Adaptive Control Model for
the Human Eye Movement System

Terry Bahill
Systems and Industrial Engineering
University of Arizona
Tucson, AZ 85721-0020
terry@sie.arizona.edu
© 1998-2004 Bahill

Open loop experiments were used to develop a linear model for a physiological system. The specific system studied was the eye movement system: however, the technique presented should generalize to other physical systems. Human smooth pursuit eye movements were measured in response to sinusoidal, step, ramp, and step-ramp target motions in the normal closed-loop condition and in the open-loop condition. The human responses were compared to the outputs of four models; the best match was provided by the K/(#tau#s + 1) model. Simulation results suggested that in the open-loop condition, the human often changed control strategy, for example by turning off the saccadic system and making no position correcting saccades, in spite of large positional errors.

In order to overcome a time delay and track with zero latency, like the human, our model for the human smooth pursuit eye movement system had to predict target velocity and compensate for system dynamics. The model accomplished this using a least mean squares prediction algorithm. To help validate the model, a sensitivity analysis and parameter estimation study were performed.

References [30, 31, 36, and 37]. This lecture on modeling is suitable for biomedical, systems and control engineers. It requires an overhead projector and a 35 mm slide projector. This lecture takes one and a half hours.