GPD-1: Generative Pre-training for Driving

TL;DR

GPD-1 is a unified autoregressive transformer model pre-trained on large-scale driving data that can perform multiple autonomous driving tasks like scene generation, prediction, and planning without additional fine-tuning.

Contribution

This paper introduces GPD-1, a novel unified pre-training framework that handles various autonomous driving tasks simultaneously using token-based scene representation and transformer architecture.

Findings

GPD-1 effectively generalizes to multiple tasks without fine-tuning.

The model achieves competitive results on scene generation and motion prediction.

Pre-training on nuPlan enhances the model's understanding of driving scenarios.

Abstract

Modeling the evolutions of driving scenarios is important for the evaluation and decision-making of autonomous driving systems. Most existing methods focus on one aspect of scene evolution such as map generation, motion prediction, and trajectory planning. In this paper, we propose a unified Generative Pre-training for Driving (GPD-1) model to accomplish all these tasks altogether without additional fine-tuning. We represent each scene with ego, agent, and map tokens and formulate autonomous driving as a unified token generation problem. We adopt the autoregressive transformer architecture and use a scene-level attention mask to enable intra-scene bi-directional interactions. For the ego and agent tokens, we propose a hierarchical positional tokenizer to effectively encode both 2D positions and headings. For the map tokens, we train a map vector-quantized autoencoder to efficiently compress ego-centric semantic maps into discrete tokens. We pre-train our GPD-1 on the large-scale nuPlan dataset and conduct extensive experiments to evaluate its effectiveness. With different prompts, our GPD-1 successfully generalizes to various tasks without finetuning, including scene generation, traffic simulation, closed-loop simulation, map prediction, and motion planning. Code: https://github.com/wzzheng/GPD.