FutureX: Enhance End-to-End Autonomous Driving via Latent Chain-of-Thought World Model

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lin, Hongbin, Yang, Yiming, Zhang, Yifan, Zheng, Chaoda, Feng, Jie, Wang, Sheng, Wang, Zhennan, Chen, Shijia, Wang, Boyang, Zhang, Yu, Liu, Xianming, Cui, Shuguang, Li, Zhen
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918245697585152
author Lin, Hongbin
Yang, Yiming
Zhang, Yifan
Zheng, Chaoda
Feng, Jie
Wang, Sheng
Wang, Zhennan
Chen, Shijia
Wang, Boyang
Zhang, Yu
Liu, Xianming
Cui, Shuguang
Li, Zhen
author_facet Lin, Hongbin
Yang, Yiming
Zhang, Yifan
Zheng, Chaoda
Feng, Jie
Wang, Sheng
Wang, Zhennan
Chen, Shijia
Wang, Boyang
Zhang, Yu
Liu, Xianming
Cui, Shuguang
Li, Zhen
contents In autonomous driving, end-to-end planners learn scene representations from raw sensor data and utilize them to generate a motion plan or control actions. However, exclusive reliance on the current scene for motion planning may result in suboptimal responses in highly dynamic traffic environments where ego actions further alter the future scene. To model the evolution of future scenes, we leverage the World Model to represent how the ego vehicle and its environment interact and change over time, which entails complex reasoning. The Chain of Thought (CoT) offers a promising solution by forecasting a sequence of future thoughts that subsequently guide trajectory refinement. In this paper, we propose FutureX, a CoT-driven pipeline that enhances end-to-end planners to perform complex motion planning via future scene latent reasoning and trajectory refinement. Specifically, the Auto-think Switch examines the current scene and decides whether additional reasoning is required to yield a higher-quality motion plan. Once FutureX enters the Thinking mode, the Latent World Model conducts a CoT-guided rollout to predict future scene representation, enabling the Summarizer Module to further refine the motion plan. Otherwise, FutureX operates in an Instant mode to generate motion plans in a forward pass for relatively simple scenes. Extensive experiments demonstrate that FutureX enhances existing methods by producing more rational motion plans and fewer collisions without compromising efficiency, thereby achieving substantial overall performance gains, e.g., 6.2 PDMS improvement for TransFuser on NAVSIM. Code will be released.
format Preprint
id arxiv_https___arxiv_org_abs_2512_11226
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FutureX: Enhance End-to-End Autonomous Driving via Latent Chain-of-Thought World Model
Lin, Hongbin
Yang, Yiming
Zhang, Yifan
Zheng, Chaoda
Feng, Jie
Wang, Sheng
Wang, Zhennan
Chen, Shijia
Wang, Boyang
Zhang, Yu
Liu, Xianming
Cui, Shuguang
Li, Zhen
Computer Vision and Pattern Recognition
In autonomous driving, end-to-end planners learn scene representations from raw sensor data and utilize them to generate a motion plan or control actions. However, exclusive reliance on the current scene for motion planning may result in suboptimal responses in highly dynamic traffic environments where ego actions further alter the future scene. To model the evolution of future scenes, we leverage the World Model to represent how the ego vehicle and its environment interact and change over time, which entails complex reasoning. The Chain of Thought (CoT) offers a promising solution by forecasting a sequence of future thoughts that subsequently guide trajectory refinement. In this paper, we propose FutureX, a CoT-driven pipeline that enhances end-to-end planners to perform complex motion planning via future scene latent reasoning and trajectory refinement. Specifically, the Auto-think Switch examines the current scene and decides whether additional reasoning is required to yield a higher-quality motion plan. Once FutureX enters the Thinking mode, the Latent World Model conducts a CoT-guided rollout to predict future scene representation, enabling the Summarizer Module to further refine the motion plan. Otherwise, FutureX operates in an Instant mode to generate motion plans in a forward pass for relatively simple scenes. Extensive experiments demonstrate that FutureX enhances existing methods by producing more rational motion plans and fewer collisions without compromising efficiency, thereby achieving substantial overall performance gains, e.g., 6.2 PDMS improvement for TransFuser on NAVSIM. Code will be released.
title FutureX: Enhance End-to-End Autonomous Driving via Latent Chain-of-Thought World Model
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.11226