The Underwater Moving Target Tracking Technology

Target tracking remains a hot science field internationally as it plays a vital role to obtain accurate target quickly and precisely in military,and even to determine the success of a campaign. Target tracking, especially for underwater target tracking, has been widely used in the systems requiring target location, velocity and quantity (including navigation, obstacle avoidance,monitoring,etc.),but there is still a big gap compared to the air target tracking,which is mature relatively. After decades of development, a great effort and achievement have been made by lots of researchers, and many theories on data correlation and filtering have been proposed to solve problems on the technology of target tracking, however, there are many difficulties remains in tracking moving targets, which is an important part in the target tracking technology.The underwater moving target tracking technology is studied in this thesis.In the first part, the thesis starts from the basic filter and introduces two common types of filter which is the Kalman Filter and Particle Filter. After establishing the tracking model,including the CV model, CA model and Singer model, an Adaptive Gaussian Model will be present.The second part is focus on the research of the data association algorithm in tracking system. The role of tracking gate in target tracking is discussed first in this part. Then nearest neighbor data association algorithm, probabilistic data association algorithm and joint probabilistic data association algorithm are studied and deduced in theory. Simulation and analysis are done to each algorithm next.The experimental data of the pool have been processed in the third part. Several statements on initialization and methods for calculating process will be made at the beginning.Then several groups of experimental data will be processed to verify the validity of the tracking algorithm proposed in the thesis.