Robust distributed target tracking method for radar networking, storage medium and equipment

Abstract

The invention discloses a radar networking-oriented robust distributed target tracking method, a storage medium and equipment, and belongs to the technical field of target tracking of multi-radar networking. The objective of the invention is to solve the problem of poor tracking effect caused by non-Gaussian noise in a target tracking scene for multi-radar networking. The method comprises the following steps: firstly, accurately modeling a state equation, a radar measurement equation and non-Gaussian flicker noise of a target; a maximum correlation entropy CKF method with variance compensation is determined based on the maximum correlation entropy and a CKF algorithm, then the state of a target is estimated, and an estimated value # imgabs0 # of the state and a corresponding estimated error covariance matrix # imgabs1 # are obtained based on estimations # imgabs2 # and # imgabs3 # obtained by all radar nodes to perform diffusion fusion with neighbor nodes by using a CI strategy.

Description

Technical Field

[0001] The present invention belongs to the technical field of target tracking in multi-radar networking, and relates to a target tracking method, storage medium and equipment for radar networking. Background Art

[0002] With the updating and iteration of science and technology and manufacturing technology, radars have gradually become diversified, intelligent and networked. Single radar systems can no longer meet the actual needs of target tracking, and multi-radar networking target tracking methods have become the core technology. Multi-radar systems make full use of the measurement information of each radar to form information resource advantages and greatly improve radar tracking capabilities. The radar network consists of multiple radars with topological structures. Compared with centralized processing methods, distributed structures have more advantages in fault tolerance, flexibility and reliability, so they become the first choice.

[0003] For actual target tracking scenarios, the state equation and measurement equation are both nonlinear, so the essence is a nonlinear state estimation problem. Designing a reasonable tracking method is the key to ensuring tracking performance. Usually, nonlinear estimation methods include extended Kalman filter (EKF), unscented Kalman filter (UKF) and cubature Kalman filter (CKF). Compared with other algorithms, CKF does not need to calculate the Jacobian matrix, and can obtain higher accuracy with fewer sampling points, and it has more advantages in stability and computational complexity. In addition, due to interference such as reflection of the target position and changes in the environment, the radar will have obvious deviations in detecting the target, and the noise will no longer satisfy the Gaussian distribution, and non-Gaussian flicker noise will appear. The above nonlinear estimation method is obtained under the Gaussian assumption. If used directly, it will cause a sharp drop in accuracy or even divergence. Therefore, it is necessary to design a more reasonable nonlinear non-Gaussian estimation method to accurately track the target. Summary of the invention

[0004] The present invention aims to solve the problem of poor tracking effect caused by non-Gaussian noise in a target tracking scenario of multi-radar networking.

[0005] A robust distributed target tracking method for radar networking includes the following steps:

[0006] Step 1: Construct the target state equation, radar tracking measurement equation and non-Gaussian flicker noise model;

[0007] The state equation of the target:

[0008] x k =s(x k-1 )+ω k-1

[0009] Among them, x k represents the state quantity at time k, s(x k-1) represents the nonlinear transfer equation of the target state vector over time, ω k-1 Gaussian white noise with zero mean, the noise variance matrix is ​​Q k-1 ;

[0010] The measurement equation for tracking a target using radar is:

[0011]

[0012] Where N is the number of radars, Indicates the measured value. is non-Gaussian flicker noise; g(x k ) is the measurement vector;

[0013] Flicker noise model, i.e. non-Gaussian flicker noise The flicker noise model is as follows:

[0014]

[0015] Among them, α (i) is a random variable between 0 and 1, used to represent the probability of flickering; and They are two types of zero mean and noise variance respectively. Gaussian noise, generally With ω k-1 Uncorrelated, take a nominal noise variance Approximate flicker noise Usually advisable

[0016] Step 2: Based on the state equation, measurement equation and non-Gaussian flicker noise model, the radar estimates the state of the target and obtains x k Estimated value of and the corresponding estimation error covariance matrix

[0017] Step 3: Estimates based on all radar nodes and Use CI strategy to perform diffusion fusion with neighboring nodes.

[0018] Furthermore, when the tracking system is two-dimensional, the measurement vector includes the slant range and azimuth of the radar station from the target, that is, When the tracking system is three-dimensional, the measurement vector includes the slant range, elevation angle and azimuth of the radar station from the target, that is, in are the position coordinates in the x, y and z directions at time k respectively.

[0019] Furthermore, based on the state equation, the measurement equation and the non-Gaussian flicker noise model, the radar estimates the state of the target including the process of time update and measurement update;

[0020] For radar i, the time update process uses CKF processing to obtain the state prediction estimate at time k. And the prediction estimation error covariance matrix at time k The measurement update process includes the following steps:

[0021]

[0022] in, Indicated by The matrix obtained by Cholesky decomposition;

[0023]

[0024]

[0025] in, represents the volume point obtained by the prediction estimate and the prediction estimate error covariance matrix, Represents the new volume point obtained through the measurement equation transfer;

[0026]

[0027] in, represents the predicted value of the measurement; represents the prediction estimation error covariance matrix of the measurement, represents the measurement nominal noise variance;

[0028] Get the state estimate at time k according to the cost function Based on Get the corresponding estimation error covariance matrix

[0029] The cost function is:

[0030]

[0031] in, represents the state estimation at time k; δ σ represents the Gaussian kernel function; x k Represents the actual value of the state quantity at time k; Indicates the actual value of the measurement at time k; operator symbol A stands for or

[0032] Furthermore, the CKF processing method is used to obtain the state prediction estimate at time k And the prediction estimation error covariance matrix at time k The process includes the following steps:

[0033]

[0034] in, Represents the estimated error covariance matrix at time k-1; Is The matrix obtained by Cholesky decomposition;

[0035]

[0036] Where n is the dimension of the state, represents the volume point obtained by the state estimation and estimation error covariance matrix, and the subscript μ represents the μth column element of the corresponding matrix. represents the state estimate of the target at time k-1; [1] μ represents the μth column of [1];

[0037]

[0038] in, represents the new volume point obtained through the nonlinear function transfer of the system state, and s(·) represents the nonlinear transfer function of the state vector;

[0039]

[0040] Where, α = 1 / m; represents the state prediction estimate at time k; represents the prediction estimation error covariance matrix at time k, Q k-1 represents the noise variance of the system.

[0041] Furthermore, the state estimation at time k is obtained according to the cost function In the process of After correction, (20) is calculated to obtain

[0042] Furthermore, the state estimation at time k is obtained according to the cost function In the process of After correction, (20) is calculated to obtain The process includes the following steps:

[0043] The new information at time k is

[0044]

[0045] in, represents the predicted value of the measurement;

[0046] Under the Gaussian noise assumption, the theoretical value of the expected innovation covariance is expressed as

[0047]

[0048] According to the principle of orthogonality of new information, (22) can be generalized to

[0049]

[0050] in

[0051]

[0052] Introducing the attenuation matrix To compensate get

[0053]

[0054] Taking the diagonal elements on both sides equal, we get

[0055]

[0056] Where n is the dimension of the state, is the variance of the actual new information, d (l) (·) means taking the element in the lth row and lth column of the matrix;

[0057] Then get

[0058]

[0059] When k = 1, Calculated as

[0060]

[0061] When k>1, Calculated as

[0062]

[0063] Among them, λ is the forgetting factor;

[0064] The variance after compensation is

[0065]

[0066] Therefore, after compensation in, represents the prediction estimation error covariance matrix of the compensated measurement;

[0067] Cross-covariance for

[0068]

[0069] By statistical linearization, the observation equation can be rewritten as

[0070]

[0071] in

[0072]

[0073] in, represents the pseudo-measurement matrix obtained by statistical linearization, represents the measurement noise obtained by statistical linearization, represents the measurement noise variance obtained by statistical linearization;

[0074] Substituting (34) into (20) yields

[0075]

[0076] Taking the derivative of (38) and setting it equal to 0, we get

[0077]

[0078] in

[0079]

[0080] Then get

[0081]

[0082] Using the prediction information once get Formula (42) can be rewritten as

[0083]

[0084] in

[0085]

[0086] Further, based on Get the corresponding estimation error covariance matrix The process includes the following steps:

[0087] From (46) we can get

[0088]

[0089] in, is the estimated error, I n represents the identity matrix with diagonal 1, and n is the dimension of the system state;

[0090] Get the state estimation error covariance matrix

[0091]

[0092] Furthermore, based on the estimates obtained from all radar nodes and The process of using CI strategy to diffuse and merge with neighboring nodes includes:

[0093]

[0094]

[0095] Wherein, trace(·) represents the matrix trace operation.

[0096] A computer storage medium stores at least one instruction, and the at least one instruction is loaded and executed by a processor to implement a robust distributed target tracking method for radar networking.

[0097] A robust distributed target tracking device for radar networking, the device comprising a processor and a memory, the memory storing at least one instruction, the at least one instruction being loaded and executed by the processor to implement the robust distributed target tracking method for radar networking.

[0098] Beneficial effects of the present invention:

[0099] The present invention firstly accurately models the state equation, radar measurement equation and non-Gaussian flicker noise of the target. Secondly, a variance compensation method is designed, and a maximum correlation entropy CKF method with variance compensation is proposed based on the maximum correlation entropy and CKF algorithm. This method can effectively deal with the problem of non-Gaussian noise of radar, and its accuracy is better than other algorithms in the prior art. Finally, the method is extended to a multi-radar networking system, and a fast CI fusion algorithm is used for diffusion fusion, which avoids the calculation of mutual covariance and significantly improves the tracking accuracy and detection range. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] Figure 1 The flowchart of robust distributed target tracking for radar networking is shown below;

[0101] Figure 2 This is the network structure diagram of four radars;

[0102] Figure 3 RMSE curves corresponding to the three methods;

[0103] Figure 4 This is the average RMSE curve of the four radars. DETAILED DESCRIPTION

[0104] Specific implementation method 1: Combination Figure 1 To explain this embodiment,

[0105] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0106] This embodiment is a robust distributed target tracking method for radar networking, comprising the following steps:

[0107] Step 1: Construct the target state equation, radar tracking measurement equation and non-Gaussian flicker noise model;

[0108] The state equation of the target is:

[0109] x k =s(x k-1 )+ω k-1 (3)

[0110] Among them, x k represents the state quantity at time k, s(x k-1 ) represents the nonlinear transfer equation of the target state vector over time, ω k-1 Gaussian white noise with zero mean, the noise variance matrix is ​​Q k-1 , used to represent the deviation caused by modeling;

[0111] Measurement Modeling for Radar Tracking:

[0112] The communication topology between radar networks is represented by an undirected connected graph G = (V, E), where V = {1, 2, ..., N} represents the nodes corresponding to the radars. represents the set of all links. If radar i and radar j can communicate with each other, then radar i and radar j are neighboring nodes. The neighboring nodes of radar i can be represented as N i ={j∈V:(j,i)∈E}∪{i}.

[0113] When radar is used to track the target, the measurement equation is:

[0114]

[0115] Where N is the number of radars, Indicates the measured value. It is non-Gaussian flicker noise;

[0116] When the tracking system is two-dimensional, the measurements include the slant range and azimuth of the radar station from the target, that is, When the tracking system is three-dimensional, the measurements include the slant range, elevation angle and azimuth of the radar station from the target, that is, in are the position coordinates in the x, y and z directions at time k respectively.

[0117] Flicker noise model, i.e. non-Gaussian flicker noise The probability density function of

[0118]

[0119] Among them, α (i) is a random variable between 0 and 1, used to represent the probability of flickering; and They are two types of zero mean and noise variance respectively. Gaussian noise, generally With ω k-1 In practice, the parameters in expression (3) cannot be accurately known, so a nominal noise variance is taken To approximate flicker noise Usually advisable

[0120] Step 2: Based on the state equation, measurement equation and non-Gaussian flicker noise model (3), a robust state estimation method is designed to obtain the state estimation of all radars for the target.

[0121] The process of radar state estimation of the target includes time update and measurement update. Among them, the time update can generally be processed by CKF. In the measurement update, the attenuation matrix is ​​used to compensate for the measurement noise variance. Based on the compensated measurement noise variance and the maximum correlation entropy criterion, each radar obtains the state estimation of the target and its error covariance matrix.

[0122] Maximum correlation entropy criterion:

[0123] For two random variables X and Y, the joint probability density function of the two is F X,Y (x,y), then its related entropy is defined as

[0124] V(X,Y)=∫∫κ(x,y)dF X,Y (x,y) (6)

[0125] Among them, x and y are variables used to represent probability density; κ(x, y) represents a positive definite kernel function.

[0126] Commonly used Gaussian kernel functions have the following forms:

[0127]

[0128] Among them, e=xy is the difference between two variables x and y, and σ is the kernel width.

[0129] F X,Y (x,y) is usually not directly available, and the correlation entropy can be estimated using a limited number of sampling points.

[0130]

[0131] Where, e(i) = x(i) - y(i); N represents the number of sampling points, δ σ represents the Gaussian kernel function; It represents the estimation of correlation entropy, and the correlation entropy reaches its maximum value when and only when X=Y.

[0132] Therefore, the estimation based on maximum correlation entropy can be expressed as

[0133]

[0134] in, For x k The estimated value of J(x k ) is about x k The cost function of .

[0135] When the radar is affected by the environment, the noise no longer satisfies the Gaussian distribution, and flicker noise will appear. Traditional nonlinear estimation methods such as CKF cannot directly handle non-Gaussian problems. Therefore, the maximum correlation entropy criterion and variance compensation strategy will be used to solve the non-Gaussian problem. Therefore, for radar i,

[0136] Time update:

[0137]

[0138] in, Represents the estimated error covariance matrix at time k-1; Is The matrix obtained by Cholesky decomposition;

[0139]

[0140] Where n is the dimension of the state, represents the volume point obtained by the state estimation and estimation error covariance matrix, and the subscript μ represents the μth column element of the corresponding matrix. Represents the state estimate of the target at time k-1;

[0141]

[0142]

[0143] Among them, [1] μ represents the μth column of [1]; represents the new volume point obtained through the nonlinear function transfer of the system state, and s(·) represents the nonlinear transfer function of the state vector;

[0144]

[0145] Where, α = 1 / m; represents the state prediction estimate at time k;

[0146]

[0147] represents the prediction estimation error covariance matrix at time k, Q k-1 represents the noise variance of the system;

[0148] Measurement Update:

[0149]

[0150] in, Indicated by The matrix obtained by Cholesky decomposition;

[0151]

[0152] in, represents the volume point obtained by the prediction estimate and the prediction estimate error covariance matrix, represents the new volume point obtained through the measurement equation transfer, g(·) is the measurement equation which is the same as the expression of formula (2);

[0153]

[0154] in, Represents the predicted value of a measurement.

[0155]

[0156] in, represents the prediction estimation error covariance matrix of the measurement, represents the new volume point obtained by transferring the radar measurement equation, represents the measurement nominal noise variance;

[0157] Define the cost function

[0158]

[0159] in, represents the state estimation at time k; δ σ represents the Gaussian kernel function; x k Represents the actual value of the state quantity at time k; Indicates the actual value of the measurement at time k; operator symbol A stands for or

[0160] When the noise is non-Gaussian, Mainly affected by the noise variance The impact of It cannot directly reflect the characteristics of abnormal measurement. Performing the correction and then calculating (22) will inevitably further improve the accuracy and robustness of the algorithm and eliminate the impact of non-Gaussian noise on the estimation performance.

[0161] The innovation at time k is defined as

[0162]

[0163] in, represents the predicted value of the measurement;

[0164] Under the Gaussian noise assumption, the theoretical value of the expected innovation covariance is expressed as

[0165]

[0166] According to the principle of orthogonality of new information, (24) can be extended to

[0167]

[0168] in

[0169]

[0170] Formula (25) is obtained under the Gaussian assumption. When the noise variance is non-Gaussian, the obtained innovation variance is With theoretical value There is an obvious deviation, so the attenuation matrix is ​​introduced To compensate To reduce the interference of abnormal observations on the system, thereby improving the robustness of the estimation. So there is

[0171]

[0172] Taking the diagonal elements on both sides equal, we can get

[0173]

[0174] Where n is the dimension of the state, is the variance of the actual new information, d (l) (·) means taking the element at the lth row and lth column of the matrix.

[0175] so

[0176]

[0177] definition So there is

[0178]

[0179] When k = 1, Calculated as

[0180]

[0181] When k>1, Calculated as

[0182]

[0183] Among them, λ is the forgetting factor, which is usually 0.95.

[0184] The variance after compensation is

[0185]

[0186] Therefore, after compensation

[0187]

[0188] in, represents the prediction estimation error covariance matrix of the compensated measurement;

[0189] The cross-covariance of x and z for

[0190]

[0191] By statistical linearization, the observation equation can be rewritten as

[0192]

[0193] in

[0194]

[0195] in, represents the pseudo-measurement matrix obtained by statistical linearization, represents the measurement noise obtained by statistical linearization, represents the measurement noise variance obtained by statistical linearization;

[0196] Substituting (34) into (22) yields

[0197]

[0198] Taking the derivative of (40) and setting it equal to 0, we get

[0199]

[0200] in

[0201]

[0202] Then you can get

[0203]

[0204] Since (44) is a binary system with both sides being about x k The function is a fixed point equation, which can be solved by Newton iteration or the following method:

[0205] Using the prediction information once Satisfactory estimation accuracy can be achieved. Formula (44) can be rewritten as

[0206]

[0207] in

[0208]

[0209] From (46) we can get

[0210]

[0211] in, is the estimated error, I n represents the identity matrix with a diagonal of 1, and n is the dimension of the system state; therefore, the state estimation error covariance matrix is

[0212]

[0213] Step 3: Get local estimates based on all radar nodes from step 2 and Use CI strategy to diffuse and merge with neighboring nodes

[0214]

[0215]

[0216] Wherein, trace(·) represents the matrix trace operation.

[0217] Therefore, the robust distributed target tracking method for radar networking proposed in the present invention can be summarized as method 1.

[0218]

[0219] Example:

[0220] In order to verify the effectiveness of the present invention, a scenario of two-dimensional radar tracking a turning target is selected, and its state equation and measurement equation are respectively

[0221]

[0222] in x k , y k , Ω k They represent the x-direction position, velocity, y-direction position, velocity and turning rate at time k respectively. A distributed network consisting of 4 sensors is used for tracking. are the locations of the sensors, which are (1000m, 300m), (1500m, 300m), (1000m, -300m), (1500m, -300m), and their topological structures are as follows: Figure 2 shown.

[0223] Simulation time T = 0.1s, initial state x0 = [1000, 20, 1000, 0, -π / 60] T , P0=diag([1000,100,1000,100,2 / 100]), q1=0.2, q2=1.75×10- 5 , Process noise ω k The variance Q k for

[0224] Q k =diag([q1M; q1M; q2T]) (54)

[0225] Measuring noise The nominal noise is Measuring noise is non-Gaussian flicker noise, that is

[0226]

[0227] In order to verify the effectiveness of the invention, the root mean square error (RMSE) of the position is selected as the performance indicator, which is defined as

[0228]

[0229] Among them, (x k ,y k )and are the true position and estimated position of the i-th node and the j-th experiment, M is the number of Monte Carlo experiments, is the simulation step size.

[0230] Taking radar 1 as the comparison object, CKF and MCKF methods are compared with the present invention. The RMSE curves of the three methods are as follows: Figure 3 As shown. Figure 3 It can be seen that the RMSE of the present invention is lower and the tracking accuracy is higher. This is because the method introduces a variance compensation strategy and a distributed fusion strategy, thereby improving the tracking accuracy. Figure 4 is the average RMSE value of the four radars, and it can be seen that all radars have achieved consistent tracking accuracy. Therefore, the present invention is more suitable for the scenario of multi-radar networking to track targets and has greater application value. Specific implementation method 2:

[0232] This embodiment is a computer storage medium, in which at least one instruction is stored. The at least one instruction is loaded and executed by a processor to implement the robust distributed target tracking method for radar networking.

[0233] It should be understood that the instructions include computer program products, software or computerized methods corresponding to any method described in the present invention; the instructions can be used to program a computer system, or other electronic device. Computer storage media may include readable media on which instructions are stored, which may include but are not limited to magnetic storage media, optical storage media; magneto-optical storage media include read-only memory ROM, random access memory RAM, erasable programmable memory (e.g., EPROM and EEPROM) and flash memory layers, or other types of media suitable for storing electronic instructions. Specific implementation method three:

[0235] This embodiment is a robust distributed target tracking device for radar networking, the device includes a processor and a memory. It should be understood that the device includes any device including a processor and a memory described in the present invention, and the device may also include other units and modules that perform display, interaction, processing, control, and other functions through signals or instructions;

[0236] At least one instruction is stored in the memory, and the at least one instruction is loaded and executed by the processor to implement the robust distributed target tracking method for radar networking.

[0237] Those skilled in the art will appreciate that at least one instruction stored is a computer program product corresponding to the method or system. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application may adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes. The scheme in the embodiment of the present application can be implemented in various computer languages, for example, object-oriented programming language Java and literal scripting language JavaScript, etc.

[0238] The present application is described with reference to the flowcharts and / or block diagrams of the methods, systems, and computer program products according to the embodiments of the present application, and may also be used for corresponding devices. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0239] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0240] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0241] Although the preferred embodiments of the present application have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0242] Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalents, the present application is also intended to include these modifications and variations.

[0243] The above calculation examples of the present invention are only used to explain the calculation model and calculation process of the present invention in detail, and are not intended to limit the implementation methods of the present invention. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present invention are still within the scope of protection of the present invention.

Claims

1. A robust distributed target tracking method for radar networking, characterized in that: The following steps are involved: Step 1: Construct the target state equation, radar tracking measurement equation and non-Gaussian flicker noise model; The state equation of the target is: x k =s(x k-1 )+ω k-1 Among them, x k represents the state quantity at time k, s(x k-1 ) represents the nonlinear transfer equation of the target state vector over time, ω k-1 Gaussian white noise with zero mean, the noise variance matrix is ​​Q k-1 ; The measurement equation for tracking a target using radar is: Where N is the number of radars, Indicates the measured value. is non-Gaussian flicker noise; g(x k ) is the measurement vector; Flicker noise model, i.e. non-Gaussian flicker noise The flicker noise model is as follows: Among them, α (i) is a random variable between 0 and 1, used to represent the probability of flickering; and They are two types of zero mean and noise variance respectively. Gaussian noise, generally With ω k-1 Uncorrelated, take a nominal noise variance Approximate flicker noise Usually advisable Step 2: Based on the state equation, measurement equation and non-Gaussian flicker noise model, the radar estimates the state of the target and obtains x k Estimated value of and the corresponding estimation error covariance matrix Step 3: Estimates based on all radar nodes and Use CI strategy to perform diffusion fusion with neighboring nodes.

2. The robust distributed target tracking method for radar networking according to claim 1, characterized in that: When the tracking system is two-dimensional, the measurement vector includes the slant range and azimuth of the radar station from the target, that is, When the tracking system is three-dimensional, the measurement vector includes the slant range, elevation angle and azimuth of the radar station from the target, that is, in are the position coordinates in the x, y and z directions at time k respectively.

3. The robust distributed target tracking method for radar networking according to claim 1, characterized in that: Based on the state equation, measurement equation and non-Gaussian flicker noise model, the radar estimates the state of the target, including the time update and measurement update process; For radar i, the time update process uses CKF processing to obtain the state prediction estimate at time k. And the prediction estimation error covariance matrix at time k The measurement update process includes the following steps: in, Indicated by The decomposed matrix; in, represents the volume point obtained by the prediction estimate and the prediction estimate error covariance matrix, Represents the new volume point obtained through the measurement equation transfer; in, represents the predicted value of the measurement; represents the prediction estimation error covariance matrix of the measurement, represents the measurement nominal noise variance; Get the state estimate at time k according to the cost function Based on Get the corresponding estimation error covariance matrix The cost function is: in, represents the state estimation at time k; δ σ represents the Gaussian kernel function; x k Represents the actual value of the state quantity at time k; Indicates the actual value of the measurement at time k; operator symbol A stands for or 4. The robust distributed target tracking method for radar networking according to claim 3 is characterized in that: The CKF processing method is used to obtain the state prediction estimate at time k And the prediction estimation error covariance matrix at time k The process includes the following steps: in, Represents the estimated error covariance matrix at time k-1; Is The decomposed matrix; Where n is the dimension of the state, represents the volume point obtained by the state estimation and estimation error covariance matrix, and the subscript μ represents the μth column element of the corresponding matrix. represents the state estimate of the target at time k-1; [1] μ represents the μth column of [1]; in, represents the new volume point obtained through the nonlinear function transfer of the system state, and s(·) represents the nonlinear transfer function of the state vector; Where, α = 1 / m; represents the state prediction estimate at time k; represents the prediction estimation error covariance matrix at time k, Q k-1 represents the noise variance of the system.

5. A robust distributed target tracking method for radar networking according to claim 3 or 4, characterized in that: Get the state estimate at time k according to the cost function In the process of After correction, (20) is calculated to obtain 6. A robust distributed target tracking method for radar networking according to claim 5, characterized in that: Get the state estimate at time k according to the cost function In the process of After correction, (20) is calculated to obtain The process includes the following steps: The new information at time k is in, represents the predicted value of the measurement; Under the Gaussian noise assumption, the theoretical value of the expected innovation covariance is expressed as According to the principle of orthogonality of new information, (22) can be generalized to in Introducing the attenuation matrix To compensate get Taking the diagonal elements on both sides equal, we get Where n is the dimension of the state, is the variance of the actual new information, d (l) (·) means taking the element in the lth row and lth column of the matrix; Then get When k = 1, Calculated as When k>1, Calculated as Among them, λ is the forgetting factor; The variance after compensation is Therefore, after compensation in, represents the prediction estimation error covariance matrix of the compensated measurement; Cross-covariance for By statistical linearization, the observation equation can be rewritten as in in, represents the pseudo-measurement matrix obtained by statistical linearization, represents the measurement noise obtained by statistical linearization, represents the measurement noise variance obtained by statistical linearization; Substituting (34) into (20) yields Taking the derivative of (38) and setting it equal to 0, we get in Then get Using the prediction information once get Formula (42) can be rewritten as in 7. The robust distributed target tracking method for radar networking according to claim 6, characterized in that: based on Get the corresponding estimation error covariance matrix The process includes the following steps: From (1), we can get in, is the estimated error, I n represents the identity matrix with diagonal 1, and n is the dimension of the system state; Get the state estimation error covariance matrix 8. The robust distributed target tracking method for radar networking according to claim 7, characterized in that: Based on the estimates obtained from all radar nodes and The process of using CI strategy to diffuse and merge with neighboring nodes includes: Wherein, trace(·) represents the matrix trace operation.

9. A computer storage medium, characterized in that The storage medium stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the robust distributed target tracking method for radar networking as described in any one of claims 1 to 8.

10. A robust distributed target tracking device for radar networking, characterized in that: The device includes a processor and a memory, wherein the memory stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement a robust distributed target tracking method for radar networking as described in any one of claims 1 to 8.

Images