A Study Of Situation Reasoning Based On Intelligent Computing

Situation assessment is the advanced fusion process in the field of information fusion.It is also the key theory and technology on the artificial intelligence research.Situation assessment is essentially a reasoning process on decision level,which tries to sense the present battlefield situation,gradually recognizes adversarial intention and planned deployment,and provides the basis and support for the command and control decision-making.It is of great significance to promote the modernization of equipment,improve the ability of intelligence analysis and enhance the effectiveness of the military.The thesis does a research on the theories of situation assessment and the reasoning algorithms.The contents of the thesis are outlined as follows: 1.The thesis studies the function for fulfilling situation assessment.A three-level function processing model which consists of situation perception,situation recognition and situation projection is set up and each level is analyzed in detail.Simultaneously,the situation ming analysis technology and the reasoning algorithms are discussed.2.For the problem of single-objective situation reasoning,a situation reasoning algorithm based on rule tree and a situation reasoning algorithm based on BP neural network are proposed.In the situation reasoning algorithm based on rule tree,the establishment of situation rule tree and the realization of reasoning algorithm are introduced emphatically.In the situation reasoning algorithm based on BP neural network,situation elements are analyzed,the track preprocessing module is added to establish the attribute set of situation elements,and the realization of reasoning algorithm is introduced.On the basis,an adaptive reasoning system is proposed by combining the rule tree and the BP neural network,which makes the situation reasoning rule change with the change of the battlefield environment and improve the accuracy of the reasoning algorithm.3.For the problem of synergistic situation reasoning,multi-agent technique is applied to situation assessment and a reasoning model based on multi-agent network is proposed.For the limitations of traditional search algorithms in dealing with multi-agent networks,the particle swarm optimization and the evolutionary algorithm are used to solve the multi-agent networks,and the advantages and disadvantages of the two algorithms are analyzed.On this basis,an improved evolutionary algorithm is adopted by combining the two algorithms.The simulation results show that the improved evolutionary algorithm not only has faster convergence speed to meet the real-time requirement of situation reasoning,but has higher reliability to satisfy the accuracy requirement of situation reasoning.