Safe Online Control-Informed Learning

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhou, Tianyu, Liang, Zihao, Lu, Zehui, Mou, Shaoshuai
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918261673689088
author Zhou, Tianyu
Liang, Zihao
Lu, Zehui
Mou, Shaoshuai
author_facet Zhou, Tianyu
Liang, Zihao
Lu, Zehui
Mou, Shaoshuai
contents This paper proposes a Safe Online Control-Informed Learning framework for safety-critical autonomous systems. The framework unifies optimal control, parameter estimation, and safety constraints into an online learning process. It employs an extended Kalman filter to incrementally update system parameters in real time, enabling robust and data-efficient adaptation under uncertainty. A softplus barrier function enforces constraint satisfaction during learning and control while eliminating the dependence on high-quality initial guesses. Theoretical analysis establishes convergence and safety guarantees, and the framework's effectiveness is demonstrated on cart-pole and robot-arm systems.
format Preprint
id arxiv_https___arxiv_org_abs_2512_13868
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Safe Online Control-Informed Learning
Zhou, Tianyu
Liang, Zihao
Lu, Zehui
Mou, Shaoshuai
Systems and Control
Machine Learning
Optimization and Control
This paper proposes a Safe Online Control-Informed Learning framework for safety-critical autonomous systems. The framework unifies optimal control, parameter estimation, and safety constraints into an online learning process. It employs an extended Kalman filter to incrementally update system parameters in real time, enabling robust and data-efficient adaptation under uncertainty. A softplus barrier function enforces constraint satisfaction during learning and control while eliminating the dependence on high-quality initial guesses. Theoretical analysis establishes convergence and safety guarantees, and the framework's effectiveness is demonstrated on cart-pole and robot-arm systems.
title Safe Online Control-Informed Learning
topic Systems and Control
Machine Learning
Optimization and Control
url https://arxiv.org/abs/2512.13868