Abstract
Response characteristics of human operators in manual pursuit tracking with auditory input are investigated. The human operator hears through his left ear a sound whose frequency varies in proportion to an external random signal. At the same time, he hears through his right ear another sound whose frequency varies in proportion to the angle of the control lever of a potentiometer. The operator controls the angle of the lever so that the frequencies of the sounds in both ears remain as close as possible. The dynamics of the human operator is studied by assuming a 'man-machine system' whose input is the external signal and whose output is the voltage of the potentiometer. A learning identification method proposed by one of the authors is used to calculate the weighting function of the man-machine system, which is displayed on a CRT screen in real time.
Results of the experiments are summarized as follows;
(1) The shapes of the weighting functions of the auditory tracking systems are classified into two modes-the stable mode corresponding to smooth control, and the unstable one corresponding to awkward control.
(2) Smooth tracking controls are further divided into fine tracking and rough tracking.
(3) There are significant differences in tracking performance between subjects.
Comparisons of the results obtained from the auditory tracking system with those of visual and combined tracking systems show that (1) the visual tracking system is more stable than the auditory system only for relatively poor trackers, and (2) the response of the combined tracking system is most stable and rapid for all subjects.
