Research On Multi-Source Fusion Target Tracking Algorithm Based On Optical Measurement

With the rapid development of science and technology,many targets include aircraft,missiles and ships have stronger maneuvering ability,more maneuvering styles and thorough countermeasures.The problem of maneuvering target tracking has been transformed from simple moving targets in a non-interference environment to complex maneuvering targets in clutter environment.Therefore,based on the optical measurement information,the multi-source fusion target tracking algorithm has been researched,the maneuvering target tracking model has been established,the tracking precision of different filtering algorithms and different target motion model have been contrast,and the Multi-source data fusion target tracking algorithm has been given.The main content is as follows:Firstly,the thesis researched the model of maneuvering target tracking,gave the target motion models and measurement models.Considering the actual situation of engineering application,the constant velocity(CV),constant acceleration(CA),Singer,"current" statistical and coordinated turn models which are commonly used is established in the thesis.According to the principle of active and passive optical sensors,the relationship between measurement and state is determined,and the measurement model is established.Moreover,a target tracking algorithm which based on the least square method is designed,and the simulation demonstrated this algorithm.Secondly,variety of nonlinear filtering algorithms is studied.By contrasting the principle of EKF,UKF and HCKF,the main steps and equations are given.In addition,by making use of the control variable method,the Monte-Carlo simulations are conducted to contrast the precision of different filtering algorithms and different target motion models.Based on the simulation results,the most precise nonlinear filtering algorithm and motion models are determined.Thirdly,the interactive multiple model(IMM)which based on the HCKF has been given in the thesis.In consideration that single model algorithm cannot track the complex maneuvering target precisely,and HCKF has the best precision in approximate nonlinear system.Therefore,the thesis embed the HCKF into the IMM method,and the precision in tracking complex maneuvering targets is improved.Finally,the IMM-Probabilistic Data Association(PDA)is researched.In consideration that PDA is good at tracking targets in clutter and IMM is good at tracking complex maneuvering target.Therefore,these two algorithms are combined together,and the IMM-PDA algorithm is given.By making use of the IMM-PDA,the tracking precision of complex maneuvering target in cluster is improved.Moreover,the simulation is conducted and the algorithm is demonstrated.