Machine-Learning-Assisted Optimization of Aircraft Trajectories Under Realistic Constraints

What is it about?

We propose to optimize aircraft 4D reference trajectories by developing an A*-variant graph search algorithm with the inclusion of weather conditions, free route airspace and other practical operational restrictions with a wide range of domain-specific heuristics. We use Machine Learning models to quickly estimate the fuel consumption when approximating the heuristic values in the A* algorithm.

Featured Image