Trajectory training of feedforward neural networks for DC motor speed control

This work discusses the use of template trained neural networks for DC motor speed control. It proposes the use of input-output trajectories to train the neural network instead of using data points as is normally done for feedforward neural networks. Experimental results show that the trajectory trained neural network can successfully perform speed control of PMDC motors when compared to neural networks trained using conventional training methods.

I. Introduction

Artificial Neural Networks (ANN) are attractive tools for use in control system applications (also known as neural control) because of characteristics like non-linearity, parallel processing, learning and adaptation and MIMO capabilities [1]. Neural control can be broadly classified in four different approaches, namely 1) template learning, 2) learning plant inversion, 3) closed loop optimization, and 4) critic systems [2]. Of these four approaches the most well developed are the closed loop optimization and critic systems. This paper looks specifically at the template learning, and proposes a new training method for PMDC motor speed control.