NaviDriveVLM: Decoupling High-Level Reasoning and Motion Planning for Autonomous Driving

TL;DR

NaviDriveVLM introduces a decoupled framework for autonomous driving that separates high-level reasoning from motion planning, improving efficiency and interpretability while maintaining strong semantic understanding.

Contribution

The paper presents a novel decoupled architecture with a large Navigator and a lightweight Driver, enhancing reasoning, reducing training costs, and providing interpretable planning.

Findings

Outperforms large VLM baselines on nuScenes benchmark

Reduces training cost compared to monolithic models

Provides explicit intermediate representations for planning

Abstract

Vision-language models (VLMs) have emerged as a promising direction for end-to-end autonomous driving (AD) by jointly modeling visual observations, driving context, and language-based reasoning. However, existing VLM-based systems face a trade-off between high-level reasoning and motion planning: large models offer strong semantic understanding but are costly to adapt for precise control, whereas small VLM models can be fine-tuned efficiently but often exhibit weaker reasoning. We propose NaviDriveVLM, a decoupled framework that separates reasoning from action generation using a large-scale Navigator and a lightweight trainable Driver. This design preserves reasoning ability, reduces training cost, and provides an explicit interpretable intermediate representation for downstream planning. Experiments on the nuScenes benchmark show that NaviDriveVLM outperforms large VLM baselines in end-to-end motion planning.