采用离散动态贝叶斯网络的直接推理方法作为基础,提出弹性变结构离散动态贝叶斯网络的概念,构建了空战目标识别的弹性变结构离散动态贝叶斯网络模型,给出了相应的推理算法,以此克服了离散静态贝叶斯网络和定结构离散动态贝叶斯网络在目标分类识别过程中出现的问题。通过仿真结果对比,表明该方法可以综合各个时刻各种被观测到的确定和不确定信息,从而更为有效的实现目标的分类和识别。 Based on the direct inference method of discrete dynamic Bayesian network, the concept of discrete dynamic Bayesian network with elastic variable structure is proposed and a discrete dynamic Bayesian network model of elastic variable structure for air combat target recognition is constructed. The corresponding reasoning Algorithm, so as to overcome the problems of discrete static Bayesian network and fixed structure discrete dynamic Bayesian network in the process of target classification and recognition. The simulation results show that the proposed method can synthesize all kinds of observed and uncertain information at different moments so as to achieve more effective classification and identification of targets.