Genetic algorithms for solving air traffic control conflicts

摘要

Shows how genetic algorithmic methods can be used to solve air traffic control (ATC) conflicts. They authors compare these methods, (in order to validate their solutions), to more classical methods, such as graph searching with A* and its related algorithms, or simulated annealing. It is shown that genetic algorithms are perfectly suited for ATC conflict problems, as they are able to give many different solutions for a given problem, and they provide an almost optimal solution in a very short amount of time, with the possibility of enhancing the solution later if more time is available, which is a critical property for ATC systems.