Learning control of an inverted pendulum using neural networks

Abstract
Several problems in control using neural networks are pointed out, and a method for using some knowledge about the properties of the controlled object is indicated. As an example, a hierarchical control system of an inverted pendulum using a neural network is proposed. The controller consists of a pendulum part driven by the neural network and a cart part generating the virtual signal. As a result, the neurocontroller can be constructed easily and can be learned rapidly. The effectiveness of the control system has been confirmed by simulation and experimental results.<>

This publication has 5 references indexed in Scilit: