Base station collinear positioning precision control system and control method

Abstract

The invention discloses a base station collinear positioning precision control method. The method comprises the following steps: S1, placing a base station; s2, equipment connection; s3, delimiting a positioning range; s4, base station label communication is carried out; s5, angle calculation; s6, distance calculation; s7, obtaining a preliminary positioning result; s8, calculating the GDOP; s9, calculating a theoretical error; and S10, fusion positioning is carried out. According to the invention, single AOA angle measurement positioning and TOF distance measurement positioning are fused, high-precision full-coverage positioning of a positioning area on one side is realized for the collinear condition of the base stations, and the problem that high-precision full coverage cannot be realized when the base stations are collinear in a single TOF positioning method is solved. According to the invention, high-precision full-coverage positioning can be realized under the condition of special base station configuration caused by scene constraints in actual positioning application, and the positioning precision in different directions can be controlled.

Description

Technical Field

[0001] This invention belongs to the field of positioning technology, specifically ultra-wideband positioning technology, and more specifically relates to a positioning accuracy control system and control method for base station collinearity. Background Technology

[0002] Current positioning technologies generally use non-collinear base stations for two-dimensional positioning, and research on positioning scenarios where positioning accuracy requirements vary in different directions, such as narrow positioning areas, is insufficient. However, the deployment of base stations in non-collinear base station positioning systems has certain requirements for the positioning area, and in some scenarios, controlling positioning accuracy in different directions is more effective and convenient than improving positioning accuracy. Taking highway road maintenance as an example, deploying base stations on the side of the highway guardrail helps ensure the safety of construction workers. Simultaneously, installing base stations on the side of the guardrail prevents them from being touched by vehicles and personnel, ensuring positioning stability. However, deploying base stations on the guardrail results in a near-linear deployment area, making conventional non-collinear base station deployment systems unsuitable. Furthermore, this scenario is a classic narrow and elongated area, with the longer side being much longer than the shorter side, and the positioning constraints for construction workers have a higher priority in the shorter side direction. Improving positioning accuracy in the shorter side direction provides greater safety assurance. Therefore, to address these issues, it is necessary to study the case of collinear base stations and propose a method to control positioning accuracy in different directions.

[0003] Similar research, such as Chinese Patent No. CN110611878A, discloses a high-precision positioning system and method based on one-dimensional deployment of base stations. It pre-sets multiple positioning base stations arranged in one-dimensional form and calculates the precise position of the positioning terminal by the time the signal arrives at each base station and the known position of the base station. However, this patent only considers the TOF positioning method and does not mention the control method for positioning accuracy in different directions. Summary of the Invention

[0004] To address the shortcomings of existing technologies, such as reduced positioning accuracy due to base station configuration limitations, and the need for positioning accuracy control methods in different directions, this invention discloses a positioning accuracy control system and method for base stations with collinearity.

[0005] The positioning accuracy control method for base station collinearity described in this invention includes the following steps:

[0006] Step S1: In the actual positioning scenario, place two or more base stations on the same straight line;

[0007] Step S2: Power on the base station, tag, and host computer to enable communication signals between base stations and between base stations and tags. Select the master base station and communication station from the base stations and enable the communication station to establish a communication connection with the host computer.

[0008] Step S3: Define the positioning range, that is, define the positioning area based on the scene and the coverage of the base station signal;

[0009] Step S4: After the tag enters the positioning range, the positioning process of base station-tag communication is carried out. The positioning process includes at least one communication between the base station and the tag; each base station and tag independently records the timestamp of each of its own signal transmission and reception.

[0010] Step S5: Calculate the label angle;

[0011] Step S6: Calculate the label distance;

[0012] Step S7: All base stations send the calculated tag distance and angle data to the main base station; the main base station processes the data and obtains the preliminary positioning location;

[0013] Step S8: Calculate GDOP, that is, calculate the GDOP value of each positioning algorithm based on the preliminary positioning results in step S7 and the base station location;

[0014] Step S9: Calculate the minimum deviation of the tag's position, that is, calculate the minimum deviation of each positioning algorithm based on the relationship between the GDOP value and the root mean square error of each positioning algorithm, and select the positioning algorithm with the smallest minimum deviation value as the solution method for the position coordinates.

[0015] Step S10: Fusion positioning;

[0016] Using the preliminary positioning result obtained in step S7 as the input data in step S10, and based on the solution method selected in step S9, the position coordinates are optimized by the weighted LM algorithm, and finally the position coordinates are output.

[0017] Preferably, the positioning algorithms in step S8 include the TOF positioning algorithm, the AOA dual-station positioning algorithm, and the AOA single-station positioning algorithm.

[0018] Preferably, in step S8, the method for calculating the distribution of GDOP is as follows:

[0019] G(1,1) and G(2,2) represent the element values ​​of matrix G at the corresponding row and column, respectively;

[0020] G=(H T H) -1 H is the coefficient matrix of the variables.

[0021] Preferably, for the TOF positioning algorithm, the planar positioning deviation...

[0022] R H =HDOP·σ d

[0023] σ d This represents the standard deviation of the ranging error distribution in the TOF positioning algorithm;

[0024] For single-station AOA positioning methods, planar positioning deviation

[0025]

[0026] HDOP d and HDOP φ σ represents the HDOP values ​​for distance and angle, respectively. d , σ φ Let represent the standard deviations of distance and angle, respectively, and d represent the minimum distance between the target and the base station.

[0027] Preferably, step S7 specifically includes:

[0028] Step S71: The main base station data is classified according to distance and angle, and the classified data is packaged and sent to the communication station;

[0029] Step S72: The communication station sends the pre-processed data to the host computer.

[0030] Step S73: The host computer calculates the positioning result of the single-station AOA positioning algorithm based on the received angle and distance data.

[0031] Preferably, step S10 includes the following steps:

[0032] Step S1001: Calculate the corresponding pre-parameter weighting matrix W according to the positioning algorithm selected in step S9;

[0033] Step S1002: Use the preliminary positioning position obtained in step S7 as the initial value θ0 for the optimization solution;

[0034] Step S1003:

[0035] Calculate the current position θ k The corresponding Jacobian matrix form of the variable coefficient matrix H k ;

[0036]

[0037] x i ,y i Let X and Y represent the X and Y coordinates of the i-th base station, respectively, and n be the total number of base stations.

[0038] The formulas for calculating the residuals of each positioning algorithm are as follows:

[0039] TOF algorithm residual

[0040] Single-site AOA algorithm residual

[0041] X BEn θ k (1), θ k (2) d n φ n Let x be the distance and angle from the tag to the nth base station, respectively. n y n Let X represent the coordinates of the nth base station. BEn θ represents the location of the nth base station. k (1), θ k (2) represents a two-dimensional vector θ k The first and second elements, i.e., θ k =(θ k (1), θ k (2)); Double vertical lines represent distance operators;

[0042] Step S1004: Set the initial value of the iteration stopping condition and the damping factor λ. Setting the iteration stopping condition includes setting the maximum number of iterations k. max The residual threshold ε and the initial value λ0 of the damping factor λ are set as shown in the following formula:

[0043] λ0=ε·max i [(H T WH) ii ]

[0044] The subscript ii indicates that the elements in the matrix within the parentheses have the same number of rows and columns, i.e., all the elements on the diagonal;

[0045] Step S1005: Threshold decision. If either of the following two conditions is met, terminate the loop and output the current θ. k As the updated position;

[0046] Condition 1.‖H k Wr(θ k )‖≤ε

[0047] Condition 2. k≥k max

[0048] If none of the conditions are met, proceed to step S1006;

[0049] Step S1006:

[0050] Calculate the step size Δθ from the residual obtained in step 1003;

[0051] (H T WH+λI)Δθ=H T Wr

[0052] Step S1007: Evaluate the step size and adjust the damping factor λ. If the drop gain ratio ρ > 0.75, then let: λ = λ / 2; if ρ < 0.25, then let: λ = 2λ.

[0053] Actual decrease

[0054] Predicted decrease

[0055] Drop-off gain ratio

[0056] Δθ T Let represent the transpose of Δθ, and n represent the total number of base stations;

[0057] Step S1008: After adjusting the damping factor λ, update the current θ. k+1 =θ k +Δθ, and repeat steps S1403 to S1008 until the threshold decision in step S1005 meets the condition.

[0058] Preferably, in step S1001, the selected positioning algorithms are the TOF positioning algorithm and the single-station AOA positioning algorithm, wherein the weighting matrix of the pre-parameters of the TOF positioning algorithm is W. TOF , σ di 2 This represents the distance variance detected by the i-th base station, where n is the total number of base stations;

[0059] The weighted matrix of the pre-parameters for the single-site AOA localization algorithm is W. AOA , σ φi 2 This represents the angle variance detected by the i-th base station;

[0060]

[0061] This is a combination of the partial derivatives of the UWB ranging observation equation with respect to the x and y directions; This is a combination of the partial derivatives of the UWB angle measurement observation equations with respect to the x and y directions;

[0062]

[0063] Preferably, let a = 0, b = 1, then:

[0064]

[0065] The present invention also discloses a positioning accuracy control system for base stations collinear, used to execute the positioning accuracy control method, including a base station that can be communicated, a tag to be located, and a host computer, wherein there are two or more base stations placed on the same straight line.

[0066] Preferably, the base station includes a UWB module and an MCU. The UWB module includes a transceiver antenna and a receiving antenna. The transceiver antenna is connected to an amplifier via a radio frequency switch, and the receiving antenna is directly connected to the amplifier. The amplifier, mixer, filter, IQ demodulator, baseband processor, and MCU are connected in sequence. The tag has the same structure as the base station, except that it does not have a receiving antenna.

[0067] The advantages of the positioning accuracy control system and method for base station collinearity described in this invention are as follows:

[0068] 1. This invention uses GDOP for accuracy comparison and switches different positioning methods according to the comparison results. It can combine the advantages of different positioning methods to achieve high-precision full-coverage positioning within the positioning area under the co-line base station configuration.

[0069] 2. By weighting the LM algorithm, this invention adjusts the information weights of different positioning base stations in a narrow positioning area, thereby controlling the positioning accuracy in different directions at the solution level and improving the adaptability of the positioning method to application scenarios.

[0070] 3. The base station deployment configuration of the method described in this invention is a collinear configuration, which provides a new positioning solution for narrow positioning environments where conventional base station configurations are difficult to deploy. Attached Figure Description

[0071] Figure 1 This is a schematic diagram of the distribution of base stations, tags, and host computers in a specific embodiment of the present invention;

[0072] Figure 2 This is a block diagram of the module composition of a base station in a specific embodiment of the present invention;

[0073] Figure 3 This is a block diagram of the module composition of the label in a specific embodiment of the present invention;

[0074] Figure 4 This is a schematic diagram of calculating the signal incident angle based on the phase difference information when the base station receives tag signals using dual antennas, according to a specific embodiment of the present invention;

[0075] Figure 5 This is a schematic diagram of a UWB two-way communication ranging method based on time of flight in a specific embodiment of the present invention;

[0076] Figure 6This is a schematic diagram of the distribution of tags and base stations during GDOP calculation in a specific embodiment of the present invention;

[0077] Figure 7 This is a schematic diagram illustrating the workflow of the tag, UWB base station, and host computer in a specific embodiment of the present invention;

[0078] Figure 8 This is a schematic diagram of the interaction process between the base station and the tag information in a specific embodiment of the present invention;

[0079] Figure 9 This is a distribution diagram of the TOF and single-station AOA positioning accuracy factors in a specific embodiment of the present invention;

[0080] Figure 10 This is a graph showing the relationship between the RMSE (Ranging Measured Range and Angle Measurement) and distance of a UWB base station in an open scene, according to a specific embodiment of the present invention.

[0081] Figure 11 This is a positioning range division diagram in a specific embodiment of the present invention;

[0082] Figure 12 This is a contour map of fusion positioning error in a specific embodiment of the present invention;

[0083] Figure 13 This is a fusion positioning error distribution diagram in a specific embodiment of the present invention;

[0084] Figure 14 This is a comparison chart of the weighted and unweighted LM algorithm solution results in a specific embodiment of the present invention;

[0085] Figure 15 This is a schematic diagram of a specific implementation of the position coordinate optimization using the weighted LM algorithm in this invention. Detailed Implementation

[0086] The specific embodiments of the present invention will be described in further detail below.

[0087] The positioning accuracy control system for base station collinearity described in this invention mainly includes: a base station employing UWB (Ultra-Wideband) positioning technology, tags, and a host computer. The base station needs to simultaneously possess ranging and angle measurement functions, as well as the ability to communicate with the host computer and other base stations; the tags can periodically transmit and respond to signals to the base station; after receiving the relevant ranging and angle measurement information sent by the base station, the host computer processes and calculates it.

[0088] The positioning accuracy control method for base station collinearity described in this invention includes the following steps:

[0089] Step S1: Place base stations, that is, place base stations in a fixed position in the actual positioning scenario. The number of base stations is an integer greater than or equal to 2. The distance between adjacent base stations can be equal or unequal. All base stations are located on the same straight line, and the number of tags is one or more.

[0090] Step S2: Device connection,

[0091] Power on the base station, tags, and host computer to establish communication signals between base stations and between base stations and tags. Select one base station as the master base station and one as the communication station. The master base station and the communication station can be the same base station or different base stations.

[0092] The function of the main base station is:

[0093] 1. Synchronize the clock information of all base stations to ensure that the clocks of all UWB base stations are synchronized;

[0094] 2. Summarize and perform preliminary processing on data from other UWB base stations;

[0095] 3. Send the pre-processed data to the communication station.

[0096] The role of the communication station is to communicate with the host computer and the main base station, and to transmit the data that has been initially processed by the main base station to the host computer.

[0097] Connect the communication station to the host computer signal, and observe on the host computer whether all connections are successful, that is, all base stations, tags and host computers have achieved communication signal connection;

[0098] If the connection is successful, the communication modules of all base stations, tags, and host computers will be activated and the process will proceed to step S3. If the connection fails, the process will check whether each device is working properly. If it is working properly, the process will repeat step S2.

[0099] Wireless connections are typically established between base stations and between tags and base stations via UWB signals. The connection between a communication station and a host computer can be established by the host computer sending radio communication signals to the communication station.

[0100] Once all connections are complete, the host computer interface can display the UWB base station and tag transmission information, proving that the connection is successful. If some base station or tag information is missing from the display, it means that the connection has failed, and step S2 should be repeated.

[0101] Step S3: Define the positioning range, that is, define the positioning area based on the scene and the coverage of the base station signal to form scene constraints on the positioning system.

[0102] The positioning area refers to the area that can be positioned based on needs and equipment conditions. The tag can be accurately positioned within the positioning area, but the tag cannot be accurately positioned outside the positioning area.

[0103] A typical positioning area is defined by using the line segment determined by the two farthest base stations as the reference side, and then constructing short sides perpendicular to the reference side at the two endpoints of the reference side. The length of the short sides can be set according to the actual application scenario.

[0104] Step S4: After the tag enters the positioning range, the positioning process of base station tag communication is carried out.

[0105] Since subsequent ranging uses the time-of-flight (TOF) calculation method, to accommodate this method, the single positioning process is divided into four segments: Segment 1: The tag transmits signals to each base station; Segment 2: Each base station transmits signals to the tag; Segment 3: The tag transmits signals back to each base station; Segment 4: Each base station sends signals back to the tag. Each base station and tag independently records the timestamps of each signal transmission and reception. For the AOA single-site algorithm, only one mutual communication is required, and the above positioning process is also applicable.

[0106] Step S5: Calculate the label angle.

[0107] After step S4 is completed, each base station performs angle calculation to obtain the angle information of the tag relative to each base station. The angle information calculation is based on the phase difference of the tag signal received by the base station array antenna.

[0108] The angle information calculation in step S5 can be based on the principle of Angle of Arrival (AOA), that is, the transmitting signal is emitted from one antenna, and the two receiving antennas are spaced half a wavelength apart. The phase difference measured by the receiving antenna can represent the time difference of the same transmitted signal arriving at the two antennas. The angle information of the transmitted signal can be calculated by using the inverse cosine function. The above process is the prior art in this field.

[0109] Step S6: Calculate the label distance.

[0110] After step S4 is completed, the base station performs distance calculation to obtain the distance between the tag and the base station. The distance information calculation used in this invention is based on the UWB bilateral bidirectional ranging method.

[0111] Preferably, the distance information calculation in step S6 can be based on the Time of Flight (TOF) principle. The timestamps obtained in step S4 are used to calculate four time lengths through the difference. The flight time of the signal from the base station to the tag is then calculated using the four time lengths to obtain the distance between the base station and the tag. The above process is the prior art in this field.

[0112] Step S7: All base stations send the calculated tag distance and angle data to the main base station, which performs preliminary processing on the data.

[0113] The so-called preliminary processing is that the main base station divides the uploaded data from different base stations into two categories according to angle and distance, so as to facilitate the subsequent calculation of the tag position using the TOF algorithm and the single-site AOA algorithm respectively; that is, based on the full coverage positioning characteristic of the single-site AOA positioning algorithm, the host computer calculates the preliminary positioning result using the single-site AOA algorithm based on the received angle data and distance data.

[0114] Step S7 may specifically include:

[0115] Step S71: The main base station data is classified according to distance and angle, and the classified data is packaged and sent to the communication station;

[0116] Step S72: The communication station sends the pre-processed data to the host computer.

[0117] Step S73: The host computer calculates the positioning result of the single-station AOA positioning algorithm based on the received angle and distance data.

[0118] The preliminary positioning result includes the preliminary positioning position θ0, and also includes the distance variance σ calculated based on the angle data φi and distance data di received by each base station. di 2 and angle variance σ φi 2 And the variable coefficient matrix H, where i represents different base stations.

[0119] Step S8: Calculate GDOP, that is, calculate the GDOP value of each positioning algorithm based on the preliminary positioning results in step S7 and the location of the UWB base station;

[0120] Calculating GDOP (Geometric Dilution Precision) involves the host computer calculating the distribution of GDOP for each positioning algorithm based on the collinear locations of each base station. The positioning algorithms mainly include TOF (Time of Flight) positioning algorithm, AOA (Angle-of-Arrival) bistational direction finding positioning algorithm, and AOA monostation positioning algorithm.

[0121] The specific calculation method for GDOP is as follows:

[0122] Suppose the coordinates of a certain base station in space are known to be (x... i ,y i ,z iLet i = 1, 2, 3, ..., and let the coordinates of a certain undetermined label l be l(x, y, z). According to the distance formula, the true geometric distance between them is r. i Here, assuming distance measurement is performed under synchronized conditions, the distance function can be obtained:

[0123]

[0124] like It is the approximate position calculated by the positioning algorithm. It has the relationship with the actual position as shown in equation (2), where Δx, Δy, and Δz are the errors in each direction.

[0125]

[0126] Substituting equation (2) into equation (1) yields equation (3):

[0127]

[0128] Put it in Taylor expansion at point yields

[0129]

[0130] The partial derivatives of the function at point p in all directions are:

[0131]

[0132] f = f i (x,y,z), Let f be the function value of f at point p.

[0133] Let the partial derivative be a xi ,a yi ,a zi Its geometric meaning represents the direction cosine of the approximate coordinates relative to the known point i, then equation (4) can be rearranged as:

[0134]

[0135] Equation (6) can be written in matrix form as shown in equation (7).

[0136] Δf=HΔX(7),

[0137] H is the coefficient matrix of the variables, ΔX = (Δx, Δy, Δz) T

[0138] make

[0139] G=(H T H) -1 (8)

[0140] Then geometric precision factor

[0141] GDOP = σ s 2 (G(1,1)+G(2,2)+G(3,3))(9) G(1,1), G(2,2), and G(3,3) represent the element values ​​of the G matrix defined by formula (8) at the corresponding row and column.

[0142] σ s The total ranging error has a component position accuracy factor:

[0143]

[0144] Horizontal precision factor

[0145]

[0146] This invention is a two-dimensional positioning system, therefore the accuracy factor only considers the horizontal accuracy factor HDOP, that is, in this invention, GDOP = HDOP.

[0147] Step S9: Calculate the minimum deviation of the tag's position, that is, calculate the minimum deviation of each positioning algorithm based on the relationship between the GDOP value and the root mean square error of each positioning algorithm, and select the positioning algorithm with the smallest minimum deviation value as the solution method for the position coordinates.

[0148] In step S9, when using the TOF ranging and positioning algorithm, the actual positioning error calculation method for the horizontal accuracy factor HDOP is as follows:

[0149] In UWB positioning systems, Time-of-Flight (TOF) ranging and positioning calculates distance by measuring the signal propagation time t. For the tag location X obtained through the positioning system... UE =[x,y] T and base station location X BE =[x i ,y i ] T The positioning equations established for (i = 1, 2, ..., n) are as follows:

[0150] d i =‖X UE -X BE ‖+δ i (12)

[0151] Where δ i This is the ranging error, which is usually assumed to have a Gaussian distribution:

[0152] δ i ~N(0,σ d 2 (13)

[0153] σ d 2 Represents the variance of the Gaussian distribution

[0154] Assume the actual position of the label is X Tr =(x0,y0) T In the real coordinates X Tr Perform a first-order Taylor expansion on equation (12):

[0155]

[0156] Organized into

[0157]

[0158] Where Δx = x - x0 and Δy = y - y0 are position errors, and the coefficients of Δx and Δy are matrix elements in the variable coefficient matrix H.

[0159] The positioning equations (14) after expanding all base stations are stacked into a matrix form:

[0160] δ d =H·δ X +δ all (15)

[0161] δ d δ represents the vector obtained by stacking the left-hand side of equation (14). all Represents all δ i The stacked matrix, δ X Let (Δx, Δy) be the vector.

[0162] The least squares estimate of the position error is:

[0163] δ X ≈(H T H) -1 H T δ d (16)

[0164] The covariance matrix of the position error is:

[0165] C = Ε[δ] X ·δ X T ]=(H T H) -1 H T ·E[δ d ·δ d T ]·H(H T H) -1 (17)

[0166] in:

[0167] E[δ d ·δ d T ]=σ d 2 I (18), where I represents the identity matrix

[0168] therefore:

[0169] C = σ d 2 (H T H) -1 (19)

[0170] According to the definition of GDOP and equation (8), the formula for calculating the component level precision factor HDOP of GDOP is:

[0171]

[0172] The planar positioning deviation is:

[0173]

[0174] Equation (21) illustrates the actual positioning error of TOF ranging and positioning, i.e., the planar positioning deviation R. H It equals the calculated HDOP value and the standard deviation σ of the ranging error distribution over that distance. d The product of.

[0175] In another positioning method, namely the single-site AOA positioning method, similarly, for a known base station coordinate in space (x... i ,y i ,z i ), i = 1, 2, 3…, where the coordinates of a certain undetermined label l are l(x, y, z). Since both distance measurement and angle measurement are involved, the observation equations can be written as equations (12) and (22), where This represents the angular error.

[0176]

[0177] Similarly, the angle observation equation is applied at the true position X. Tr =(x0,y0) T Linearization:

[0178]

[0179] dio represents a known base station (x) i ,y i ,z i The distance between ) and the undetermined label l(x,y,z), where Δx and Δy represent the minimum quantities when unfolded.

[0180] The observation equations for stacking all n base stations are given by equation (23), which can be rewritten in matrix form as shown in equation (24):

[0181]

[0182] in:

[0183]

[0184] The variable coefficient matrix H is:

[0185]

[0186] Indicates all A stacked matrix.

[0187] The error covariance matrix is:

[0188]

[0189] σ d , σ φ These represent the standard deviations of distance and angle, respectively.

[0190] According to the definition of HDOP, the planar positioning error of a single-station AOA algorithm is:

[0191]

[0192] HDOP d and HDOP φ The HDOP values ​​represent distance and angle, respectively, with d representing distance.

[0193] For the bi-station AOA algorithm, the analysis shows that the planar positioning deviation is much greater than that of the TOF algorithm and the mono-station AOA algorithm. Therefore, this invention does not use the bi-station AOA algorithm.

[0194] Standard deviations σ of distance and angle d With σ φ All values ​​are actual measured values ​​of ranging and angle measurement errors from the UWB base stations used. Different base station equipment or different scenarios may result in different standard deviations, leading to variations in the error magnitude of different positioning algorithms. In this specific embodiment, calculations show that the Time-of-Flight (TOF) algorithm and the single-station AOA algorithm have relatively small deviations.

[0195] Step S10: Fusion positioning, that is, the host computer uses the weighted LM algorithm to perform fusion positioning based on the preliminary positioning results in step S7, so as to control the positioning accuracy during the solution process.

[0196] In step S10, the solution algorithm for the fusion positioning of TOF positioning and single-station AOA is a weighted LM algorithm, where the weighting matrix of the pre-parameters for TOF positioning is shown in equation (29), σ di 2 Let represent the distance variance detected by the i-th base station, and n be the total number of base stations.

[0197] The weighted matrix of the pre-parameters for single-station AOA positioning is shown in equation (30), σ φi 2 This represents the angular variance detected by the i-th base station.

[0198]

[0199] The core iterative formula of the weighted LM algorithm is shown in equation (31).

[0200] (H T WH+λI)Δθ=H T Wr(31)

[0201] In the formula, I is the identity matrix, λ represents the damping factor, and W is the W... AOA or W TOF Δθ is the iteration step size.

[0202] Specifically, step S10 includes the following steps:

[0203] Step S1001: If the positioning algorithm selected in step S9 is the TOF positioning algorithm or the single-station AOA positioning algorithm, then calculate the weighted matrix W of the pre-parameters, that is, calculate the weighted matrix corresponding to the positioning algorithm selected in step S9 according to equation (29) and equation (30); if the selected algorithm is the TOF positioning algorithm, then use equation (29) to calculate it; if the selected algorithm is the single-station AOA positioning algorithm, then use equation (30) to calculate it.

[0204] Step S1002: Use the initial positioning position obtained in step S7 as the initial value θ0 for optimization calculation. Obviously, the initial value θ0 is a two-dimensional vector that includes two-dimensional coordinates.

[0205] Step S1003:

[0206] Calculate the current position θ using formula (26) k The corresponding Jacobian matrix form of the variable coefficient matrix H k ;

[0207] The formulas for calculating the residuals of each positioning algorithm are shown in equations (37) to (38):

[0208] TOF algorithm residual

[0209] Single-site AOA algorithm residual

[0210] X BEn θ k (1), θ k (2) d n φ n Let x be the distance and angle from the tag to the nth base station, respectively. n y n Let X represent the coordinates of the nth base station. BEn θ represents the location of the nth base station. k (1), θ k (2) represents a two-dimensional vector θ k The first and second elements, i.e., θ k =(θ k (1), θ k (2)); Double vertical lines represent distance operators.

[0211] Step S1004: Set the initial value of the iteration stopping condition and the damping factor λ. Setting the iteration stopping condition includes setting the maximum number of iterations k. max The residual threshold ε and the initial value λ0 of the damping factor λ are set as shown in equation (36):

[0212] λ0=ε·max i [(H T WH) ii (36)

[0213] The subscript ii indicates that the elements in the matrix with the same number of rows and columns are all the elements on the diagonal. Equation (36) means taking the maximum value of the elements on the diagonal of the matrix.

[0214] Step S1005: Threshold decision. If either of the two conditions in equation (39) or equation (40) is met, the loop terminates and the current θ is output. k As the updated position;

[0215] ‖H k Wr(θ k )‖≤ε (39)

[0216] k≥k max (40)

[0217] If not satisfied, proceed to step S1006;

[0218] Step S1006:

[0219] The step size Δθ is calculated from the residual obtained in step 1003 according to formula (31);

[0220] (H TWH+λI)Δθ=H T Wr(31)

[0221] The resulting step size Δθ is a two-dimensional vector;

[0222] Step S1007: Evaluate the step size and adjust the damping factor λ, with the corresponding formulas as shown in equations (41) to (43). If the drop gain ratio ρ > 0.75, then let: λ = λ / 2; if ρ < 0.25, then let: λ = 2λ.

[0223] Actual decrease

[0224] Predicted decrease

[0225] Drop-off gain ratio

[0226] Δθ T Let represent the transpose of Δθ, and n represent the total number of base stations;

[0227] Step S1008: After adjusting the damping factor λ, update the current θ. k+1 =θ k +Δθ, and repeat steps S1403 to S1008 until the threshold decision in step S1005 meets the condition.

[0228] A specific process of step S10 is as follows: Figure 15 As shown.

[0229] Finally, the host computer processes the data and outputs the tag location information.

[0230] In equation (29) The equation is a combination of the partial derivatives of the UWB ranging observation equation with respect to the x and y directions. The values ​​of a and b can be proportionally varied to control the accuracy in different directions. Specifically, to improve the accuracy in the x direction, the value of a / b is increased; conversely, to improve the accuracy in the y direction, the value of a / b is decreased, as shown in equation (32). In equation (30)... Similarly, the combination of the partial derivatives of the UWB angle measurement observation equation with respect to the x and y directions is shown in equation (33).

[0231]

[0232] The idea behind determining the overall weighted matrix is ​​to use the reciprocal of the measurement variance as a basis and the magnitude of the partial derivatives of the observation equation with respect to the x and y directions as control factors to achieve control over the positioning accuracy in different directions. In the highway road maintenance scenario described in this invention, due to the overall elongated characteristic where the length of the long side is much greater than the length of the short side, the accuracy requirement in the short side direction is higher. Therefore, with the UWB base station connection line as the x-axis direction, we can set a = 0 and b = 1 to achieve the maximum improvement in y-axis accuracy, as shown in equations (34) and (35).

[0233]

[0234] This invention is applicable to areas with special requirements for base station deployment configuration, such as underground mines and highways. A specific embodiment is given below.

[0235] Each tag and each base station uses a UWB module with a multi-antenna interface, enabling simultaneous ranging and angle measurement. The base station has dual external antennas with a 0.02m spacing, and the UWB signal modulation carrier frequency is 7.163GHz. Before ranging begins, the base station and tag must be pre-paired, the master base station and communication station must be pre-selected, and the communication station and host computer, as well as the slave base station and master base station, must be pre-paired. The tag sends a UWB signal and wakes up the UWB base station, which receives the signal and verifies the tag's identity. Communication between the UWB base station and the tag continues. Angle and distance information are calculated from the data obtained from the UWB base station, and the GDOP is further calculated. Based on these calculations, the positioning result is given.

[0236] Communication between any UWB base station and a tag, and between UWB base stations themselves, follows the communication protocol specified in the IEEE 802.15.4z standard. Information transmitted between UWB base stations and tags, and between UWB base stations, generally uses data frames defined by IEEE 802.15.4z. These data frames typically consist of three parts: a frame header (MAC Header, MHR), a payload (MAC Payload), and a frame footer (MAC Footer, MFR). The frame header comprises frame control information, a frame sequence number, and an address. The payload is of variable length and, in this invention, includes a ranging field (poll and response signals for bidirectional ranging) and an angle field (phase reference signal). The frame footer is a 16-bit checksum sequence of the frame header and payload data. The receiving base station calculates the Time-of-Flight (TOF) by measuring the difference between the timestamps of the received ranging fields and calculates the phase difference between the two antennas by extracting the IQ data from the phase reference signal. The phase reference signal is typically an unmodulated carrier pulse or a known modulation sequence.

[0237] Figure 1 A specific structure of the positioning system is given. For example... Figure 1As shown, the positioning system includes one UWB base station group and one tag 20. The tag 20 can be a physical tag specifically matched with the base station, or a mobile terminal, such as a smartphone, that implements tag functionality through software loading. A host computer 30 is used to process the measurement results of the base station group. The UWB base station group includes four base stations with known locations: a main base station 11, a communication station 12, a first slave base station 10, and a second slave base station 13. Figure 1 A two-dimensional XY coordinate system is established using the straight line where the four collinear base stations are located as the X-axis.

[0238] The main base station 11 is used to receive UWB information from other base stations and perform preliminary processing. The communication station 12 receives the preliminary processing information from the main base station 11 and sends the preliminary processing results to the host computer 30 via Bluetooth Low Energy (BLE). The host computer 30 further processes the results to obtain the positioning results.

[0239] Figure 2 The block diagram of the base station module in this embodiment is shown, including a UWB module and an MCU. The UWB module includes a transceiver antenna 1 and a receiving antenna 2. The transceiver antenna 1 is connected to an amplifier through an RF switch, and the receiving antenna 2 is directly connected to the amplifier. The amplifier, mixer, filter, IQ demodulator, baseband processor and MCU are connected in sequence.

[0240] Transceiver antenna 1 is responsible for transmitting and receiving UWB pulse signals for TOF ranging and basic communication. Receiving antenna 2 only receives signals and forms a baseline with transceiver antenna 1 to calculate the phase difference and obtain the AOA (Angle of Arrival). The RF switch is used to switch the transmission and reception modes of transceiver antenna 1. The amplifier is a low-noise amplifier at the receiving end to amplify the UWB signal from the antenna, and a power amplifier at the transmitting end to enhance the power of the transmitted signal. The mixer is used to up-convert the RF and baseband signals. The filter is used for bandpass filtering to suppress out-of-band interference. The RF switch, amplifier, mixer, and filter constitute the RF front end of the UWB module. The IQ demodulator works at the receiving end to decompose the received UWB signal into in-phase (I) and quadrature (Q) components, retaining carrier phase information for calculating the phase difference.

[0241] The baseband processor mainly includes a digital correlator for detecting the precise arrival time of UWB pulses; a timing control unit for managing frame synchronization and timestamp recording; a MAC layer controller for processing the frame format of the IEEE 802.15.4z standard; a CIR module for performing multipath analysis, assisting in ranging and angle measurement; and a phase difference calculation unit for calculating phase differences. The baseband processor transmits the calculated and analyzed data to the MCU via a serial peripheral interface (SPI).

[0242] The clock management module provides the necessary clock signals; the power supply module provides the power required for operation. It should be understood that... Figure 2 The diagram shown is a schematic. In an actual base station, there are two receiving links at the receiving end and one transmitting link at the transmitting end. The IQ demodulator is not working during transmission.

[0243] Figure 3 A specific embodiment of the label described in this invention is shown, such as... Figure 3 As shown, with Figure 2 The components shown are different: Figure 3 The middle tag has only one transceiver antenna, which enables signal transmission and reception.

[0244] Figure 4 This paper demonstrates a method for measuring the incident angle of UWB signals based on the Angle of Arrival (AOA) principle. Figure 4 As shown, the antenna array of a UWB base station has two antenna elements spaced d apart, with a wavelength of... If a UWB signal is incident on the antenna array from the θ direction, then the phase difference between the outputs of the two antenna elements is... for

[0245]

[0246] The incident angle θ of the UWB signal is

[0247]

[0248] Figure 5 This paper illustrates a UWB bilateral bidirectional ranging principle based on Time of Flight (TOF). It calculates distance by measuring the round-trip time (TOF) of the UWB signal between the UWB base station and the tag. The tag initiates ranging, the base station receives and replies with the ranging result, the tag receives the ranging reply and retransmits the ranging result to the UWB base station, and the base station receives the ranging result again. Both the base station and the tag record the timestamps of transmission and reception using the clock generator of their respective UWB modules to obtain the time interval between transmission and reception.

[0249] For example, let t1 be the time interval between the tag's first transmission and first reception of the signal, t2 be the time interval between the tag's first reception and second transmission of the signal, t3 be the time interval between the UWB base station's first reception and first transmission of the signal, t4 be the time interval between the UWB base station's first transmission and second reception of the signal, and t5 be the flight time of the UWB signal. TOF It can be calculated as follows:

[0250] t TOF =(t1*t4-t2*t3) / (t1+t2+t3+t4).

[0251] Then, the distance d between the UWB base station and the tag can be calculated by multiplying the flight time and the electromagnetic wave propagation speed: d = c * t. TOF c represents the speed of light.

[0252] Figure 6 This illustrates a specific implementation of the GDOP calculation method used in this invention, such as... Figure 6 As shown, there are four base stations BS1 to BS4 and one tag MS. A two-dimensional XY coordinate system is established with the straight line where each base station is located as the X-axis.

[0253] The location coordinates of the base station and the tag are respectively

[0254]

[0255] Then the distance d between base station BSi (i = 1, 2, 3, 4) and tag MS i for

[0256]

[0257] Construct the Jacobian matrix H, where the element values ​​H(i,j) of H are calculated as follows:

[0258]

[0259] The matrix G is obtained as follows:

[0260]

[0261] In one embodiment, considering the TOF timing error t1, the matrix G is obtained as follows:

[0262] G=(H T (t1E) -1 H) -1

[0263] Where E is the identity matrix.

[0264] The GDOP value at the MS position is calculated as follows:

[0265]

[0266] Figure 7 The workflow between the base station, tag, and host computer is shown, including steps S3 to S15.

[0267] Figure 8 The diagram illustrates a specific implementation of information interaction between a tag and a UWB base station. The tag initiates communication with the base station, collecting and measuring angle and distance information via UWB signals. The base station sends an end signal to the tag to indicate the end of communication.

[0268] Figure 9 Shows the distribution of dilution factors of the TOF positioning method and the single-station AOA positioning method in an embodiment of the present invention. The values of three contour lines are shown in the TOF dilution factor distribution diagram. Figure 9 In it, LEVEL represents the HDOP value; it can be seen that in a part of the area above the middle two base stations perpendicular to the connection line of the positioning base stations, the positioning accuracy of using TOF positioning is relatively high. As the point to be positioned deviates from the vertical line, the value of its dilution factor shows an obvious upward trend, and a positioning blind area appears when approaching the two side edges of the positioning area.

[0269] It can be seen from the dilution factor distribution diagram of single-station AOA positioning that the values of the dilution factors of single-station AOA positioning and TOF positioning are both relatively low in a part of the area above the middle two base stations perpendicular to the base station line. However, the single-station AOA positioning only has a situation of large dilution factor values in the middle position between the two base stations, and there is no positioning blind area problem of TOF positioning. Obviously, the stability of the positioning accuracy is better than that of TOF positioning.

[0270] Figure 10 Shows the relationship between the root mean square error (RMSE) of range measurement and angle measurement of a base station in an open scenario in an embodiment of the present invention. In the research on range measurement performance, since it is an open scenario and the influence of non-line-of-sight (NLOS) is not considered, the RMSE distribution from 1 m to 30 m is calculated and simulated, with a sampling point set every 1 m, and 1000 Monte Carlo simulations are performed at each sampling point. It can be seen that without the influence of non-line-of-sight conditions, the change of range measurement RMSE at different distances is not large, and distance basically has no influence on the RMSE of range measurement.

[0271] In the research on angle measurement performance, the carrier frequency is set to 7.163 GHz and the antenna spacing is 0.02 m. In an open environment, the AOA angle measurement RMSE at different distances is tested. The simulation distance is set from 1 m to 20 m with a sampling point set every 1 m, and 1000 Monte Carlo simulations are performed at each sampling point. It can be seen that the RMSE of AOA angle measurement fluctuates continuously with distance and shows an upward trend as a whole.

[0272] Figure 11 Shows the regional division diagram of the positioning range in a specific embodiment of the present invention. In the figure, the white area is the area where the TOF error ≤ AOA error, and the black area is the area where the AOA error < TOF error. That is, if the preliminary positioning solution position is located in the white area, TOF solution optimization is performed starting from the preliminary position. If the preliminary positioning solution position is located in the black area, the position can be directly output.

[0273] Figure 12The diagram shows a contour map of fused positioning error in one embodiment of the present invention. Figure 9 The calculation of accuracy factor values ​​for TOF positioning and single-station AOA positioning. Figure 10 The relationship between the RMSE of the distance and angle measurement and the distance is obtained through simulation calculations, as shown below. Figure 12 The positioning error contour map shown below; Figure 13 This is a diagram showing the distribution of positioning errors using the fusion positioning method. Figure 13 The darker areas represent locations with positioning errors exceeding one meter. Considering the actual application scenario, these areas are located near highway guardrails, where the accuracy requirements are not high and the safety risks are minimal. Therefore, their impact on the safety of the positioning system is negligible and acceptable. Furthermore, the calculated average positioning error for all points within this area is 0.548m, which is within 0.5m of the line connecting the cones. Figure 12 The average positioning error within the y-axis range of 6.75m to 7.25m is 0.276m. The positioning accuracy is highest within the entire positioning range when construction workers are near the conical barrel, which matches the actual scenario requirements. It can be seen that the overall positioning accuracy is high and can achieve full coverage within the positioning range.

[0274] Figure 14 The diagram shows a comparison of the weighted and unweighted LM algorithm solution results in one embodiment of the present invention. The base station location coordinates corresponding to the two diagrams are (0, 0), (24, 0), and (48, 0), and the tag location coordinates are (20, 5). The left diagram shows the distribution of positioning accuracy in ten solutions using a confidence ellipse when unweighted, and the right diagram shows the distribution of positioning accuracy in ten solutions using a confidence ellipse after weighting. It is clear that by weighting, the positioning accuracy in the direction perpendicular to the base station connection is improved while slightly reducing the positioning accuracy in the direction parallel to the base station connection. This is more in line with the safety requirements of the narrow positioning area in actual highway maintenance construction.

[0275] The foregoing descriptions are preferred embodiments of the present invention. Unless there is a clear contradiction between the preferred embodiments or a prerequisite for a particular preferred embodiment, the preferred embodiments can be arbitrarily combined and used. The embodiments and specific parameters described are only for clearly illustrating the inventor's invention verification process and are not intended to limit the scope of patent protection of the present invention. The scope of patent protection of the present invention shall still be determined by its claims. Similarly, any equivalent structural changes made based on the description and drawings of the present invention shall also be included within the scope of protection of the present invention.

Claims

1. A method for controlling positioning accuracy of base stations collinearity, characterized in that... It includes the following steps: Step S1: In the actual positioning scenario, place two or more base stations on the same straight line; Step S2: Power on the base station, tag, and host computer to enable communication signals between base stations and between base stations and tags. Select the master base station and communication station from the base stations and enable the communication station to establish a communication connection with the host computer. Step S3: Define the positioning range, that is, define the positioning area based on the scene and the coverage of the base station signal; Step S4: After the tag enters the positioning range, the positioning process of base station-tag communication is carried out. The positioning process includes at least one communication between the base station and the tag; each base station and tag independently records the timestamp of each of its own signal transmission and reception. Step S5: Calculate the label angle; Step S6: Calculate the label distance; Step S7: All base stations send the calculated tag distance and angle data to the main base station; the main base station processes the data and obtains the preliminary positioning location; Step S8: Calculate GDOP, that is, calculate the GDOP value of each positioning algorithm based on the preliminary positioning results in step S7 and the base station location; Step S9: Calculate the minimum deviation of the tag's position, that is, calculate the minimum deviation of each positioning algorithm based on the relationship between the GDOP value and the root mean square error of each positioning algorithm, and select the positioning algorithm with the smallest minimum deviation value as the solution method for the position coordinates. Step S10: Fusion positioning; Using the preliminary positioning result obtained in step S7 as the input data in step S10, and based on the solution method selected in step S9, the position coordinates are optimized by the weighted LM algorithm, and finally the position coordinates are output.

2. The positioning accuracy control method for base station collinearity as described in claim 1, characterized in that... The positioning algorithms in step S8 include the TOF positioning algorithm, the AOA dual-station positioning algorithm, and the AOA single-station positioning algorithm.

3. The positioning accuracy control method for base station collinearity as described in claim 1, characterized in that... In step S8, the distribution calculation method for GDOP is as follows: G(1,1) and G(2,2) represent the element values ​​of matrix G at the corresponding row and column, respectively; G=(H T H) -1 H is the coefficient matrix of the variables.

4. The positioning accuracy control method for base station collinearity as described in claim 2, characterized in that... , For the TOF positioning algorithm, planar positioning deviation R H =HDOP·σ d σ d This represents the standard deviation of the ranging error distribution in the TOF positioning algorithm; For single-station AOA positioning methods, planar positioning deviation HDOP d and HDOP φ σ represents the HDOP values ​​for distance and angle, respectively. d , σ φ Let represent the standard deviations of distance and angle, respectively, and d represent the minimum distance between the target and the base station.

5. The positioning accuracy control method for base station collinearity as described in claim 1, characterized in that... Step S7 specifically includes: Step S71: The main base station data is classified according to distance and angle, and the classified data is packaged and sent to the communication station; Step S72: The communication station sends the pre-processed data to the host computer. Step S73: The host computer calculates the positioning result of the single-station AOA positioning algorithm based on the received angle and distance data.

6. The positioning accuracy control method for base station collinearity as described in claim 1, characterized in that... Step S10 includes the following steps: Step S1001: Calculate the corresponding pre-parameter weighting matrix W according to the positioning algorithm selected in step S9; Step S1002: Use the preliminary positioning position obtained in step S7 as the initial value θ0 for the optimization solution; Step S1003: Calculate the current position θ k The corresponding Jacobian matrix form of the variable coefficient matrix H k ; x i ,y i Let X and Y represent the X and Y coordinates of the i-th base station, respectively, and n be the total number of base stations. The formulas for calculating the residuals of each positioning algorithm are as follows: TOF algorithm residual Single-site AOA algorithm residual X BEn θ k (1), θ k (2) d n φ n Let x be the distance and angle from the tag to the nth base station, respectively. n y n Let X represent the coordinates of the nth base station. BEn θ represents the location of the nth base station. k (1), θ k (2) represents a two-dimensional vector θ k The first and second elements, i.e., θ k =(θ k (1), θ k (2)); Double vertical lines represent distance operators; Step S1004: Set the initial value of the iteration stopping condition and the damping factor λ. Setting the iteration stopping condition includes setting the maximum number of iterations k. max The residual threshold ε and the initial value λ0 of the damping factor λ are set as shown in the following formula: λ0=ε·max i [(H T (WH) ii ] The subscript ii indicates that the elements in the matrix within the parentheses have the same number of rows and columns, i.e., all the elements on the diagonal; Step S1005: Threshold decision. If either of the following two conditions is met, terminate the loop and output the current θ. k As the updated position; Condition 1. ‖H k Wr(θ k )‖≤ε Condition 2. k≥k max If none of the conditions are met, proceed to step S1006; Step S1006: Calculate the step size Δθ from the residual obtained in step 1003; (H T WH+λI)Δθ=H T Wr Step S1007: Evaluate the step size and adjust the damping factor λ. If the drop gain ratio ρ > 0.75, then let: λ = λ / 2; if ρ < 0.25, then let: λ = 2λ. Actual decrease Predicted decrease Drop-off gain ratio Δθ T Let represent the transpose of Δθ, and n represent the total number of base stations; Step S1008: After adjusting the damping factor λ, update the current θ. k+1 =θ k +Δθ, and repeat steps S1403 to S1008 until the threshold decision in step S1005 meets the condition.

7. The positioning accuracy control method for base station collinearity as described in claim 6, characterized in that... In step S1001, the selected positioning algorithms are the TOF positioning algorithm and the single-station AOA positioning algorithm, wherein the weighting matrix of the pre-parameters of the TOF positioning algorithm is W. TOF , σ di 2 This represents the distance variance detected by the i-th base station, where n is the total number of base stations; The weighted matrix of the pre-parameters for the single-site AOA localization algorithm is W. AOA , σ φi 2 This represents the angle variance detected by the i-th base station; This is a combination of the partial derivatives of the UWB ranging observation equation with respect to the x and y directions; This is a combination of the partial derivatives of the UWB angle measurement observation equations with respect to the x and y directions; 8. The positioning accuracy control method for base station collinearity as described in claim 7, characterized in that... Let a = 0, b = 1, then:

9. A positioning accuracy control system for base stations collinearity, characterized in that, The method for performing the positioning accuracy control method as described in any one of claims 1 to 8 includes a base station that can be communicated, a tag to be positioned, and a host computer, wherein there are two or more base stations placed on the same straight line.

10. The positioning accuracy control system for base station collinearity as described in claim 9, characterized in that, The base station includes a UWB module and an MCU. The UWB module includes a transceiver antenna and a receiving antenna. The transceiver antenna is connected to an amplifier via a radio frequency switch, and the receiving antenna is directly connected to the amplifier. The amplifier, mixer, filter, IQ demodulator, baseband processor, and MCU are connected in sequence. The tag has the same structure as the base station except that it does not have a receiving antenna.

Images