WO2015021903A1 - 目标定位方法及系统 - Google Patents

目标定位方法及系统 Download PDF

Info

Publication number
WO2015021903A1
WO2015021903A1 PCT/CN2014/084163 CN2014084163W WO2015021903A1 WO 2015021903 A1 WO2015021903 A1 WO 2015021903A1 CN 2014084163 W CN2014084163 W CN 2014084163W WO 2015021903 A1 WO2015021903 A1 WO 2015021903A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
sampling points
conditional probability
probability density
sampling point
Prior art date
Application number
PCT/CN2014/084163
Other languages
English (en)
French (fr)
Inventor
陆晓欢
王新珩
贾尚杰
Original Assignee
无锡知谷网络科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 无锡知谷网络科技有限公司 filed Critical 无锡知谷网络科技有限公司
Priority to KR1020167006590A priority Critical patent/KR101912233B1/ko
Priority to JP2016533803A priority patent/JP2016539333A/ja
Priority to US14/911,865 priority patent/US20160195401A1/en
Priority to EP14835947.4A priority patent/EP3034998A4/en
Publication of WO2015021903A1 publication Critical patent/WO2015021903A1/zh

Links

Classifications

    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20Instruments for performing navigational calculations
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01CMEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
    • G01C21/00Navigation; Navigational instruments not provided for in groups G01C1/00 - G01C19/00
    • G01C21/20Instruments for performing navigational calculations
    • G01C21/206Instruments for performing navigational calculations specially adapted for indoor navigation
    • GPHYSICS
    • G01MEASURING; TESTING
    • G01SRADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S5/00Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations
    • G01S5/02Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves
    • G01S5/0278Position-fixing by co-ordinating two or more direction or position line determinations; Position-fixing by co-ordinating two or more distance determinations using radio waves involving statistical or probabilistic considerations
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10Complex mathematical operations
    • G06F17/18Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis

Definitions

  • the invention relates to the field of indoor positioning and simulation, and in particular to a target positioning method and system. Background technique
  • Indoor positioning technology is a technology for acquiring location information of people and objects in a room. Based on location information, it can provide users with a variety of services, and has broad application prospects in both military and civilian applications.
  • Existing indoor positioning methods include infrared indoor positioning technology, ultrasonic positioning technology, Bluetooth technology, radio frequency identification technology, and wireless sensor network technology. Since the implementation of the above positioning technology relies on the collection of various types of wireless information, when the wireless information of the detection environment is relatively large and the situation is complicated, the existing indoor positioning method will not work normally, thereby affecting the target to be treated. Positioning. Summary of the invention
  • an aspect of the present invention provides a target positioning method for determining a current position of a target to be positioned by determining a probability that a target to be positioned is at a sampling point position by measuring the current electromagnetic field.
  • a target positioning method provided according to an aspect of the present invention includes:
  • Generating a plurality of sampling points in the target activity range map obtaining a conditional probability density of the target to be located at the plurality of sampling points according to an initial electromagnetic field strength of the plurality of sampling points and a current electromagnetic field strength of the target to be positioned;
  • the conditional probability density is used to resample the sample point; and the target position to be located is obtained according to the updated sample point coordinate value.
  • the step of acquiring a conditional probability density of the target to be located at the plurality of sampling points comprises: updating initial coordinates of the plurality of sampling points according to a Monte Carlo method of motion; The initial electromagnetic field strength and the current electromagnetic field strength of the updated plurality of sampling points are obtained by the Monte Carlo method to obtain the conditional probability density of the target to be positioned at the plurality of sampling points.
  • the step of generating a plurality of sampling points in a target range of activity map Including: generating multiple sampling points by a random function in the target activity range map.
  • the step of generating a plurality of sampling points in the active range map of the target comprises: generating, in the target active range map, a plurality of sampling points according to the historical sampling point information and the basic series algorithm BSAS.
  • the step of updating initial coordinates of the plurality of sampling points according to the Monte Carlo method of motion comprises: operating a model according to a Monte Carlo method + ⁇ Updates the two-dimensional initial coordinates of the plurality of sampling points.
  • the initial electromagnetic field strength and the current electromagnetic field strength of the updated plurality of sampling points are used to obtain a conditional probability density of the target to be located at the plurality of sampling points by a Monte Carlo measurement model.
  • the step of resampling the sampling point according to the conditional probability density comprises: correcting the sampling according to a random adoption function according to a set sampling number, a conditional probability density of the sampling point The point is resampled to obtain the current sample point.
  • the step of acquiring the target position to be located according to the updated sample point coordinate value comprises: determining the target position according to the updated sample point coordinate value and the conditional probability density according to the updated sample point
  • the two-dimensional coordinates and the corresponding conditional probability density m are obtained according to the public to be positioned.
  • the indoor positioning technology can be solved by the detection environment in the above prior art.
  • a target positioning system including:
  • a sampling point generating module configured to generate a plurality of samples in the target active range map, *;
  • a conditional probability density acquiring module configured to obtain a pending based on an initial electromagnetic field strength of the plurality of sampling points and a current electromagnetic field strength of the target to be positioned a conditional probability density of the bit target at the plurality of sampling points;
  • a resampling module configured to resample the sample point according to the conditional probability density
  • the positioning module is configured to obtain a target location to be located according to the updated sampling point coordinate value.
  • conditional probability density acquisition module includes: a Monte Carlo method model unit configured to update initial coordinates of the plurality of sampling points according to a Monte Carlo method of motion;
  • the Monte Carlo method is configured to measure a conditional probability density of the target to be located at the plurality of sampling points by using a Monte Carlo method to measure the initial electromagnetic field strength and the current electromagnetic field strength of the plurality of sample points.
  • the sampling point generating module includes:
  • Randomly generating a sampling point unit configured to generate a plurality of sampling points by a random function in the target active range map
  • the historical sampling point updating unit is configured to generate a plurality of sampling points according to the historical sampling point information and the basic series algorithm BSAS in the target active range map.
  • the present invention can at least have the following advantages:
  • the BSAS clustering method can effectively solve the problem of effective particle loss in Monte Carlo method.
  • FIG. 1 is a schematic diagram of steps of a target positioning method in an embodiment of the present invention.
  • FIG. 2 is a schematic view of an indoor magnetic field in an initial state in an embodiment of the present invention
  • FIG. 3 is a schematic diagram of an indoor plane of an initial state in the embodiment of the present invention.
  • FIG. 4 is a schematic plan view of a sampling point initially generated in an embodiment of the present invention
  • FIG. 5 is a schematic diagram of steps of a method for generating a sampling point according to an embodiment of the present invention
  • FIG. 7 is a schematic plan view showing a sampling point that has not been resampled and updated in an embodiment of the present invention.
  • FIG. 8 is a schematic plan view showing a sampling point resampled and updated in an embodiment of the present invention.
  • FIG. 9 is a schematic diagram of a composition of a target positioning system of an electromagnetic field in an embodiment of the present invention.
  • the target positioning method includes:
  • Step S101 Perform on-the-spot measurement on the indoor location where the target is located, and obtain an indoor plan view.
  • the initial magnetic field in the room is checked by a magnetometer to obtain an initial magnetic field map in the room.
  • 2 is a schematic diagram showing an indoor plane of an initial state in an embodiment of the present invention, in which the target "A" is the target to be positioned, and the "B" boundary is the indoor activity area of the target "A", when the target "A" is absent
  • the magnetic field at each point in the "B” boundary is measured by a magnetometer to obtain the initial magnetic field map of the "B” boundary, as shown in Fig. 3.
  • Step S102 uniformly generating a plurality of sampling points on the indoor plan view (shown in FIG. 2) obtained in step S101.
  • sampling points can be generated on the indoor floor plan (ie, Figure 2) using the different methods described below.
  • the rand() function can be used to generate a sampling point matrix composed of random numbers of "uniform distribution" or other probability distribution types in the x, y directions, respectively. For example, n sample points are randomly generated on the indoor plan (Fig. 3).
  • Figure 4 It is a schematic plan view of the initial generation of sampling points in an embodiment of the present invention.
  • the basic sequence algorithm BSAS can be used to initialize the historical sample points, including the following steps:
  • Step S1022 Calculate the current Euler distance between the plurality of historical sample points extracted in step S1021 and the initial cluster by the formula " ⁇ 1 ) ' ⁇ 1111111 ⁇ 1 )' ⁇ .
  • the threshold value ⁇ is 2 m
  • A1 is set as the reference sampling point, wherein the Euler distance between the initial cluster 1 and A3 is lm, so the threshold value is less than 2 m.
  • the historical sampling point A3 can be classified into the initial cluster 1.
  • the Euler distance between the initial cluster 1 and A2 is 3 m, which is greater than the threshold of 2 m, so that the historical sampling point A2 cannot be included in the initial cluster, and the cluster 2 is additionally created.
  • Step S1024 After all the sampling points are clustered, if the number of sampling points in one of the clusters is greater than the number n of the set cluster particles, then p particles are selected from the cluster to represent the cluster, The next np particles are reinitialized. The re-initialization process randomly generates n-p uniformly distributed sampling points on the indoor plan through the rand function in the initial positioning.
  • a cluster with 80 historical sampling points has been generated.
  • the initial cluster 1 is preset. Obtain the Euler distance of a cluster of known historical sample points.
  • the threshold value ⁇ is set to 2 to 3 m, and can also be set according to the accuracy requirements. Thereafter, based on a threshold value of 2 to 3 m, a judgment is made for each of the historical sampling points in the cluster. For example, if the historical sampling point is compared with the initial adoption point as lm, then within the threshold value , it is classified into the initial cluster 1. The historical sampling point is 4m compared with the initial adoption point. In addition to the threshold ⁇ , a new cluster 2 is created.
  • clusters Judging sequentially, for example, four clusters have been created based on the historical sampling point cluster, wherein cluster 1 (10 sampling points), cluster 2 (40 sampling points), cluster 3 (20 sampling points) ), cluster 4 (10 sample points).
  • cluster 1 (10 sampling points)
  • cluster 2 40 sampling points
  • cluster 3 (20 sampling points)
  • cluster 4 (10 sample points).
  • 40 sampling points in the cluster 2 exceed the set value 30
  • 30 particles are selected from the cluster 2 to represent the cluster, and 10 sample points are left to be regrouped.
  • Step S1025 The sampling points in all the clusters determined in the above step S1024 are determined as the current sampling point(s).
  • Step S103 First, the initial coordinates of the 80 sampling points generated in step S102 are updated according to the Monte Carlo method of motion.
  • Monte Carlo motion model which means that for each particle 1, the particle state is updated at an arbitrary speed at an arbitrary angle to obtain a new set of sample values ⁇ , that is, new position sample values of 80 sample points.
  • the sampling point A is randomly moved from the initial position 11 to 12 by the Monte Carlo motion model (one of any random moving positions shown as the initial position 11).
  • the advantage of the Monte Carlo motion model to simulate the target to be located is that it can give a realistic simulation of human behavior or robot motion, so that the target location is more accurate.
  • the probability density corresponding to the sampled point after moving is ⁇ 12 ⁇ , where the probability density ⁇ (
  • the current direction angle ⁇ is randomly generated, and the current position X, y is added according to the formula (X ⁇ ) W to update the position.
  • the magnetometer obtains the current magnetic field strength value of the target to be located in the room, that is, obtains the measured value ⁇ .
  • the measurement model is used to indicate the conditional probability density when the target is measured at the state x t as z t.
  • the Gaussian process can be specifically used to perform the operation, and h map is the initial magnetic field map value.
  • the magnetic field value Ux can be obtained by the magnetic field map search, and the current target to be positioned is measured as z, and p(z
  • Point si the current weight value.
  • the sampling point si changes through the position coordinates of the motion model.
  • the same magnetic field value hx, y ) also changes, and the current weight value is obtained by using the Gaussian formula according to the measured value z of the current target to be positioned.
  • the current weight value is multiplied by the weight value of the previous moment to obtain the weight value of S1, that is, the conditional probability density.
  • Step S104 Resampling each sampling point according to a random sampling function according to the set sampling number and the conditional probability density of each sampling point obtained in S103.
  • the random sampling function can be selected from the randsample ( ) function in matlab software. Get the current sampling point after filtering. As shown in Figure 7, there are 500 sample points sl-s500, each of which has a corresponding normalized weight value. The 500 sample points are resampled using the randsample function, and the weight value of each point determines the probability that the point will be sampled. For 500 samplings, the probability that the points with significant weight are sampled is large. After resampling, the distribution of 500 sample points will change. The resampling results are shown in Figure 8.
  • Step S105 According to the two-dimensional coordinate value of the current sampling point updated in step S104, the position of the target to be positioned on the indoor plane map can be directly determined. As shown in the sampling point shown in Fig. 8, the central area of the dense position where the point is used in the figure can be determined as the position of the target to be positioned.
  • conditional probability density m corresponding to the sampling point combined with the formula Get the 2D coordinates of the current target. Thereby making the positioning of the target more accurate.
  • Positioning can be achieved by obtaining a flat map and a magnetic field map plus a common electromagnetic sensor.
  • the Monte Carlo method adopted by the invention is simple, flexible and easy to implement, has good real-time performance and stable performance; and the BSAS clustering method can effectively solve the problem of effective particle loss in Monte Carlo method.
  • the method of the invention has low requirements on computer hardware.
  • the dynamic map method can be used to dynamically segment the map to ensure that the running time is within an acceptable range.
  • an embodiment of the present invention further provides a target positioning system, including:
  • the sample point generation module 201 is configured to generate a plurality of sample points in the target activity range map.
  • the conditional probability density acquisition module 202 is configured to obtain a conditional probability density of the target to be located at a plurality of sampling points according to the initial electromagnetic field strength of the plurality of sampling points and the current electromagnetic field strength of the target to be positioned.
  • the resampling module 203 is configured to resample the sample points according to the conditional probability density. Among them, the resampling function is the randsample function.
  • the positioning module 204 is configured to obtain a target location to be located according to the updated sampling point coordinate value. Specifically, the method includes: obtaining a target position to be located according to the two-dimensional coordinates of the updated sampling point. Or according to the updated two-dimensional coordinates of the sampling point and the corresponding conditional probability density m, according to the formula
  • the method includes:
  • the Monte Carlo motion model unit 2021 is configured to update the initial coordinates of the plurality of sampling points according to the Monte Carlo motion model.
  • the Monte Carlo method of action is
  • the Monte Carlo method model unit 2022 is configured to pass the Monte Carlo based on the initial electromagnetic field strength of the updated plurality of sampling points and the current electromagnetic field strength of the target to be positioned.
  • the Luofa measurement model obtains the conditional probability density of the target to be located at multiple sampling points. The specific step is, for each sampling point,
  • the sampling point generating module 201 includes:
  • the historical sampling point updating unit 2012 is configured to generate a plurality of sampling points according to the historical sampling point information and the basic series algorithm BSAS in the target active range map.

Landscapes

  • Engineering & Computer Science (AREA)
  • Remote Sensing (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Automation & Control Theory (AREA)
  • Data Mining & Analysis (AREA)
  • Probability & Statistics with Applications (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Theoretical Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Operations Research (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Algebra (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Databases & Information Systems (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Biology (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)
  • Radar Systems Or Details Thereof (AREA)
  • Environmental & Geological Engineering (AREA)
  • General Life Sciences & Earth Sciences (AREA)
  • Geology (AREA)
  • Image Analysis (AREA)
  • Measurement Of Radiation (AREA)

Abstract

公开了目标定位方法和系统,包括:在目标的活动范围图中生成多个采样点;根据多个采样点的初始电磁场强度及待定位目标的当前电磁场强度,获取待定位目标在多个采样点的条件概率密度;根据条件概率密度对采样点进行重采样更新;根据更新后的采样点的坐标值获取待定位目标的位置。本发明可以解决室内定位受检测环境影响而不能精确定位的问题,从而使得对室内目标的定位或跟踪更准确。

Description

目标定位方法及系统 技术领域
本发明涉及室内定位和仿真领域, 特别涉及目标定位方法及系统。 背景技术
室内定位技术是一种获取室内的人和物位置信息的技术。 以位置信息 为基础, 可以为使用者提供多种服务, 在军事和民用领域都有着广阔的应 用前景。 现有的室内定位方法有红外线室内定位技术、 超声波定位技术、 蓝牙技术、 射频识别技术和无线传感器网络技术等。 由于上述定位技术的 实现均依赖于对各类无线信息的采集, 因此, 当检测环境的无线信息干扰 较大, 情况较复杂时, 现有的室内定位方法将无法正常工作, 从而影响对 待定位目标的定位。 发明内容
针对现有技术中的缺陷, 本发明一方面提供了目标定位方法, 通过对当 前电磁场的测量, 确定待定位目标处于采样点位置的概率, 从而获得待定 位目标的当前位置。
根据本发明的一方面提供的目标定位方法包括:
在目标活动范围图中生成多个采样点; 根据所述多个采样点的初始电磁 场强度及待定位目标的当前电磁场强度, 获取待定位目标在所述多个采样 点的条件概率密度; 根据所述条件概率密度对所述采样点进行重采样更新; 根据所述更新后的采样点坐标值获取待定位目标位置。
在一些实施方式中, 所述获取待定位目标在所述多个采样点的条件概率 密度的步骤包括: 根据蒙特卡罗法动作模型对所述多个采样点的初始坐标 进行更新; 才艮据更新后的多个采样点的初始电磁场强度及当前电磁场强度, 通过蒙特卡罗法测量模型获取待定位目标在所述多个采样点的条件概率密 度。
在一些实施方式中, 所述在目标的活动范围图中生成多个采样点的步骤 包括:在目标活动范围图中, 通过随机函数生成多个采样点。
在一些实施方式中, 所述在目标的活动范围图内生成多个采样点的步骤 包括:在目标活动范围图中, 根据历史采样点信息及基本系列算法 BSAS生 成多个采样点。
在一些实施方式中, 所述根据历史采样点信息及基本系列算法 BSAS生 成多个采样点的步骤包括: 建立初始聚群 m=l, Cm={xm} ; 根据所述任意两 个历史采样点间的欧拉距离 (1)' ) = 011111≤^^(1)' ) , 设定门限值 Θ ; 根据 所述门限值 Θ 及参考采样点, 依次判断所述多个历史采样点是否属于所述 初始聚群 ¾ = u{x(i)} , 若属于, 则加入所述初始聚群, 若不属于, 则创建 新聚群 m=m+l, Cm={x(l)}; 若所述聚群的采样点数量大于设定聚群粒子数 量, 则根据所述设定聚群粒子数量对所述聚群的采样点进行提取; 将所述 聚群的采样点确定为多个采样点。 在一些实施方式中, 所述根据蒙特卡罗法动作模型对所述多个采样点 的初始坐标进行更新的步骤包括: 根据蒙特卡罗法动作模型
Figure imgf000003_0001
+ ^^更新所述多个采样点的二维初始坐标。 在一些实施方式中, 所述才艮据更新后的多个采样点的初始电磁场强度 及当前电磁场强度, 通过蒙特卡罗法测量模型获取待定位目标在所述多个 采样点的条件概率密度的步骤包括:对每一采样点,将当前磁场强度值 Zt加 入蒙特卡罗法测量模型"^ = m " ^ ) } , 获取每一采样点的条件概率值 m\ = pizt \ A )。 在一些实施方式中, 所述根据所述条件概率密度对所述采样点进行重 采样更新的步骤包括: 根据设定取样次数、 所述采样点的条件概率密度, 根据随机采用函数对所述采样点进行重采样更新, 获取当前采样点。
在一些实施方式中, 所述根据所述更新后的采样点坐标值获取待定位目 标位置的步骤包括: 根据所述更新后的采样点坐标值及条件概率密度确定 目标位置根据更新后的采样点的二维坐标及对应条件概率密度 m, 根据公 获取待定位目标位置。
Figure imgf000003_0002
根据上述方法, 可解决上述现有技术中, 室内定位技术受检测环境影响 而不能精确定位的问题。
根据本发明另一方面, 还提供了目标定位系统, 包括:
采样点生成模块, 配置为在目标活动范围图中生成多个采样 , *; 条件概率密度获取模块, 配置为根据所述多个采样点的初始电磁场强度 及待定位目标的当前电磁场强度, 获取待定位目标在所述多个采样点的条 件概率密度;
重采样模块, 配置为根据所述条件概率密度对所述采样点进行重采样更 新;
定位模块, 配置为根据所述更新后的采样点坐标值获取待定位目标位 置。
在一些实施方式中, 在所述条件概率密度获取模块中, 包括: 蒙特卡罗法动作模型单元, 配置为根据蒙特卡罗法动作模型对所述多个 采样点的初始坐标进行更新;
蒙特卡罗法测量模型单元, 配置为根据更新后的多个采样点的初始电磁 场强度及当前电磁场强度, 通过蒙特卡罗法测量模型获取待定位目标在所 述多个采样点的条件概率密度。
在一些实施方式中, 所述采样点生成模块中包括:
随机生成采样点单元, 配置为在目标活动范围图中, 通过随机函数生成 多个采样点; 或
历史采样点更新单元, 配置为在目标活动范围图中, 根据历史采样点信 息及基本系列算法 BSAS生成多个采样点。
与现有技术相比, 本发明可至少具有如下优点:
1 ) 避免了复杂的软硬件需求限制, 只要获得平面地图和磁场地图加上 普通的电磁传感器便可实现定位。
2) 采用的蒙特卡罗法模型简单, 使用灵活易实现, 实时性好, 性能稳 定性高。
3 ) 通过采用的 BSAS聚类方法能够有效解决蒙特卡罗法有效粒子流失 的问题。
4) 对计算机硬件要求不高, 节约了系统投入。 附图说明
图 1 为本发明 -种实施方式中目标定位方法的步骤示意图
图 2为本发明 -种实施方式中初始状态的室内磁场示意图
图 3为本发明 -种实施方式中初始; 态的室内平面示意图
图 4为本发明 -种实施方式中初始生成采样点后的平面示意图; 图 5为本发明 -种实施方式中采样点生成方法的步骤示意图; 图 6为本发明 -种实施方式中采样点在平面图上的移动过程示意图 图 7为本发明 -种实施方式中采样点未重采样更新时的平面示意图 图 8为本发明 -种实施方式中采样点已重采样更新时的平面示意图
9为本发明 -种实施方式中电磁场的目标定位系统的组成示意图 具体实施方式
下面结合附图对发明作进一步详细的说明。
图 1 为本发明一种实施方式的基于电磁场的目标定位方法的步骤示意 图。 如图 1所示, 该目标定位方法包括:
步骤 S101 : 对待定位目标所处的室内进行实地测量, 获得室内平面图。 在室内中无待定位目标时, 通过磁强计对室内的初始磁场进行检查, 获得 室内初始磁场地图。 图 2 显示了本发明一种实施方式中初始状态的室内平 面示意图, 其中目标" A"为待定位目标, "B"边界范围内为目标" A"的室内活 动区域, 当目标 "A"不在" B"边界范围内时, 通过磁强计测定" B"边界范围内 的各点磁场, 获取" B" 边界的初始磁场地图, 如图 3所示。
步骤 S102 : 在步骤 S101 中所获得的室内平面图 (如图 2所示) 上, 均 匀生成多个采样点。 根据当前所处的不同目标定位阶段 (即初始定位状态 或跟踪定位状态), 可以采用下述不同的方法在室内平面图 (即图 2) 上进 行采样点的生成。
若当前状态为初始定位状态, 即没有可参考的历史采样点时, 可利用 rand()函数分别在 x,y 方向生成呈"均匀分布"或其他概率分布类型的随机数 组成的采样点矩阵。 如在室内平面图 (图 3) 上随机生成 n个采样点。 图 4 是本发明一种实施方式中初始生成采样点后的平面示意图。
考虑到目标跟踪的连续性, 若当前状态为跟踪定位状态 (跟踪定位状 态为,对待定位目标进行连续定位跟踪的定位状态),即之前的跟踪步骤(或 上一跟踪时段) 中已产生历史采样点, 则如图 5 所示, 可采用基本序列算 法 BSAS对历史采样点进行初始化更新, 包括如下步骤:
步骤 S1021 :根据设定门限值 Θ对历史采样点进行抽取, 建立初始聚群 m=l, Cm={Xm} ; 如: 设定门限值 Θ为 2m, 现有的历史聚群包括 Al、 A2、 A3、 A4四个历史采样点, 其中, Al、 A4之间的距离为 2m, 从而 Al、 A4 组成初始聚群 1。
步骤 S1022 : 通过公式"^1)'^^1111111 ^^1)'^依次计算步骤 S1021抽取 后的多个历史采样点与初始聚群之间的当前欧拉距离。 如在初始聚群中共 有 Al、 A2及 A3三个历史采样点, 其中, 初始聚群 1与 A2之间的欧拉距 离为 3m、 初始聚群 1与 A3之间的欧拉距离为 lm。
步骤 S1023 : 判断各历史采样点与初始聚群之间的欧拉距离,是否小于 设定门限值 Θ , 若是, 则加入初始聚群, 若否, 则创建新聚群 m=m+l, Cm={xw}。 如步骤 S1022中的例子, 当门限值 Θ为 2m时, 将 A1设定为参 考采样点, 其中, 由于初始聚群 1与 A3之间的欧拉距离为 lm, 因此小于 2m的门限值, 从而历史采样点 A3可归入初始聚群 1 中。初始聚群 1与 A2 之间的欧拉距离为 3m, 大于 2m的门限值, 从而历史采样点 A2不能列入 初始聚群中, 并另创建聚群 2。
步骤 S1024 :在对所有采样点都进行聚群分类后,若其中某一聚群中的 采样点数大于设定聚群粒子数量 n, 则从该聚群中选中 p 个粒子代表该聚 群, 剩下 n-p 个粒子进行重新初始化。 该重新初始化过程同初始定位中通 过 rand函数在室内平面图上随机生成 n-p个均匀分布的采样点。
例如: 已产生具有 80个历史采样点的聚群。 首先, 预设定初始聚群 1。 获得已知历史采样点聚群的欧拉距离。 设定门限值 Θ为为 2〜3m, 也可根 据精度要求设定。 之后, 根据 2〜3m的门限值, 对于该聚群中的每一个历 史采样点进行判断。 如, 历史采样点与初始采用点比较为 lm, 则在门限值 Θ之内, 则归类到初始聚群 1 中。 历史采样点与初始采用点比较为 4m, 则 在门限值 Θ之外, 则创建新聚群 2。 依次判断, 例如, 根据该历史采样点聚 群已创建了 4个聚群,其中,聚群 1 (10个采样点)、 聚群 2 (40个采样点)、 聚群 3 (20个采样点)、 聚群 4 (10个采样点)。 其中聚群 2中的 40个采样 点超过了设定值 30,则从该聚群 2中选中 30个粒子代表该聚群, 剩下 10个 采样点重新进行聚群分类。
步骤 S1025: 将上述步骤 S1024 中所确定的所有聚群中的采样点, 确 定为当前采样点 (多个)。
在上述实施方式中, 首先获取了粒子收敛程度的信息, 然后决定用来 代表该区域粒子的数目, 通过从旧的粒子集中选取一定比例的粒子然后进 行重新初始化。 从而通过上述方式, 可以成功解决有效粒子流失的问题, 提高定位成功率。
步骤 S103: 首先, 根据蒙特卡罗法动作模型对步骤 S102 所生成的 80 个采样点的初始坐标进行更新。 采用 蒙特卡罗法动作模型
Figure imgf000007_0001
, 其中 表示对于每一个粒子 1 , 以任意角度 任意速度运动更新粒子状态, 得到一组新的采样值 ^, 即 80个采样点的新 的位置采样值。如图 6, 采样点在室内平面图上的移动过程所示, 采样点 A 通过蒙特卡罗法动作模型,从初始位置 11随机移动到 12 (图示为初始位置 11的任意随机移动位置中的一个),通过蒙特卡罗法动作模型进行待定位目 标模拟的优势在于, 可对人类行为或机器人运动给予较为真实的仿真模拟, 从而使目标定位更为准确。 移动后采样点所对应的概率密度为 ^^12^ , 其 中, 概率密度为 ρ( |Ζί1〕可基于上一轮迭代得到的后验概率 ρ( ι'— 运用 一个运动模型 P(^l^-i, -i)来获得。如对于采样点 S1, 随机生成当前方向角 θ ,根据公式 (X^)W 分别对当前位置 X, y进行相加来更新位 置。 之后, 通过磁强计获取室内待定位目标的当前的磁场强度值, 即获取 测量值 ^。 采用测量模型 表示当目标在状态 xt下测量值为 zt的条件概 率密度。对每一个采样点 4 ,将测量数据 Zt代入测量模型 ^ -ι ¾Ι(¾'Λ) } 后计算出每个粒子的权重" ^ ^ ^), 即条件概率密度, 条件概率密度代 表该采样点在该位置上的概率大小。 其中, 可具体采用高斯过程 严 - 进行运算, hmap为初始磁场地图 值。 如: 对于采样点 sl, 根据其位置坐标可通过磁场地图查找获得磁场值 Ux, ,当前待定位目标的测量值为 z, 利用如上高斯公式可获得 p(z|x,y), 这就是采样点 si 当前的权重值。 下一时刻, 采样点 si经过运动模型位置坐 标会发生变化。 同理磁场值 h x, y)也跟着变化, 根据当前待定位目标的测 量值 z再利用高斯公式求出当前权重值。 根据" ^) =0( I ( '¾)W)公式, 将当 前权重值与上一时刻的权重值相乘, 得到 S1的权重值, 即条件概率密度。 依此迭代下去。
步骤 S104 : 根据设定采样次数、 S103中所获得的各采样点的条件概率 密度, 根据随机采样函数对各采样点进行重采样。 其中, 随机采样函数可 选用 matlab软件中的 randsample ( ) 函数。 筛选后获取当前采样点。 如图 7所示, 设有 500个采样点 sl-s500,每个采样点都有对应的归一化后的权重 值。 采用 randsample函数对这 500个采样点进行重采样, 每个点的权重值 决定该点被抽样的概率。 进行 500次抽样, 权重大的点被抽样的概率大。 重采样结束后, 500个采样点的分布将会改变。 重采样结果如图 8所示。
步骤 S105 : 根据步骤 S104中所更新的当前采样点的二维坐标值, 可直 接确定待定位目标在室内平面地图上的位置。 如图 8 中所示采样点, 可将 图示中采用点的密集位置的中心区域确定为待定位目标的位置。
采样点所对应的条件概率密度 m , 并结合公式
Figure imgf000008_0001
获取当前的目标的二维坐标。 从而使目标的定位更 为准确。
由此可知, 上述方案避免了复杂的软硬件需求限制。 只要获得平面地 图和磁场地图加上普通的电磁传感器便可实现定位。 本发明所采用的蒙特 卡罗法模型简单, 灵活易实现, 实时性好, 性能较稳定; 采用的 BSAS 聚 类方法能够有效解决蒙特卡罗法有效粒子流失的问题。 本发明中方法对计 算机硬件要求不高, 如遇大数据量地图可用动态地图法对地图进行动态分 割, 保证运行时间在可接受范围内。 同时, 如图 9所示, 本发明的一种实施方式中还提供了目标定位系统, 包括:
采样点生成模块 201 , 配置为在目标活动范围图中生成多个采样点。 条件概率密度获取模块 202, 配置为根据多个采样点的初始电磁场强度 及待定位目标的当前电磁场强度, 获取待定位目标在多个采样点的条件概 率密度。
重采样模块 203, 配置为根据条件概率密度对采样点进行重采样更新。 其中, 重采样函数为 randsample函数。
定位模块 204,配置为根据更新后的采样点坐标值获取待定位目标位置。 具体包括: 根据更新后的采样点的二维坐标获取待定位目标位置。 或根据 更新后的采样点的二维坐标及对应条件概率密度 m , 根据公式
(¾y)es*~te =-,S>w (¾y> 获取待定位目标位置。 其中, 在条件概率密度获取模块 202中, 包括:
蒙特卡罗法动作模型单元 2021 ,配置为根据蒙特卡罗法动作模型对多个 采样 点 的 初 始 坐 标 进行 更新 。 蒙 特 卡 罗 法 动 作 模 型 为
( ,¾)(") ^^,^ + ^)。 蒙特卡罗法测量模型单元 2022,配置为根据更新后的多个采样点的初始 电磁场强度及待定位目标的当前电磁场强度, 通过蒙特卡罗法测量模型获 取待定位目标在多个采样点的条件概率密度。 具体步骤为, 对每一采样点,
(n) _ (n) ( I ( {^)
将当前磁场强度值 加入蒙特卡罗法测量模型 - mt-i ^ ^yt ) , 获取每 一采样点的条件概率值 = /7(Zz |0。
其中, 在采样点生成模块 201 中包括:
随机生成采样点单元 2011, 配置为在目标活动范围图中, 通过随机函 数生成多个采样点;
历史采样点更新单元 2012, 配置为在目标活动范围图中, 根据历史采 样点信息及基本系列算法 BSAS 生成多个采样点。 具体步骤为: 根据设定 门限值 Θ建立初始聚群 m=l, Cm={xm}; 判断历史采样点与初始聚群之间的 欧拉距离 = ^I^^XW' 是否小于设定门限值 θ, 若是, 则加入初始 聚群, 若否, 则创建新聚群 m=m+l, Cm={x(l)}; 若聚群的采样点数量大于 设定聚群粒子数量, 则根据设定聚群粒子数量对聚群的采样点进行提取; 将聚群的采样点确定为多个采样点。
该系统的具体操作方式在前文已详细说明, 不再赘述。
以上所述的仅是本发明的一些实施方式。 对于本领域的普通技术人员 来说, 在不脱离本发明创造构思的前提下, 还可以做出若干变形和改进, 这些都属于本发明的保护范围。

Claims

权利要求书
1、 目标定位方法, 包括:
在目标的活动范围图中生成多个采样 , *;
才艮据所述多个采样点的初始电磁场强度及待定位目标的当前电磁场强 度, 获取待定位目标在所述多个采样点的条件概率密度;
根据所述条件概率密度对所述采样点进行重采样更新;
根据所述更新后的采样点的坐标值获取待定位目标的位置。
2、 根据权利要求 1 所述的方法, 其中所述获取待定位目标在所述多个 采样点的条件概率密度的步骤包括:
根据蒙特卡罗法动作模型对所述多个采样点的初始坐标进行更新; 才艮据更新后的多个采样点的初始电磁场强度及当前电磁场强度, 通过蒙 特卡罗法测量模型获取待定位目标在所述多个采样点的条件概率密度。
3、 根据权利要求 1 或 2所述的方法, 其中在目标的活动范围图中, 通 过随机函数生成多个采样点。
4、 根据权利要求 1 所述的方法, 其中所述在目标的活动范围图内生成 多个采样点的步骤包括:
在目标的活动范围图中, 根据历史采样点信息及基本系列算法 BSAS 生成多个采样点。
5、 根据权利要求 4 所述的方法, 其中所述根据历史采样点信息及基本 系列算法 BSAS生成多个采样点的步骤包括:
根据设定门限值 Θ建立初始聚群 m=l, Cm={xm};
判 断 所述历 史采样点 与 所述初始 聚群之 间 的 欧拉距 离
^^C^ min^^xd^)是否小于所述设定门限值 Θ , 若是, 则加入所述初 始聚群, 若否, 则创建新聚群 m=m+l, Cm={x(l)}; 若所述聚群的采样点数量大于设定聚群粒子数量, 则根据所述设定聚 群粒子数量对所述聚群的采样点进行提取;
将所述聚群的采样点确定为多个采样点。
6、 根据权利要求 2所述的方法, 其中所述根据蒙特卡罗法动作模型对 所述多个采样点的初始坐标进行更新的步骤包括:
根据蒙特卡罗法动作模型( ,WW =(¾-b^-i)(M) + ^(M))更新所述多个采样 点的二维初始坐标。
7、 根据权利要求 2所述的方法, 其中所述根据更新后的多个采样点的 初始电磁场强度及当前电磁场强度, 通过蒙特卡罗法测量模型获取待定位 目标在所述多个采样点的条件概率密度的步骤包括:
对每一采样点, 将当前磁场强度值 加入蒙特卡罗法测量模型 ^) =^' Ι (Χ"^)(Μ)) , 获取每一采样点的条件概率值^' = ^ ^)。
8、 根据权利要求 1所述的方法, 其中所述根据所述条件概率密度对所 述采样点进行重采样更新的步骤包括:
根据设定的采样次数、 所述采样点的条件概率密度, 根据随机采样函 数对所述采样点进行重采样更新, 获取当前采样点。
9、 根据权利要求 1所述的方法, 其中所述根据更新后的采样点坐标值 获取待定位目标位置的步骤包括:
根据更新后的采样点的二维坐标获取待定位目标的位置; 或
根据更新后的采样点的二维坐标及对应条件概率密度 m , 根据公式
^y ^ =^ \^ 获取待定位目标位置。
10、 目标定位系统, 包括:
采样点生成模块, 配置为在目标的活动范围图中生成多个采样点; 条件概率密度获取模块, 配置为根据所述多个采样点的初始电磁场强度 及待定位目标的当前电磁场强度, 获取待定位目标在所述多个采样点的条 件概率密度;
重采样模块, 配置为根据所述条件概率密度对所述采样点进行重采样更 新;
定位模块, 配置为根据所述更新后的采样点坐标值获取待定位目标位 置。
11、 根据权利要求 10 所述的系统, 其中在所述条件概率密度获取模块 中, 包括:
蒙特卡罗法动作模型单元, 配置为根据蒙特卡罗法动作模型对所述多个 采样点的初始坐标进行更新;
蒙特卡罗法测量模型单元, 配置为根据更新后的多个采样点的初始电磁 场强度及当前电磁场强度, 通过蒙特卡罗法测量模型获取待定位目标在所 述多个采样点的条件概率密度。
12、 根据权利要求 10所述的系统, 其中所述采样点生成模块中包括: 随机生成采样点单元, 配置为在目标的活动范围图中, 通过随机函数生 成多个采样点; 或
历史采样点更新单元, 配置为在目标的活动范围图中, 根据历史采样点 信息及基本系列算法 BSAS生成多个采样点。
PCT/CN2014/084163 2013-08-12 2014-08-12 目标定位方法及系统 WO2015021903A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
KR1020167006590A KR101912233B1 (ko) 2013-08-12 2014-08-12 물체의 위치를 결정하기 위한 시스템 및 방법
JP2016533803A JP2016539333A (ja) 2013-08-12 2014-08-12 ターゲットの位置決め方法及びシステム
US14/911,865 US20160195401A1 (en) 2013-08-12 2014-08-12 Method and system for locating an object
EP14835947.4A EP3034998A4 (en) 2013-08-12 2014-08-12 Target positioning method and system

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201310350209.6A CN104375117B (zh) 2013-08-12 2013-08-12 目标定位方法及系统
CN201310350209.6 2013-08-12

Publications (1)

Publication Number Publication Date
WO2015021903A1 true WO2015021903A1 (zh) 2015-02-19

Family

ID=52468051

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/084163 WO2015021903A1 (zh) 2013-08-12 2014-08-12 目标定位方法及系统

Country Status (6)

Country Link
US (1) US20160195401A1 (zh)
EP (1) EP3034998A4 (zh)
JP (1) JP2016539333A (zh)
KR (1) KR101912233B1 (zh)
CN (1) CN104375117B (zh)
WO (1) WO2015021903A1 (zh)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105118015A (zh) * 2015-09-21 2015-12-02 无锡知谷网络科技有限公司 用于公共场所的信息提示方法及移动服务终端
CN107132504B (zh) * 2016-02-29 2020-12-22 富士通株式会社 基于粒子滤波的定位追踪装置、方法及电子设备
CN106289242B (zh) * 2016-07-18 2019-03-22 北京方位捷讯科技有限公司 基于地磁的粒子滤波定位方法及装置
US10650621B1 (en) 2016-09-13 2020-05-12 Iocurrents, Inc. Interfacing with a vehicular controller area network
CN106441279B (zh) * 2016-12-08 2019-03-29 速感科技(北京)有限公司 基于自主定位和边缘探索的机器人定位方法、系统
CN106767828A (zh) * 2016-12-29 2017-05-31 南京邮电大学 一种手机室内定位解决方法
CN107976184A (zh) * 2017-02-15 2018-05-01 北京金坤科创技术有限公司 一种基于地磁信号极值检测与特征匹配的定位方法
US10739318B2 (en) * 2017-04-19 2020-08-11 Baker Hughes, A Ge Company, Llc Detection system including sensors and method of operating such
CN107132521B (zh) * 2017-05-16 2019-12-06 哈尔滨工程大学 一种bslam中地形匹配结果正确性判别方法
CN107526801A (zh) * 2017-08-21 2017-12-29 浙江理工大学 一种基于布朗桥的移动对象跟随模式挖掘方法
CN108399377B (zh) * 2018-02-08 2022-04-08 北京理工大学 一种基于模式分类的光学定位方法
CN109579824B (zh) * 2018-10-31 2022-12-27 重庆邮电大学 一种融入二维码信息的自适应蒙特卡诺定位方法
CN112268561A (zh) * 2020-10-12 2021-01-26 西北工业大学 一种融合磁场信息的机器人蒙特卡罗定位方法

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101476860A (zh) * 2009-01-22 2009-07-08 中国科学院电工研究所 一种高背景磁场下的磁定位方法及装置
CN101556154A (zh) * 2008-10-13 2009-10-14 美新半导体(无锡)有限公司 定位及路径地图生成系统及其数据采集分析方法
CN103175529A (zh) * 2013-03-01 2013-06-26 上海美迪索科电子科技有限公司 基于室内磁场特征辅助的行人惯性定位系统

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPH11298946A (ja) * 1998-04-09 1999-10-29 Locus:Kk 位置決めシステム
US6944465B2 (en) * 1998-09-22 2005-09-13 Polaris Wireless, Inc. Estimating the location of a mobile unit based on the elimination of improbable locations
US7733224B2 (en) * 2006-06-30 2010-06-08 Bao Tran Mesh network personal emergency response appliance
JP4032126B1 (ja) * 2006-07-05 2008-01-16 国立大学法人北陸先端科学技術大学院大学 移動体位置推定システム、及び、移動体位置推定方法
CN101520328B (zh) * 2009-04-01 2010-08-25 西北工业大学 地磁场线图匹配自主导航方法
WO2011144967A1 (en) * 2010-05-19 2011-11-24 Nokia Corporation Extended fingerprint generation
CN101867943A (zh) * 2010-06-23 2010-10-20 哈尔滨工业大学 基于粒子滤波算法的wlan室内跟踪方法
US8498811B2 (en) 2011-02-09 2013-07-30 SenionLab AB Method and device for indoor positioning
EP2817785B1 (en) * 2012-02-23 2019-05-15 Charles D. Huston System and method for creating an environment and for sharing a location based experience in an environment
CN102743176A (zh) * 2012-06-30 2012-10-24 安翰光电技术(武汉)有限公司 基于双磁场传感器阵列平面远程目标定位方法及系统

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101556154A (zh) * 2008-10-13 2009-10-14 美新半导体(无锡)有限公司 定位及路径地图生成系统及其数据采集分析方法
CN101476860A (zh) * 2009-01-22 2009-07-08 中国科学院电工研究所 一种高背景磁场下的磁定位方法及装置
CN103175529A (zh) * 2013-03-01 2013-06-26 上海美迪索科电子科技有限公司 基于室内磁场特征辅助的行人惯性定位系统

Also Published As

Publication number Publication date
EP3034998A1 (en) 2016-06-22
JP2016539333A (ja) 2016-12-15
CN104375117A (zh) 2015-02-25
CN104375117B (zh) 2018-05-04
KR101912233B1 (ko) 2018-10-26
US20160195401A1 (en) 2016-07-07
EP3034998A4 (en) 2017-05-10
KR20160042126A (ko) 2016-04-18

Similar Documents

Publication Publication Date Title
WO2015021903A1 (zh) 目标定位方法及系统
Xie et al. A reliability-augmented particle filter for magnetic fingerprinting based indoor localization on smartphone
Shu et al. Gradient-based fingerprinting for indoor localization and tracking
Jiang et al. Hallway based automatic indoor floorplan construction using room fingerprints
CN105044662B (zh) 一种基于wifi信号强度的指纹聚类多点联合室内定位方法
Wang et al. Toward accurate device-free wireless localization with a saddle surface model
CN108632761B (zh) 一种基于粒子滤波算法的室内定位方法
CN108919177B (zh) 一种基于虚拟信源估计与轨迹校正的定位地图构建方法
Zickler et al. RSS-based relative localization and tethering for moving robots in unknown environments
CN105554873B (zh) 一种基于pso-ga-rbf-hop的无线传感器网络定位算法
CN101820676B (zh) 传感器节点定位方法
CN105101090B (zh) 一种环境监测无线传感网的节点定位方法
Wendeberg et al. Anchor-free TDOA self-localization
Lim et al. Distributed localization for anisotropic sensor networks
Chen et al. CWDV-Hop: A hybrid localization algorithm with distance-weight DV-Hop and CSO for wireless sensor networks
Xiang et al. $ Tack: $ Learning towards contextual and ephemeral indoor localization with crowdsourcing
Wu et al. Cooperative motion parameter estimation using RSS measurements in robotic sensor networks
Chen et al. A connectivity weighting DV_Hop localization algorithm using modified artificial bee Colony optimization
Cao et al. Comparison of particle swarm optimization algorithms in wireless sensor network node localization
Wang et al. Virtual force-directed particle swarm optimization for dynamic deployment in wireless sensor networks
CN110493731A (zh) 移动轨迹获取方法、装置、存储介质以及设备
Ding et al. OGI-SLAM2: A hybrid map SLAM framework grounded in inertial-based SLAM
He et al. An efficient range-free localization algorithm for wireless sensor networks
Zhipeng et al. An improved hop size estimation for DV-hop localization algorithm in wireless sensor networks
CN114245291A (zh) 参考节点虚拟化为未知节点的测距定位方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14835947

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2016533803

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 14911865

Country of ref document: US

WWE Wipo information: entry into national phase

Ref document number: 2014835947

Country of ref document: EP

ENP Entry into the national phase

Ref document number: 20167006590

Country of ref document: KR

Kind code of ref document: A