This contribution considers a new realization of the cognitive stabilizer, which is an adaptive stabilization control method based on a cognition-based framework. It is assumed, that the model of the system to be controlled is unknown. Only the knowledge about the system inputs, outputs, and equilibrium points are the preliminaries assumed within this approach. A new improved realization of the cognitive stabilizer is designed in this contribution using 1) a neural network estimating suitable inputs according to the desired outputs, 2) Lyapunov stability criterion according to a certain Lyapunov function, and 3) an optimization method to determine the desired system outputs with respect to the system energy. The proposed cognitive stabilizer is able to stabilize an unknown nonlinear MIMO system at arbitrary equilibrium point of it. Suitable control input can be designed automatically to guarantee the stability of motion of the system during the whole process although the changing of the system behavior or the environment. Numerical examples are shown to demonstrate the successful application and performance of this method.
In this paper, a cognitive stabilizer concept is introduced. The framework acts as an adaptive discrete control approach. The aim of the cognitive stabilizer is to stabilize a specific class of unknown nonlinear MIMO systems. The cognitive stabilizer is able to gain useful local knowledge of the system assumed as unknown. The approach is able to define autonomously suitable control inputs to stabilize the system. The system class to be considered is described by the following assumptions: unknown input/output behavior, fully controllable, stable zero dynamics, and measured state vector. The cognitive stabilizer is realized by its four main modules: (1) "perception and interpretation" using system identifier for the system local dynamic online identification and multi-step-ahead prediction;(2) "expert knowledge" relating to the quadratic stability criterion to guarantee the stability of the considered motion of the controlled system; (3) "planning" to generate a suitable control input sequence according to a certain cost function; (4) "execution" to generate the optimal control input in a corresponding feedback form. Each module can be realized using different methods. Two realizations will be stated in this paper. Using the cognitive stabilizer, the control goal can be achieved efficiently without an individual control design process for different kinds of unknown systems. Numerical examples (e.g., a chaotic nonlinear MIMO system-Lorenz system) demonstrate the successful application of the proposed methods.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.