Multi-station The Bearings Passive Tracking Particle Filter Algorithm

Bearings-only passive tracking is an important branch of passive localization and tracking technique. It is to estimate target motion parameters from bearing measurements and complete target locating. In sonar, infrared and navigation of the military field, bearings-only passive tracking system has important theoretical and applied value, and has become the hot area in nonlinear estimation area.Compared with the single-station tracking system, multi-station tracking system has anti-jamming ability and strong survivability. Because of the need for multi-station work at the same time and data transmission, they will face inherent technical difficulties in multi-sensor data fusion. In addition, the traditional recursive filtering algorithms are difficult to realize the precision positioning and tracking, so a practical, rapid, and high-accuracy tracking algorithm is preferred. According to this, this paper focuses on the study of bearings-only passive tracking system in the aspects of the target location, time alignment and nonlinear filtering algorithm.First of all, a method of target location is proposed based on bearings-only multi-station tracking system. Because of the data pre-processing technical difficulties in multi-station tracking system, two commonly used time alignment algorithms are provided. In order to correct time drift errors, a new time registration project is presented based on the interpolation method.Then, the main nonlinear filtering algorithms, as well as the extensive research of particle filter at home and abroad, are studied based on Bayesian framework. The results of simulation give their different characteristics and apply ranges. And the appropriate algorithm is selected for the passive tracking system.This paper analyses the reason of degeneration phenomenon existing in particle filter and considers choosing a reasonable proposal distribution and re-sampling algorithm. And a new adaptive particle algorithm is presented to resist the influence of the state disturbances. The validity of the algorithm has been proved by theoretical ratiocination and digital simulation.Finally, the digital simulation result shows the feasibility of the proposed algorithm, and can supply a reference for the application in engineering.