Review of control models for human pilot behavior

Abstract

Pilot modeling techniques, have played a crucial role in manned aviation and have also, as a consequence, generated major advances in human control behavior research. This paper presents a review of control models for human pilot behavior. The purposes of the models are to analyze the characteristics of the pilot-aircraft system, and to provide valuable guidance in the flight control system design. Existing human pilot models are classified into three types. The first category of models, based on control theory, can only be used to describe the control functions of human pilot. The typical models include McRuer quasi-linear models and optimal control models. Then the principle of human pilot control behavior was revealed from the physiological perspective, the Hess structural model being the most common one. Hosman's descriptive model and biodynamic model are also summarized. The paper also proposes that the development of artificial intelligence technology has enabled the study of nonlinear characteristics of pilot behavior in manual control. Models based on intelligence techniques are briefly introduced, e.g., fuzzy control models, neural network models. Finally, the paper outlines several typical and future applications of the techniques in the pilot modeling field. However, as the presented discussion is limited to a small area of pilot modeling, many other challenges remain open for further research and development.