ABSTRACT
This chapter describes a model for the active perception and control of visually-driven egomotion. The model is founded in modern estimation and control theory, and is motivated by past and current studies of piloted flight control, particularly low-level terrain-following (TF) flight, an intense and demanding visually-driven task. To provide a rationale for the overall model structure presented here, we first described five candidate submodels we consider central to understanding TF flight. We then describe an overall model architecture which serves to integrate these submodels, and which provides the structure for a more generic estimation/control model of egomotion. We then close with three applications of the model, to demonstrate the model’s utility in understanding visually-driven active control behavior.
