基于空战环境认知的态势评估

Situation assessment based on air combat environment cognition

针对现有的态势评估法存在忽略飞行员认知行为的问题,提出一种面向环境认知的态势评估模型。分析了航迹信息的序列性特点,利用当前采集到的航迹信息,通过自适应矩估计法优化的长短期记忆神经网络对敌方航迹预测,不仅提高了航迹预测的精度,而且Adam优化算法动态调整学习率,避免了参数求解陷入局部最优,也在一定程度上提高了算法的收敛速度。通过模仿飞行员的认知行为,将航迹预测的态势信息纳入态势评估的内容,建立基于时间序列的态势评估模型,使态势评估结果更具合理性。采用空战训练中记录的数据进行仿真,结果表明所提方法的评估结果与空战实际相吻合。

Aiming at the problem of ignoring pilot cognitive behavior in situation assessment methods,a situation assessment model based on environmental cognition is proposed.This method analyzes the sequential characteristics of the track information,and uses the currently collected track information to predict the enemy track by the Long-short term memory networks optimized by the adaptive moment estimation method,which not only improves the accuracy of the track prediction,and Adam vector optimizes algorithm dynamic adjustment,it not only avoids the parameters fall into a local optimum,but also improves the convergence speed of the algorithm to a certain extent.By imitating the cognitive behavior of pilots,the situation information of track prediction is incorporated into the content of situation assessment,and a situation assessment model based on time series is established to make the situation assessment results more reasonable.The simulation results using real training air combat data show that the evaluation results of the proposed method are consistent with the actual air combat.