Flight Path Plan Research Based On Genetic And Simulated Annealing Algorithm

In the modern war, low altitude,super low altitude break defense by the UAVs(unmanned aerial vehicle) and the LACM(low altitude cruise missile) ,and the defense of them has become a focal point of the attacker and the defensive. The main component of the mission planning of the low altitude aircraft is to choose a flight path which makes the vehicle has the best survivability and strike effectiveness: the flight path must satisfy the terrain following,threats avoiding and the navigation requests, to make sure that the low altitude flight vehicle has the higher battlefield survivability and strike accuracy. Flight path planning is a very important part of the mission planning system, so the research on the flight path planning algorithms is well worth both on theoretic and practical.The GA(Genetic Algorithms) is a very effective method to deal with large scale and complex problems. The biggest feature of GA is the parallelism and the global searching, so it's very suitable for the multi-goals problems like flight path planning. But the obvious weakness of the basic GA is that in the beginning, the searching is fast but very slow in the ending, moreover, it always falling into local optimum. Because of the above reasons, there were varieties of improved three dimensional flight path planning methods based on GA. But they were difference from the presentation of planning environment,the method of deal with constrain condition,the produce of air line and the performance of the calculation.This Paper compared some of the basic flight path planning algorithms. And corresponding to the low-altitude aircraft flight path plan problem, I adopted an 3-demension flight path plan algorithm which combined the Simulated Annealing Algorithm and the Genetic Algorithm. In the modeling process, we integrated the DEM(digital elevation model) data and every kind of threats and the aircraft's flight performance . Based on the characteristics of low attitude penetration and the aircraft's flexibilities, we designed an effective genetic operator and use the advantage of the Simulated Annealing which can close to the global optimum solution, to suppress the Genetic Algorithm's weakness that liable to converges to the local optimum solution.At the last part of this paper, we did the simulation using the real DEM data and the improving GA method that I have raised in the beginning of this paper. The result indicates that this algorithm's suitability is very well and could satisfy the requiring of terrain following and threat avoiding in low attitude penetration missions.