Threat Estimation Method Based on Dynamic Bayesian Network

Abstract

The invention discloses a threat estimation method based on a dynamic Bayesian network, and relates to the technical field of situation estimation. The implementation steps of this method are: 1. Collecting and sorting data, extracting situation elements; 2. Synthesizing multiple situation elements, establishing a dynamic Bayesian network topology; 3. Learning and setting network parameters; 4. Realizing fast approximation according to Markov property Bayesian inference; 5. Merge the probabilities of each threat level into a continuous threat index and a discrete threat level; 6. Output threat estimation results. The invention can synthesize multiple situation elements, perform reasonable and intelligent reasoning and analysis, realize the quantitative and qualitative estimation of the threat dynamics of blue square group targets, and can be used in situation estimation and command and control systems.

Description

technical field [0001] The invention belongs to the technical field of situation estimation, in particular to a threat estimation method based on a dynamic Bayesian network, which can be used in situation estimation and command and control systems. Background technique [0002] Today's regional conflicts are characterized by diverse objects and complex environments. Faced with a sharp increase in the amount of observation data, if manual processing is still relied on, the timeliness and consistency are difficult to meet actual needs. Therefore, it is necessary to use the advantages of computer storage and computing to deal with a large number of recurring regular situations, so as to reduce the workload of the commander and enable him to grasp the real-time dynamics more quickly and effectively. Among them, the threat analysis is based on the extracted situational elements, reasoning and analysis of the threat level of the blue party in the environment, so as to provide a re...

AI Technical Summary

Problems solved by technology

The defects of the existing methods mainly include: (a) only consider the static relationship between each factor and the threat, and fail to reflect the continuous change characteristics of the threat in time; (b) estimate the threat between a single target, and the actual target is usually Tasks are executed in the form of formation group targets, the threat between individual targets does not take into account the number of targets, and it is difficult to effectively reflect the real threat situation; (c) traditional threat level estimation, usually when the probability of a certain level is greater than the threshold, will its final result

However, when the probabilities of each level are relatively close, it is difficult to set a reasonable threshold for this method, resulting in unstable output

Images