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

arXiv: https://arxiv.org/abs/2505.07855 · HTML: https://mlsanity.com/paper/2505.07855/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Generative simulation for driving, one of 9 directions in End-to-end autonomous driving.
- Builds on 14 of its 25 references that are in the map.
- First posted on arXiv 8 May 2025.

## Placed in

- End-to-end autonomous driving / [Generative simulation for driving](https://mlsanity.com/driving/generative-simulation-for-driving/)

## Builds on (most cited first, 14 of 14)

- [Semantic Scene Completion from a Single Depth Image](https://mlsanity.com/paper/1611.08974/) (2016)
- [Occupancy Flow Fields for Motion Forecasting in Autonomous Driving](https://mlsanity.com/paper/2203.03875/) (2022)
- [Dynamic Occupancy Grid Prediction for Urban Autonomous Driving: A Deep Learning Approach with Fully Automatic Labeling](https://mlsanity.com/paper/1705.08781/) (2017)
- [A Random Finite Set Approach for Dynamic Occupancy Grid Maps with Real-Time Application](https://mlsanity.com/paper/1605.02406/) (2016)
- [Occupancy Grids: A Stochastic Spatial Representation for Active Robot Perception](https://mlsanity.com/paper/1304.1098/) (2013)
- [Occupancy Prediction-Guided Neural Planner for Autonomous Driving](https://mlsanity.com/paper/2305.03303/) (2023)
- [OFMPNet: Deep End-to-End Model for Occupancy and Flow Prediction in Urban Environment](https://mlsanity.com/paper/2404.02263/) (2024)
- [A Hybrid Rule-Based and Data-Driven Approach to Driver Modeling through Particle Filtering](https://mlsanity.com/paper/2108.12820/) (2021)
- [A Multi-Task Recurrent Neural Network for End-to-End Dynamic Occupancy Grid Mapping](https://mlsanity.com/paper/2202.04461/) (2022)
- [Enhance Planning with Physics-informed Safety Controller for End-to-end Autonomous Driving](https://mlsanity.com/paper/2405.00316/) (2024)
- [MixNet: Structured Deep Neural Motion Prediction for Autonomous Racing](https://mlsanity.com/paper/2208.01862/) (2022)
- [Bayesian Learning of Occupancy Grids](https://mlsanity.com/paper/1911.07915/) (2019)
- [A Dynamic Motion Planning Framework for Autonomous Driving in Urban Environments](https://mlsanity.com/paper/1912.04458/) (2019)
- [Transitional Grid Maps: Joint Modeling of Static and Dynamic Occupancy](https://mlsanity.com/paper/2401.06518/) (2024)

## Abstract (arXiv)

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.