Flight Trajectory Prediction Based on Hybrid- Recurrent Networks

The development of future technologies for the National Airspace System (NAS) will be reliant on a new communications infrastructure capable of managing the limited available spectrum for communications among aircraft and ground systems. Emerging approaches to autonomous allocation of aviation spectrum mostly rely on machine learning techniques, where 4D (longitude, latitude, altitude, time) trajectory prediction is an important data input to enable real-time resource allocation. This study explores and evaluates effective data sources and deep recurrent neural network techniques when determining flight trajectories. Specifically, data are collected and evaluated in a 14-day period. Sources of data include NASA Sherlock Data Warehouse, MIT Lincoln Labs Corridor Integrated Weather Service (CIWS), and assorted NOAA weather datasets. Deep learning models for 4D predictions all utilize a hybrid-recurrent technique. A baseline model is considered via the convolutional-recurrent design from the existing literature. The modified design considers Gated Recurrent Units (GRU), Independently Recurrent Neural Networks (IndRNN), and stand-alone self-attention layers. Results indicate the effectiveness of Long-Short Term Memory (LSTM) and GRU cells for state-of-the-art data processing (interpolation). Attention mechanisms provide notable performance improvements to convolutional layers and may extend dimensional capabilities of a learning model. Finally, NOAA measurements provide only a supplemental value by improving predictions along a flight altitude; however, tailored products such as CIWS Echo Top remain the most holistically accurate.