Natural Gradient Descent for Control

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Esmzad, Ramin, Yaghmaie, Farnaz Adib, Modares, Hamidreza
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915187766853632
author Esmzad, Ramin
Yaghmaie, Farnaz Adib
Modares, Hamidreza
author_facet Esmzad, Ramin
Yaghmaie, Farnaz Adib
Modares, Hamidreza
contents This paper bridges optimization and control, and presents a novel closed-loop control framework based on natural gradient descent, offering a trajectory-oriented alternative to traditional cost-function tuning. By leveraging the Fisher Information Matrix, we formulate a preconditioned gradient descent update that explicitly shapes system trajectories. We show that, in sharp contrast to traditional controllers, our approach provides flexibility to shape the system's low-level behavior. To this end, the proposed method parameterizes closed-loop dynamics in terms of stationary covariance and an unknown cost function, providing a geometric interpretation of control adjustments. We establish theoretical stability conditions. The simulation results on a rotary inverted pendulum benchmark highlight the advantages of natural gradient descent in trajectory shaping.
format Preprint
id arxiv_https___arxiv_org_abs_2503_06070
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Natural Gradient Descent for Control
Esmzad, Ramin
Yaghmaie, Farnaz Adib
Modares, Hamidreza
Systems and Control
Optimization and Control
This paper bridges optimization and control, and presents a novel closed-loop control framework based on natural gradient descent, offering a trajectory-oriented alternative to traditional cost-function tuning. By leveraging the Fisher Information Matrix, we formulate a preconditioned gradient descent update that explicitly shapes system trajectories. We show that, in sharp contrast to traditional controllers, our approach provides flexibility to shape the system's low-level behavior. To this end, the proposed method parameterizes closed-loop dynamics in terms of stationary covariance and an unknown cost function, providing a geometric interpretation of control adjustments. We establish theoretical stability conditions. The simulation results on a rotary inverted pendulum benchmark highlight the advantages of natural gradient descent in trajectory shaping.
title Natural Gradient Descent for Control
topic Systems and Control
Optimization and Control
url https://arxiv.org/abs/2503.06070