ON THE DESIGN OF ESTIMATORS WITH HIGH BREAKDOWN POINTS FOR OUTLIER IDENTIFICATION IN TRIANGULATION NETWORKS

For detection of gross errors in processing triangulation networks, th is paper introduces the principle of designing robust estimators with high breakdown points based on the median approach. Three examples pre sented in the paper show how to form the combinations of observations while considering geometrical constraints necessary for computing the median estimate, and how to calculate the breakdown points as a measur e of global reliability of the estimate.