摘要
针对D-S证据理论应用于态势估计过程中,基本概率赋值函数难以确定的问题,提出一种基于神经网络获取基本概率赋值函数再进行证据组合的方法。首先,在传感器探测和0级处理、1级处理的基础上,将各时刻传感器所探测目标的状态、属性、类别、事件等提取出来;然后,经过预处理后利用BP神经网络进行敌目标意图分类识别;最后,将神经网络的输出作为基本概率赋值,采用D-S证据理论来完成决策级融合,实现对海战场的战术态势估计。通过算例仿真,结果表明该方法是一种有效可行的态势估计方法。
Considering the difficulty in determining the Basic Probability Assignment Function(BPAF)in the application of evidence the- ory in naval battlefield situation assessment, a new method of evidence combination based on neural network acquisition probability assign- ment function is proposed. Firstly, the state, properties, categories and events of the probe target are extracted based on the sensor detection, 0 processing and 1 processing. Secondly, the intentions of the enemy target are classified and identified with BP neural network after the pre treatment. Finally the outputs of the neural network are as the basic probability assignment. D-S evidence theory is used to complete the decision level fusion to estimate the battlefield situation. In simulations,the result shows that the method is an effective and feasible method for naval battlefield situation assessment.
出处
《舰船电子工程》
2013年第8期50-52,134,共4页
Ship Electronic Engineering
关键词
态势估计
证据理论
神经网络
信息融合
situation assessment, D-S evidence theory, neural network, information fusion