DriveGPT: Scaling Autoregressive Behavior Models for Driving

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Huang, Xin, Wolff, Eric M., Vernaza, Paul, Phan-Minh, Tung, Chen, Hongge, Hayden, David S., Edmonds, Mark, Pierce, Brian, Chen, Xinxin, Jacob, Pratik Elias, Chen, Xiaobai, Tairbekov, Chingiz, Agarwal, Pratik, Gao, Tianshi, Chai, Yuning, Srinivasa, Siddhartha
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909598507597824
author Huang, Xin
Wolff, Eric M.
Vernaza, Paul
Phan-Minh, Tung
Chen, Hongge
Hayden, David S.
Edmonds, Mark
Pierce, Brian
Chen, Xinxin
Jacob, Pratik Elias
Chen, Xiaobai
Tairbekov, Chingiz
Agarwal, Pratik
Gao, Tianshi
Chai, Yuning
Srinivasa, Siddhartha
author_facet Huang, Xin
Wolff, Eric M.
Vernaza, Paul
Phan-Minh, Tung
Chen, Hongge
Hayden, David S.
Edmonds, Mark
Pierce, Brian
Chen, Xinxin
Jacob, Pratik Elias
Chen, Xiaobai
Tairbekov, Chingiz
Agarwal, Pratik
Gao, Tianshi
Chai, Yuning
Srinivasa, Siddhartha
contents We present DriveGPT, a scalable behavior model for autonomous driving. We model driving as a sequential decision-making task, and learn a transformer model to predict future agent states as tokens in an autoregressive fashion. We scale up our model parameters and training data by multiple orders of magnitude, enabling us to explore the scaling properties in terms of dataset size, model parameters, and compute. We evaluate DriveGPT across different scales in a planning task, through both quantitative metrics and qualitative examples, including closed-loop driving in complex real-world scenarios. In a separate prediction task, DriveGPT outperforms state-of-the-art baselines and exhibits improved performance by pretraining on a large-scale dataset, further validating the benefits of data scaling.
format Preprint
id arxiv_https___arxiv_org_abs_2412_14415
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DriveGPT: Scaling Autoregressive Behavior Models for Driving
Huang, Xin
Wolff, Eric M.
Vernaza, Paul
Phan-Minh, Tung
Chen, Hongge
Hayden, David S.
Edmonds, Mark
Pierce, Brian
Chen, Xinxin
Jacob, Pratik Elias
Chen, Xiaobai
Tairbekov, Chingiz
Agarwal, Pratik
Gao, Tianshi
Chai, Yuning
Srinivasa, Siddhartha
Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Robotics
We present DriveGPT, a scalable behavior model for autonomous driving. We model driving as a sequential decision-making task, and learn a transformer model to predict future agent states as tokens in an autoregressive fashion. We scale up our model parameters and training data by multiple orders of magnitude, enabling us to explore the scaling properties in terms of dataset size, model parameters, and compute. We evaluate DriveGPT across different scales in a planning task, through both quantitative metrics and qualitative examples, including closed-loop driving in complex real-world scenarios. In a separate prediction task, DriveGPT outperforms state-of-the-art baselines and exhibits improved performance by pretraining on a large-scale dataset, further validating the benefits of data scaling.
title DriveGPT: Scaling Autoregressive Behavior Models for Driving
topic Machine Learning
Artificial Intelligence
Computer Vision and Pattern Recognition
Robotics
url https://arxiv.org/abs/2412.14415