Unlocking air traffic flow prediction through microscopic aircraft-state modeling
Researchers have developed AeroSense, a novel approach to predicting air traffic flow in terminal airspace. Unlike traditional methods that aggregate traffic data into time series, AeroSense models future flow directly from the instantaneous states of individual aircraft, derived from ADS-B trajectories. This microscopic aircraft-state modeling preserves fine-grained dynamics and accommodates varying traffic densities without relying on historical look-back windows. Experiments on real-world data demonstrate AeroSense's superior predictive accuracy and robustness, particularly during high-density traffic periods, suggesting it as a promising alternative to conventional forecasting techniques. AI
IMPACT This approach could enhance the efficiency and safety of air traffic management by providing more accurate short-term flow predictions.