A Maneuvering Target Tracking Method Based on Improved srckf Strong Tracking Filter

Abstract

The invention relates to a maneuvering target tracking method for improving SRCKF strong tracking filtering, and belongs to the technical field of data association and tracking. Including the following steps: S1: Establish motion model and observation model according to target motion and satellite observation; S2: Filter initialization; S3: Filter prediction; S4: Perform first measurement update; S5: Calculate with time factor The fading factor of S5 is used to make a threshold judgment; S6: Perform measurement update again according to the threshold judgment result of S5. The beneficial effects of the present invention are as follows: firstly, there is no need to establish a relatively accurate target motion model, and only the motion model when the target is in normal motion can be used to achieve the tracking effect on the maneuvering target; secondly, compared with the general adaptive filtering method , when the target is not maneuvering, the tracking effect of the present invention is equivalent to that of ordinary filtering methods that do not consider maneuvering, and is less disturbed by noise, and the tracking result is more robust.

Description

technical field [0001] The invention relates to a mobile target tracking method for improving square-root volumetric Kalman filter (Square-root Cubature Kalman Filter, SRCKF) strong tracking filtering for optical satellite observation, and belongs to the technical field of data association and tracking. Background technique [0002] In recent years, optical image satellites have always been the focus of research in various countries due to their advantages such as wide observation area, not easily restricted by national borders, and electromagnetic interference. In optical satellite observation, state estimation and situational awareness of space targets are also important. A major focus of the research. [0003] However, under the existing conditions, the movement patterns of space targets are becoming more and more complex, and behaviors such as throwing bait, changing orbits, and splitting guides will occur, so they will deviate from the original movement patterns and pro...

AI Technical Summary

Problems solved by technology

[0003] However, under the existing conditions, the movement patterns of space targets are becoming more and more complex, and behaviors such as throwing bait, changing orbits, and splitting guides will occur, so they will deviate from the original movement patterns and produce certain maneuvering phenomena, and then estimate the state of the target and situational awareness creates greater difficulties

[0004] At the same time, the existing maneuvering target tracking methods are generally divided into two categories. The first category is to model the target motion first, and then use the interactive multi-models (IMM) method to track the target. Such methods need to establish a relatively accurate target motion model, and often consume a lot of time in the process of model conversion. Spatial objects throwing bait, orbit change, and diversion are often short-term behaviors, and their motion models are relatively difficult to establish. Therefore, this method is not suitable for tracking space targets

The second category is to use the idea of ​​adaptive filtering. The strong tracking filtering method based on the square root volumetric Kalman filter is one of them. This method does not change the motion model, but changes the filtering so that it can track the target However, this kind of method is easy to cause large errors in the early stage of target observation, and the robustness to observation noise is not good, so it is not suitable for the whole process of space target tracking. The present invention is mainly aimed at such The shortcomings of the method are improved, so that it can adapt to the situation where the target does not maneuver and make its tracking results more robust

Images