A maximum likelihood estimation approach for image based target localization via small unmanned aerial vehicle

Abstract

To improve the accuracy and the robustness of the image based target localization for the small unmanned aerial vehicle (UAV), a maximum likelihood estimation (MLE) approach is proposed. A Monte Carlo method is used for estimating the error information of the tradition localization method. After retrieving the distribution parameters from the Monte Carlo simulations, the maximum likelihood estimation is then applied to acquire the final estimation result based on two traditional localization measurement results. Flying tests show that the MLE method could achieve a better result than the traditional method and a significant improvement on the robustness.