Modelling and identification of eye–hand dynamics

Abstract

Human behaviour in manual control especially in transport systems has some non-linearities such as saturation, delay dead band and so on. In this paper, the standard model of human operator is analysed from the viewpoint of system identification. Therefore, some procedures for the identification of the human behaviour model are developed from available data. These procedures are validated by application according to experimental data from laboratory experiments with human tracking of the various stimuli for a step, sinusoidal and pseudo-random target movement. The simulation experiment results had shown that the simple linear model of human behaviour describes it for a wide area of behaviour. Used linear model of the operator is suitable only for a step response type of target movement. To track a time-varying input signal, a non-linear model needs to be used. The paper concludes with a recommendation for further identification methods for a non-linear model of the human operator.