, Kongming Ma, Zhanyu |
| author_facet | Diao, Muxi Yang, Lele Yin, Hongbo Wang, Zhexu Wang, Yejie Tian, Daxin Liang, Kongming Ma, Zhanyu |
| contents | Effective autonomous driving hinges on robust reasoning across perception, prediction, planning, and behavior. However, conventional end-to-end models fail to generalize in complex scenarios due to the lack of structured reasoning. While recent vision-language models (VLMs) have been applied to driving tasks, they typically rely on isolated modules and static supervision, limiting their ability to support multi-stage decision-making. We present AutoDriveRL, a unified training framework that formulates autonomous driving as a structured reasoning process over four core tasks. Each task is independently modeled as a vision-language QA problem and optimized using task-specific reward models, enabling fine-grained reinforcement signals at different reasoning stages. Within this framework, we train DriveRX, a cross-task reasoning VLM designed for multi-stage decision-making. DriveRX achieves strong performance on the public benchmark, outperforming GPT-4o in behavior reasoning and demonstrating robustness under complex or corrupted driving conditions. DriveRX serves as a high-level semantic reasoning backbone, producing structured stage-wise reasoning chains that enhance decision consistency. These outputs also provide high-quality supervisory signals for annotation and downstream planning/control models. We release the AutoDriveRL framework and DriveRX to support future research. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_20665 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DriveRX: A Vision-Language Reasoning Model for Cross-Task Autonomous Driving Diao, Muxi Yang, Lele Yin, Hongbo Wang, Zhexu Wang, Yejie Tian, Daxin Liang, Kongming Ma, Zhanyu Computer Vision and Pattern Recognition Effective autonomous driving hinges on robust reasoning across perception, prediction, planning, and behavior. However, conventional end-to-end models fail to generalize in complex scenarios due to the lack of structured reasoning. While recent vision-language models (VLMs) have been applied to driving tasks, they typically rely on isolated modules and static supervision, limiting their ability to support multi-stage decision-making. We present AutoDriveRL, a unified training framework that formulates autonomous driving as a structured reasoning process over four core tasks. Each task is independently modeled as a vision-language QA problem and optimized using task-specific reward models, enabling fine-grained reinforcement signals at different reasoning stages. Within this framework, we train DriveRX, a cross-task reasoning VLM designed for multi-stage decision-making. DriveRX achieves strong performance on the public benchmark, outperforming GPT-4o in behavior reasoning and demonstrating robustness under complex or corrupted driving conditions. DriveRX serves as a high-level semantic reasoning backbone, producing structured stage-wise reasoning chains that enhance decision consistency. These outputs also provide high-quality supervisory signals for annotation and downstream planning/control models. We release the AutoDriveRL framework and DriveRX to support future research. |
| title | DriveRX: A Vision-Language Reasoning Model for Cross-Task Autonomous Driving |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.20665 |