Awesome-VLA4AD

KO | EN

gitlite — search

#javascript #python #hacktoberfest #typescript #llm #go #golang #machine-learning #linux #java #pytorch #security #mcp #open-source

(★ 614)

Vision–Language–Action models for Autonomous Driving (VLA4AD) resources, serving as the companion repository to the survey paper “A Survey on Vision–Language–Action Models for Autonomous Driving”.

파일 탐색기

최종 버전 다운로드 (.zip)

- ADRIVERI.pdf
- AutoVLA.pdf
- Overview_VLA4AD_New.png
- Paradigm_Comparison.png
- Progress_VLA4AD.png
- table1.png
- table1VLA4AD.png
A_Survey_on_VLA4AD.pdf
LICENSE.txt
README.md

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

<div align="center">
  <h1><b> Awesome Vision–Language–Action Models for Autonomous Driving 🚗 </b></h1>
  </div>
  
<div align="center">
  
[![arXiv](https://img.shields.io/badge/arXiv-2506.24044-B31B1B.svg?style=flat&logo=arxiv&logoColor=white)](https://arxiv.org/abs/2506.24044) [![GitHub stars](https://img.shields.io/github/stars/JohnsonJiang1996/Awesome-VLA4AD?style=social)](https://github.com/JohnsonJiang1996/Awesome-VLA4AD/stargazers) [![GitHub forks](https://img.shields.io/github/forks/JohnsonJiang1996/Awesome-VLA4AD?style=social)](https://github.com/JohnsonJiang1996/Awesome-VLA4AD/network)
<a href="https://github.com/JohnsonJiang1996/Awesome-VLA4AD/issues"><img src="https://img.shields.io/github/issues/JohnsonJiang1996/Awesome-VLA4AD" alt="Issues Badge"/></a>
<a href="https://github.com/JohnsonJiang1996/Awesome-VLA4AD/blob/main/LICENSE.txt"><img src="https://img.shields.io/github/license/JohnsonJiang1996/Awesome-VLA4AD" alt="License Badge"/></a>
</div>

**[<a href="https://drive.google.com/file/d/1JZz7Y8qfwwP9NBcLhs-tyWCbr7Uk-AJP/view?usp=sharing">Survey Slides</a>]**
**[<a href="https://mp.weixin.qq.com/s/VTMYuAAFYNHX2t8lamcw3Q">自动驾驶之心解读</a>]**
**[<a href="https://mp.weixin.qq.com/s/DwSDgrNNX_EnsuJPEW6xoQ">具身进化中文解读</a>]**
**[<a href="https://mp.weixin.qq.com/s/pq_PHr6GJu3BvzoKRDfLxw">焉知汽车中文解读</a>]**
**[<a href="https://zhuanlan.zhihu.com/p/1923912633192523592">知乎中文解读</a>]**

</div>

Welcome to **Awesome VLA4AD**—a curated, continuously updated collection of research papers and resources on Vision–Language–Action models for Autonomous Driving (VLA4AD). This repository tracks the latest advances in VLA4AD, from explanatory perception modules to end-to-end reasoning and control architectures.

Our latest survey is [here](https://arxiv.org/abs/2506.24044). We invite your feedback and discussion!

⭐️ **Follow & Star** to stay up to date!  
🤝 **Contributions welcome**—if you know of new papers, datasets, or tools, please open an issue or submit a PR.  
📬 **Questions or suggestions?** Reach us at **sicong.jiang@mail.mcgill.ca** or **qka23@mails.tsinghua.edu.cn**.

---

## 📜 Citation

If this project is useful in your work, we'd appreciate a star 🌟 and a citation of our survey.
```
@article{jiang2025survey,
  title={A Survey on Vision-Language-Action Models for Autonomous Driving},
  author={Jiang, Sicong and Huang, Zilin and Qian, Kangan and Luo, Ziang and Zhu, Tianze and Zhong, Yang and Tang, Yihong and Kong, Menglin and Wang, Yunlong and Jiao, Siwen and others},
  journal={arXiv preprint arXiv:2506.24044},
  year={2025}
}
```
---

## 📚 Table of Contents
- [🔥 Motivation & Paradigm Shift](#-motivation--paradigm-shift)  
- [🚀 Overview of VLA4AD](#-overview-of-vla4ad)  
- [🏆 Awesome VLA4AD Papers](#-awesome-vla4ad-papers)  
  - [1️⃣ Pre-VLA: VLM as Explainers](#1%EF%B8%8F%E2%83%A3-pre-vla-vlm-as-explainers)  
  - [2️⃣ Modular VLA4AD](#2%EF%B8%8F%E2%83%A3-modular-vla4ad)  
  - [3️⃣ End-to-End VLA4AD](#3%EF%B8%8F%E2%83%A3-end-to-end-vla4ad)  
  - [4️⃣ Reasoning-Augmented VLA4AD](#4%EF%B8%8F%E2%83%A3-reasoning-augmented-vla4ad)  
- [📊 Datasets & Benchmarks](#-datasets--benchmarks)  
- [⚙️ Installation & Usage](#%EF%B8%8F-installation--usage)

---
## 🔥 Motivation & Paradigm Shift

The development of autonomous driving has progressed from modular pipelines to fully integrated systems. This survey summarizes the latest advances into three core paradigms:

* **End-to-End AD:** Direct sensor-to-control mapping—efficient but opaque and weak on rare scenarios.  
  * **Flow:** Sensors → Network → Actions

* **VLMs for AD:** Adds language reasoning—boosts explainability but doesn’t drive the vehicle.  
  * **Flow:** Sensors → VLM → Answers

* **VLA for AD:** Unifies vision, language, and control in one policy—understands instructions, reasons, acts, and explains.  
  * **Flow:** Sensors → Multimodal Encoder → LLM/VLM → Decoder → Actions

![Driving Paradigms Comparison](https://raw.githubusercontent.com/JohnsonJiang1996/Awesome-VLA4AD/main/figs/Paradigm_Comparison.png)  
*Figure 1. (a) conventional end-to-end AD, (b) vision-language models as explainers, (c) full Vision–Language–Action systems.*

---

## 🚀 Overview of VLA4AD

A typical VLA4AD model follows an “Input–Process–Output” flow, unifying environment perception, instruction understanding, and vehicle control.

![Overview of VLA4AD](https://raw.githubusercontent.com/JohnsonJiang1996/Awesome-VLA4AD/main/figs/Overview_VLA4AD_New.png)  
*Figure 2. Overview of VLA4AD, integrating vision, language, and action modules.*

A snapshot of the field’s evolution through four successive stages—from VLM-as-explainer to augmented, reasoning-centric agents:

![Progress of VLA Models for AD](https://raw.githubusercontent.com/JohnsonJiang1996/Awesome-VLA4AD/main/figs/Progress_VLA4AD.png)  
*Figure 3. Progression of VLA4AD models: (1) VLMs as passive explainers; (2) Modular VLA with intermediate representations; (3) End-to-end VLA mapping sensors directly to actions; (4) Augmented VLA with long-horizon reasoning and tool use.*

The following table shows the representative models of VLA4AD and their inside modules:

![T](https://raw.githubusercontent.com/JohnsonJiang1996/Awesome-VLA4AD/main/figs/table1VLA4AD.png)  
*Table 1. Representative VLA4AD Models (2023–2025). Sensor Inputs: Single = single forward-facing camera input; Multi = multi-view
camera input; State = vehicle state information & other sensor input. Outputs: LLC= low-level control, Traj.= future trajectory, Multi.=
multiple tasks such as perception, prediction or planning.*

---

## 🏆 Awesome VLA4AD Papers

### 1️⃣ Pre-VLA: VLM as Explainers
| Model          | Year | Key Features                   | Link                                                             |
| -------------- | ---- | ---------------------- | ---------------------------------------------------------------- |
| DriveGPT-4     | 2023 | Scene Narration, QA    | https://arxiv.org/abs/2310.01412 / [Code](https://tonyxuqaq.github.io/projects/DriveGPT4/)|
| TS-VLM         | 2025 | Text-guided Attention  | https://arxiv.org/abs/2505.12670 / [Code](https://github.com/AiX-Lab-UWO/TS-VLM)|
| DynRsl-VLM     | 2025 | Adaptive Resolution    | https://arxiv.org/abs/2503.11265                                  |

### 2️⃣ Modular VLA4AD
| Model          | Year |  Key Features                      | Link                                                             |
| -------------- | ---- | ---------------------- | ---------------------------------------------------------------- |
| RAG-Driver     | 2024 | Retrieval-Augmented   |https://arxiv.org/abs/2402.10828 / [Code](https://github.com/YuanJianhao508/RAG-Driver)|
| OpenDriveVLA   | 2025 | Language-guided Planning | https://arxiv.org/abs/2503.23463 / [Code](https://github.com/DriveVLA/OpenDriveVLA) |
| DriveMoE       | 2025 | Expert Routing         | https://arxiv.org/abs/2505.16278 / [Code](https://github.com/Thinklab-SJTU/DriveMoE) |
| LangCoop       | 2025 | V2V Coordination       | https://arxiv.org/abs/2504.13406 / [Code](https://github.com/taco-group/LangCoop)  |
| SafeAuto       | 2025 | Rule-based Safety      | https://arxiv.org/abs/2503.00211 / [Code](https://github.com/AI-secure/SafeAuto)  |
| ReCogDrive     | 2025 | Diffusion + RL      |https://arxiv.org/abs/2506.08052 / [Code](https://github.com/xiaomi-research/recogdrive/)  |

### 3️⃣ End-to-End VLA4AD
| Model          | Year |  Key Features                      | Link                                                             |
| -------------- | ---- | ---------------------- | ---------------------------------------------------------------- |
| ADriver-I      | 2023 | Diffusion-based World Model | https://arxiv.org/abs/2311.13549                                  |
| EMMA           | 2024 | Detection + Planning   | https://arxiv.org/abs/2410.23262 / [Code](https://github.com/taco-group/OpenEMMA)  |
| CoVLA-Agent    | 2024 | Caption + Trajectory   | https://arxiv.org/abs/2408.10845 / [Code](https://turingmotors.github.io/covla-ad/)     |
| SimLingo       | 2025 | Action Dreaming        | https://arxiv.org/abs/2503.09594 / [Code](https://github.com/RenzKa/simlingo) |
| DiffVLA        | 2025 | Sparse-Dense Diffusion | https://arxiv.org/abs/2505.19381                                 |
| S4-Driver        | 2025 | Sparse 3D Representation | https://arxiv.org/abs/2505.24139                                 |

### 4️⃣ Reasoning-Augmented VLA4AD
| Model          | Year |  Key Features                     | Link                                                             |
| -------------- | ---- | ---------------------- | ---------------------------------------------------------------- |
| ORION          | 2025 | Memory + Rationales    | https://arxiv.org/abs/2503.19755 / [Code](https://xiaomi-mlab.github.io/Orion/)              |
| Impromptu-VLA  | 2025 | CoT-Aligned Planning   | https://arxiv.org/abs/2505.23757 / [Code](https://github.com/ahydchh/Impromptu-VLA)           |
| FSDrive       | 2025 | Visual Reasoning     | https://arxiv.org/abs/2505.17685 / [Code](https://github.com/MIV-XJTU/FSDrive)           |
| AutoVLA        | 2025 | Drive Tokens + CoT     | https://arxiv.org/abs/2506.13757 / [Code](https://github.com/ucla-mobility/AutoVLA)           |
|Drive-R1        | 2025 | CoT-Aligned Planning   | https://arxiv.org/abs/2506.18234            |
|Alpamayo-R1     | 2025 | Chain of Causation Reasoning | https://arxiv.org/abs/2511.00088      |
---

## 📊 Datasets & Benchmarks

| Name                   | Year | Modality                   | Task                      | URL                                       |
| ---------------------- | ---- | -------------------------- | ------------------------- | ----------------------------------------- |
| BDD100K / BDD-X        | 2018 | Video + Rationales         | Captioning, QA            | [bdd-data.berkeley.edu](http://bdd-data.berkeley.edu/)            |
| nuScenes               | 2020 | Camera, LiDAR, Radar       | Detection, QA             | www.nuscenes.org                 |
| Bench2Drive            | 2024 | CARLA Simulator            | Closed-loop Driving       | [Github](https://github.com/Thinklab-SJTU/Bench2Drive) |
| Reason2Drive           | 2024 | Video–QA                   | CoT-Chain Consistency     | [Github](https://github.com/fudan-zvg/Reason2Drive)        |
| Impromptu-VLA Dataset  | 2025 | Video + QA + Traj          | Corner-Case Testing       |[Github](https://github.com/ahydchh/Impromptu-VLA)|
| NuInteract              | 2025 | Multi-view QA              | 3D QA                     |[Github](https://github.com/zc-zhao/DriveMonkey)|
| DriveAction            | 2025 | In-the-wild QA             | High-level Actions        |       [HuggingFace](https://huggingface.co/datasets/LiAuto-DriveAction/drive-action)          |

---

## ⚙️ Installation & Usage

```bash
git clone https://github.com/JohnsonJiang1996/Awesome-VLA4AD.git
cd Awesome-VLA4AD
# Browse papers, datasets & code samples in each folder
```

---

Awesome-VLA4AD

파일 탐색기

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Awesome-VLA4AD

파일 탐색기

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README