Enregistré dans:
Détails bibliographiques
Auteurs principaux: Broadbent, Nicholas Drake, Weber, Trey, Mori, Daiki, Gerdes, J. Christian
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:https://arxiv.org/abs/2407.13760
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866917727075041280
author Broadbent, Nicholas Drake
Weber, Trey
Mori, Daiki
Gerdes, J. Christian
author_facet Broadbent, Nicholas Drake
Weber, Trey
Mori, Daiki
Gerdes, J. Christian
contents Automated drifting presents a challenge problem for vehicle control, requiring models and control algorithms that can precisely handle nonlinear, coupled tire forces at the friction limits. We present a neural network architecture for predicting front tire lateral force as a drop-in replacement for physics-based approaches. With a full-scale automated vehicle purpose-built for the drifting application, we deploy these models in a nonlinear model predictive controller tuned for tracking a reference drifting trajectory, for direct comparisons of model performance. The neural network tire model exhibits significantly improved path tracking performance over the brush tire model in cases where front-axle braking force is applied, suggesting the neural network's ability to express previously unmodeled, latent dynamics in the drifting condition.
format Preprint
id arxiv_https___arxiv_org_abs_2407_13760
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Neural Network Tire Force Modeling for Automated Drifting
Broadbent, Nicholas Drake
Weber, Trey
Mori, Daiki
Gerdes, J. Christian
Systems and Control
Artificial Intelligence
Automated drifting presents a challenge problem for vehicle control, requiring models and control algorithms that can precisely handle nonlinear, coupled tire forces at the friction limits. We present a neural network architecture for predicting front tire lateral force as a drop-in replacement for physics-based approaches. With a full-scale automated vehicle purpose-built for the drifting application, we deploy these models in a nonlinear model predictive controller tuned for tracking a reference drifting trajectory, for direct comparisons of model performance. The neural network tire model exhibits significantly improved path tracking performance over the brush tire model in cases where front-axle braking force is applied, suggesting the neural network's ability to express previously unmodeled, latent dynamics in the drifting condition.
title Neural Network Tire Force Modeling for Automated Drifting
topic Systems and Control
Artificial Intelligence
url https://arxiv.org/abs/2407.13760