Research On Trajectory Planning Of High Survival Possibility For Unmanned Combat Air Vehicle

In modern warfare, Unmanned Combat Air Vehicle (UCAV) will face the threat of air-defense systems, which consist of early warning radar, interceptor aircraft, surface to air missiles, anti-aircraft artillery and other weapons systems. In particular, as the rapid development of remote sensing and detection technologies, the detection range of air defense systems, shooting accuracy and anti-jamming capability are rapidly improved, UCAV will face increasingly serious threat. It is very important to produce low-visibility trajectory, so that the survival probability of UCAV in the defense penetration combat is improved.Reduce the detection probability by enemy radar can improve the survival probability of defense penetration effectively. In this thesis, two approaches have been proposed to produce feasible trajectories so that UCAV can reduce the chances of being detected. One is the method of trajectory optimization, which enables UCAV to face the enemy radars with the RCS as small as possible. The second approach is the one which considers battlefield terrain environment, thus the generated trajectory is distributed in the blind area of the enemy radar.The route planning issues with high survival rate are analyzed. The planning optimization goals, including radar detection range, radar detection probability, are determined. The restriction conditions of planning, including the performance bound of the aircraft and battlefield environment restriction, are established. Optimization objectives and quantitative calculation methods of constraints are both analyzed for further modeling calculation.The method of trajectory optimization is studied. The applications of genetic evolutionary algorithm in the route planning are analyzed. The definitions of coding mode, selection operator, cross-operator and mutation are also analyzed. The optimization objectives of low-visibility trajectory and constraints are modeled. The genetic algorithm-based UCAV route planning algorithm is designed and realized. A series of typical examples are designed to test the feasibility of the algorithm.The terrain environment of the battlefield is studied, and the low altitude penetration route planning method, which is based neural networks, is put forward. The spread order of potential field is determined. The numerical potential field based on neural network is set up to achieve a pseudo-parallel method. In this method, the price and the safety of flight are comprehensively considered, and the potential field with single peak is established to ensure the global optimum of the trajectory.Finally, an experimental system is established to test the trajectory. In order to ensure the correctness of the planned trajectory, the paper presents the use of visualization tools to check it. On this basis, an improved method of interpolation by the use of B-spline-face is put forward to interpolate among discrete detection points, and thus a continuous RCS detection range surface is formed so as to improve test accuracy.