Research On Air Target Situation Assessment Method Based On Trajectory Prediction

The battlefield environment is changing rapidly.For incoming air targets,how to obtain useful information from massive data to help commanders perceive the battlefield situation in a timely and accurate manner and grasp the battlefield initiative has become an urgent problem to be solved.The perception system learns the intention and threat of the target in advance based on the radar measurement data,and assists commanders in making decisions.It has important military value and research significance.In order to obtain the battlefield situation quickly and accurately,this paper studies air targets from three aspects: trajectory prediction,intention recognition,and threat assessment.First,predict the trajectory of the incoming air target.When the target is maneuvering,the prediction accuracy of the traditional unbiased gray scale theory is greatly reduced.The error data set of the unbiased gray-scale prediction model is established according to a large amount of flight trajectory data,and the BP neural network is used to train the data set to obtain the prediction error network model.The unbiased grayscale prediction model is used to predict the trajectory of the air target at the current two moments to obtain the initial value of the position at a time in the future,plus the BP neural network to obtain the trajectory prediction error,and obtain the final trajectory prediction value of the target,and compare the unbiased The prediction accuracy of the gray prediction model and the improved unbiased gray BP neural network prediction model proves that the improved algorithm reduces the prediction error.Secondly,infer the target intention,establish a multi-input single-output Mamdani fuzzy inference model,formulate the target intention fuzzy inference rules,and build a fuzzy inference system.Based on the predicted trajectory,parameters such as speed,distance,height,and steering trend are selected as the indicators of the fuzzy inference system,and the fuzzy inference membership function of each index is determined according to the target combat attributes,and the fuzzy inference rules formulated can be accurate through the fuzzy inference system.Infer the intent of the air target.Then,threat assessment is performed on the target,and the threat assessment algorithm based on the improved dynamic Bayesian network is studied.Select the corresponding threat index through the predicted trajectory information,construct the index Gaussian membership function to reduce the subjectivity of the value of the target attribute state,determine the dynamic Bayesian network node and value status,and consider the weight of the target’s individual attribute to the threat degree The influence of calculation is discussed,based on the weight of information entropy-dynamic Bayesian network threat assessment calculation.The simulation experiment compares the target threat degree obtained by the improved algorithm and the traditional algorithm,and proves that the result obtained by the improved dynamic Bayesian network is more in line with the battlefield situation.When the target’s single index probability state increases,the target’s threat degree will also increase.Finally,an air target situation assessment system is designed based on QT software.The system includes three modules: trajectory prediction,intent identification,and threat assessment.The target’s trajectory and the simulation results of the three modules are shown to verify the overall performance of the situation assessment system.The system improves the commanders’ real-time perception of the air battlefield,and provides an important reference for target allocation and firepower strike decisions.