A Neural Network Solution for Fixed-Final Time Optimal Control of Nonlinear Systems

Abstract

We consider the use of neural networks and Hamilton-Jacobi-Bellman equations towards obtaining fixed-final time optimal control laws in the input nonlinear systems. The method is based on Kronecker matrix methods along with neural network approximation over a compact set to solve a time-varying Hamilton-Jacobi-Bellman equation. The result is a neural network feedback controller that has time-varying coefficients found by a priori offline tuning. Convergence results are shown. The results of this paper are demonstrated on two examples.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jun 01, 2006
Accession Number
ADA455430

Entities

People

  • Frank L. Lewis
  • Murad Abu-khalaf
  • Tao Cheng

Organizations

  • University of Texas at Arlington

Tags

Communities of Interest

  • Energy and Power Technologies

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Convergence
  • Differential Equations
  • Electronic Mail
  • Equations
  • Hilbert Space
  • Information Operations
  • Least Squares Method
  • Linear Systems
  • Neural Networks
  • Nonlinear Systems
  • Partial Differential Equations
  • Residuals
  • Riccati Equation
  • Steady State

Fields of Study

  • Mathematics

Readers

  • Computer Networking
  • Finite Element Method (FEM) for solving Partial Differential Equations (PDEs)
  • Radio communications and signal processing.

Technology Areas

  • AI & ML
  • AI & ML - Machine Learning Algorithms
  • AI & ML - Neural Networks