Optimal exploratory paths for a mobile rover

We consider the problem of maximizing the localization accuracy of a mobile vehicle, based on triangulation measurements derived from optical data. The problem is intrinsically nonlinear, as the linear approximation of the system is not observable. This implies that the choice of inputs (i.e., the path followed) may affect the quality of observations made, and ultimately the localization accuracy. We consider the problem of finding the most informative exploratory path of the given length for a rover (modeled as a point in the plane) with optical triangulation information.