Model Based Vision for Aircraft Position Determination

Passive vision techniques to estimate aircraft position during landing can be developed using a runway model, images acquired by onboard imaging sensor, orientation information provided by the inertial navigation system and the position estimate provided by devices such as the global positioning system. In this paper expressions are derived for the sensitivity of point features to errors in the position and orientation of the sensor. Using these, the sensitivity of the sensor image to aircraft position and orientation errors along a typical landing glide path is studied. A least squares technique based on this sensitivity analysis is described for the correction of position and orientation estimates.