Double-station multi-target tracking method and device based on WMFS and UKF

Abstract

The invention provides a double-station multi-target tracking method and device based on WMFS and UKF, and the method comprises the steps: constructing a WMFS model, and reconstructing a state transfer function through a pre-trained WMFS model; combining the optimized WMFS model with the UKF model to obtain a double-station multi-target tracking model; and tracking a target by using the double-station multi-target tracking model. According to the method, the advantages of the WMFS in the aspects of processing system uncertainty and complex modeling are fully utilized, in the WMFS-UKF derivation process, an unscented (UT) frame is adopted to solve the nonlinear measurement problem, meanwhile, a pre-trained WMFS is used for reconstructing a state conversion function, and through the historical state and the expected output of a target, the nonlinear measurement problem is solved. The WM fuzzy inference system realizes more accurate state prediction and accurate covariance estimation, improves the performance and stability of target tracking, and verifies the effectiveness and robustness of the algorithm in a target tracking scene.

Description

Technical Field

[0001] The invention belongs to the field of sensor technology, and in particular relates to a dual-station multi-target tracking method and device based on WMFS and UKF. Background Art

[0002] PASSIVE positioning and tracking technology has the characteristics of strong concealment, strong anti-interference ability, and wide detection range, which can effectively improve the system's survivability, combat capability, and detection capability. Common passive positioning methods include direction of arrival (DOA), frequency difference of arrival (FDOA), and time difference of arrival (TDOA).

[0003] Among them, DOA is a relatively simple and reliable passive positioning and tracking method. The target position is determined by measuring the arrival time or arrival angle of the target signal on different receivers. DOA can be traced back to submarine sonar tracking, in which the position and velocity of the target are determined based on the azimuth obtained by sonar measurement. However, the traditional DOA method can only provide position information and shows limitations when tracking highly maneuverable targets. In order to solve this problem, the Direction of Arrival (DOAD) method with Doppler is proposed.

[0004] The Doppler (DOAD) method uses the Doppler frequency shift of the target signal to obtain velocity information, thereby improving tracking performance. DOAD is a relatively new passive positioning and tracking method with high research significance and practical value. It is a passive positioning and tracking method with high research significance and practical value in this field. In the field of target tracking, dealing with nonlinear measurements is a major challenge in filtering problems, especially complex measurements such as azimuth-Doppler. The classic Kalman filter (KF) performs well in simple linear systems with Gaussian noise, but performs poorly in terms of nonlinearity of measurements, which may reduce the estimation accuracy. To solve this problem, various filtering algorithms have been introduced to better handle nonlinear measurements.

[0005] Existing technologies such as the classic Kalman filter (KF) work well when dealing with linear systems and Gaussian noise, but their estimation accuracy drops significantly in the case of nonlinear measurements, especially when dealing with complex multi-target tracking, which becomes an obvious bottleneck. To improve this, methods such as the extended Kalman filter (EKF) have been proposed, however, these methods still have limitations when dealing with highly maneuverable targets, especially when the measurement noise is large, the tracking stability and accuracy will drop significantly. Summary of the invention

[0006] The present application embodiment provides a dual-station multi-target tracking method based on WMFS and UKF, including: Build a WMFS model and use the pre-trained WMFS model to reconstruct the state transfer function; The optimized WMFS model is combined with the UKF model to obtain a dual-station multi-target tracking model; The dual-station multi-target tracking model is used to track the target.

[0007] Furthermore, the construction of the WMFS model includes: The WMFS model is trained using historical data, and the fuzzy rules and parameters of the trained WMFS model are determined. The WMFS model is a multi-layer WMFS, each layer of WMFS uses the output results of the upper layer as the input data of this layer, and the last layer of WMFS integrates the results.

[0008] Furthermore, the reconstructing the state transfer function using the pre-trained WMFS model includes: Using WMFS to generate state estimation and uncertainty information through internally defined fuzzy rules, and integrating the state estimation and uncertainty information into the framework of UKF to predict and update the target state; Through the propagation and transformation of Sigma points, the target state and covariance are estimated.

[0009] Furthermore, the estimation of the target state and covariance through the propagation and conversion of Sigma points includes: The signal state is determined according to the state equation, measurement equation, dual-station measurement equation and observation equation of the unscented Kalman filter.

[0010] Generate sigma points according to the signal state and transform the sigma point set using the state function; Compute a priori estimates of the states and covariances of the transformed sigma points, and estimate the sigma points from the prior states and covariances; Transform the sigma points by the measurement function and compute the prior estimates and covariances.

[0011] Furthermore, the reconstructed WMFS model is combined with the UKF model to obtain a dual-station multi-target tracking model, including: Determine the input and output data pairs of the Wang-Mendel fuzzy model; Determine the weights and weighted output parameters of the fuzzy set according to the input-output data pairs; The nonlinear state function of the UKF model is calculated according to the weights of the fuzzy sets and the weight output parameters.

[0012] The present invention provides a dual-station multi-target tracking device based on WMFS and UKF, comprising: An acquisition module is used to build a WMFS model and reconstruct the state transfer function using the pre-trained WMFS model; A processing module is used to combine the optimized WMFS model with the UKF model to obtain a dual-station multi-target tracking model; The execution module is used to track the target using the dual-station multi-target tracking model.

[0013] Furthermore, the construction of the WMFS model includes: The first acquisition module is used to train the WMFS model using historical data and determine the fuzzy rules and parameters of the trained WMFS model, wherein the WMFS model is a multi-layer WMFS, each layer of WMFS uses the output results of the upper layer as the input data of this layer, and the last layer of WMFS integrates the results.

[0014] Furthermore, the acquisition module includes: The second acquisition module is used to generate state estimation and uncertainty information by using WMFS through internally defined fuzzy rules, and integrate the state estimation and uncertainty information into the framework of UKF to predict and update the target state; The first processing module is used to estimate the target state and covariance through propagation and transformation of Sigma points.

[0015] Furthermore, the first processing submodule includes: The second processing submodule is used to determine the signal state according to the state equation, measurement equation, dual-station measurement equation and observation equation of the unscented Kalman filter.

[0016] The third processing submodule is used to generate sigma points according to the signal state and transform the sigma point set using the state function; a fourth processing submodule for calculating a priori estimates of the states and covariances of the transformed sigma points and estimating the sigma points from the prior states and covariances; The first processing submodule is used to transform the sigma point through the measurement function and calculate the prior estimate and covariance.

[0017] Furthermore, the processing module includes: The second processing submodule is used to determine the input and output data pairs of the Wang-Mendel fuzzy model; A third processing submodule, used for determining the weight and weight output parameter of the fuzzy set according to the input and output data pairs; The fourth processing submodule is used to calculate the nonlinear state function of the UKF model according to the weights of the fuzzy sets and the weight output parameters.

[0018] The present invention provides a computer device, including a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the dual-station multi-target tracking method based on WMFS and UKF as described above.

[0019] The present invention provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, enables the one or more processors to perform the steps of the dual-station multi-target tracking method based on WMFS and UKF as described above.

[0020] In order to solve the problems of excessive estimation error and filter instability caused by process noise uncertainty and system model uncertainty, a non-gain Kalman filter (UKF) algorithm based on Wang-Mendel fuzzy system (WMFS) is proposed for Doppler dual-station multi-target tracking algorithm. The algorithm fully utilizes the advantages of WMFS in dealing with system uncertainty and complex modeling. In the derivation process of WMFS-UKF, the non-gain transformation (UT) framework is adopted to solve the nonlinear measurement problem, and the pre-trained WMFS is used to reconstruct the state transition function. By utilizing the historical state of the target and the expected output, the WM fuzzy inference system achieves more accurate state prediction and precise covariance estimation, which greatly improves the performance and stability of target tracking. The simulation experiment verifies the effectiveness and robustness of the algorithm in the target tracking scenario. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0022] Figure 1 A flowchart of a dual-station multi-target tracking method based on WMFS and UKF provided in an embodiment of the present application; Figure 2 A schematic diagram of a Wang-Mendel fuzzy system provided for an embodiment of the present application; Figure 3This is the flow chart of the deep fuzzy reasoning system based on WMFS. DETAILED DESCRIPTION

[0023] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0024] like Figure 1 The present invention provides a dual-station multi-target tracking method based on WMFS and UKF, including: S1, build a WMFS model, and use the pre-trained WMFS model to reconstruct the state transfer function; The WMFS model is trained using historical data, and the fuzzy rules and parameters of the trained WMFS model are determined. The state estimation and uncertainty information are generated by WMFS through the internally defined fuzzy rules, and the state estimation and uncertainty information are integrated into the UKF framework to predict and update the target state. The WMFS model is a multi-layer WMFS, and each layer of WMFS uses the output results of the upper layer as the input data of this layer. The last layer of WMFS integrates the results to predict the target state.

[0025] Among them, historical state data and Sigma points used to characterize state estimation and uncertainty.

[0026] like Figure 2 As shown in Figure 1, the Wang-Mendel fuzzy system (WMFS) is constructed using the Wang-Mendel method, which designs low-dimensional fuzzy systems in a bottom-up, layer-by-layer manner, and finally forms the entire WMFS model. The size of the moving window m is to control the number of inputs to the low-dimensional fuzzy system, and m can be 3, 4, or 5. Figure 2 As shown, the vector of WMFS input is This vector has a high dimension, where The output is Y L At the first level (l = 1, 2, ..., L-1), there are several fuzzy systems Its output is recorded as And as the input of the l+1th layer. In the Lth layer, there is only one fuzzy system WMFS L , which combines the output of the L-1th layer into the final output x L .

[0027] definition Fuzzy Set , fuzzy system The rules are as follows:

[0028] x and y represent input and output respectively, is a fuzzy set defined on the output variable. This set represents the output of the fuzzy set based on the input variable.

[0029] Fuzzy Sets It can be expressed by a fuzzy membership function. In this embodiment, a triangular membership function is used. The output is:

[0030] parameter Fuzzy Set Center, The key to fuzzy system is to determine the parameters , Representations are membership functions of fuzzy sets, which define the degree to which the input variables correspond to the fuzzy set.

[0031] S2, combining the reconstructed WMFS model with the UKF model to obtain a dual-station multi-target tracking model; In this embodiment, the reconstructed WMFS model is combined with the unscented Kalman filter for dual-station azimuth and Doppler target tracking. Specifically: Step 1: Determine the signal state according to the state equation, measurement equation, dual-station measurement equation and observation equation of the unscented Kalman filter.

[0032] Define a nonlinear system whose state equation and measurement equation are:

[0033]

[0034] in, represents the process noise that affects the system state, represents the measurement noise affecting the observations.

[0035] The double-station measurement equation is:

[0036] The observation equation is as follows:

[0037]

[0038] Among them, the state of observation station i=1,2 at time K is .

[0039] Step 2: Generate sigma points according to the signal state, and transform the sigma point set using the state function; from The state and covariance at each moment generate 2n+1 sigma points:

[0040] in, Represents a scaling parameter used to control the diffusion degree of Sigma points. represents the covariance matrix.

[0041] Calculate the weight of each 2n+1 sigma point:

[0042] Transform the sigma point set using the state function:

[0043] Step 3: Calculate the prior estimates of the state and covariance of the transformed sigma point, and estimate the sigma point from the previous state and covariance; The prior estimates of the states and covariances are:

[0044]

[0045] Generate 2n+1 sigma points from the previous state and covariance estimate:

[0046] Step 4: Transform the sigma point by measuring the function and calculate the prior estimate and covariance; Transform the sigma point set by the measurement function:

[0047] The a priori estimates and covariances of the measurements are:

[0048]

[0049]

[0050]

[0051]

[0052]

[0053] In the embodiment of the present invention, the reconstructed WMFS model is combined with the UKF model to obtain a dual-station multi-target tracking model, and its process is as follows: Figure 3As shown, specifically including: Step 1: Determine the input and output data pairs of the Wang-Mendel fuzzy model; Given N input / output data pairs:

[0054] Input reorganization: Use sliding window method to divide the input vector into T-2 segments. Window size setting is 3, and the moving step is set to 1.

[0055] enter:

[0056] For design The input / output data pairs are:

[0057] for For each input, consider Fuzzy Sets , select the endpoints as:

[0058] Step 2: determining the weight and weight output parameter of the fuzzy set according to the input and output data pairs; For each input / output data pair starting from k, determine the fuzzy set Obtained at Fuzzy Sets The maximum membership value in

[0059] Update weights and weight output parameters

[0060]

[0061] Step 3: Calculate the nonlinear state function of the UKF model according to the weights of the fuzzy sets and the weight output parameters.

[0062] Determining fuzzy systems Parameters in

[0063]

[0064] say Cell , is overwritten by the data, and defines:

[0065] Defining neighborhoods For cells whose distance does not exceed 1.

[0066]

[0067] Calculate the average C(k)

[0068] Calculate the new support set , and the calculated arrive

[0069]

[0070] in As a new input / output data pair, a level 1 fuzzy system is designed

[0071]

[0072] The nonlinear state function is defined as:

[0073] The technical solution of the present invention improves the performance of a dual-station multi-target tracking system in processing nonlinear and uncertain problems by combining WMFS and UKF.

[0074] S3. Tracking the target using the dual-station multi-target tracking model.

[0075] The non-gain Kalman filter (UKF) algorithm based on the Wang-Mendel fuzzy system (WMFS) proposed in this invention is used for the Doppler dual-station multi-target tracking algorithm. The algorithm makes full use of the advantages of WMFS in dealing with system uncertainty and complex modeling. In the derivation process of WMFS-UKF, the untraceable transformation (UT) framework is adopted to solve the nonlinear measurement problem, and the pre-trained WMFS is used to reconstruct the state transition function. By utilizing the historical state of the target and the expected output, the WM fuzzy inference system achieves more accurate state prediction and precise covariance estimation, which greatly improves the performance and stability of target tracking. The simulation experiment verifies the effectiveness and robustness of the algorithm in the target tracking scenario.

[0076] The present invention provides a dual-station multi-target tracking device based on WMFS and UKF, comprising: an acquisition module, used for building a WMFS model, and reconstructing a state transfer function by using a pre-trained WMFS model; a processing module, used for combining an optimized WMFS model with a UKF model to obtain a dual-station multi-target tracking model; and an execution module, used for tracking a target by using the dual-station multi-target tracking model.

[0077] In some embodiments, the construction of the WMFS model includes: a first acquisition module, which is used to train the WMFS model using historical data and determine the fuzzy rules and parameters of the trained WMFS model, wherein the WMFS model is a multi-layer WMFS, each layer of WMFS uses the output results of the upper layer as the input data of this layer, and the last layer of WMFS integrates the results.

[0078] In some embodiments, the acquisition module includes: a second acquisition module for using WMFS to generate state estimation and uncertainty information through internally defined fuzzy rules, and fusing the state estimation and the uncertainty information into the framework of UKF to predict and update the target state; The first processing module is used to estimate the target state and covariance through propagation and transformation of Sigma points.

[0079] In some embodiments, the first processing submodule includes: a second processing submodule for determining the signal state according to the state equation, measurement equation, two-station measurement equation and observation equation of the unscented Kalman filter; a third processing submodule for generating sigma points according to the signal state and transforming the sigma point set with the state function; a fourth processing submodule for calculating the a priori estimate of the state and covariance of the transformed sigma point and estimating the sigma point from the previous state and covariance; and a first processing submodule for converting the sigma point through the measurement function and calculating the a priori estimate and covariance.

[0080] In some embodiments, the processing module includes: a second processing submodule, used to determine the input-output data pairs of the Wang-Mendel fuzzy model; a third processing submodule, used to determine the weights and weight output parameters of the fuzzy set based on the input-output data pairs; and a fourth processing submodule, used to calculate the nonlinear state function of the UKF model based on the weights and weight output parameters of the fuzzy set.

[0081] The non-gain Kalman filter (UKF) algorithm based on the Wang-Mendel fuzzy system (WMFS) proposed in this invention is used for the Doppler dual-station multi-target tracking algorithm. The algorithm makes full use of the advantages of WMFS in dealing with system uncertainty and complex modeling. In the derivation process of WMFS-UKF, the non-gain transformation (UT) framework is adopted to solve the nonlinear measurement problem, and the pre-trained WMFS is used to reconstruct the state transition function. By utilizing the historical state of the target and the expected output, the WM fuzzy inference system achieves more accurate state prediction and precise covariance estimation, which greatly improves the performance and stability of target tracking. The simulation experiment verifies the effectiveness and robustness of the algorithm in the target tracking scenario.

[0082] To solve the above technical problems, an embodiment of the present invention also provides a computer device. The computer device includes a processor, a non-volatile storage medium, a memory and a network interface connected through a system bus. Among them, the non-volatile storage medium of the computer device stores an operating system, a database and a computer-readable instruction, and the database may store a control information sequence. When the computer-readable instruction is executed by the processor, the processor can implement a dual-station multi-target tracking method based on WMFS and UKF. The processor of the computer device is used to provide computing and control capabilities to support the operation of the entire computer device. The memory of the computer device may store computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor may execute a dual-station multi-target tracking method based on WMFS and UKF. The network interface of the computer device is used to connect and communicate with the terminal. In this embodiment, the processor is used to execute the specific contents of the acquisition module and the processing module, and the memory stores the program code and various types of data required to execute the above modules. The network interface is used to transmit data between the user terminal or the server. The memory in this embodiment stores the program code and data required to execute all sub-modules in the image processing method, and the server can call the program code and data of the server to execute the functions of all sub-modules.

[0083] The present invention also provides a storage medium storing computer-readable instructions. When the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of a dual-station multi-target tracking method based on WMFS and UKF described in any of the above embodiments.

[0084] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0085] It should be understood that, although the steps in the flowchart of the accompanying drawings are displayed in sequence as indicated by the arrows, these steps are not necessarily executed in sequence in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least a part of the steps in the flowchart of the accompanying drawings may include multiple sub-steps or multiple stages, and these sub-steps or stages are not necessarily executed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be executed in turn or alternately with other steps or at least a part of the sub-steps or stages of other steps.

[0086] The above descriptions are only some embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

[0087] The above content is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application.

Claims

1. A dual-station multi-target tracking method based on WMFS and UKF, characterized in that: include: Build a WMFS model and use the pre-trained WMFS model to reconstruct the state transfer function; The optimized WMFS model is combined with the UKF model to obtain a dual-station multi-target tracking model; The dual-station multi-target tracking model is used to track the target.

2. The method according to claim 1, characterized in that The WMFS model is constructed, including: The WMFS model is trained using historical data, and the fuzzy rules and parameters of the trained WMFS model are determined. The WMFS model is a multi-layer WMFS, each layer of WMFS uses the output results of the upper layer as the input data of this layer, and the last layer of WMFS integrates the results.

3. The method according to claim 1, characterized in that The reconstructing the state transfer function by using the pre-trained WMFS model includes: Using WMFS to generate state estimation and uncertainty information through internally defined fuzzy rules, and integrating the state estimation and uncertainty information into the framework of UKF to predict and update the target state; Through the propagation and transformation of Sigma points, the target state and covariance are estimated.

4. The method according to claim 3, characterized in that The target state and covariance are estimated through the propagation and transformation of Sigma points, including: The signal state is determined according to the state equation, measurement equation, dual-station measurement equation and observation equation of the unscented Kalman filter. Generate sigma points according to the signal state and transform the sigma point set using the state function; Compute a priori estimates of the states and covariances of the transformed sigma points, and estimate the sigma points from the prior states and covariances; Transform the sigma points by the measurement function and compute the prior estimates and covariances.

5. The method according to claim 4, characterized in that The dual-station multi-target tracking model is obtained by combining the reconstructed WMFS model with the UKF model, including: Determine the input and output data pairs of the Wang-Mendel fuzzy model; Determine the weights and weighted output parameters of the fuzzy set according to the input-output data pairs; The nonlinear state function of the UKF model is calculated according to the weights of the fuzzy sets and the weight output parameters.

6. A dual-station multi-target tracking device based on WMFS and UKF, characterized in that: include: An acquisition module is used to build a WMFS model and reconstruct the state transfer function using the pre-trained WMFS model; A processing module is used to combine the optimized WMFS model with the UKF model to obtain a dual-station multi-target tracking model; The execution module is used to track the target using the dual-station multi-target tracking model.

7. The device according to claim 6, characterized in that The WMFS model is constructed, including: The first acquisition module is used to train the WMFS model using historical data and determine the fuzzy rules and parameters of the trained WMFS model, wherein the WMFS model is a multi-layer WMFS, each layer of WMFS uses the output results of the upper layer as the input data of this layer, and the last layer of WMFS integrates the results.

8. The device according to claim 6, characterized in that The acquisition module comprises: The second acquisition module is used to generate state estimation and uncertainty information by using WMFS through internally defined fuzzy rules, and integrate the state estimation and uncertainty information into the framework of UKF to predict and update the target state; The first processing module is used to estimate the target state and covariance through propagation and transformation of Sigma points.

9. A computer device comprising a memory and a processor, wherein the memory stores computer-readable instructions, and when the computer-readable instructions are executed by the processor, the processor executes the steps of the dual-station multi-target tracking method based on WMFS and UKF as described in any one of claims 1 to 5.

10. A storage medium storing computer-readable instructions, wherein when the computer-readable instructions are executed by one or more processors, the one or more processors execute the steps of the dual-station multi-target tracking method based on WMFS and UKF as described in any one of claims 1 to 5.

Images