Advanced trajectory modeling: Use of aircraft-derived data in ground automation

In the current air traffic management environment, flight trajectories with varying prediction fidelities are used by ground and aircraft automation systems to manage aircraft movement from gate-to-gate. These automation systems utilize multiple information sources, including static look up tables for aircraft characteristics and performance parameters, to each generate their trajectory predictions. As a result, there are differences in the trajectory predictions produced by different systems. Differences in predicted trajectories inherently lead to inefficiencies for both the airspace users and air traffic management. The Advanced Trajectory Modeling project leveraged innovations in the digital environment and aviation technologies to explore the potential of air-ground information exchanges to enable the use of flight-specific information to improve the ground automation trajectory modeling capabilities. The project activities reported here focused on using aircraft-derived data for ground automation trajectory predictions in support of arrival metering operations. A small set of data was derived that performance-based Flight Management Systems (FMS) could provide and that trajectory predictors could use; namely, aircraft mass, top of descent location, and descent speed schedule. A second small data set was derived that geometric-based FMS could provide and trajectory predictors could use: ground speed, flight path angle, and descent speed at the time of vertical navigation mode engagement. An additional, more comprehensive data set was also used, a Trajectory and Speed Profile, that was similar to the Extended Projected Profile report augmented with the descent speed profile. The trajectory predictors of Time Based Flow Management (TBFM) and En Route Automation Modernization (ERAM) were used in this research. The project team conducted simulations in which these aircraft-derived data were provided by operational FMS and incorporated into the ERAM and TBFM trajectory predictors. The aircraft-derived data had a sizable impact on the accuracy of ground automation trajectory predictions when compared with trajectory predictions computed without the aircraft-derived data, improving top of descent location prediction accuracy as well as estimated time of arrival at the meter fix. This paper discusses the use case scenarios, simulation architecture environment, analysis methodologies, assumptions and limitations, and results of this research.