An Adaptive Target Tracking Method Based on Multi-filter Information Fusion

Abstract

The invention discloses an adaptive target tracking method based on multi-filter information fusion, comprising the following steps: step 1 constructing a multi-Kalman filter data fusion tracking model; step 2 judging whether the adaptive Kalman filter needs to update parameters; Step 3 calculates the uncertain information of multiple fixed parameter Kalman filters; Step 4 converts the uncertain information of multiple fixed parameter Kalman filters into evidence; Step 5 converts the fused evidence into probability; Step 6 updates from Parameters to fit the Kalman filter. The present invention adds a Kalman filter with parameter self-adaptive adjustment on the basis of multiple Kalman filters, so it has better anti-interference performance; and uses evidence theory to fuse uncertain information of new information to generate probability values; Multiple Kalman filter parameters are used as reference values, combined with probability values ​​to update the parameters of the adaptive Kalman filter, which improves the accuracy of target tracking in the interference environment.

Description

technical field [0001] The invention belongs to the technical field of target tracking, and in particular relates to a fuzzy Kalman filter target tracking method improved by evidence theory. Background technique [0002] With the advent of the Internet of Everything era, all kinds of electronic devices and sensors have spread in every corner of life, work and work. Combining technologies such as communication and artificial intelligence, people's lives have been greatly improved. At the same time, the upgrading of manufacturing industry and intelligent manufacturing have brought about the update and iteration of industrial production equipment. Automatically produced equipment often needs to have the ability to identify and track items; Applications such as video surveillance and face recognition have become more intelligent, and these applications have also made target recognition and tracking more popular. Therefore, the development of technology has greatly enriched the ...

AI Technical Summary

Problems solved by technology

However, in the process of target tracking, the sensor is affected by a series of environmental uncertainties such as noise interference in a complex environment, and there may be inconsistencies and conflicts between multi-source data, which will affect the performance of the sensor and cause the observed data to have inconsistencies. However, it is difficult to directly integrate multi-source data. These uncertainties will bring troubles to target recognition and tracking under multi-source data. Therefore, when tracking targets, it is necessary to combine uncertainty theory to deal with uncertain information.

Images