Lyapunov-Based Deep Residual Neural Network (ResNet) Adaptive Control

Deep Neural Network (DNN)-based controllers have emerged as a tool to compensate for unstructured uncertainties in nonlinear dynamical systems. A recent breakthrough in the adaptive control literature provides a Lyapunov-based approach to derive weight adaptation laws for each layer of a fully-conne...

Full description

Saved in:
Bibliographic Details
Main Authors: Omkar Sudhir Patil, Duc M. Le, Emily J. Griffis, Warren E. Dixon
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/11059245/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850115725632471040
author Omkar Sudhir Patil
Duc M. Le
Emily J. Griffis
Warren E. Dixon
author_facet Omkar Sudhir Patil
Duc M. Le
Emily J. Griffis
Warren E. Dixon
author_sort Omkar Sudhir Patil
collection DOAJ
description Deep Neural Network (DNN)-based controllers have emerged as a tool to compensate for unstructured uncertainties in nonlinear dynamical systems. A recent breakthrough in the adaptive control literature provides a Lyapunov-based approach to derive weight adaptation laws for each layer of a fully-connected feedforward DNN-based adaptive controller. However, deriving weight adaptation laws from a Lyapunov-based analysis remains an open problem for deep residual neural networks (ResNets). This paper provides the first result on Lyapunov-derived weight adaptation for a ResNet-based adaptive controller. A nonsmooth Lyapunov-based analysis is provided to guarantee asymptotic tracking error convergence. Comparative Monte Carlo simulations are provided to demonstrate the performance of the developed ResNet-based adaptive controller. The ResNet-based adaptive controller shows a 64% improvement in the tracking and function approximation performance, in comparison to a fully-connected DNN-based adaptive controller.
format Article
id doaj-art-31ba6b48edbb44799e526a6b68ba64d6
institution OA Journals
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-31ba6b48edbb44799e526a6b68ba64d62025-08-20T02:36:30ZengIEEEIEEE Access2169-35362025-01-011311794311795210.1109/ACCESS.2025.358425311059245Lyapunov-Based Deep Residual Neural Network (ResNet) Adaptive ControlOmkar Sudhir Patil0https://orcid.org/0000-0002-3820-2025Duc M. Le1https://orcid.org/0000-0003-2891-0439Emily J. Griffis2https://orcid.org/0009-0004-7684-1444Warren E. Dixon3https://orcid.org/0000-0002-5091-181XDepartment of Mechanical and Aerospace Engineering, University of Florida, Gainesville, FL, USADepartment of Mechanical and Aerospace Engineering, University of Florida, Gainesville, FL, USADepartment of Mechanical and Aerospace Engineering, University of Florida, Gainesville, FL, USADepartment of Mechanical and Aerospace Engineering, University of Florida, Gainesville, FL, USADeep Neural Network (DNN)-based controllers have emerged as a tool to compensate for unstructured uncertainties in nonlinear dynamical systems. A recent breakthrough in the adaptive control literature provides a Lyapunov-based approach to derive weight adaptation laws for each layer of a fully-connected feedforward DNN-based adaptive controller. However, deriving weight adaptation laws from a Lyapunov-based analysis remains an open problem for deep residual neural networks (ResNets). This paper provides the first result on Lyapunov-derived weight adaptation for a ResNet-based adaptive controller. A nonsmooth Lyapunov-based analysis is provided to guarantee asymptotic tracking error convergence. Comparative Monte Carlo simulations are provided to demonstrate the performance of the developed ResNet-based adaptive controller. The ResNet-based adaptive controller shows a 64% improvement in the tracking and function approximation performance, in comparison to a fully-connected DNN-based adaptive controller.https://ieeexplore.ieee.org/document/11059245/Deep neural networksResNetsadaptive controlLyapunov-based methods
spellingShingle Omkar Sudhir Patil
Duc M. Le
Emily J. Griffis
Warren E. Dixon
Lyapunov-Based Deep Residual Neural Network (ResNet) Adaptive Control
IEEE Access
Deep neural networks
ResNets
adaptive control
Lyapunov-based methods
title Lyapunov-Based Deep Residual Neural Network (ResNet) Adaptive Control
title_full Lyapunov-Based Deep Residual Neural Network (ResNet) Adaptive Control
title_fullStr Lyapunov-Based Deep Residual Neural Network (ResNet) Adaptive Control
title_full_unstemmed Lyapunov-Based Deep Residual Neural Network (ResNet) Adaptive Control
title_short Lyapunov-Based Deep Residual Neural Network (ResNet) Adaptive Control
title_sort lyapunov based deep residual neural network resnet adaptive control
topic Deep neural networks
ResNets
adaptive control
Lyapunov-based methods
url https://ieeexplore.ieee.org/document/11059245/
work_keys_str_mv AT omkarsudhirpatil lyapunovbaseddeepresidualneuralnetworkresnetadaptivecontrol
AT ducmle lyapunovbaseddeepresidualneuralnetworkresnetadaptivecontrol
AT emilyjgriffis lyapunovbaseddeepresidualneuralnetworkresnetadaptivecontrol
AT warrenedixon lyapunovbaseddeepresidualneuralnetworkresnetadaptivecontrol