Taming High-Dimensional Dynamics: Learning Optimal Projections onto Spectral Submanifolds

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Buurmeijer, Hugo, Pabon, Luis A., Alora, John Irvin, Kaundinya, Roshan S., Haller, George, Pavone, Marco
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918135451353088
author Buurmeijer, Hugo
Pabon, Luis A.
Alora, John Irvin
Kaundinya, Roshan S.
Haller, George
Pavone, Marco
author_facet Buurmeijer, Hugo
Pabon, Luis A.
Alora, John Irvin
Kaundinya, Roshan S.
Haller, George
Pavone, Marco
contents High-dimensional nonlinear systems pose considerable challenges for modeling and control across many domains, from fluid mechanics to advanced robotics. Such systems are typically approximated with reduced-order models, which often rely on orthogonal projections, a simplification that may lead to large prediction errors. In this work, we derive optimality of fiber-aligned projections onto spectral submanifolds, preserving the nonlinear geometric structure and minimizing long-term prediction error. We propose a data-driven procedure to learn these projections from trajectories and demonstrate its effectiveness through a 180-dimensional robotic system. Our reduced-order models achieve up to fivefold improvement in trajectory tracking accuracy under model predictive control compared to the state of the art.
format Preprint
id arxiv_https___arxiv_org_abs_2504_03157
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Taming High-Dimensional Dynamics: Learning Optimal Projections onto Spectral Submanifolds
Buurmeijer, Hugo
Pabon, Luis A.
Alora, John Irvin
Kaundinya, Roshan S.
Haller, George
Pavone, Marco
Systems and Control
Robotics
High-dimensional nonlinear systems pose considerable challenges for modeling and control across many domains, from fluid mechanics to advanced robotics. Such systems are typically approximated with reduced-order models, which often rely on orthogonal projections, a simplification that may lead to large prediction errors. In this work, we derive optimality of fiber-aligned projections onto spectral submanifolds, preserving the nonlinear geometric structure and minimizing long-term prediction error. We propose a data-driven procedure to learn these projections from trajectories and demonstrate its effectiveness through a 180-dimensional robotic system. Our reduced-order models achieve up to fivefold improvement in trajectory tracking accuracy under model predictive control compared to the state of the art.
title Taming High-Dimensional Dynamics: Learning Optimal Projections onto Spectral Submanifolds
topic Systems and Control
Robotics
url https://arxiv.org/abs/2504.03157