CN111447549B - Non-uniform UWB positioning error set network construction method and positioning error modeling method - Google Patents
Non-uniform UWB positioning error set network construction method and positioning error modeling method Download PDFInfo
- Publication number
- CN111447549B CN111447549B CN201911425919.4A CN201911425919A CN111447549B CN 111447549 B CN111447549 B CN 111447549B CN 201911425919 A CN201911425919 A CN 201911425919A CN 111447549 B CN111447549 B CN 111447549B
- Authority
- CN
- China
- Prior art keywords
- positioning error
- positioning
- data acquisition
- data
- uniform
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/021—Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/02—Services making use of location information
- H04W4/023—Services making use of location information using mutual or relative location information between multiple location based services [LBS] targets or of distance thresholds
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W4/00—Services specially adapted for wireless communication networks; Facilities therefor
- H04W4/30—Services specially adapted for particular environments, situations or purposes
- H04W4/33—Services specially adapted for particular environments, situations or purposes for indoor environments, e.g. buildings
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
Abstract
The invention provides a non-uniform UWB positioning error set network construction method and a positioning error modeling method. The construction method of the non-uniform UWB positioning error set network comprises the steps of establishing a uniform acquisition network, uniformly acquiring network positioning error data and establishing the non-uniform acquisition network. The positioning error modeling method is used for modeling by acquiring the average positioning error, covariance matrix and weight of all the data acquisition points and combining two-dimensional Gaussian distribution to form a non-uniform UWB positioning error grid model. The invention can solve the technical problem of positioning interference of UWB in fixed structured indoor environment in the prior art, and can accurately perform wireless positioning.
Description
Technical Field
The invention belongs to the field of indoor positioning navigation, particularly relates to UWB wireless positioning related technology, and particularly relates to a non-uniform UWB positioning error set network construction method and a positioning error modeling method.
Background
Wireless positioning methods have been a hotspot in indoor positioning research. The present UWB (Ultra Wide-Band) technology has a great promotion effect on the wireless mobile positioning field, the positioning precision of the UWB technology is greatly improved compared with other wireless network positioning methods, and meanwhile, the cost and the data processing amount are far smaller than those of expensive positioning methods such as laser positioning and the like. The UWB has the advantages of low cost, multipath interference resistance, strong penetration capacity and the like in positioning, so that the UWB can be applied to positioning and tracking of static or moving objects and personnel, and can provide very accurate positioning precision. However, as a wireless positioning method, UWB still has certain limitations, and its positioning accuracy is also affected by time synchronization, multipath effect, sensor layout, NLOS, signal penetration, and the like, so that further data processing is required for UWB positioning data to improve its positioning accuracy and stability.
Disclosure of Invention
The invention aims to provide a non-uniform UWB positioning error set network construction method and a positioning error modeling method, which can solve the technical problem of positioning interference of UWB in a fixed structured indoor environment in the prior art and can accurately perform wireless positioning.
In order to achieve the above object, the present invention provides a method for constructing a non-uniform UWB positioning error set network, comprising the following steps:
s10, establishing a uniform acquisition network, wherein at least three ultra-wide band label nodes are arranged in a closed space, a plurality of data acquisition points are uniformly distributed in the closed space in an array manner, and the data acquisition points are connected in the closed space to form a meshed uniform acquisition network;
s20, uniformly acquiring network data, acquiring multiple groups of positioning data at each data acquisition point of the uniformly acquiring network, and calculating the average positioning error of each data acquisition point; and
s30, establishing a non-uniform acquisition network, redistributing the data acquisition points according to the average positioning error, wherein the distribution density of the data acquisition points is positively correlated with the average positioning error.
Further, after the step of uniformly collecting network data collecting S20 and before the step of establishing a non-uniform collecting network S30, the method further comprises:
s22, subdividing the average positioning error value, and interpolating between the data acquisition points by using a cubic spline interpolation method to subdivide the error precision of each data acquisition point;
in the step S30, redistributing the data acquisition points according to the error precision, wherein the distribution density of the data acquisition points is positively correlated with the error precision.
Further, before the step S22, the step S22 of dividing the average positioning error value further includes:
s21, establishing a coarse-precision positioning error network model, connecting data acquisition points in the closed space to form a gridded plane map, and labeling an average positioning error value on each data acquisition point in the plane map to form the coarse-precision positioning error network model.
Further, after the step S22, the method further includes:
s23, establishing a rough fitting positioning error network model, and forming a rough fitting positioning error network model through data fitting according to the positioning data measurement result on each data acquisition point and the overall trend of the rough precision positioning error network model.
Further, after the step S23 of establishing a rough-fit positioning error network model, the method further includes:
s24, calculating a positioning error gradient, namely calculating the positioning error gradient between each data acquisition point and the surrounding data acquisition points through a Robert operator;
in the step S30, calculating an error change rate according to the gradient of the positioning error value, and redistributing the data acquisition points according to the error change rate.
Further, after the step of calculating a positioning error gradient S24, the method further includes:
s25, establishing a rough fitting positioning error gradient network model, and converting the rough fitting positioning error gradient network model to form the rough fitting positioning error gradient network model by taking the Robert gradient of the adjacent data acquisition points as the positioning error change rate.
Further, before the step S30 of establishing a non-uniform acquisition network, the method further includes:
s26, building an integrated positioning error network model, and combining the rough fitting positioning error network model and the rough fitting positioning error gradient network model to form the integrated positioning error network model.
Further, after the step S30 of establishing a non-uniform acquisition network, the method further includes:
s40, establishing a positioning error probability model of the non-uniform acquisition grid, which specifically comprises
S41, acquiring data of the non-uniform acquisition network, namely acquiring multiple groups of positioning data at each data acquisition point of the non-uniform acquisition network;
s42, clustering positioning data, namely clustering according to the discrete distribution condition of the multiple groups of positioning data of each data acquisition point to make the positioning data of each cluster obey two-dimensional normal distribution;
s43, calculating and obtaining the positioning error probability density of each data acquisition point by weighted summation; and
s44, modeling is carried out through two-dimensional Gaussian distribution of the positioning error probability density of each cluster of each data acquisition point, and a positioning error probability model of the non-uniform acquisition network is formed.
Further, the step S43 of calculating and obtaining the probability density of positioning error of each data acquisition point by weighted summation specifically includes the steps of:
s431, calculating and obtaining the positioning error probability density of each cluster of each data acquisition point;
s432, calculating the weight of the positioning data of each cluster, wherein the weight is the proportion of the number of the positioning data in each cluster to the total number of the positioning data of the data acquisition point; and
and S433, calculating and obtaining the positioning error probability density of each data acquisition point by weighting and summing the positioning error probability densities of all the clusters.
The invention also provides a positioning error modeling method, which comprises the following steps:
s50, constructing a non-uniform acquisition network by adopting the non-uniform UWB positioning error set network construction method;
s51, acquiring the positioning error of each cluster of each data acquisition point;
s52, calculating and acquiring the average positioning error of each cluster of each data acquisition point;
s53, calculating and acquiring a covariance matrix of each cluster of each data acquisition point;
s54, calculating the weight of the positioning data of each cluster, wherein the weight is the proportion of the number of the positioning data in each cluster to the total number of the positioning data of the data acquisition point;
s55, sorting and obtaining the average positioning error, covariance matrix and weight of all the data acquisition points; and
and S56, modeling by combining the data acquired by each data acquisition point with two-dimensional Gaussian distribution to form a non-uniform UWB positioning error grid model.
The invention has the beneficial effects that the invention provides a non-uniform UWB positioning error set network construction method and a positioning error modeling method, which can solve the technical problem of positioning interference of UWB in a fixed structured indoor environment in the prior art and can accurately perform wireless positioning.
Drawings
FIG. 1 is a schematic structural view of the enclosed space in an embodiment of the present invention;
FIG. 2 is a flow chart of a method for constructing a non-uniform UWB positioning error set network according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the non-uniform acquisition network according to an embodiment of the present invention;
FIG. 4 is a coarse positioning error network model according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a coarse fit positioning error network model according to an embodiment of the present invention;
FIG. 6 is a rough fit localization error gradient network model according to an embodiment of the present disclosure;
FIG. 7 is a flowchart of the steps for establishing a probability model of localization errors for a non-uniform acquisition grid in an embodiment of the present invention;
FIG. 8 is a flowchart of the step of calculating the probability density of the positioning error for each data acquisition point according to the weighted summation in the embodiment of the present invention;
FIG. 9 is a flowchart of a positioning error modeling method according to an embodiment of the present invention.
The various components in the figures are numbered as follows:
10. the system comprises a closed space, 20 ultra-wideband tag nodes, 30 data acquisition points.
Detailed Description
The preferred embodiments of the present invention will be described in full hereinafter with reference to the accompanying drawings, for the technical content thereof to be more clearly understood. The present invention may be embodied in many different forms of embodiments and its scope is not limited to the embodiments set forth herein.
In the drawings, elements having the same structure are denoted by the same reference numerals, and elements having similar structure or function are denoted by the same reference numerals throughout. Directional phrases used in this disclosure, such as, for example, upper, lower, front, rear, left, right, inner, outer, upper, lower, side, top, bottom, front, rear, end, etc., are used in the drawings only for the purpose of explaining and illustrating the present invention and are not intended to limit the scope of the present invention.
When certain components are described as being "on" another component, the components can be directly on the other component; there may also be an intermediate member disposed on the intermediate member and the intermediate member disposed on the other member. When an element is referred to as being "mounted to" or "connected to" another element, they may be directly "mounted to" or "connected to" the other element or indirectly "mounted to" or "connected to" the other element through an intermediate element.
As shown in fig. 1, an embodiment of the present invention provides a method for constructing a non-uniform UWB positioning error set network, where a closed space 10 is selected as a map, and at least one ultra-wide band labeled node 20, also called UWB anchor node, is preferably 3, and is distributed in the closed space 10 in a triangular manner. In order to obtain sufficient map prior information, it is necessary to set enough data acquisition points 30 uniformly distributed in the map and establish a database with a large data volume for storing signal data. In fig. 1, circles represent the data acquisition points 30, triangles represent the locations of the ultra-wide tagged nodes 20, one data coarse acquisition data point is located every 2m of space, and a total of 25 data acquisition points 30 are located in approximately 8 x 10m of space.
As shown in fig. 1 and fig. 2, the method for constructing the non-uniform UWB positioning error set network includes the following steps:
s10, establishing a uniform acquisition network, wherein at least three ultra-wideband label nodes 20 are arranged in a closed space 10, a plurality of data acquisition points 30 are uniformly distributed in the closed space 10 in an array manner, and the data acquisition points 30 are connected in the closed space 10 to form a meshed uniform acquisition network for measurement and positioning and convenient for forming a network model for data analysis;
s20, acquiring uniformly network data, acquiring multiple groups of positioning data, preferably 100 positioning data, at each data acquisition point 30 of the uniformly network, so that in the construction of the preliminary error distribution in the previous stage, a total of 2500 data are acquired, and the average positioning error of each data acquisition point 30 is calculated; the average positioning error is an average value of the positioning errors of the multiple groups of positioning data of each data acquisition point 30; after the data acquisition points 30 with coarse accuracy are acquired, the preliminary distribution of the positioning accuracy in the closed space 10 can be obtained, and the average value of the positioning error of each data acquisition point 30 can be expressed in a mode of the average value of the positioning error, namely the average value of the positioning error of each measurement point can be obtained; and
s30, establishing a non-uniform acquisition network, redistributing the data acquisition points 30 according to the average positioning error, wherein the distribution density of the data acquisition points 30 is positively correlated with the average positioning error.
Since the cause of the UWB positioning error is mainly the influence of NLOS, multipath, and the like, the error distribution of a certain position is relatively fixed in an indoor environment where the position is relatively fixed. Because the UWB positioning errors are distributed differently in each indoor area, in areas with higher accuracy, the UWB positioning accuracy difference of each data acquisition point 30 is smaller, and the positioning error accuracy can be described with similar accuracy, so that the data acquisition grids in the area can be relatively sparse to reduce the data acquisition amount when the error map is created. In the area with relatively large error variation (such as the existence of NLOS boundary), the measurement precision is improved by improving the density of the data acquisition points 30, so as to describe the positioning error in the map more accurately.
As shown in fig. 3, it is a schematic diagram of the structure of the non-uniform acquisition network. Wherein the distribution density of the data acquisition points 30 in the step S30 of establishing a non-uniform acquisition network comprises at least three density levels. In the embodiment, the large positioning error grids are arranged in the high-precision (i.e., small error) area, and the fine grids are arranged in the low-precision (large error) area, so that the number of the data acquisition points 30 is reduced on the premise of ensuring the positioning precision as much as possible, and the efficiency of establishing the error grids is improved.
As shown in fig. 2, in this embodiment, after the step S20 of uniformly collecting network data, the method further includes:
s21, establishing a coarse-precision positioning error network model, connecting data acquisition points 30 in the closed space 10 to form a gridded plane map, and labeling an average positioning error value on each data acquisition point 30 in the plane map to form the coarse-precision positioning error network model.
As shown in fig. 4, an error network model is positioned for the coarse precision.
As shown in fig. 2, in this embodiment, after the step S21 of establishing the coarse-precision positioning error network model, the method further includes:
s22, subdividing the average positioning error value, interpolating between the data acquisition points 30 by a cubic spline interpolation method, so as to subdivide the error precision of each data acquisition point 30;
in the step S30, redistributing the data collection points 30 according to the error accuracy, wherein the distribution density of the data collection points 30 is positively correlated with the error accuracy.
As shown in fig. 2, in this embodiment, after step S22, the method further includes:
s23, establishing a rough fitting positioning error network model, and forming a rough fitting positioning error network model through data fitting according to the positioning data measurement result on each data acquisition point 30 and the overall trend of the rough precision positioning error network model.
As shown in fig. 5, an error network model is positioned for the coarse fit.
As shown in fig. 2, in this embodiment, after the step S23 of establishing a rough-fit positioning error network model, the method further includes:
s24, calculating a positioning error gradient, namely calculating the positioning error gradient between each data acquisition point and the surrounding data acquisition points 30 through a Robert operator; that is, calculating the Robert gradient (Robert gradient) between each of the data acquisition points 30 and the surrounding data acquisition points 30, and expressing the rate of change of the error between the point and the surrounding points by the Robert gradient;
in the step S30, calculating an error change rate according to the gradient of the positioning error value, and redistributing the data collection points 30 according to the error change rate, wherein the distribution density of the data collection points 30 is positively correlated to the magnitude of the robert gradient.
As shown in fig. 2, in this embodiment, after the step S24 of calculating the positioning error gradient, the method further includes:
s25, establishing a rough fitting positioning error gradient network model, and converting the rough fitting positioning error gradient network model to form the rough fitting positioning error gradient network model by taking the Robert gradient of the adjacent data acquisition points 30 as the positioning error change rate. Wherein the positioning error change rate is a gradient change rule of each data acquisition point 30 position.
As shown in fig. 6, an error gradient network model is positioned for the coarse fit.
In this embodiment, before the step S30 of establishing the non-uniform acquisition network, the method further includes:
s26, building an integrated positioning error network model, and combining the rough fitting positioning error network model and the rough fitting positioning error gradient network model to form the integrated positioning error network model.
The step of building a non-uniform acquisition network S30 is included after the step of building an integrated positioning error network model S26. Through the processing of the UWB coarse positioning data, the predicted positioning precision of each grid point in the indoor environment and the error change gradient of the grid point are obtained. For the establishment of the subsequent UWB fine positioning error grid, the establishment of the data acquisition point needs to consider the following conditions:
(1) at a position with higher positioning precision and smaller error change rate, fewer data acquisition nodes can be arranged, namely the overall error distribution in the range can be approximately represented;
(2) in a position with high positioning accuracy and a large error change rate (such as the case of NLOS), more acquisition nodes need to be arranged in the local part to describe the influence of the NLOS in the range as much as possible;
(3) at a position with lower positioning precision and smaller error change rate, relatively more data acquisition nodes can be arranged so as to improve the description of the area on the error as much as possible;
(4) the most dense data acquisition nodes should be set at the positions with lower positioning accuracy and larger error change rate.
Based on the above-described situations established for data acquisition points, it can be seen that the positioning accuracy and the rate of change of the positioning accuracy in a certain area determine how many data acquisition points are arranged in the area (i.e., the density of the data acquisition grid). And combining the established position error and the error change gradient, and integrating the error and the error gradient of each point to make the decision of collecting the point density.
And when the distribution of error data acquisition points in the positioning error grid is constructed, integrating the fitted error information and the error gradient information to divide the heterogeneity of the error map data acquisition grid. Wherein in the region where the sum of the error and the error gradient is small, it is divided into a grid of 20 x 20 cm; dividing the error and the error gradient into 15 × 15cm areas; in the region where the sum of the error and the error gradient is large, the grid is divided into 10 x 10cm grids. In the process of accurate UWB positioning error modeling acquisition, the intermediate point of each grid is used as a data acquisition point of the area represented by the grid, forming an error data acquisition grid as shown in fig. 3.
In this embodiment, after the step S30 of establishing the non-uniform acquisition network, the method further includes:
and S40, establishing a positioning error probability model of the non-uniform acquisition grid.
As shown in fig. 7, the step S40 of establishing the probability model of the localization error of the non-uniform acquisition grid specifically includes the steps of:
s41, acquiring data of the non-uniform acquisition network, namely acquiring multiple groups of positioning data at each data acquisition point 30 of the non-uniform acquisition network; the number of UWB positioning error data acquisitions at each data acquisition point 30 is preferably 500;
s42, clustering positioning data, namely clustering according to the discrete distribution condition of the multiple groups of positioning data of each data acquisition point 30, so that the positioning data of each cluster obey two-dimensional normal distribution; when the UWB positioning error data are multi-cluster, the error information can be regarded as multi-cluster two-dimensional normal distribution, and is also a common error distribution form in UWB positioning error data acquisition, the number of clusters is generally 2-3, and the condition of a small number of more clusters of two-dimensional normal distribution also exists;
s43, calculating and obtaining the positioning error probability density of each data acquisition point 30 by weighted summation; and
s44, modeling is carried out through two-dimensional Gaussian distribution of the positioning error probability density of each cluster of each data acquisition point 30, and a positioning error probability model of the non-uniform acquisition grid is formed.
As shown in fig. 8, in the present embodiment, the step S43 of calculating the probability density of positioning error for acquiring each data acquisition point 30 by weighted summation specifically includes steps S431 to S433.
S431, calculating and obtaining the positioning error probability density of each cluster of each data acquisition point 30;
first, define the error vector of the current UWB positioning data:
from the error vector, the point (x) can be obtainedn,yn) The measured data set may be expressed as:
wherein z isx,tAnd zy,tFor the t-th acquired positioning data in the plurality of sets of positioning data of the point, 0<t≤500；
By the above method, x in each data acquisition Point 30n＝[x,y]TThe error of 500 collected groups of data is calculated: e.g. of the typet＝zt-xn(ii) a Thereby obtaining the error e of each positioning data of the data acquisition point 30t。
According to the positioning data acquired at the above-mentioned set data acquisition points 30 in the present invention, the above-mentioned positioning error data acquired at the acquisition points is also modeled by a single-cluster or multi-cluster two-dimensional gaussian distribution. The probability density function for the positioning error data of the data acquisition points 30 can be expressed as:
wherein i represents the ith cluster, mu, in the error data acquired by the acquisition pointx,i,μy,iRespectively, the average values of the positioning errors in the ith cluster of data (x, y) direction;
are the variance of the positioning error in the (x, y) direction of the ith cluster of data, respectively, and ρx,y,iAre correlation coefficients describing the linear relationship of the positioning errors in the (x, y) direction of the ith cluster of data with each other.
E is as described abovex,eyIs denoted by e, i.e.
The above formula can be expressed as
Wherein, muiIs a vector representing the mean of the error data of the ith cluster, andia covariance matrix for the cluster of error data, of the form:
s432, calculating the weight of the positioning data of each cluster, wherein the weight is the proportion of the number of the positioning data in each cluster to the total number of the positioning data of the data acquisition point 30;
the method for calculating the weight of each cluster of positioning data comprises the following steps:
wherein N isn,iThe number of ith cluster data in 500 measurement data is counted.
S433, calculating and obtaining the positioning error probability density of each data acquisition point 30 by weighting and summing the positioning error probability densities of all clusters;
through the two-dimensional Gaussian model description of the positioning error result distribution of the measuring points, the following corresponding x in the data acquisition points 30 can be obtainednThe probability density function of the positioning error data of (1):
the abbreviation is:
wherein, InIs represented by xnNumber of clusters, omega, of measured data at a measurement pointn,iDenotes xnThe weight of the ith cluster of data at the measurement point.
As shown in FIG. 9, the present invention further provides a positioning error modeling method, which comprises steps S50-S56.
S50, constructing a non-uniform acquisition network by adopting the non-uniform UWB positioning error set network construction method.
S51, an average positioning error for each cluster that acquired each data acquisition point 30 is calculated, which may be obtained based on the probability density distribution p (e | x).
S52, calculating and acquiring the average positioning error of each cluster of each data acquisition point 30;
at data acquisition Point 30xnIs the mean value mu of the ith cluster of positioning error datan,iCan be expressed as:
where N represents the number of error data measured at that point, i.e., 500; e.g. of the typen,iIndicates the ith cluster in the error data set (e)x,ey) Is/are as followsError data.
S53, calculating and acquiring a covariance matrix of each cluster of each data acquisition point 30;
covariance matrix sigma of i-cluster datan,iCan be expressed as:
s54, calculating the weight of the positioning data of each cluster, wherein the weight is the proportion of the number of the positioning data in each cluster to the total number of the positioning data of the data acquisition point 30;
ωn,irepresenting the measurement point xnThe weight of the ith cluster data is used for weight representation in the subsequent probability density function calculation in the same way as the above calculation method of the weight of each cluster positioning data.
S55, sorting and obtaining the average positioning error, covariance matrix and weight of all the data acquisition points 30;
in summary, the coordinates x for a grid set according to acquisition points in the gridnThe neatly acquired UWB positioning error may be expressed as:
s56, data acquired by each data acquisition point 30
Modeling is carried out by combining two-dimensional Gaussian distribution to form a non-uniform UWB positioning error grid model. The distribution of UWB positioning errors on the whole map grid can be obtained and used in the subsequent positioning process to improve the UWB positioning accuracy.
The invention has the beneficial effects that the invention provides a non-uniform UWB positioning error set network construction method and a positioning error modeling method, which can solve the technical problem of positioning interference of UWB in a fixed structured indoor environment in the prior art and can accurately perform wireless positioning.
The above description is only of the preferred embodiments of the present invention to make it clear for those skilled in the art how to practice the present invention, and these embodiments do not limit the scope of the present invention. It will be apparent to those skilled in the art that various modifications and enhancements can be made without departing from the principles of the invention, and such modifications and enhancements are intended to be included within the scope of the invention.
Claims (9)
1. A non-uniform UWB positioning error set network construction method is characterized by comprising the following steps:
s10, establishing a uniform acquisition network, wherein at least three ultra-wide band label nodes are arranged in a closed space, a plurality of data acquisition points are uniformly distributed in the closed space in an array manner, and the data acquisition points are connected in the closed space to form a meshed uniform acquisition network;
s20, uniformly acquiring network data, acquiring multiple groups of positioning data at each data acquisition point of the uniformly acquiring network, and calculating the average positioning error of each data acquisition point; and
and S30, establishing a non-uniform acquisition network, redistributing the data acquisition points according to the average positioning error, wherein the distribution density of the data acquisition points is determined by the positioning accuracy of the area where the data acquisition points are located and the change rate of the positioning accuracy.
2. The method of constructing a non-uniform UWB positioning error set network of claim 1 further comprising, after the step of uniformly collecting network data collection S20 and before the step of establishing a non-uniform collection network S30:
s22, subdividing the average positioning error value, namely, interpolating between the data acquisition points by using a cubic spline interpolation method to subdivide the average positioning error value of each data acquisition point;
in the step S30, the data acquisition points are redistributed according to the average positioning error value, and the distribution density of the data acquisition points is determined by the positioning accuracy of the area where the data acquisition points are located and the variation rate of the positioning accuracy.
3. The method of constructing a non-uniform UWB positioning error set network of claim 2 further comprising, prior to the step S22 of subdividing the average positioning error values:
s21, establishing a coarse-precision positioning error network model, connecting data acquisition points in the closed space to form a gridded plane map, and labeling an average positioning error value on each data acquisition point in the plane map to form the coarse-precision positioning error network model.
4. The method of constructing a non-uniform UWB positioning error set network of claim 3 further comprising, after the step S22 of subdividing the average positioning error values:
s23, establishing a rough fitting positioning error network model, and forming a rough fitting positioning error network model through data fitting according to the positioning data measurement result on each data acquisition point and the overall trend of the rough precision positioning error network model.
5. The method of constructing a non-uniform UWB positioning error set network of claim 4 further comprising, after the step of establishing a coarse-fit positioning error network model S23:
s24, calculating a positioning error gradient, namely calculating the positioning error gradient between each data acquisition point and the surrounding data acquisition points through a Robert operator;
in the step S30, calculating an error change rate according to the gradient of the positioning error value, and redistributing the data acquisition points according to the error change rate.
6. The method of constructing a non-uniform UWB positioning error set network of claim 5 further comprising, after said step of computing a positioning error gradient S24:
s25, establishing a rough fitting positioning error gradient network model, and converting the rough fitting positioning error gradient network model to form the rough fitting positioning error gradient network model by taking the Robert gradient of the adjacent data acquisition points as the positioning error change rate.
7. The method for constructing a non-uniform UWB positioning error set network according to claim 6, further comprising, before the step S30 of establishing a non-uniform acquisition network:
s26, building an integrated positioning error network model, and combining the rough fitting positioning error network model and the rough fitting positioning error gradient network model to form the integrated positioning error network model.
8. The method for constructing a non-uniform UWB positioning error set network according to claim 1, further comprising, after the step S30 of establishing a non-uniform acquisition network:
s40, establishing a positioning error probability model of the non-uniform acquisition grid, which specifically comprises
S41, acquiring data of the non-uniform acquisition network, namely acquiring multiple groups of positioning data at each data acquisition point of the non-uniform acquisition network;
s42, clustering positioning data, namely clustering according to the discrete distribution condition of the multiple groups of positioning data of each data acquisition point to make the positioning data of each cluster obey two-dimensional normal distribution;
s43, calculating and obtaining the positioning error probability density of each data acquisition point by weighted summation; and
s44, modeling is carried out through two-dimensional normal distribution of the positioning error probability density of each cluster of each data acquisition point, and a positioning error probability model of the non-uniform acquisition network is formed;
wherein the step S43 of calculating and obtaining the probability density of the positioning error of each data acquisition point by weighted summation specifically includes:
s431, calculating and obtaining the positioning error probability density of each cluster of each data acquisition point;
s432, calculating the weight of the positioning data of each cluster, wherein the weight is the proportion of the number of the positioning data in each cluster to the total number of the positioning data of the data acquisition point; and
and S433, calculating and obtaining the positioning error probability density of each data acquisition point by weighting and summing the positioning error probability densities of all the clusters.
9. A method of modeling positioning error, comprising the steps of:
s50, constructing a non-uniform acquisition network by adopting the non-uniform UWB positioning error set network construction method of claim 8;
s51, acquiring the positioning error of each cluster of each data acquisition point;
s52, calculating and acquiring the average positioning error of each cluster of each data acquisition point;
s53, calculating and acquiring a covariance matrix of each cluster of each data acquisition point;
s54, calculating the weight of the positioning data of each cluster, wherein the weight is the proportion of the number of the positioning data in each cluster to the total number of the positioning data of the data acquisition point;
s55, sorting and obtaining the average positioning error, covariance matrix and weight of all the data acquisition points; coordinates x for a grid set according to acquisition points in the gridnAnd arranging the acquired non-uniform UWB positioning error as
And
s56, data acquired by each data acquisition point
Modeling is carried out by combining two-dimensional normal distribution, and a non-uniform UWB positioning error grid model is formed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911425919.4A CN111447549B (en) | 2019-12-31 | 2019-12-31 | Non-uniform UWB positioning error set network construction method and positioning error modeling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911425919.4A CN111447549B (en) | 2019-12-31 | 2019-12-31 | Non-uniform UWB positioning error set network construction method and positioning error modeling method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111447549A CN111447549A (en) | 2020-07-24 |
| CN111447549B true CN111447549B (en) | 2021-06-15 |
Family
ID=71653946
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911425919.4A Active CN111447549B (en) | 2019-12-31 | 2019-12-31 | Non-uniform UWB positioning error set network construction method and positioning error modeling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111447549B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107979817A (en) * | 2017-11-24 | 2018-05-01 | 王红军 | A kind of mobile terminal two dimension fingerprint positioning method |
| CN109362036A (en) * | 2018-10-17 | 2019-02-19 | 桂林电子科技大学 | A kind of multi-modal indoor orientation method combined based on image with WIFI |
| CN109640251A (en) * | 2018-12-10 | 2019-04-16 | 北京众行智能科技有限公司 | A kind of indoor localization method and device |
| CN110418283A (en) * | 2019-06-25 | 2019-11-05 | 北京航空航天大学 | Location fingerprint base establishing method based on WiFi data |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103747419B (en) * | 2014-01-15 | 2017-09-08 | 福建师范大学 | A kind of indoor orientation method based on signal strength difference and dynamic linear interpolation |
| CN104507097A (en) * | 2014-12-19 | 2015-04-08 | 上海交通大学 | Semi-supervised training method based on WiFi (wireless fidelity) position fingerprints |
-
2019
- 2019-12-31 CN CN201911425919.4A patent/CN111447549B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107979817A (en) * | 2017-11-24 | 2018-05-01 | 王红军 | A kind of mobile terminal two dimension fingerprint positioning method |
| CN109362036A (en) * | 2018-10-17 | 2019-02-19 | 桂林电子科技大学 | A kind of multi-modal indoor orientation method combined based on image with WIFI |
| CN109640251A (en) * | 2018-12-10 | 2019-04-16 | 北京众行智能科技有限公司 | A kind of indoor localization method and device |
| CN110418283A (en) * | 2019-06-25 | 2019-11-05 | 北京航空航天大学 | Location fingerprint base establishing method based on WiFi data |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111447549A (en) | 2020-07-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107180450B (en) | DEM-based river valley cross section morphology algorithm | |
| CN104038901B (en) | Indoor positioning method for reducing fingerprint data acquisition workload | |
| CN105118090A (en) | Adaptive point-cloud filtering method for complex terrain structure | |
| CN103747419A (en) | Indoor positioning method based on signal intensity difference values and dynamic linear interpolation | |
| CN103256914B (en) | A kind of method and system calculating silt arrester inundated area based on DEM | |
| CN103428629A (en) | Hybrid location implementation method and system | |
| CN103249144A (en) | C-type-based wireless sensor network node location method | |
| CN107179525A (en) | A kind of location fingerprint construction method of the Kriging regression based on Thiessen polygon | |
| CN110032939A (en) | A kind of remote sensing time series data approximating method based on gauss hybrid models | |
| CN101873605A (en) | Adaptive method for classifying communication environments in network planning | |
| CN106570936A (en) | Grid DEM (digital elevation model) data-based equidistant weight interpolation encryption method | |
| CN111447549B (en) | Non-uniform UWB positioning error set network construction method and positioning error modeling method | |
| CN113139337A (en) | Partitioned interpolation processing method and device for lake terrain simulation | |
| CN115442887B (en) | Indoor positioning method based on RSSI of cellular network | |
| CN104635206B (en) | A kind of method and device of wireless location | |
| CN113301503B (en) | Weak coverage area detection method and device | |
| CN110426562B (en) | High-precision lightning three-dimensional positioning method based on layered search and distance space projection | |
| CN111898819B (en) | Space grid dividing method and device | |
| US20150308809A1 (en) | Apparatus and method for calculating ground surface roughness | |
| CN113935658A (en) | Mountain land utilization change evaluation method based on terrain gradient | |
| CN112015784A (en) | Wind condition data mining method and device, wind measuring device and data mining equipment | |
| CN113743027A (en) | Method and device for drawing wind resource map based on CFD technology | |
| CN101964024B (en) | Method for fast determining gas phase unstructured grid of solid particle | |
| CN105844086A (en) | Arrangement structure of pressure sensor | |
| CN116628411B (en) | High-precision flow online monitoring intelligent method based on full-sense fusion |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |