WIT Press


Neural Network Application To Optimal Control Of Nonlinear Systems

Price

Free (open access)

Paper DOI

10.2495/OP010321

Volume

54

Pages

10

Published

2001

Size

797 kb

Author(s)

J. Kasac & B. Novakovic

Abstract

Neural network application to optimal control of nonlinear systems i 'j J. Kasac & B. Novakovic ^Institute for Defense Studies, Research and Development, Croatia ^Faculty of Mechanical Engineering and Naval Architecture, University of Zagreb, Croatia Abstract This paper presents the derivation of the numerical algorithm for optimal control of nonlinear multivariable systems with control and state vectors constraints. The algorithm derivation is based on the backpropagation- through-time (BPTT) algorithm which is used as a learning algorithm for recurrent neural networks. This approach is not based on Lagrange mul- tiplier techniques and the calculus of variations. The derived algorithm is used for the control of the cooperative work of two robots with two degrees of freedom. The main problem is the determination of the control vectors of robots for the transfer of rigid load from the initial state to the final one in a fixed time while maintaining constant

Keywords