Einige Inhalte dieser Anwendung sind momentan nicht verfügbar.
Wenn diese Situation weiterhin besteht, kontaktieren Sie uns bitte unterFeedback&Kontakt
1. (CN104318059) Target tracking method applied to nonlinear gauss system and tracking system

Amt : China
Anmeldenummer: 201410494854.X Anmeldedatum: 24.09.2014
Veröffentlichungsnummer: 104318059 Veröffentlichungsdatum: 28.01.2015
Erteilungsnummer: 104318059 Erteilungsdatum: 02.02.2018
Veröffentlichungsart : B
IPC:
G06F 19/00
G Physik
06
Datenverarbeitung; Rechnen; Zählen
F
Elektrische digitale Datenverarbeitung
19
Digitale Rechen- oder Datenverarbeitungsanlagen oder -verfahren, besonders angepasst an spezielle Anwendungen
Anmelder: 深圳大学
Erfinder: 刘宗香
陈飞
谢维信
李良群
Vertreter: 深圳市恒申知识产权事务所(普通合伙) 44312
Prioritätsdaten:
Titel: (EN) Target tracking method applied to nonlinear gauss system and tracking system
(ZH) 用于非线性高斯系统的目标跟踪方法和跟踪系统
Zusammenfassung: front page image
(EN) The invention is applied to the field of multi-sensor information fusion, and provides a target tracking method applied to a nonlinear gauss system and a tracking system. The method comprises the following steps: firstly predicating the marginal distribution and the existence probability of each target at the current time according to the marginal distribution and the existence probability of each target at the previous time, appointing corresponding marginal distribution and existence probability for a new target at the current time, obtaining updated marginal distribution and existence probability of each target at the current time through the position measurement at the current time and a Bayesian rule, cutting and combining updated marginal distribution related to each target, so as to obtain the marginal distribution and the existence probability of each target at the current time, finally cutting marginal distribution with the existence probability being smaller than a first threshold valve, taking remaining marginal distribution and existence probability of each target after cutting as input values of recursion at the next time, and meanwhile, extracting marginal distribution with the existence probability being greater than a second threshold valve as output values of the current time.
(ZH) 本发明适用于多传感器信息融合领域,提供了一种用于非线性高斯系统的目标跟踪方法和跟踪系统。所述方法步骤如下:首先根据前一时刻各目标的边缘分布及其存在概率,预测当前时刻各目标的边缘分布及其存在概率,并为当前时刻新生的目标指定相应的边缘分布及其存在概率。再结合当前时刻的位置测量,利用贝叶斯规则得到当前时刻各目标的更新边缘分布及其存在概率,进而对与各目标相关的更新边缘分布进行裁减与合并,得到当前时刻各目标的边缘分布及其存在概率,最后裁减掉存在概率小于第一阈值的边缘分布,并将裁减后余下的各边缘分布及其存在概率作为下一时刻递归的输入,同时,提取存在概率大于第二阈值的边缘分布作为当前时刻的输出。