A New Kind Dynamic Bayesian Network and Application to Threat1 Assessment
Liu Zhe · Fire Control and Command Control · 2014
Aiming at the shortcoming of traditional Bayesian network that used to threat assessment,a new threat assessment method based on varied- structure interval probability DBN(VSIP-DBN) is proposed. The basic idea and definition of VSIP-DBN is proposed,the inference algorithm is also given. The threat assessment VSIP-DBN model can be constructed in view of the air combat situation and air combat capability. The observation data can be used to infer using the VSIPDBN inference algorithms,and the inference result is just the threat level. The former methods of threat assessment must get the accurate observation data of the target and the accurate model parameter,the simulation results show that the threat assessment model based on VSIP-DBN can get the threat level even with inaccurate model parameter and incidental observation error.