A Time-Varying Approach to the Modeling of Human Control Remnant

Previous models for human operator control performance have shown that such control possesses stochastic characteristics. Even with a repetitive task and a highly trained and motivated subject, there is a clear variability of output (control actions) by the subject. This variability, termed remnant, has in the past been modeled as observation and motor noise associated with the operator. In the present work, a new approach is taken: remnant is modeled as a time-varying operator characteristic associated with changes in the relative weighting of various performance measures and lack of understanding of the dynamics of the plant to be controlled. The model uses as a foundation the well-developed theory of linear state regulators, and produces results which agree well with several experiments.