Abstract | ||
---|---|---|
The research on knowledge representation and reasoning methods of battlefield situation assessment is of great military value. The Bayesian network describes causalities among variables by graph model, which provides an effective reasoning method for uncertain information based on Bayesian probability theory. The battlefield environment contains a lot of uncertainties. Different battlefield environments may contain different numbers and types of combat entities. Therefore, it is impossible to determine a Bayesian network that satisfies all conditions in advance. The normal Bayesian network model lacks scalability and reusability. Aiming at the battlefield situation assessment problem, this paper proposes a multi-entity Bayesian network modeling method, which divides the problems into target recognition and intention recognition problems, and considers the target radiation source characteristics in the target recognition problem to improve the model's feature fusion ability. The relationships between different entities are described as Bayesian network segments, and the Bayesian network segments are combined to form a multi-entity Bayesian network. So the modeling process is simplified by using the good scalability of multi-entity Bayesian networks. At last, the feasibility and effectiveness of the proposed method are verified by a situational simulation modeling example. |
Year | DOI | Venue |
---|---|---|
2020 | 10.1109/ICCA51439.2020.9264559 | 2020 IEEE 16TH INTERNATIONAL CONFERENCE ON CONTROL & AUTOMATION (ICCA) |
Keywords | DocType | ISSN |
Bayesian network, multi-entity, target recognition, situation assessment | Conference | 1948-3449 |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
6 |
Name | Order | Citations | PageRank |
---|---|---|---|
Guoqing Shi | 1 | 3 | 2.04 |
Junwei Pu | 2 | 0 | 0.34 |
Lin Zhang | 3 | 91 | 24.95 |
Xiutang Geng | 4 | 0 | 0.34 |
Qing Zhou | 5 | 454 | 49.57 |
Yahang Zhao | 6 | 0 | 0.34 |