A Physics-informed End-to-End Occupancy Framework for Motion Planning of Autonomous Vehicles

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shen, Shuqi, Yang, Junjie, Lu, Hongliang, Zhong, Hui, Zhang, Qiming, Zheng, Xinhu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910991275524096
author Shen, Shuqi
Yang, Junjie
Lu, Hongliang
Zhong, Hui
Zhang, Qiming
Zheng, Xinhu
author_facet Shen, Shuqi
Yang, Junjie
Lu, Hongliang
Zhong, Hui
Zhang, Qiming
Zheng, Xinhu
contents Accurate and interpretable motion planning is essential for autonomous vehicles (AVs) navigating complex and uncertain environments. While recent end-to-end occupancy prediction methods have improved environmental understanding, they typically lack explicit physical constraints, limiting safety and generalization. In this paper, we propose a unified end-to-end framework that integrates verifiable physical rules into the occupancy learning process. Specifically, we embed artificial potential fields (APF) as physics-informed guidance during network training to ensure that predicted occupancy maps are both data-efficient and physically plausible. Our architecture combines convolutional and recurrent neural networks to capture spatial and temporal dependencies while preserving model flexibility. Experimental results demonstrate that our method improves task completion rate, safety margins, and planning efficiency across diverse driving scenarios, confirming its potential for reliable deployment in real-world AV systems.
format Preprint
id arxiv_https___arxiv_org_abs_2505_07855
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Physics-informed End-to-End Occupancy Framework for Motion Planning of Autonomous Vehicles
Shen, Shuqi
Yang, Junjie
Lu, Hongliang
Zhong, Hui
Zhang, Qiming
Zheng, Xinhu
Robotics
Accurate and interpretable motion planning is essential for autonomous vehicles (AVs) navigating complex and uncertain environments. While recent end-to-end occupancy prediction methods have improved environmental understanding, they typically lack explicit physical constraints, limiting safety and generalization. In this paper, we propose a unified end-to-end framework that integrates verifiable physical rules into the occupancy learning process. Specifically, we embed artificial potential fields (APF) as physics-informed guidance during network training to ensure that predicted occupancy maps are both data-efficient and physically plausible. Our architecture combines convolutional and recurrent neural networks to capture spatial and temporal dependencies while preserving model flexibility. Experimental results demonstrate that our method improves task completion rate, safety margins, and planning efficiency across diverse driving scenarios, confirming its potential for reliable deployment in real-world AV systems.
title A Physics-informed End-to-End Occupancy Framework for Motion Planning of Autonomous Vehicles
topic Robotics
url https://arxiv.org/abs/2505.07855