Application of system identification to modeling the human controller under stress conditions

This short paper discusses the development of realistic models for human controllers in man-machine systems that can be used to predict changes in overall system performance under stress conditions. An input-output pursuit-type feed-forward model for the human controller is identified. Parameter-identification under normal and stress conditions is carried out using a maximum likelihood procedure. Results presented show that the identification algorithm is very successful in estimating the parameters of the model. However, it is not possible to make significant inferences about performance decrements due to stress because of an insufficient data base.