An iteratively updated multi-sensor gmphd adaptive fusion method

Abstract

The invention discloses an iteratively updated multi-sensor GMPHD self-adaptive fusion method. The method first constructs an asynchronous multi-sensor multi-target tracking scene and constructs a multi-sensor iteratively updated self-adaptive fusion framework; and applies a Gaussian mixture to each sensor respectively. The PHD filtering algorithm filters and estimates the historical estimated information and the measured values ​​obtained by itself; then sorts and fuses the sensors, performs branch pruning and merging operations on the mixed Gaussian information filtered by each sensor, and outputs the target estimated information; then feeds back the output to Each sensor is used as the input of the next moment. The invention has a clear configuration structure and a small amount of calculation, and can be widely used in the field of multi-target tracking. The method can improve the estimation accuracy of multi-sensors for targets in a monitoring area in a dense clutter environment, and maintain the tracking process.

Description

technical field [0001] The present invention relates to the field of multi-sensor multi-target tracking in a dense clutter environment, and relates to a multi-sensor adaptive fusion multi-target tracking method based on PHD filtering, which is used to solve multi-target tracking in a dense clutter environment and improve the monitoring space The tracking quality of the unknown target in the medium can achieve high precision and stable tracking effect. Background technique [0002] In a multi-sensor tracking system, data fusion technology needs to fuse data from multiple sensors to obtain state estimation of the target, which can improve the performance of the tracking system. However, with the increase in the number of targets and the complexity of data association, multi-sensor multi-target tracking technology also faces many challenges. So far, domestic and foreign researchers have proposed many data fusion algorithms, mainly including two types: sensor-level fusion and f...

AI Technical Summary

Problems solved by technology

Mahler pointed out that since multi-sensors have a certain amount of data loss in each iterative update, in principle, the fusion results of multi-sensor GMPHD based on iterative updates are order-sensitive, that is, different iterative update orders will result in different fusion results.

Despite this, the current research on the fusion sequence of multisensor GMPHD is relatively scarce.

Images