|M.Sc Student||Ben-Yishay Lior|
|Subject||Minimizing the Number of Affected Aircraft in Optimal Air-|
|Department||Department of Aerospace Engineering||Supervisors||Professor Gil Iosilevskii|
|Professor Moshe Idan|
|Full Thesis text|
Automatic air-traffic control managing aircraft prior and during flight is one of the main issues taken care of in the world of civilian flights all over the world. Currently the air traffic reaches a few thousands of aircraft airborne simultaneously above Europe or US, with predictions that it will grow to an order of ten thousands in the very near future. It is expected that all this traffic will be optimized well in advanced based on the best available weather forecast, providing a conflict-free flight plan for each aircraft. This is a time consuming process that may take hours of computing time. Flight plan deviations, caused, e.g., by the difference between the actual and predicted weather, equipment malfunctions, operational and airport delays, may create in-flight conflicts that need to be resolved in a matter of minutes. Due to the high computational complexity, re-optimizing the entire traffic is not practical for this task. In this work we suggest a rapid, suboptimal trajectory re-planning algorithm. Its efficiency is attained by re-computing new trajectories only for a small, possibly minimal, subset of aircraft still resolving all the conflicts. Finding optimal trajectories only for those aircraft drastically reduces the computational time. Depending on the size of this set, it can usually be carried out in the matter of minutes, even at the current computing capabilities.