Autonomous Vehicle Path Planning by Searching With Differentiable Simulation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Nachkov, Asen, Zaech, Jan-Nico, Paudel, Danda Pani, Wang, Xi, Van Gool, Luc
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915634381586432
author Nachkov, Asen
Zaech, Jan-Nico
Paudel, Danda Pani
Wang, Xi
Van Gool, Luc
author_facet Nachkov, Asen
Zaech, Jan-Nico
Paudel, Danda Pani
Wang, Xi
Van Gool, Luc
contents Planning allows an agent to safely refine its actions before executing them in the real world. In autonomous driving, this is crucial to avoid collisions and navigate in complex, dense traffic scenarios. One way to plan is to search for the best action sequence. However, this is challenging when all necessary components - policy, next-state predictor, and critic - have to be learned. Here we propose Differentiable Simulation for Search (DSS), a framework that leverages the differentiable simulator Waymax as both a next state predictor and a critic. It relies on the simulator's hardcoded dynamics, making state predictions highly accurate, while utilizing the simulator's differentiability to effectively search across action sequences. Our DSS agent optimizes its actions using gradient descent over imagined future trajectories. We show experimentally that DSS - the combination of planning gradients and stochastic search - significantly improves tracking and path planning accuracy compared to sequence prediction, imitation learning, model-free RL, and other planning methods.
format Preprint
id arxiv_https___arxiv_org_abs_2511_11043
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Autonomous Vehicle Path Planning by Searching With Differentiable Simulation
Nachkov, Asen
Zaech, Jan-Nico
Paudel, Danda Pani
Wang, Xi
Van Gool, Luc
Artificial Intelligence
Robotics
Planning allows an agent to safely refine its actions before executing them in the real world. In autonomous driving, this is crucial to avoid collisions and navigate in complex, dense traffic scenarios. One way to plan is to search for the best action sequence. However, this is challenging when all necessary components - policy, next-state predictor, and critic - have to be learned. Here we propose Differentiable Simulation for Search (DSS), a framework that leverages the differentiable simulator Waymax as both a next state predictor and a critic. It relies on the simulator's hardcoded dynamics, making state predictions highly accurate, while utilizing the simulator's differentiability to effectively search across action sequences. Our DSS agent optimizes its actions using gradient descent over imagined future trajectories. We show experimentally that DSS - the combination of planning gradients and stochastic search - significantly improves tracking and path planning accuracy compared to sequence prediction, imitation learning, model-free RL, and other planning methods.
title Autonomous Vehicle Path Planning by Searching With Differentiable Simulation
topic Artificial Intelligence
Robotics
url https://arxiv.org/abs/2511.11043