基于改进Elman神经网络的红外被动测距算法研究

为了满足现代战争的需求,对于空中目标的距离精确估计显得愈发重要,由于目标的红外辐射在大气传播过程中的衰减,目标在不同波长的辐射会随着传输距离的改变而发生变化,故不同的波长的辐射里包含了目标的距离信息。基于以上原理,本文利用美国空军大气传输软件Modtran生成空中目标在不同波段的大气透过率数据,利用经纬仪获得目标的天顶角,最后建立基于改进Elman神经网络的被动测距模型。仿真结果表明本文算法能够有效提高对于空中目标距离估计的精确性。

Passive Ranging Algorithm Based on Improved Elman Neural Network

To meet the needs of modern warfare, accurate estimation of the distance of targets in air is becoming increasingly important. Owing to the attenuation of infrared radiation of targets during atmospheric propagation, the radiation of targets at different wavelengths change with the change in transmission distance. Therefore, the radiation of different wavelengths contains information regarding the distance of the targets. Based on the above-mentioned principles, this paper utilizes the Modtran, which is a software used to transfer atmospheric in MARCOR to generate the atmospheric transmittance data of airborne targets in different bands to obtain the zenith angle of the targets using a theodolite, and to establish a passive ranging model based on an improvised form of the Elman neural network. The simulation results indicate that the algorithm proposed in this paper can effectively enhance the accuracy of target distance estimation in air.