In this paper the human control output in the compensatory tracking task is studied under the assumption that it consists of two components: the input and the system component. The former is assumed to be a continuous function of time generated by the input model in operator's brain, while the latter is an intermittent component which takes place, depending on the current state of the error.
In order to clarify characteristics of these components, the control output is subdivided into two phases by comparison of the control output and the ideal control which is capable of nullifying the error.
The quantitative changes of these components in the course of subject's training, plant dynamics variation and input frequency variation are studied experimentally. The following results are obtained:
(1) With the untrained subjects the system component is dominant, while with the trained the input one is.
(2) The input component is well approximating to the ideal control. However the accuracy is dependent on the plant dynamics and the input frequency.
(3) The system component is also influenced by the plant dynamics and the input frequency, so as to be able to compensate for the insufficiency of the input component.
View full abstract