基于多元时间序列融合的飞行动作识别方法

针对飞行动作数据随机性强与长度不一致的问题,提出通过减小动态时间规整(DTW)算法的搜索空间,并定义不同特征参数贡献度的概念,实现对飞行数据的多元时间序列融合,从而完成对战术机动动作的识别。通过引入预分类和细分类结合的方式,对动作数据进行预处理,然后根据改进的动态时间规整(WDTW)算法对待测数据进行识别。仿真实验表明,相比传统DTW算法,WDTW算法通过降低算法复杂度,识别计算时间变化明显;对核密度与精准度系数的分析表明识别准确率亦有所提高。实验结果验证了所提方法的准确性。

A Flight Action Recognition Based on Multivariate Time Series Fusion

Aimed at the problems that the randomness and the inconsistent length of flight motion data are not unanimous, this paper proposes a method of reducing the search space of dynamic time warping (DTW) algorithm and defining the contribution of different feature parameters. Then the flight action can be recognized by multivariate time series fusion of flight data. The preprocessing of the action data is introduced by combining the pre classification and fine classification, then the improved dynamic time warping (WDTW) algorithm is used to identify the measured data. The simulation results show that compared with the traditional DTW algorithm, the WDTW algorithm reduces the complexity of the algorithm, and a change of the computation time is obvious; Finally, according to the analysis of the nuclear density and the precision coefficient, the recognition accuracy is also improved. The accuracy and innovation of the proposed method are valid.