Determination and visualization of uncertainties in 4D-trajectory prediction

Future air traffic management (ATM) relies on trajectory forecasting for higher safety, efficiency, and environmental friendliness. The quality of trajectory predictions is subject to inaccuracies from a variety of sources [1]. This impreciseness in performance and environmental parameters cause inaccuracies in the calculations of flight trajectories. Therefore, the implementation of future trajectory-based air traffic management procedures requires decision support tools as well as reporting tools to analyze and predict uncertainties in actual flight paths. This paper presents a concept to detect uncertainties based on air traffic simulations, and to visualize the results in a quantitative (diagrams) as well as a qualitative way (geospatial imaging). For the representation of scalar errors, charts generated by JFreeChart-Tool [2] are suitable. NASA World Wind [3] is used for the geographical visualization of trajectories and their errors. Within this paper we describe the process of evaluating the accuracy of a trajectory concerning the predicted flight path. Furthermore, the paper provides an overview of the most important factors for trajectory prediction, which are taken into account for the theoretical concept of the paper. The selected factors are meteorological parameters (wind, temperature), aircraft performance (weight, speed), and navigation performance. To compute the uncertainty of trajectory prediction we develop a metric that classifies the errors by four dimensions. For the investigation of the effect of parametric errors a simulation with over 2,000 flights under various environmental conditions has been implemented. For the validation of the method the study applies error analysis to the simulation results. With the developed tool it is possible to analyze the ramifications of uncertain prediction of trajectory on a single flight, as well as on the complex ATM system.