基于改进灰狼算法优化的UKF在空战轨迹预测中的应用

Application of UKF Optimized by Improved Gray Wolf Algorithm in Air Combat Trajectory Prediction

针对空战轨迹预测精度不高的问题,提出一种由改进灰狼算法优化的无迹卡尔曼滤波算法(Unscented Kalman Filter,UKF)。首先介绍了UKF算法并对其进行了分析;其次针对传统灰狼优化算法(Grey Wolf Optimizer,GWO)的不足,提出使用基于适应度值的动态权重、根据适应度值变化率调节的自适应控制参数调整、反向多倍中心对称变异策略的改进措施,形成改进的灰狼优化算法(IGWO),并运用IGWO算法对UKF中的滤波参数Q、R进行实时优化;最后,运用所提出的IGWO-UKF算法与GWO-UKF、UKF、BP神经网络算法一起对空战轨迹进行预测仿真,形成三维轨迹图、各方向轨迹图与相对误差图。结果表明,所提算法误差小、精度高,具有良好的预测效果。

Aiming at the problem of low precision of air combat trajectory prediction,an Unscented Kalman Filter optimized by improved gray wolf algorithm is proposed.Firstly,UKF algorithm is introduced and analyzed.Secondly,aiming at the shortcomings of the traditional Grey Wolf Optimizer algorithm,an improved Grey Wolf Optimizer(IGWO)algorithm is formed by using the dynamic weight based on the fitness value,the adaptive control parameter is adjusted according to the change rate of the fitness value,and the improvement measures of the inverse multiple central symmetry mutation strategy,and the filtering parameters Q and R in UKF are optimized in real time by using the IGWO algorithm.Finally,the IGWO-UKF algorithm,GWO-UKF,UKF and BP neural network algorithm proposed are used to predict and simulate the air combat trajectory,forming three-dimensional trajectory map,trajectory map in all directions and relative error map.The results show that the algorithm proposed has small error,high precision and good prediction effect.