Universal design framework for intelligent target tracking method

Abstract

The invention discloses a universal design framework for an intelligent target tracking method, belongs to the field of radar multi-target tracking, and mainly solves the problems that an existing target tracking method is simple in model, low in complexity, poor in universality and incapable of learning. The universal design framework includes: collecting radar detection data and actual target data under various scenes, and building an original training sample database; performing data extraction, preprocessing and feature extraction aiming at universal framework technical links such as trackprediction, point-track association and track filtering to generate corresponding training data sets; using a sequence prediction method to solve the track prediction problem, using a classificationjudgement method to solve the point-track association problem, using a model parameter learning method to solve the track filtering problem, and using the artificial intelligence technology and the training datasets to generate a target intelligent target tracking algorithm adaptive to a specific radar through preliminary training and multiple combined training. By the universal design framework which is wide in application range, many in adaptive scenes and good in practical effect, the target tracking algorithm is generated through automatic training.

Description

technical field [0001] The invention belongs to the field of radar multi-target tracking, provides a general design framework for target intelligent tracking methods, relates to the construction, learning and generation of target tracking methods, and is suitable for radar data processing systems. Background technique [0002] Target tracking is the core key technology of radar data processing, which is specifically defined as establishing the correspondence between each frame of radar measurement data and different real targets, and further obtaining the target's motion trajectory and motion parameters through filtering estimation, and finally realizing the target individual Continuous mastering in real time. Since the concept of radar target tracking was proposed in 1955, after years of development, radar target tracking technology has formed a relatively complete system and method, including multiple technical links such as track initiation, track filtering, and point nav...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to propose a general design framework for target intelligent tracking methods, aiming to solve the problems of existing target tracking methods with simple models, low complexity, poor adaptability and inability to learn

Images