Research On The Technologies Of Passive Target Automatic Tracking And Fusion

With significant changes in modern tracking environment, stealth to anti-stealth, combat toagainst anti-measures, strong maneuver, high clutter, low detection rates and low threshold andother problems, making tracking systems a strong challenge. In order to observe targets’ morecomprehensive and accurate information, a variety of complex applications of multi-sensor datafusion technologies are bound to be the key way in future wars. Sensor multi-target tracking isthe key technology in data fusion. It combines sensor information together, estimated targetmotion state, against a single sensor to produce superior tracking performance. In this paper,multi-sensor data fusion was considered, it focuses on automatically tracking multiple targets bymultiple hypothesis tracking technology and other data fusion technologies. Sonar is the mosteffective means for underwater detection. However, multi-path, dispersion, and surfaceunevenness uneven seabed boundaries and other marine environmental characteristics, willgreatly influence underwater acoustic propagation; and, accompanying by the emergence of newquiet submarines and other equipment, detected target will vanish soon, sometimes the bearing isvarying quickly, so that the target detection becomes more difficult. These effects require thesonar distance, accuracy and other performance indicators increase accordingly. The traditionalsonar is functionally independent that has little relations among different sonars, such an isolatedsonar system cannot meet the actual needs. Taking full advantage of multi-sensor-based datafusion treatment, it can not only improve the performance of the entire sonar system, but alsothere may be many tasks to achieve a result which single sonar measurements are insufficientand difficult to complete.However, the application of multi-array sonar makes data processingburden become even heavier. As a result, to obtain great benefits by multi-arrays in reasonableapplications, we need to study the development of sonar data fusion techniques.Through reading a lot of foreign papers, based on the data fusion algorithm, in this paper,we study single array multi-target detection and automatic tracking in passive sonar system.According to track fusion algorithm, this paper gives a multi-array associated track fusionscheme. Furthermore, a passive automatic tracking and fusion system was presented. Bytheoretical analysis, numerical simulation and real data analysis, the research work involves:(1) Gives the principle of the multiple hypothesis tracking (MHT) algorithm, based on thenumerical simulation, given the multiple hypothesis tracking plain bearing at low thresholdcondition automatic multi-target tracking results, and gives real data results.(2) Combining gray correlation theory and dual front track fusion algorithm, the multi-arrayfusion algorithm is given through the target track, multi-array associated with multi-target track fusion methods. Theoretical simulation results are given in multi-target tracking fusion systems.According to the target’s position, course, distance, target characteristic parameters and otherinformation, the multi-target fusion system output fusion results, while gives the systemperformance analysis.(3) Combined with single array target detection and automatic tracking algorithm has beenstudied and associated multi-target track fusion array system proposed submarine passive sonartarget tracking system. Each of the system to achieve a single sonar array of independent targetdetection and automatic tracking function, and then get the ministries sonar array to track thetarget position, course, distance, target characteristic parameters and other information into themulti-array target aircraft trace associated fusion system, fusion algorithm by using graycorrelation theory and dual front track, multi-target sound track associated with sonar array, theoutput track correlation results and the associated track the target has been achieved, givenmulti-objective situation map.Through these studies above, we can conclude:(1) Under low threshold target conditions, through simulation and data analysis, bearingsonly multiple hypothesis tracking method not only gets better detection capabilities, but also hasbetter tracking result for cross-track and newfound target in automatic tracking.(2) Through simulation and performance analysis, injecting the target position, course,distance, target characteristic parameters and other information together into the fusion systemgives better tracking results through multi-array data.(3) The passive sonar target tracking and fusion system based on frank array and towed linearray gives a complete result. This tracking and fusion system can effectively improve thepassive tracking accuracy, the reliability of the data, and can obtain the estimation results whichis more accurate and completed than those of a single array.