Doe-1: Closed-Loop Autonomous Driving with Large World Model

TL;DR

Doe-1 introduces a unified closed-loop framework for autonomous driving that leverages a large world model to improve perception, prediction, and planning through autoregressive token generation, demonstrating effectiveness on the nuScenes dataset.

Contribution

The paper presents a novel large world model (Doe-1) that unifies perception, prediction, and planning in autonomous driving using a multi-modal transformer and token-based representation.

Findings

Effective in visual question-answering tasks.

Generates accurate action-conditioned video.

Improves motion planning performance.

Abstract

End-to-end autonomous driving has received increasing attention due to its potential to learn from large amounts of data. However, most existing methods are still open-loop and suffer from weak scalability, lack of high-order interactions, and inefficient decision-making. In this paper, we explore a closed-loop framework for autonomous driving and propose a large Driving wOrld modEl (Doe-1) for unified perception, prediction, and planning. We formulate autonomous driving as a next-token generation problem and use multi-modal tokens to accomplish different tasks. Specifically, we use free-form texts (i.e., scene descriptions) for perception and generate future predictions directly in the RGB space with image tokens. For planning, we employ a position-aware tokenizer to effectively encode action into discrete tokens. We train a multi-modal transformer to autoregressively generate perception, prediction, and planning tokens in an end-to-end and unified manner. Experiments on the widely used nuScenes dataset demonstrate the effectiveness of Doe-1 in various tasks including visual question-answering, action-conditioned video generation, and motion planning. Code: https://github.com/wzzheng/Doe.