Trajectory Prediction Accuracy Report: User Request Evaluation Tool (URET)/Center-TRACON Automation System (CTAS)

摘要

This report presents the results of an independent analysis of the accuracy of the trajectory modelers implemented in me User Request Evaluation Tool (URET) and Center-TRACON Automation System (CTAS) prototypes. These results are based on the completion of the first phase of a planned two phased effort. As originally envisioned, efforts during Phase I would develop a generic methodology to measure the trajectory prediction accuracy of any decision support tool (DST), which would be validated by applying it to CTAS and URET based on their currently adapted sites. In Phase II, the methodology would be applied to URET and CTAS adapted to a common site and supplied with the same scenario. As such, the results from Phase I would have provided a common set of results based on the same site and scenario, allowing a comparison of the two trajectory modelers to be made, in support of research into the performance requirements for a common en route trajectory model. Due to funding cuts, this task was curtailed to the completion of Phase I. The results from this phase do provide the FAA with an independent set of scenario-based trajectory accuracy statistics for each DST, but they cannot be used to compare the two DSTs due to the confounding site-specific factors. A methodology was developed and CTAS and URET were measured based on one scenario each from their currently adapted sites (Fort Worth and Indianapolis, respectively). The Phase I study measured the spatial error between trajectory predictions versus the Host Computer System (HCS) track position reports, which were assumed to be the ground truth location of the aircraft. The spatial error consisted of horizontal and vertical errors. The horizontal error was further partitioned into two geometric components, lateral and longitudinal errors, representing the cross track and along track prediction errors. The focus of the analysis was on the overall trajectory accuracy of each DST, not on individual errors.