Two New Methods for the Optimal Control of Nonlinear Systems using Neural Networks
The topic of nonlinear control design has attracted particular attention to satisfy the demanding requirements of recent real-world applications. In this article, a neural network based controller which optimizes a finite horizon quadratic cost function is developed for a class of nonlinear systems with unknown dynamics. Two new types of controllers with different iterative schemes are introduced to converge to the optimal trajectories. To apply such controllers, the system is first modeled using neural networks with back-propogation learning mentod. Both the controllers require the Jacobian matrices of the system state-equations which are obtained directly from the neural network learning process. To test the two control methods, a nonlinear sample system and a physical nonlinear system, a vibrating plate, are used.
H. Hu and L. Acar, "Two New Methods for the Optimal Control of Nonlinear Systems using Neural Networks," Intelligent Engineering Systems Through Artificial Neural Networks, vol. 8, pp. 585-590, American Society of Mechanical Engineers (ASME), Nov 1998.
Artificial Neural Networks in Engineering Conference, ANNIE '98 (1998: Nov. 1-4, St. Louis, MO)
Electrical and Computer Engineering
Keywords and Phrases
Backpropagation; Functions; Iterative methods; Learning systems; Mathematical models; Neural networks; Optimization; Plates (structural components); Finite horizon quadratic cost function; Jacobian matrix; Optimal control; Vibrating plate; Optimal control systems
International Standard Book Number (ISBN)
Article - Conference proceedings
© 1998 American Society of Mechanical Engineers (ASME), All rights reserved.