High-precision ultra-wideband positioning method and system based on clustering optimization algorithm

Abstract

The invention discloses a high-precision ultra-wideband positioning method and system based on a clustering optimization algorithm, the method is based on a base station system, and the base station system comprises a label base station, four reference base stations and a processor. The distance between the label base station and the reference base station is calculated through a bilateral bidirectional distance measurement method, and the three-dimensional coordinates of the label base station are preliminarily obtained through a time arrival positioning method. Further, a K-means clustering process is optimized by combining a particle swarm optimization algorithm, and spatial division of coordinate data is realized; for each classification space, parameters of a Kalman filtering function are adaptively adjusted, and preliminary optimization coordinates are obtained through filtering processing; and finally, an error correction model is introduced to compensate a coordinate residual error, and an error value is superposed to an optimization result, so that the positioning error is remarkably reduced. The method effectively improves the accuracy and stability of ultra-wideband positioning through a multi-stage optimization and correction mechanism, and is suitable for an indoor high-precision positioning scene.

Description

Technical Field

[0001] This invention relates to the field of communications, and specifically to a high-precision ultra-wideband positioning method and system based on a clustering optimization algorithm. Background Technology

[0002] In recent years, with the rise of the Internet of Things (IoT) field, ultra-wideband (UWB) technology has ushered in more development opportunities. As a wireless communication technology, UWB has many advantages, including insensitivity to channel fading, low transmitted signal power spectral density, low intercept rate, low system complexity, and the ability to provide accuracy down to several centimeters. Furthermore, compared to other positioning technologies such as WiFi, Bluetooth, and ZigBee, UWB offers superior security, high positioning accuracy, large system capacity, low power consumption, and strong anti-interference capabilities. In addition to enabling communication between devices, UWB can also be applied to indoor ranging and indoor positioning.

[0003] In the existing technology, during communication between devices, there may be a certain error between the calculated coordinates of the tag base station and the actual coordinates due to some physical reasons. This error is generally caused by insufficient accuracy of the device itself or errors in the parameters used to calculate the coordinates.

[0004] Therefore, there is a need for a method or apparatus that can solve the problem of accurately calculating the three-dimensional coordinates of tag base stations in existing technologies. Summary of the Invention

[0005] To address the aforementioned shortcomings in the existing technology, this invention provides a high-precision ultra-wideband positioning method and system based on a clustering optimization algorithm.

[0006] To achieve the above-mentioned objectives, the technical solution adopted by this invention is as follows: A high-precision ultra-wideband positioning method based on clustering optimization algorithm includes the following steps: Construct a base station system, which includes: a tag base station, multiple reference base stations, and a processor. The calculated distance between the tag base station and the reference base station is calculated using a two-way bidirectional ranging method; The calculated coordinates of the tag base station are calculated using a time-of-arrival positioning method; The K-means clustering optimization algorithm based on particle swarm optimization is used to cluster the coordinates of the tag base station. The parameters of the Kalman filter function in each classification space are optimized using the particle swarm optimization algorithm. The calculated coordinates of the tag base station are substituted into the optimized Kalman filter function to obtain the optimized tag base station coordinates; The tag base station coordinates are further optimized using an error correction model to obtain the optimal positioning coordinates of the tag base station.

[0007] Furthermore, the step of clustering the tag base station coordinates using the K-means clustering optimization algorithm based on particle swarm optimization specifically includes: Initialize the parameters in the particle swarm optimization algorithm and the mean point in the K-means clustering algorithm, wherein the parameters include the number of particle populations, the maximum number of iterations, particle positions and velocities, mean point coordinates, and the number of classes; Calculate the distance between the mean point coordinates and the calculated coordinates of each tag base station, perform clustering criteria calculation, and classify according to the distance and the number of categories; The particle swarm optimization algorithm is used to optimize the coordinates of the mean point and reclassify the calculated coordinates of the tag base stations so that the distance between the calculated coordinates of the tag base stations and the mean point in each class is similar, thus obtaining the optimal classification result.

[0008] Furthermore, the step of optimizing the cluster centers using the particle swarm optimization algorithm specifically includes: Initialize the parameters of the particle swarm optimization algorithm, wherein the parameters include the learning factor, inertia weight, maximum number of iterations, search space dimension, and initial number of particles in the swarm; Substitute the parameters into the clustering function to obtain the fitness of the particles; Compare the fitness of all particles and take the local optimum of the particle with the lowest fitness as the global optimum. The fitness corresponding to the global optimal solution is compared with the preset precision. If the fitness is not greater than the preset precision, the parameters in the global optimal solution vector are used as the cluster center parameters of the optimized clustering function.

[0009] Furthermore, the step of the processor using the particle swarm optimization algorithm to optimize the parameters in the Kalman filter function of each classification space specifically includes: Initialize the parameters in the Kalman filter function, wherein the parameters include process noise Q, measurement noise R, initial coordinates X of the tag base station, and estimated covariance P parameter; The calculated coordinates are substituted into the Kalman filter function for filtering optimization to obtain the optimized coordinates, and the difference between the preset actual coordinates and the optimized coordinates is calculated. If the difference is greater than a preset difference threshold, then the parameters of the Kalman filter function are optimized using a particle swarm optimization algorithm.

[0010] Furthermore, if the difference is greater than a preset difference threshold, the step of optimizing the parameters of the Kalman filter function using the particle swarm optimization algorithm specifically includes: Initialize the parameters of the particle swarm optimization algorithm, wherein the parameters include the learning factor, inertia weight, maximum number of iterations, search space dimension, and initial number of particles in the swarm; Substitute the parameters into the Kalman filter function to obtain the predicted coordinates; Substituting the predicted coordinates into the optimization criteria yields the particle's fitness. Compare the fitness of all particles and take the local optimum of the particle with the lowest fitness as the global optimum. The fitness corresponding to the global optimal solution is compared with the preset accuracy. If the fitness is not greater than the preset accuracy, the parameters in the global optimal solution vector are used as the parameters in the optimized Kalman filter function.

[0011] Furthermore, the Kalman gain value of the Kalman filter function is calculated as follows:

[0012] in, This represents the Kalman gain value at the i-th iteration. Let be the covariance of the (i-1)th iteration, and R represent the measurement noise; The optimized coordinates of the Kalman filter function are:

[0013] Where X(i) is the optimal estimated value of the i-th time, K(i) is the Kalman gain value of the i-th time, data(i) is the calculated coordinate of the i-th time, and X(i-1) is the optimal estimated value of the (i-1)-th time. The covariance matrix of the Kalman filter function is:

[0014] Where P(i) is the covariance value of the i-th iteration, P(i-1) is the covariance value of the (i-1)-th iteration, K(i) is the Kalman gain value of the i-th iteration, and Q is the process noise.

[0015] Furthermore, the optimization criterion is as follows:

[0016] Where sum represents the sum of squared errors. This represents the data before filter optimization. This represents the target coordinate data, i.e., the preset actual coordinates, where N is the number of data points.

[0017] Furthermore, the method for establishing the error correction model is as follows: Calculate the error between the calculated coordinates of the tag base station and the preset actual coordinates, and fit the error result and the calculated coordinates into a first-order function, expressed as:

[0018] Δ x Δ y Δ z The error of the calculated coordinates in the x, y, and z axes; The predicted coordinates optimized by the Kalman filter function are substituted into the error correction model to obtain the error result. The error result is then superimposed onto the predicted coordinates optimized by the Kalman filter function to obtain the final optimized coordinates.

[0019] A high-precision ultra-wideband positioning system based on clustering optimization algorithm includes a first calculation module, a second calculation module, a third calculation module, an optimization module, and a fourth calculation module. The first calculation module is used to calculate the distance between the tag base station and the reference base station using a two-sided bidirectional ranging method; The second calculation module is used to calculate the coordinates of the tag base station using a time-of-arrival positioning method; The third calculation module is used to calculate the coordinates of the tag base station and the distance between the cluster center point; The optimization module is used to optimize the parameters in the Kalman filter function using the particle swarm optimization algorithm. The fourth calculation module is used to substitute the calculated coordinates into the Kalman filter function to obtain the optimized tag base station coordinates.

[0020] The present invention has the following beneficial effects: A high-precision ultra-wideband positioning method based on clustering optimization algorithms and error correction models is proposed. The method is based on a base station system, which includes one tag base station, four reference base stations, and a processor. The method includes: calculating the computational distance between the tag base station and the reference base stations using a bilateral bidirectional ranging method; calculating the computational coordinates of the tag base station using a time-of-arrival positioning method based on computer theory; optimizing the mean points of a K-means clustering algorithm using a particle swarm optimization algorithm; substituting the computational coordinates into the K-means clustering optimization algorithm to obtain the classification results of the tag base station's computational coordinates; optimizing the Kalman filter function parameters of each classification space using a particle swarm optimization algorithm; substituting the tag base station's computational coordinates into the Kalman filter function after updating and optimizing the parameters to obtain the optimized tag base station coordinates; and substituting the optimized coordinates into an error correction model to calculate the error of each computational coordinate relative to the preset actual coordinates, adding the error to the output coordinates of the Kalman filter function to further reduce random errors and obtain the optimal computational coordinates of the tag base station. The optimized coordinates, compared to the computational coordinates, have significantly improved accuracy, thus solving the problem of accurately calculating the coordinates of tag base stations in existing technologies. Attached Figure Description

[0021] Figure 1 This application provides an illustration of an application scenario. Figure 2 A flowchart illustrating a high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model, provided for an embodiment of this application; Figure 3 A schematic diagram of the clustering optimization algorithm flow for a high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model, provided in this application embodiment. Figure 4 A flowchart illustrating an intelligent optimization method for the filtering function of each classification space in another high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model, provided for an embodiment of this application. Figure 5 A flowchart illustrating the parameter optimization of the filter function in another high-precision ultra-wideband positioning method based on clustering optimization algorithm and error correction model provided in this application embodiment; Figure 6 A flowchart illustrating the error correction model of a high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model, provided in this application embodiment; Figure 7 An error diagram between the calculated coordinates of the tag base station obtained by the traditional optimization method and the preset actual coordinates; Figure 8 This is an error diagram between the optimized coordinates and actual coordinates of the tag base station obtained by the high-precision ultra-wideband positioning method based on clustering optimization algorithm and error correction model of this application. Figure 9 This is a schematic diagram of a high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model, provided as an embodiment of this application. Detailed Implementation

[0022] The specific embodiments of the present invention are described below to enable those skilled in the art to understand the present invention. However, it should be understood that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, various changes are obvious as long as they are within the spirit and scope of the present invention as defined and determined by the appended claims. All inventions utilizing the concept of the present invention are protected.

[0023] Figure 1 This application provides an illustration of an application scenario, such as... Figure 1 As shown, the method in this application can be applied to Figure 1 The electronic device 10 shown. For example... Figure 1 As shown, the electronic device 10 may include: tag base station 1, reference base station 1, reference base station 2, reference base station 3, reference base station 4, and processor 1.

[0024] Tag base station 1, reference base station 1, reference base station 2, reference base station 3, and reference base station 4 are electrically connected to each other directly or indirectly to realize data transmission or interaction. The processor 1 controls and processes the signals transmitted between tag base station 1, reference base station 1, reference base station 2, reference base station 3, and reference base station 4.

[0025] Understandable. Figure 1 The structure shown is for illustrative purposes only; the electronic device 10 may also include components that are more advanced than those shown. Figure 1 The more or fewer components shown, or having the same Figure 1 The different configurations shown. Figure 1 The components shown can be implemented using hardware, software, or a combination thereof.

[0026] Based on the above, this application also provides a computer-readable storage medium, which includes a computer program that, when executed, controls the electronic device 10 containing the computer-readable storage medium to perform the following method.

[0027] To make the implementation process of this invention clearer, a detailed description will be provided below in conjunction with the accompanying drawings.

[0028] Figure 2 A flowchart illustrating a high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model, provided for embodiments of this application; Figure 2 As shown, this application provides a high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model. The method is based on a base station system, which includes: a tag base station, a reference base station, and a processor. The method includes: S101. Calculate the distance between the tag base station and the reference base station using the bilateral two-way ranging method.

[0029] Optionally, the bilateral two-way ranging method is as follows: (1) TOF indicates the time it takes for a signal to travel through the air; This represents the difference between the time when the tag base station receives the permission signal and the time when the tag base station sends the request signal; This represents the difference between the time the tag base station sends a data packet and the time the tag base station receives the permission signal to send the packet. This represents the difference between the time when the reference base station sends the permission signal and the time when the reference base station receives the request signal; This represents the difference between the time when the reference base station receives the data packet and the time when the reference base station issues a signal allowing transmission. The time unit representing the ultra-wideband chip is a fixed value; The formula for calculating the distance between the tag base station and the reference base station is: Where c represents the speed of electromagnetic wave propagation, and d represents the calculated distance between the tag base station and the reference base station.

[0030] The flight time of the signal between the tag base station and the reference base station is calculated using Formula 1 above. Based on the flight time and the signal transmission speed, the calculated distance between the tag base station and the reference base station is calculated. Since the signal may be subject to interference when flying between the tag base station and the reference base station, the calculated distance may have a deviation. Therefore, it is necessary to correct and optimize the deviation of the calculated distance to obtain the optimized distance between the tag base station and the reference base station.

[0031] S102. Calculate the tag base station coordinates using a time-of-arrival positioning method.

[0032] Optionally, the time-arrival-based positioning method is as follows: assuming , , , The coordinates of the reference base stations 1, 2, ..., n are given, and the coordinates of the tag are given. Then the coordinates of the intersection point of the n spheres are the coordinates of the tag node device, which can be determined by solving the following formula.

[0033]

[0034] (2)

[0035] Due to the complexity of the indoor environment, equation (2) is linearized to obtain:

[0036]

[0037]

[0038] (3) The above formula can be expressed as: (4) in , ,

[0039] Finally, based on the least squares algorithm, the formula for calculating the label can be expressed as: (5) S103. Use the K-means clustering algorithm based on particle swarm optimization to cluster the calculated coordinates.

[0040] Figure 3 A flowchart illustrating another high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model provided in this application embodiment; as shown. Figure 3 As shown, optionally, the step of using the parameters of the cluster mean points of the particle swarm optimization algorithm and the K-means clustering algorithm specifically includes: S201. Initialize the parameters of the particle swarm optimization algorithm and the cluster mean points of the K-means clustering algorithm.

[0041] Initialize the parameters in the particle swarm optimization algorithm, that is, give the value of each parameter in the particle swarm optimization algorithm. The parameters of the particle swarm optimization algorithm include learning factor, inertia weight, maximum number of iterations, search space dimension, and initial number of particles in the swarm.

[0042] Optionally, the cluster points of this K-means clustering algorithm are the coordinates that divide the categories.

[0043] S202, Clustering Criterion Calculation.

[0044] Optionally, the clustering criteria of the clustering algorithm are: (6) x This represents the calculated coordinates of the tag base station. c i For the first i Cluster centers (mean points). f To calculate the dimensions of the coordinates.

[0045] S203. Use the particle swarm optimization algorithm to optimize the mean point for classification.

[0046] Optionally, the particle swarm optimization algorithm is as follows: (7) (8) (9) in: and Let these represent the position vector and velocity vector of the i-th particle, respectively. Represents the inertia weights. Indicates the local optimal position. Indicates the globally optimal position. and It is a constant. and It is a random number between [0,1]. Indicates the current iteration number. Indicates the maximum number of iterations. and These represent the maximum and minimum values ​​of the inertia weight, respectively.

[0047] S204. The clustering algorithm obtains the optimal clustering result.

[0048] S104. Optimize the Kalman filter function parameters for each classification space using the particle swarm optimization algorithm. Figure 4 A flowchart illustrating another high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model provided in this application embodiment; as shown. Figure 4 As shown, optionally, the steps of the processor to optimize the parameters in the Kalman filter function of each classification space using the particle swarm optimization algorithm specifically include: S301. Initialize the parameters in the Kalman filter function.

[0049] The parameters of the Kalman filter function are preset, including process noise Q, measurement noise R, initial position X of the signal transmitting base station, and estimated covariance P. That is, this step includes setting the Q, R, X, and P parameters of the Kalman filter function. The process of setting the parameters of the Kalman filter function is called initializing the parameters of the Kalman filter function. The range of the given parameters is set according to actual needs and is not specifically limited here.

[0050] S202. Substitute the calculated coordinates into the Kalman filter function for filtering optimization to obtain the optimized coordinates, and calculate the difference between the preset actual coordinates and the optimized coordinates.

[0051] Optionally, the Kalman gain of the Kalman filter function is calculated as follows: (10) in, This represents the Kalman gain value at the i-th iteration. Let be the covariance of the (i-1)th iteration, and R represent the measurement noise; The optimized distance of the Kalman filter function is: (11) Where X(i) is the optimal estimated value of the i-th time, K(i) is the Kalman gain value of the i-th time, data(i) is the measurement value of the i-th time, and X(i-1) is the optimal estimated value of the (i-1)-th time. The covariance matrix of the Kalman filter function is: (12) Where P(i) is the covariance value of the i-th iteration, P(i-1) is the covariance value of the (i-1)-th iteration, K(i) is the Kalman gain value of the i-th iteration, and Q is the process noise.

[0052] Substitute the calculated coordinates into the above formula (11) and combine them with the parameters of the initialized Kalman filter function to obtain the optimized coordinates. Calculate the difference between the preset actual coordinates and the optimized coordinates. Generally, use the preset actual coordinates to subtract the optimized coordinates and then calculate the absolute value of the difference. It should be noted that the preset actual coordinates are preset and are not specifically limited here.

[0053] S203. If the difference is greater than the preset difference threshold, the parameters of the Kalman filter function are optimized using the particle swarm optimization algorithm.

[0054] The preset difference threshold is set according to actual needs. It should be noted that the smaller the preset difference threshold, the higher the accuracy of the final calculated tag base station and reference base station. If the difference is not greater than the preset difference threshold, it means that the error is very small, the optimization stops, and the optimized coordinate value is obtained.

[0055] Figure 5 A flowchart illustrating another high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model provided in this application embodiment; as shown. Figure 5 As shown, optionally, the step of optimizing the parameters of the Kalman filter function using the particle swarm optimization algorithm if the difference is greater than a preset difference threshold specifically includes: S401. Initialize the parameters of the particle swarm optimization algorithm.

[0056] Optionally, the particle swarm optimization algorithm is as follows: (13) (14) (15) in: and Let these represent the position vector and velocity vector of the i-th particle, respectively. Represents the inertia weights. Indicates the local optimal position. Indicates the globally optimal position. and It is a constant. and It is a random number between [0,1]. Indicates the current iteration number. Indicates the maximum number of iterations. and These represent the maximum and minimum values ​​of the inertia weight, respectively.

[0057] Initialize the parameters in the particle swarm optimization algorithm, that is, give the value of each parameter in the particle swarm optimization algorithm. The parameters of the particle swarm optimization algorithm include learning factor, inertia weight, maximum number of iterations, search space dimension, and initial number of particles in the swarm.

[0058] S402. Substitute the parameters into the Kalman filter function to obtain the predicted coordinates.

[0059] The parameters of the initialized Kalman filter function are substituted into the Kalman filter function, and the predicted coordinates are obtained through calculation.

[0060] S403. Substitute the predicted coordinates into the optimization criteria to obtain the particle's fitness.

[0061] Optionally, the optimization criterion is: (16) Where sum represents the sum of squared errors. This represents the data before filter optimization. This represents the target coordinate data, which is the preset actual coordinates, and N is the number of data points.

[0062] The predicted coordinates obtained in step S402 are then substituted into the optimization criterion to calculate the fitness. It should be noted that the result output by this optimization criterion is called the fitness.

[0063] S404. Compare the fitness of all particles and take the local optimum of the particle with the lowest fitness as the global optimum.

[0064] Given a local optimum and a global optimum for a particle, all particles in the particle swarm optimization algorithm are put into the particle swarm optimization algorithm for iteration. During the overshoot of the iteration, the velocity of each particle is calculated using the above formula (13). When the calculated velocity exceeds the range, it is increased to the boundary. That is, when the calculated velocity is greater than the preset velocity threshold, the particle is discarded as an inaccurate result. The fitness calculated for each particle is compared with the preset local optimum. If the fitness of the particle is less than the local optimum, the fitness of the particle is taken as the local optimum. At the same time, the local optimum is compared with the global optimum. If the local optimum is less than the global optimum, the local optimum is taken as the global optimum. That is, after the iteration is completed, the global optimum is the value with the smallest fitness among all particles.

[0065] Furthermore, during the iteration process, the weights are used with inertial decreasing, and the calculation method is as follows: (17) in, Represents the inertia weights. Indicates the current iteration number. Indicates the maximum number of iterations. and These represent the maximum and minimum values ​​of the inertia weight, respectively.

[0066] S405. Compare the fitness corresponding to the global optimal solution with the preset accuracy. If the fitness is not greater than the preset accuracy, then use the parameters in the global optimal solution vector as the parameters in the optimized Kalman filter function.

[0067] This preset precision is designed to ensure the accuracy of the parameters and, consequently, the accuracy of the calculated tag base station coordinates. The smaller the preset precision, the higher the accuracy of the calculated tag base station coordinates. The fitness corresponding to the global optimal solution is compared with this preset precision. If the fitness corresponding to the global optimal solution is less than the preset precision, then the parameters in the vector of the global optimal solution are used as the parameters in the optimized Kalman filter function. It should be noted that the vector of the global optimal solution is a four-dimensional vector, and the parameters placed therein include Q, R, and P.

[0068] S103. Substitute the calculated coordinates in each classification space into the Kalman filter function of each space to obtain the optimized coordinates of the tag base station.

[0069] The calculated tag base station coordinates are substituted into the optimized Kalman filter function, i.e., the above formulas (10), (11) and (12), and the optimized tag base station coordinates are obtained after calculation.

[0070] S106. Use the error correction model to further optimize the tag base station coordinates and obtain the optimal positioning coordinates.

[0071] Figure 6 A flowchart illustrating another high-precision ultra-wideband positioning method based on a clustering optimization algorithm and an error correction model provided in this application embodiment; as shown. Figure 5 As shown, optionally, the steps for reducing the random error of the tag base station optimized coordinates using an error correction model specifically include: S501, the calculation error between the coordinates calculated based on the tag base station and the preset actual coordinates.

[0072] The error result is obtained by subtracting the calculated coordinates of the tag base station obtained in the experiment from the preset actual coordinates.

[0073] S502. Establish a first-order error model based on the error and calculated coordinates.

[0074] The error model is established by fitting the error result and the calculated coordinates into a first-order function based on the error between the calculated coordinates of the tag base station and the preset actual coordinates.

[0075] Optionally, the first-order function model is: (18) S503. Substitute the optimized coordinates output by the Kalman filter function into the first-order error model to solve for the random errors on the x / y / z axes.

[0076] S504. The x / y / z axis errors output by the error correction model are superimposed onto the optimized coordinates output by the Kalman function to obtain the final optimized coordinates.

[0077] The first-order error correction model established based on the error between the calculated coordinates and the preset actual coordinates of the tag base station is as follows:

[0078] The core parameters of the K-means clustering algorithm are the number of classes K and the coordinates of the mean point of each class. Calculating the optimal combination of coordinates has a great impact on reducing random positioning errors. It is hoped that the optimal mean point can be solved by the particle swarm optimization algorithm to obtain the optimal combination of classes. The process of optimizing the cluster mean point parameters of the K-means clustering algorithm using the particle swarm optimization algorithm is as follows: (1) Initialize the parameters of the particle swarm optimization algorithm, as shown in Table 1. Table 1 shows the initialization parameters of the particle swarm optimization algorithm.

[0079] Table 1 Initialization parameters of the particle swarm optimization algorithm

[0080] (2) The mean point optimization criterion is constructed using the Euclidean distance formula, and its expression is as follows:

[0081] in, x This represents the calculated coordinates of the tag base station. c i For the first i Cluster centers (mean points). f To calculate the dimensions of the coordinates; (3) Based on the performance index of the constructed optimization criteria, the mean point coordinates are optimized using a population optimization algorithm; (4) Determine whether the optimal mean point has been obtained. If so, complete the optimization design; otherwise, return to step (3).

[0082] In this embodiment, with a classification number of 3, the mean point coordinates of each classification space are KI=(0.377, 0.494, 1.743)m, KII=(1.659, 0.15, 1.115)m, and KIII=(1.457, 0.875, 0.765)m.

[0083] The parameters of the Kalman filter function include process noise Q, measurement noise R, and error covariance P. The optimal combination of these three parameters has a significant impact on the filtering effect. It is hoped that the optimal combination of values ​​can be solved by a population optimization algorithm. Particle swarm optimization (PSO) is a commonly used population optimization algorithm with advantages such as simple structure and ease of implementation. The established neural network model and optimization criteria are integrated into the PSO algorithm. The design parameters of the Kalman filter function are optimized using a PSO algorithm with linearly decreasing inertia weights. The PSO algorithm with linearly decreasing inertia weights can be described by the above formulas (13), (14), and (15). The process of optimizing the design parameters of the Kalman filter function using the PSO algorithm with linearly decreasing inertia weights is as follows: (1) Initialize the parameters of the particle swarm optimization algorithm, as shown in Table 2. Table 2 shows the initialization parameters of the particle swarm optimization algorithm.

[0084] Table 2. Parameters for initializing the particle swarm optimization algorithm

[0085] (2) Based on the parameters of the filtering function, an optimization criterion is constructed using the performance index of the sum of squared errors, and its expression is as follows:

[0086] Where sum represents the sum of squared errors. This represents the data before filter optimization. This represents the target coordinate data, which is the preset actual coordinates, and N is the number of data points. (3) Based on the performance index of the constructed optimization criteria, the parameters of the filtering function are optimized using the population optimization algorithm; (4) Determine whether the parameters of the optimal Kalman filter function have been obtained. If so, complete the optimization design; otherwise, return to step (3).

[0087] In this embodiment, according to the desired performance indicators, the parameters of the Kalman filter functions for the KI, KII, and KIII class spaces were optimized using the above-mentioned method. The optimized process noise Q, measurement noise R, and error covariance P are as follows: KI: P=diag(200,198.98,-5.49), Q=diag(2.34,6.46,5.73), R=diag(0.15,0.02,-0.04); KII: P=diag(-199.99,2.02,2.34), Q=diag(-0.37,-2.29,-5.88), R=diag(0.02,-0.33,0.28); KIII: P=diag(5.17,2.85,1.48), Q=diag(-4.54,-4.09,-3.09), R=diag(-1.89,-0.11,0.24).

[0088] Based on the parameters of the optimized filtering function and the random error output by the error correction model, the sum of squared ranging errors of traditional optimization and intelligent optimization were simulated using MATLAB software. The results are shown in Table 3. Table 3 Comparison of Results

[0089] As shown in Table 3, the sum of squared positioning errors obtained using the proposed method is smaller than that obtained using the traditional optimization method. This indicates that the proposed method can obtain the optimal design parameters of the filter function and reduce random errors in the positioning space, effectively improving the filtering function of the ultra-wideband positioning filter function and enhancing positioning accuracy.

[0090] Figure 7 An error diagram between the calculated coordinates of the tag base station obtained by the traditional optimization method and the preset actual coordinates; Figure 8 This is an error map between the optimized coordinates of the tag base station and the preset actual coordinates obtained by the high-precision ultra-wideband positioning method based on clustering optimization and error correction model of this application; Figure 7 and Figure 8 As shown, after applying the positioning method of this application, the maximum positioning error on each coordinate axis is reduced from 26.37 cm to 20.55 cm, which demonstrates that our proposed high-precision ultra-wideband positioning method based on clustering optimization algorithm and error correction model is effective.

[0091] Figure 9 A schematic diagram of a high-precision ultra-wideband positioning method system based on clustering optimization algorithm and error correction model provided in this application embodiment; as shown Figure 7 As shown, this application provides a high-precision ultra-wideband positioning system based on clustering optimization algorithm and error correction model. The system includes: a first calculation module, a second calculation module, a third calculation model, an optimization module, and a fourth calculation module. The first calculation module is used to calculate the distance between the tag base station and the reference base station using the bilateral two-way ranging method; The second calculation module is used to calculate the coordinates of the tag base station using a time-of-arrival positioning method; The third calculation module is used to calculate the coordinates of the tag base station and the distance between the cluster center point; The optimization module is used to optimize the parameters in the Kalman filter function using the particle swarm optimization algorithm; The fourth calculation module is used to substitute the calculated coordinates into the Kalman filter function to obtain the optimized tag base station coordinates.

[0092] Optionally, this optimization module is specifically used for: Initialize the parameters in the Kalman filter function, including process noise Q, measurement noise R, and estimated covariance P. The calculated coordinates are substituted into the Kalman filter function for filtering optimization to obtain the optimized coordinates, and the difference between the preset actual coordinates and the optimized coordinates is calculated. If the difference is greater than the preset difference threshold, the parameters of the Kalman filter function are optimized using the particle swarm optimization algorithm.

[0093] Optionally, this optimization module is also used for: Initialize the parameters of the particle swarm optimization algorithm, including the learning factor, inertia weight, maximum number of iterations, search space dimension, and initial number of particles in the swarm. Substitute the parameters into the Kalman filter function to obtain the predicted coordinates; Substituting the predicted coordinates into the optimization criteria yields the particle's fitness. Compare the fitness of all particles and take the local optimum of the particle with the lowest fitness as the global optimum. The fitness corresponding to the global optimum is compared with the preset accuracy. If the fitness is not greater than the preset accuracy, the parameters in the global optimum vector are used as the parameters in the optimized Kalman filter function.

[0094] Optionally, this optimization module is also used for: Initialize the parameters of the particle swarm optimization algorithm, including the learning factor, inertia weight, maximum number of iterations, search space dimension, and initial number of particles in the swarm. Substitute the parameters into the K-means clustering function to obtain the predicted cluster coordinates; Substituting the predicted cluster coordinates into the optimization criterion yields the particle fitness. Compare the fitness of all particles and take the local optimum of the particle with the lowest fitness as the global optimum. The fitness corresponding to the global optimum is compared with the preset precision. If the fitness is not greater than the preset precision, the parameters in the global optimum vector are used as the coordinate parameters of the mean point of the optimized clustering function.

[0095] This application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-mentioned intelligent optimization method for improving the accuracy of ultra-wideband ranging.

[0096] This application provides a computer-readable storage medium, which includes a computer program. When the computer program is executed, it controls the electronic device containing the computer-readable storage medium to perform the above-mentioned intelligent optimization method for improving the accuracy of ultra-wideband ranging.

[0097] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0098] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0099] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0100] Specific embodiments have been used to illustrate the principles and implementation methods of this invention. The descriptions of the embodiments above are only for the purpose of helping to understand the method and core ideas of this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

[0101] Those skilled in the art will recognize that the embodiments described herein are intended to help the reader understand the principles of the invention, and should be understood that the scope of protection of the invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific modifications and combinations based on the technical teachings disclosed in this invention without departing from the spirit of the invention, and these modifications and combinations are still within the scope of protection of this invention.

Claims

1. A high-precision ultra-wideband positioning method based on a clustering optimization algorithm, characterized in that, The method comprises the following steps: constructing a base station system comprising a tag base station, a plurality of reference base stations and a processor, calculating the calculated distance between the tag base station and the reference base station using a two-lateral two-way ranging method; calculating the calculated coordinates of the tag base station using a time-of-arrival positioning method; clustering and dividing the tag base station coordinates using a K-means clustering optimization algorithm based on a particle swarm optimization algorithm; optimizing the Kalman filter function parameters of each classification space using a particle swarm optimization algorithm; bringing the tag base station calculated coordinates into the Kalman filter function after optimization of the parameters to obtain the optimized tag base station coordinates; further optimizing the tag base station coordinates using an error correction model to obtain the optimal positioning coordinates of the tag base station.

2. The high-precision ultra-wideband positioning method based on clustering optimization algorithm according to claim 1, characterized in that, The step of clustering and dividing the tag base station coordinates using a K-means clustering optimization algorithm based on a particle swarm optimization algorithm specifically comprises: initializing the parameters in the particle swarm optimization algorithm and the mean points of the K-means clustering algorithm, wherein the parameters include the number of particle populations, the maximum number of iterations, particle positions and speeds, mean point coordinates, and the number of classifications; calculating the distance between the mean point coordinates and each tag base station calculated coordinate, performing clustering criterion calculation, and classifying according to the distance and the number of classifications; optimizing the mean point coordinates using a particle swarm optimization algorithm, reclassifying the tag base station calculated coordinates, making the distance between the tag base station calculated coordinates in each class and the mean point similar, and obtaining the optimal classification result.

3. The high-precision ultra-wideband positioning method based on a clustering optimization algorithm according to claim 2, characterized in that, The step of optimizing the cluster centers using a particle swarm optimization algorithm specifically comprises: initializing the parameters of the particle swarm optimization algorithm, wherein the parameters include a learning factor, an inertia weight, a maximum number of iterations, a search space dimension, and the number of particles in the initialized population; substituting the parameters into the clustering function to obtain the fitness of the particles; comparing the corresponding fitness of all particles, and taking the local optimal solution corresponding to the particle with the minimum fitness as the global optimal solution; comparing the fitness corresponding to the global optimal solution with a preset precision, and if the fitness is not greater than the preset precision, taking the parameters in the global optimal solution vector as the cluster center parameters of the optimized clustering function.

4. The high-precision ultra-wideband positioning method based on clustering optimization algorithm according to claim 3, characterized in that, The step of the processor optimizing the parameters in the Kalman filter function of each classification space using a particle swarm optimization algorithm specifically comprises: initializing the parameters in the Kalman filter function, wherein the parameters include process noise Q, measurement noise R, the initial coordinates X of the tag base station, and the estimation covariance P parameters; substituting the calculated coordinates into the Kalman filter function for filtering optimization to obtain the optimized coordinates, and calculating the difference between the preset actual coordinates and the optimized coordinates; if the difference is greater than a preset difference threshold, optimizing the parameters of the Kalman filter function using a particle swarm optimization algorithm.

5. The high-precision ultra-wideband positioning method based on clustering optimization algorithm according to claim 4, characterized in that, The step of optimizing the parameters of the Kalman filter function using a particle swarm optimization algorithm if the difference is greater than a preset difference threshold specifically comprises: initializing the parameters of the particle swarm optimization algorithm, wherein the parameters include a learning factor, an inertia weight, a maximum number of iterations, a search space dimension, and the number of particles in the initialized population; substituting the parameters into the Kalman filter function to obtain the predicted coordinates; Substitute the predicted coordinates into an optimization criterion to obtain a fitness of the particle; Compare the corresponding fitness of all particles, and take the local optimal solution corresponding to the particle with the minimum fitness as a global optimal solution; Compare the fitness corresponding to the global optimal solution with a preset accuracy, and if the fitness is not greater than the preset accuracy, take the parameters in the global optimal solution vector as parameters in the optimized Kalman filter function.

6. The high-precision ultra-wideband positioning method based on clustering optimization algorithm according to claim 5, characterized in that, The calculation method of the Kalman gain value of the Kalman filter function is: wherein, denotes the Kalman gain value at the i-th time, is the covariance at the (i-1)-th time, and R represents the measurement noise; The optimized coordinates of the Kalman filter function are: Wherein, X(i) is the i-th estimated optimal value, K(i) is the i-th Kalman gain value, data(i) is the i-th calculated coordinates, X(i-1) is the (i-1)-th estimated optimal value; The covariance matrix of the Kalman filter function is: Wherein, P(i) is the i-th covariance value, P(i-1) is the (i-1)-th covariance value, K(i) is the i-th Kalman gain value, and Q is the process noise.

7. The high-precision ultra-wideband positioning method based on a clustering optimization algorithm according to claim 6, characterized in that, The optimization criterion is: Wherein, sum represents error sum of squares, Data before filter optimization, Target coordinate data, i.e. preset actual coordinate, N is data number.

8. The high-precision ultra-wideband positioning method based on clustering optimization algorithm according to claim 6, characterized in that, The establishment method of the error correction model is: Calculate the error between the calculated coordinates of the tag base station and the preset actual coordinates, fit the error result and the calculated coordinates into a first-order function, and express it as: Δ x , Δ y , Δ z error of the calculated coordinates in x, y, z axis direction; Substitute the predicted coordinates of the Kalman filter function into the error correction model to obtain the error result, and superimpose the error result on the predicted coordinates of the Kalman filter function to obtain the final optimized coordinates.

9. Positioning system based on the positioning method according to any one of claims 1 to 8, characterized in that The positioning system comprises a first calculation module, a second calculation module, a third calculation module, an optimization module, and a fourth calculation module. The first calculation module is used to calculate the calculated distance between the tag base station and the reference base station by using the double bilateral and bidirectional ranging method; The second calculation module is used to calculate the calculated coordinates of the tag base station by using the time-of-arrival positioning method; The third calculation module is used to calculate the distance between the coordinates of the tag base station and the cluster center point; The optimization module is used to optimize the parameters in the Kalman filter function by using the particle swarm optimization algorithm; The fourth calculation module is used to substitute the calculated coordinates into the Kalman filter function to obtain the optimized coordinates of the tag base station.

Images