Hierarchical simulation for complex domains: air traffic flow management

摘要: A key element in the continuing growth of air traffic is increased use automation. The Next Generation (Next-Gen) Air Traffic System will include automated decision support systems and satellite navigation that let pilots know precise locations other aircraft around them. This Next-Gen suggestion system can assist making good decisions when they have to direct themselves. However, effective automation critical achieving capacity safety goals System. In this paper we show evolutionary algorithms be used achieve it not feasible a standard algorithm learning approach such detailed simulation. Therefore, apply hierarchical simulation an congestion problem where agents must reach destination while avoiding separation violations. Due dynamic nature problem, need learn fast. low fidelity for their destination, high employing technology assurance. increases convergence rate, leads better performing solution, lowers computational complexity by up 50 times.