Abstract

An output feedback (OF)-based control scheme utilizing both a scalable multilayer neural network (MNN) observer and actor–critic MNN via integral reinforcement learning (IRL)/adaptive dynamics programming (ADP) approach for a class of nonlinear systems with output constraints is introduced. The proposed observer, critic, and actor MNN weight updates are derived using a singular value decomposition (SVD) of MNN activation function gradient along with output error, Bellman and control input errors, respectively. Next, the approach incorporates continual learning (CL), utilizing a penalty function in the weight update laws for both actor–critic MNNs to consolidate knowledge from previous tasks and enhance learning in new tasks using estimated states across each layer in order to improve performance. The output constraints are addressed using the Karush–Kuhn–Tucker (KKT) conditions by utilizing the barrier Lyapunov functions (BLFs), which ensure the system output remains within a safe set at all times. Finally, the efficacy of the safety aware OF tracking control is demonstrated through empirical tests on a two-link robotic manipulator example which shows an 80% performance improvement as compared to recent literature.

Department(s)

Electrical and Computer Engineering

Second Department

Computer Science

Publication Status

Early Access

Comments

Office of Naval Research, Grant N00014-24-1-2338

Keywords and Phrases

Multilayer neural networks (MNNs); multitasking learning; optimal control; output constraint; output feedback (OF)

International Standard Serial Number (ISSN)

2168-2232; 2168-2216

Document Type

Article - Journal

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 2026 Institute of Electrical and Electronics Engineers, All rights reserved.

Publication Date

01 Jan 2026

Download

Full Text Link

Share

 
COinS