Impact of Data-Link on Ground-Based Trajectory Prediction Accuracy for Continuous Descent Arrivals

The purpose of this paper is to assess the improvements in trajectory prediction accuracy that result from enhancing the quality of the input data and supporting models of a groundbased Trajectory Predictor (TP). Focus is on the prediction of continuous descent trajectories where improvements to the quality of TP input data and models have been applied in a step by step manner in order to analyse the sensitivity of the prediction accuracy to each of the enhancements. As a first static enhancement aircraft intent within the Terminal Area is accurately modelled with use of the Aircraft Intent Description Language (AIDL) based on standard arrival and airline procedures. The dynamic enhancements are related to information likely to become available to ground-based systems through future flight planning and data-link standards like the Flight Information eXchange Model (FIXM) and Automatic Dependent Surveillance Contract (ADS-C) Extended Predicted Profile (EPP) respectively. This information includes descent speed schedule, aircraft mass, and geometric descent profile. The study used data from actual continuous descents performed by Future Air Navigation Systems (FANS) 1/A data-link equipped aircraft into Melbourne, Australia. While the current FANS-1/A standard does not directly provide information as descent speed schedule and mass as part of the down-linked trajectory information from the Flight Management System (FMS), these parameters were successfully inferred. Results indicated that descent speed schedule is most significant for temporal accuracy; the additional use of the aircraft mass mainly improved the prediction of the Top of Descent position. A technique was demonstrated to fully synchronise the geometric descent profile between air and ground predicted trajectories, while also improving on the prediction of the time dimension with respect to the original down-linked data from the FMS. While future data-link standards as EPP will provide key information as descent speed schedules directly without the need for inferring, it was shown how FANS-1/A can be successfully used to generate ground-based predictions of higher accuracy than the FMS. This is important both for quick deployment but also for the transition period where mixed equipage will be the rule.