DSDrive: Distilling Large Language Model for Lightweight End-to-End Autonomous Driving with Unified Reasoning and Planning
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912366808006656 |
|---|---|
| author | Liu, Wenru Liu, Pei Ma, Jun |
| author_facet | Liu, Wenru Liu, Pei Ma, Jun |
| contents | We present DSDrive, a streamlined end-to-end paradigm tailored for integrating the reasoning and planning of autonomous vehicles into a unified framework. DSDrive leverages a compact LLM that employs a distillation method to preserve the enhanced reasoning capabilities of a larger-sized vision language model (VLM). To effectively align the reasoning and planning tasks, a waypoint-driven dual-head coordination module is further developed, which synchronizes dataset structures, optimization objectives, and the learning process. By integrating these tasks into a unified framework, DSDrive anchors on the planning results while incorporating detailed reasoning insights, thereby enhancing the interpretability and reliability of the end-to-end pipeline. DSDrive has been thoroughly tested in closed-loop simulations, where it performs on par with benchmark models and even outperforms in many key metrics, all while being more compact in size. Additionally, the computational efficiency of DSDrive (as reflected in its time and memory requirements during inference) has been significantly enhanced. Evidently thus, this work brings promising aspects and underscores the potential of lightweight systems in delivering interpretable and efficient solutions for AD. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_05360 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DSDrive: Distilling Large Language Model for Lightweight End-to-End Autonomous Driving with Unified Reasoning and Planning Liu, Wenru Liu, Pei Ma, Jun Robotics We present DSDrive, a streamlined end-to-end paradigm tailored for integrating the reasoning and planning of autonomous vehicles into a unified framework. DSDrive leverages a compact LLM that employs a distillation method to preserve the enhanced reasoning capabilities of a larger-sized vision language model (VLM). To effectively align the reasoning and planning tasks, a waypoint-driven dual-head coordination module is further developed, which synchronizes dataset structures, optimization objectives, and the learning process. By integrating these tasks into a unified framework, DSDrive anchors on the planning results while incorporating detailed reasoning insights, thereby enhancing the interpretability and reliability of the end-to-end pipeline. DSDrive has been thoroughly tested in closed-loop simulations, where it performs on par with benchmark models and even outperforms in many key metrics, all while being more compact in size. Additionally, the computational efficiency of DSDrive (as reflected in its time and memory requirements during inference) has been significantly enhanced. Evidently thus, this work brings promising aspects and underscores the potential of lightweight systems in delivering interpretable and efficient solutions for AD. |
| title | DSDrive: Distilling Large Language Model for Lightweight End-to-End Autonomous Driving with Unified Reasoning and Planning |
| topic | Robotics |
| url | https://arxiv.org/abs/2505.05360 |