基于半监督学习的舰艇目标威胁判断研究

Research on the Target Threat Evaluation of Warships Based on Semi-supervised Learning

舰艇目标威胁判断是根据目标态势判断对舰艇的威胁程度,是指挥辅助决策的重要依据之一。传统的目标威胁判断方法存在信息利用率低、判断时效性差等不足。针对舰艇作战信息环境的特点,引入半监督模糊识别理论,建立舰艇防御作战中目标威胁等级判断模型。在训练阶段,利用积累的典型目标态势样例、离线学习确定目标威胁判断模型的参数;在测试阶段,利用离线学习好的模型,实时在线判断目标的威胁等级;最后通过示例说明了半监督模糊识别方法在舰艇目标威胁判断方面的可行性。

Threat evaluation of targets for warships,one of the important criteria for command aid decision making,is the judgement of threat level based on target situation. Traditional methods for estimating submarine threats are insufficient in utilizing information and lack of effectiveness. In this paper,the semi-supervised fuzzy recognition method is introduced based on the combat information environment of warships to build a threat level judgement model for warship defense operations. In the training phase,the typical samples of target situation are used to identify the parameters of the target threat judgement model. In the testing phase,the trained model offline can predict targets’ threat level online. Finally,the validity of the semi-supervised fuzzy recognition method in threat evaluation of targets for warships is illustrated with some examples.