Title

Non-linear System Control Using Path Generating Neural Networks

Abstract

An increasing amount of literature is available on the control of non-linear systems. For many of these systems, linear design techniques are ineffective. However, a non-linear augmentation of the linear control is sometimes able to perform the control. Determining the augmenting signal is a formidable task. However, recent work in the area of path planning has given rise to some useful tools. These tools can be used to design a complete controller. The final design takes advantage of the mapping abilities of neural networks to perform path planning. A path planning algorithm, taken from previously published work, was used to generate training data for the neural network. The results of simulation tests show that the controller composed of a linear gain and a path planning neural network was capable of controlling the non-linear system. The results are for a small class of non-trivial trajectories. Further work is planned to generalize the results.

Department(s)

Electrical and Computer Engineering

Sponsor(s)

National Science Foundation (U.S.)

Keywords and Phrases

Control; Non-Linear Augmentation; Non-Linear Systems

Document Type

Article - Conference proceedings

Document Version

Citation

File Type

text

Language(s)

English

Rights

© 1998 American Society of Mechanical Engineers (ASME), All rights reserved.


Share

 
COinS