Research On Key Technology In Situation Assessment

In recent years, modern war is becoming more and more complex, wide and huge. These require C4ISR system of the capability for information fusion, which could enable the commander to form appropriate perception, timely and exactly to understand battle field situation, to make decisions rapidly. Situation assessment(SA) is the key technique for high level processing of information fusion, which is the level 2 fusion in JDL model. Situation assessment is the ongoing process of inferring relevant information about forces of concern in a battlefield situation to achieve situation awareness, which is needed by the campaign commanders or analysts to support decision-making. The analysis is made of data from sensors and other sources in SA for building an accurate and timely picture of the battlefield situation, including location, movements, and deployment and intension of enemy force. The main content of the dissertation is outlined as follows:1. On the basis of analyzing the definition of the situation assessment, the main function and the method are discussed of the situation awareness. According to the different type of event in the battlefield, the event detection methods are presented, including the template-based method and the approach based on fuzzy logic. The situation knowledge presentation model based on Bayesian networks is proposed in the paper and the analysis is made of target classification.2. According to the different battlefield environments, the template-based method and the approach based on fuzzy belief networks for force aggregation and classification are presented in the paper. In the template-based method, the nearest-neighbor clustering algorithm is employed to merge targets into groups by position and direction and the corresponding templates are constructed to classify each level group according to the hierarchy for force group. For the ground force group classification, the fuzzy belief networks is utilized to represent the composition and structure of various types of groups, and then fuzzy logic inference is employed to infer the type of group with its attributes.3. The analysis is made of plan recognition in situation assessment. The method for plan recognition based on hierarchical Bayesian networks is proposed. The hierarchical Bayesian networks is utilized to represent the plan. According to the event and target action, the hierarchical Bayesian networks for plan recognition is dynamically constructed. In order to solve the inference for dynamic Bayesian networks, the approach of adding virtue node is proposed. In the approach, the inference result of replacement Bayesian networks is considered as the uncertainty evidence of corresponding node in the higher networks.4. The parameter learning of Bayesian networks is a key problem in the application of Bayesian networks. The problem of parameter learning in situation assessment is discussed in the paper. The method based on Noisy-or gates model in learning conditional probabilities is proposed. In the method, the domain knowledge could be fully exploited , which is useful for the situation assessment system.