Research On Detection And Tracking Of Surface Target For Salvage

The research on target detection and tracking method for Salvage has very important application and theoretical significance to achieve efficient and safe rescue ships to achieve the goal of the rescue of the surface. Measurement data which is obtained by the sensor that relative position of the rescue ship on the surface of the water surface need to be processed to remove noise and so on, in order to provide more accurate input for the dynamic positioning control system. Because the occurrence of sea area is poor sea conditions, the relative position of the failure target of the crash is lost in real-time. In this paper, some key problems in on detection and tracking of surface target for Salvage are studied.Firstly, through the knowledge of the reference coordinate system and surface ship kinematics, it established spatial kinematics equation of surface ship. The marine environment is established, and the mathematical model of wind, wave and current is established. The mathematical model of the target of the rescue ship and the water surface is simulated and verified. The accuracy of the mathematical model of the target of rescue ship and water surface is verified.Secondly, the detection scheme of the target is presented. The sensor coordinate system is established. The performance of RadaScan Mini relative position measurement system is verified by experiments. Reliability and measurement accuracy of RadaScan Mini relative position measurement system are verified. Then, the azimuth estimation scheme of the target is presented, and the method of polar coordinate method, pure azimuth angle method and pure distance method based on the relative position measurement system based on RadaScan Mini is proposed. Through theoretical analysis and verification tests, it is proved that these three kinds of target estimation schemes are reasonable and feasible. Finally, the failure of the measurement system which is likely to occur in the surface target measurement system is analyzed.Thirdly, sensor measurement datas by wavelet analysis threshold denoise processing.Filtering estimation of data containing "colored noise" by particle filter, and the simulation results show that it is effective. Finally, the feasibility and reliability of the algorithm are verified by the test data of the relative position measurement system based on RadaScan Mini.Finally, in the sensor group occurred failure of part of the sensor measurement data by improved based on confidence distance data fusion algorithm will fail the sensor measurement data to eliminate out, finally realize a short time in the sensor failure of water crashed target tracking. In the case of complete failure of the measurement data in the sensor group, a water surface target tracking algorithm based on data fitting is proposed. And through the verification based on Voyager IV of the autonomous mobile platform of the comprehensive test, finally get a short time in the sensor failure, respectively, through improved based on confidence distance data fusion algorithm and based on data fitting surface target tracking algorithm implements the water crash moving target tracking.