A method for improving accuracy of UWB positioning data

Abstract

This invention relates to a method for improving the accuracy of UWB positioning data, belonging to the field of precise positioning technology in mines. Specifically, the method involves: setting a noise threshold and an empirical medium constant at the positioning receiver, while the positioning system receives the positioning signal; determining whether multipath effects exist in the received signal; if multipath effects exist, calculating the probability PT of no direct path extraction, and using PT to determine whether a direct path is not extracted due to an excessively high noise threshold; the positioning algorithm performs positioning calculations on the successfully extracted direct paths to obtain positioning coordinates (x0, y0, z0); and feeding the positioning coordinates (x0, y0, z0) into a Kalman filter algorithm for filtering to obtain the filtered result (x0, y0, z0). n+1 ,y n+1 ,z n+1 By comparing (x0, y0, z0) and (x n+1 ,y n+1 ,z n+1 The system determines whether the current positioning data is reliable; if the positioning data is reliable, the positioning value is optimized using Kalman filtering, and the current positioning ends. This invention can improve the accuracy and reliability of UWB positioning data, and improve the precision of underground mine positioning.

Description

Technical Field

[0001] This invention belongs to the field of precise positioning technology in mines, and relates to underground personnel management, particularly to a method for improving the accuracy of UWB positioning data. Background Technology

[0002] Currently, UWB-based precise positioning systems offer long coverage distances and high positioning accuracy in open environments, with relatively accurate and reliable positioning data. However, in underground coal mines, due to factors such as protruding coal walls, narrow tunnels, and severe obstruction, the positioning signal received by the receiver may exhibit multiple transmission paths, including direct transmission, reflection, and refraction (e.g.,...). Figure 1 As shown, the direct path is the straight-line transmission path of the positioning signal between the base station and the terminal. The signal flight time through this path can be used to obtain the accurate distance between the base station and the terminal, which is the path that needs to be extracted during the positioning process. Reflected and refracted signals are interference signals, which will cause inaccurate positioning results and are data that needs to be filtered out during the positioning process. Therefore, when performing UWB positioning in coal mines, the wireless signal has a multipath effect in the transmission space. Especially in the confined space of coal mines where the walls (coal walls) are uneven, the multipath effect is particularly serious, leading to large fluctuations and inaccuracies in the positioning data. This results in large positioning errors and seriously affects the reliability of the positioning data.

[0003] Therefore, it is necessary to design direct path extraction methods and data optimization methods for the complex scenarios in coal mines to ensure the high accuracy and reliability of positioning data. Summary of the Invention

[0004] In view of this, the purpose of this invention is to provide a method for improving the accuracy of UWB positioning data by extracting and optimizing the direct path to ensure high accuracy and reliability of the positioning data.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for improving the accuracy of UWB positioning data, characterized by the following steps:

[0007] S1. Before starting the positioning, first initialize and set the noise threshold and empirical medium constant of the UWB positioning device; the noise threshold is used to shield the noise generated by the hardware and the environmental noise, and the empirical medium constant verifies the correctness of the path extraction.

[0008] S2. After the positioning signal is received, the multipath effect probability POS of the positioning signal is calculated based on the multipath effect probability calculation model established by the extracted direct path, maximum peak position and other parameters, and it is determined whether there is a multipath effect in this positioning.

[0009] S3. If there is a multipath effect in the received signal, calculate the correct probability PT of the extracted direct path signal based on the extraction probability model of the direct path established by parameters such as the number of signal peaks and the length of the analysis window. Use PT to determine whether there is a direct path that has not been extracted due to the noise threshold being set too high.

[0010] S4. Use the TOF / TDOA positioning algorithm to calculate the positioning coordinates (x0, y0, z0) of the successfully extracted direct path.

[0011] S5. Input the positioning coordinates (x0, y0, z0) into the Kalman filter algorithm model with state parameters such as the target's movement speed and positioning time, and calculate the result using filtering. n+1 ,y n+1 ,z n+1 By comparing (x0, y0, z0) and (x n+1 ,y n+1 ,z n+1 Determine whether the location data is reliable.

[0012] S6. If the positioning data is reliable, optimize the positioning value through Kalman filtering and end the positioning process.

[0013] Furthermore, in step S2, the peak signal above the noise threshold before the maximum peak of the received signal is taken as the direct path signal, the maximum peak signal in the received signal is taken as the multipath signal, the difference between the position of the direct path signal and the position of the multipath signal is set as DT, and the POS value is obtained by comparing DT with the empirical medium constant.

[0014] Furthermore, the empirical medium constants include B, C, K0, and K1;

[0015] (1) When DT≤B, POS=0, there is no multipath signal in the received signal, proceed to step S4;

[0016] (2) When DT>B&&DT≤C, the value of POS needs to be further calculated for analysis;

[0017] (3) When DT > C, POS = 1, direct path extraction failed, return to step S1.

[0018] Furthermore, when DT > B and DT ≤ C, the POS value is calculated using the following formula:

[0019] POS = K0 * DT - K1

[0020] If POS ≤ 0.5, the direct path extraction is successful, and proceed to step S4; if POS > 0.5, it is impossible to determine whether a multipath effect exists, and proceed directly to step S3.

[0021] Furthermore, in step S3, PT is calculated using the following formula:

[0022] PT = NPA / ((LAW-1) / 2)

[0023] In the formula, NPA represents the number of peaks in the received signal, and LAW represents the duration of the received signal;

[0024] If PT>A, there are unextracted direct path signals, and it is necessary to return to step S1 to reset the noise threshold parameters;

[0025] If PT≤A, then the direct path signal extraction was successful.

[0026] Furthermore, in step S5, during Kalman filtering, a chi-square test is introduced to locate filtering errors; the chi-square test is introduced when calculating the Kalman gain, as follows:

[0027] During the Kalman update process, let Y vn+1 =(R n+1 +H n+1 P n+1|n H T ) -1 residual The chi-square test is expressed as:

[0028]

[0029] In the above formula, R represents the covariance matrix of measurement noise, H represents the state observation matrix, P represents the state covariance, and Z represents the measurement vector.

[0030] Furthermore, when When the positioning value is incorrect, Kalman filtering is not performed; when If the positioning value is correct, Kalman filtering is performed.

[0031] further, To avoid large measurement and estimation errors affecting the filtering results, the Kalman gain is improved by adding an adjustment factor:

[0032]

[0033] In the formula, thresh and α are both experimental values;

[0034] The filtered result is as follows:

[0035]

[0036] In the formula, This represents the state parameter at the current moment.

[0037] Furthermore, in step S5, if (x0, y0, z0) and (xn+1 ,y n+1 ,z n+1 If the difference between (x0, y0, z0) and (x...) is greater than M, then the location data is unreliable, and return to step S1; if the difference between (x0, y0, z0) and (x...) is greater than M, then the location data is unreliable, and return to step S1; n+1 ,y n+1 ,z n+1 If the difference between the two values ​​is less than M, then the location data is reliable.

[0038] Where M = V * T * KM; and V represents the average velocity, T represents the coordinates (x0, y0, z0) and (x... n+1 ,y n+1 ,z n+1 The acquisition time difference, where KM represents the threshold coefficient.

[0039] The beneficial effects of this invention are as follows: This invention designs a direct path extraction algorithm for multiple transmission paths, directly extracts the direct path signal from the positioning signal for positioning calculation, and at the same time uses Kalman filtering to optimize the calculation results, thereby improving the accuracy and reliability of the positioning data.

[0040] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0041] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0042] Figure 1 This is a schematic diagram of the signal transmission path during the positioning process between the base station and the terminal.

[0043] Figure 2 This is an overall flowchart of the present invention;

[0044] Figure 3 This is a schematic diagram of the received signal;

[0045] Figure 4 This is a diagram illustrating an unidentified direct path. Detailed Implementation

[0046] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0047] This invention addresses the issue of accurately extracting the direct transmission path from various transmission paths such as direct transmission, reflection, and refraction, and proposes a method to improve the accuracy of positioning data by performing positioning calculations and filtering optimizations on the direct transmission path. Figure 1 As shown, the method is as follows:

[0048] I. Localization Solution Based on Direct Path Extraction

[0049] 1. Multipath environment assessment and direct path extraction

[0050] (1) First, parameters such as the noise threshold, reception threshold, and empirical medium constant of the positioning receiver are set. The noise threshold is used to shield noise generated by the hardware and environmental noise, which facilitates the analysis and extraction of the direct path. This invention considers the received signal below the noise threshold as noise, which is not considered a valid positioning signal; the received signal above the noise threshold is a valid signal. When there is no multipath effect, the valid signal only contains the direct path signal. When there is a multipath effect in a complex space, the valid signal includes the direct path signal, reflection and refraction signals. It is necessary to correctly analyze and extract the direct path signal to achieve accurate positioning calculation.

[0051] (2) Secondly, when multipath effects exist in the wireless transmission space, the receiver will receive electromagnetic wave signals that are amplified due to the multipath effect. In this invention, the peak signal above the noise threshold before the maximum peak of the received signal is considered the direct path signal, and the maximum peak signal in the received signal is the multipath signal, such as... Figure 2 As shown, the position difference between the direct path signal and the multipath signal is set as DT.

[0052] DT = |FTP-PPT|

[0053] Where FTP represents the peak position of the direct path signal, and PPT represents the maximum peak position in the received signal. The empirical medium constants defined in this invention include B, C, K0, and K1. The POS value is obtained by comparing DT with the empirical medium constants.

[0054] 1) When DT≤B, POS=0, there is no multipath signal in the received signal, which means that the extracted direct path is correct and the positioning solution can be performed directly;

[0055] 2) When DT>B&&DT≤C, it is assumed that there is a multipath effect in the received signal, and the value of POS needs to be further calculated for analysis;

[0056] The POS value is calculated using the following formula:

[0057] POS = K0 * DT - K1

[0058] If the calculated POS ≤ 0.5, the direct path extraction is considered successful, and the location solution is performed; if POS > 0.5, it is impossible to determine whether a multipath effect exists, and the subsequent two steps of analysis and processing are required to determine whether the direct path has been extracted.

[0059] 3) When DT > C, POS = 1, the direct path extraction failed, and signal reception needs to be restarted.

[0060] 2. Determine if there are any unextracted direct paths.

[0061] like Figure 4 As shown, if the direct path signal strength in the received signal is severely attenuated, the reflected signal in the received signal may be identified as the direct path, while the actual direct path signal is filtered out due to an excessively high noise threshold. Therefore, a corresponding solution needs to be designed for this situation.

[0062] Set the direct path to extract error parameter PT:

[0063] PT = NPA / ((LAW-1) / 2)

[0064] In the formula, NPA represents the number of peaks in the received signal, and LAW represents the duration of the received signal. If PT>A, it is considered that the received signal may have unextracted direct path signals due to an excessively high noise threshold setting, and the noise threshold parameters need to be reset. If PT≤A, it is considered that the direct path signal has been extracted, and the positioning calculation is performed.

[0065] 3. Adaptive positioning solution

[0066] The extracted direct path flight time is used to calculate the coordinates (x0, y0, z0) by using the TOF (Time of Flight) algorithm and the TDOA (Time Difference of Arrival) algorithm.

[0067] II. Kalman Filter Correction

[0068] The Kalman filter algorithm is used to perform secondary optimization on the positioning solution data. The positioning value is estimated or corrected by Kalman filtering, and the chi-square test is introduced to filter out erroneous positioning values.

[0069] 1. Kalman Filter Algorithm

[0070] (1) Predicting the state equation:

[0071]

[0072] (2) Predicting the covariance matrix:

[0073] P n+1|n =AP n|n A T +Q

[0074] (3) Calculate the Kalman gain:

[0075] K n+1|n =P n+1|n H T (R n+1 +H n+1 P n+1|n H T ) -1

[0076] (4) Status correction:

[0077]

[0078] (5) Calculate the minimum mean square matrix:

[0079] P n+1|n+1 =(IK n+1 H)P n+1|n

[0080] In the above formula, x n|n Let x be the state vector from the previous time step. n+1|n Let A be the current state vector, B be the state transition matrix, and u be the control input matrix. n+1 The initial state vector is 0 (default is 0), Q is the process noise covariance matrix, P is the state covariance, Z is the measurement vector, H is the state observation matrix, K is the Kalman gain, and R is the measurement noise covariance matrix.

[0081] 2. Introduce chi-square test when adjusting Kalman filter gain.

[0082] During the Kalman update process, let Y vn+1 =(R n+1 +H n+1 P n+1|n H T ) -1 residual The chi-square test is expressed as:

[0083]

[0084] when When the location value is incorrect, Kalman filtering is not performed; when... When the positioning value is correct, Kalman filtering is performed. To avoid large measurement and estimation errors affecting the filtering result, the Kalman gain is improved by adding an adjustment factor.

[0085]

[0086] In the formula, thresh and α are experimental values, and the filtering result is as follows:

[0087]

[0088] 3. Closed-loop feedback

[0089] The location value (x) after Kalman filtering n+1 ,y n+1 ,z n+1 The value is compared with (x0, y0, z0). If the difference between the two is greater than M, the current positioning value is considered unreliable, and the process returns to the initial step to modify the parameter settings. If the difference between the two is less than M, the current positioning value is considered reliable, and Kalman filtering is used to optimize the positioning value.

[0090] Where M is the difference threshold, M = V * T * KM; where V represents the average velocity, calculated based on the average velocity over the 10 minutes prior to the positioning time, and T represents the coordinates (x0, y0, z0) and (x... n+1 ,y n+1 ,z n+1 The acquisition time difference, where KM represents the threshold coefficient.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for improving the accuracy of UWB positioning data, characterized in that: The method includes the following steps: S1. Before starting the positioning, first initialize the noise threshold and empirical medium constant of the UWB positioning device; the noise threshold is used to shield the noise generated by the hardware and the environmental noise, and the empirical medium constant verifies the correctness of the path extraction. S2. After the positioning signal is received, calculate the multipath effect probability POS of the positioning signal according to the multipath effect probability calculation model, and determine whether the positioning has a multipath effect. Specifically, the peak signal above the noise threshold before the maximum peak of the received signal is taken as the direct path signal, and the maximum peak signal in the received signal is taken as the multipath signal. The difference between the position of the direct path signal and the position of the multipath signal is set as... DT ,pass DT The POS value is obtained by comparing it with the empirical medium constant; The empirical medium constant includes B, C, K 0、 K 1; (1) When POS=0, there is no multipath signal in the received signal, proceed to step S4; (2) At this point, further calculation of the POS value is needed for analysis; through the formula Calculate the POS value; if calculated If the direct path extraction is successful, proceed to step S4; if If the multipath effect is not present, it cannot be determined and the process proceeds directly to step S3. (3) When POS=1, direct path extraction failed, return to step S1; S3. If there is a multipath effect in the received signal, calculate the probability PT of the extracted direct path signal being correct, and use PT to determine whether there is a direct path that has not been extracted due to the noise threshold being set too high. S4. Use the TOF / TDOA positioning algorithm to calculate the positioning coordinates from the successfully extracted direct path. ; S5. Position coordinates The Kalman filter algorithm model is fed into the filter calculation to obtain... By comparison and Determine whether the location data is reliable; S6. If the positioning data is reliable, optimize the positioning value through Kalman filtering and end the positioning process.

2. The method for improving the accuracy of UWB positioning data according to claim 1, characterized in that: In step S3, PT is calculated using the following formula: In the formula, NPA represents the number of peaks in the received signal, and LAW represents the duration of the received signal; If PT>A, there are unextracted direct path signals, and it is necessary to return to step S1 to reset the noise threshold parameters; like If so, the direct path signal extraction is successful.

3. The method for improving the accuracy of UWB positioning data according to claim 1, characterized in that: In step S5, during Kalman filtering, a chi-square test is introduced to locate filtering errors. The chi-square test is introduced when calculating the Kalman gain, as follows: During the Kalman update process, let residual Then the chi-square test is expressed as: In the above formula, Represents the measurement noise covariance matrix. Represents the state observation matrix, Represents the state covariance matrix. This represents the measurement vector.

4. The method for improving the accuracy of UWB positioning data according to claim 3, characterized in that: when When the positioning value is incorrect, Kalman filtering is not performed; when If the positioning value is correct, Kalman filtering is performed.

5. The method for improving the accuracy of UWB positioning data according to claim 4, characterized in that: To avoid large measurement and estimation errors affecting the filtering results, the Kalman gain is improved by adding an adjustment factor: In the formula, and All values ​​are experimental values; The filtered result is as follows: In the formula, This represents the current state parameter.

6. A method for improving the accuracy of UWB positioning data according to claim 1, characterized in that: In step S5, if and If the difference between them is greater than M, then the location data is unreliable, and return to step S1; if and If the difference between them is less than M, then the location data is reliable; in, In the formula, V Indicates average speed. T Representing coordinates and The time difference in acquisition KM This represents the threshold coefficient.

Images