Information Maps for Active Sensor Control

This paper outlines our current progress in active sensor control. We consider the problem of controlling a nonlinear observation system observing data implicitly related to parameters of interest. We show how linear estimation theory can be applied to this problem, and develop the notion of an information map showing the information expected from sensor viewpoints. We discuss the robustness of these techniques, and propose a method to enhance their robustness. We expect these maps to be useful in active sensor control.