Comparison of point matching algorithms for the UAV aerial refueling problem

This paper describes the results of a study focused on the evaluation of the performance of specific algorithms for the “point matching” task within the general problem of applying machine vision-based control laws for the problem of aerial refueling (AR) for UAVs. Two different point matching algorithms for the identification of the corner-points of the tanker are proposed. A detailed study of the algorithms is performed with special emphasis on the correct matching and required computational effort. The results show the importance of a correct point-matching scheme for obtaining accurate pose estimation; furthermore, the analysis highlights the tradeoffs involved in the selection of the appropriate algorithms.