Kestirimci fırlatmaya uygunluk bölgesi kullanarak olasılıksal yol haritaları ve voronoi diyagramları ile yapay zeka tabanlı hava-yer görev planlama

In this thesis, dynamic air-to-surface mission planning strategies based on probabilistic roadmaps and Voronoi diagrams using predictive launch acceptability region approach are proposed for opportunity targets in order to strengthen decision support capabilities of aircraft. Air-to-surface missions are planned in ground support systems and loaded to aircraft before the mission begins. This means that all the waypoints which should be followed during an air-to-surface mission are planned according to various threats and geographical formations. However, opportunity targets sometimes endanger aircraft safety because pilots may be obliged to deviate from planned waypoints in order to destroy the target which is unexpectedly appeared. First of all, threats on battlefields are modeled by ellipsoids, and geographical formations are simulated by geoTIFFs. Then, predictive launch acceptability region queries are modeled, and a strategy is developed to designate a release state. In the first proposed method, a probabilistic roadmap algorithm with Dubins path distance is developed to form a connected graph that will connect the start and the goal states in collision-free space. In the second proposed method, Voronoi diagram is generated according to threats in order to generate a roadmap. The shortest path between the start and the goal state in generated roadmaps is derived by Dijkstra's shortest path algorithm for both of the proposed methods. For Voronoi diagram-based method, a typical algorithm is developed in order to optimize the output of Dijkstra's shortest path algorithm. The optimized path is enhanced according to geographical formations by extracting the maximum envelope of elevation profile of the path using Hilbert transform. Finally, proposed methods are analyzed in terms of convergence rate, mean trajectory length, elapsed time and compared with each other. For the probabilistic roadmap-based method, minimum trajectory length is observed as 6882.8 m, convergence rate is observed between 94% and 100% with number of samples is greater than 6000 and the maximum permitted length of Dubins path between two samples is greater than 450 m, and minimum execution time is observed as 62 s. Mean trajectory length, average execution time, and convergence rate are observed as 7192.6 m, 0.60 s, and 100%, respectively for Voronoi Diagram-based method. Results show that dynamic mission planning can be accomplished for opportunity targets using a predicted release state with a sub-optimal trajectory, admissible elapsed time, and full convergence rate.