A Kalman tracking method, device, equipment and medium for radio direction finding

Abstract

The application belongs to the technical field of radio detection and sensing, and relates to a Kalman tracking method, device, equipment and medium for radio direction finding. The method comprises the following steps: establishing a main tracking chain and a backup tracking chain for radio direction finding; obtaining current observation data; performing model prediction on the state of each mode of the main tracking chain and the state of each mode of the backup tracking chain; calculating the likelihood probability of the current observation data belonging to each mode of the main tracking chain and the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain, and calculating the posterior probability of the current observation data belonging to each mode of the main tracking chain and each mode of the backup tracking chain; updating the state of each mode of the main tracking chain and the state of each mode of the backup tracking chain and traversing, and taking the mean value corresponding to the mode with the maximum heat as the tracking information of the current observation and outputting. The method can effectively resist outliers and adapt to the maneuverability of unmanned aerial vehicles.

Description

Technical Field

[0001] This application relates to the field of radio detection and sensing technology, and in particular to a Kalman tracking method, apparatus, device and medium for radio direction finding. Background Technology

[0002] Passive detection, identification, and localization of drones are primarily achieved by passively receiving and processing the electromagnetic signals emitted by the drones.

[0003] In existing technologies, the main direction finding methods for radio direction finding include amplitude comparison, AOA (Angle of Arrival), and TDOA (Time Difference of Arrival), which calculate the direction of arrival of the radiation source by utilizing the amplitude difference, phase difference, and time difference of radio signals arriving at the antenna array.

[0004] However, the aforementioned UAV radio direction finding methods may produce outliers due to noise, association errors, and direction finding ambiguity, and these outliers may occur continuously. In addition, UAVs are highly maneuverable, and their angle changes may vary significantly over a period of time, which the aforementioned methods cannot effectively address. Summary of the Invention

[0005] Therefore, it is necessary to provide a Kalman tracking method, apparatus, device, and medium for radio direction finding to address the aforementioned technical problems. This method can effectively combat outliers in UAV radio direction finding and can adapt to the maneuverability of UAVs while accurately eliminating outliers.

[0006] A Kalman tracing method based on radio direction finding includes:

[0007] Two tracking chains for radio direction finding are established and initialized separately to obtain the main tracking chain and the backup tracking chain. The main tracking chain corresponds to the main mode set, and the backup tracking chain corresponds to the backup mode set. Both the main mode set and the backup mode set include multiple modes, and the state of each mode includes heat and mean.

[0008] Obtain the data from the current observation;

[0009] When it is determined that the current observation is not the first observation, model prediction is performed on the modal states of the main tracking chain and the backup tracking chain to obtain the predicted states of each mode of the main tracking chain and the predicted states of each mode of the backup tracking chain.

[0010] Based on the predicted states of each modality in the main tracking chain and each modality in the backup tracking chain, calculate the likelihood probability of the current observed data belonging to each modality in the main tracking chain, the likelihood probability of the current observed data belonging to each modality in the backup tracking chain, the prior probability of each modality in the main tracking chain, and the prior probability of each modality in the backup tracking chain. Also calculate the posterior probability of the current observed data belonging to each modality in the main tracking chain and the posterior probability of the current observed data belonging to each modality in the backup tracking chain.

[0011] Based on the posterior probability of the current observation data belonging to each mode of the main tracking chain, update the state of each mode of the main tracking chain; based on the posterior probability of the current observation data belonging to each mode of the backup tracking chain, update the state of each mode of the backup tracking chain.

[0012] Iterate through the modal states of the updated main tracking chain and the backup tracking chain, and output the mean value corresponding to the mode with the highest popularity as the current observation tracking information.

[0013] In one embodiment, after updating the state of each modality of the main tracking chain and updating the state of each modality of the backup tracking chain, the method further includes:

[0014] Update the divergence between the main tracking chain and the backup tracking chain based on the updated modal states of the main tracking chain and the updated modal states of the backup tracking chain.

[0015] When the divergence meets the preset first condition, the updated backup tracking chain is copied to the updated main tracking chain to obtain the new main tracking chain; when the divergence does not meet the preset first condition, the updated main tracking chain is used as the new main tracking chain.

[0016] Acquire the data for the next observation, and based on the new main tracking chain and the updated backup tracking chain, perform model predictions and output the tracking information for the next observation.

[0017] In one embodiment, it also includes:

[0018] When it is determined that the current observation is the first observation, a new mode is established and initialized;

[0019] Based on the newly initialized modality, update the main modality set of the main tracking chain and the backup modality set of the backup tracking chain;

[0020] Acquire the data for the next observation, and based on the updated master mode set and the updated backup mode set, perform model prediction and output the tracking information for the next observation.

[0021] In one embodiment, it also includes:

[0022] Based on the likelihood probability of the current observation data belonging to each mode of the main tracking chain, the likelihood probability of the current observation data belonging to each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain, calculate the posterior probability of the current observation data belonging to the new mode.

[0023] When the posterior probability of the current observed data belonging to the new mode satisfies the preset second condition, the new mode is established and initialized;

[0024] Update the main mode set of the main tracking chain and the backup mode set of the backup tracking chain according to the new mode after initialization;

[0025] Acquire the data for the next observation, and based on the updated master mode set and the updated backup mode set, perform model prediction and output the tracking information for the next observation.

[0026] In one embodiment, when it is determined that the current observation is not the first observation, model prediction is performed on the modal states of the main tracking chain and the modal states of the backup tracking chain to obtain the predicted states of each modality of the main tracking chain and the predicted states of each modality of the backup tracking chain, including:

[0027] The modal states of the main tracing chain and the backup tracing chain are represented as follows:

[0028] {n k ,μ k ,Σ k}

[0029]

[0030] In the formula, n k To track the heat of the k-th mode of the chain, μ k To track the mean of the k-th mode of the chain, Σ k To track the variance of the k-th mode of the tracking chain, θ represents the orientation of the tracking chain. To track the azimuth angular velocity of the chain;

[0031] The state transition equation is:

[0032]

[0033]

[0034]

[0035] In the formula, X i Here is the state update equation, where i is the i-th observation data, and η is the η-th observation data. i The state transition uncertainty follows a Gaussian distribution with mean 0 and variance Q. The variance of the positional shift is unknown. The variance of the transfer uncertainty of the azimuth angular velocity;

[0036] Model prediction is performed based on the modal states and state transition equations of the tracking chain. This process iterates through all modes of the tracking chain. If n k =0, skip the current mode; otherwise, the predicted states of each mode in the tracking chain are:

[0037]

[0038] In the formula, To track the predicted popularity of the k-th mode of the chain, To track the predicted mean of the k-th mode of the chain, γ is the forgetting factor, representing the variance of the k-th mode in the tracking chain.

[0039] In one embodiment, based on the predicted states of each modality in the main tracking chain and the predicted states of each modality in the backup tracking chain, the likelihood probability of the current observation data belonging to each modality in the main tracking chain, the likelihood probability of the current observation data belonging to each modality in the backup tracking chain, the prior probability of each modality in the main tracking chain, and the prior probability of each modality in the backup tracking chain are calculated, including:

[0040] Based on the predicted states of each mode in the main tracking chain and each mode in the backup tracking chain, calculate the likelihood probability of the current observation data belonging to each mode in the main tracking chain and the likelihood probability of the current observation data belonging to each mode in the backup tracking chain:

[0041] like but,

[0042] P(Y i |k)=0

[0043] Y i =θ i +ε i =BX i +ε i

[0044] B = [1 0]

[0045] otherwise,

[0046]

[0047]

[0048] In the formula, P(Y) i |k) represents the likelihood probability that the current observed data belongs to each mode of the tracking chain, Y i For the measurement equation, ε i The noise is the azimuth observation noise, and N is the Gaussian probability distribution function. To observe the residual covariance;

[0049] Based on the predicted states of each modality in the main tracking chain and each modality in the backup tracking chain, calculate the prior probabilities of each modality in the main tracking chain and each modality in the backup tracking chain:

[0050]

[0051] In the formula, To track the prior probabilities of each mode in the chain, α is the Dirichlet distribution parameter.

[0052] In one embodiment, calculating the posterior probability of the current observation data belonging to each modality of the main tracking chain and the posterior probability of the current observation data belonging to each modality of the backup tracking chain includes:

[0053] Based on the likelihood probabilities of the current observation data belonging to each mode of the main tracking chain, the likelihood probabilities of the current observation data belonging to each mode of the backup tracking chain, the prior probabilities of each mode of the main tracking chain, and the prior probabilities of each mode of the backup tracking chain, calculate the posterior probabilities of the current observation data belonging to each mode of the tracking chain:

[0054]

[0055]

[0056]

[0057] In the formula, P k The posterior probability of the current observed data belonging to each mode of the tracking chain. Let P(Y) be the prior probability of the new mode. i |new) represents the likelihood distribution of the new mode to which the current observation data belongs.

[0058] A Kalman tracking device for radio direction finding, comprising:

[0059] The module is used to establish two tracking chains for radio direction finding and initialize them respectively to obtain the main tracking chain and the backup tracking chain. The main tracking chain corresponds to the main mode set, and the backup tracking chain corresponds to the backup mode set. Both the main mode set and the backup mode set include multiple modes, and the state of each mode includes heat and mean.

[0060] The acquisition module is used to acquire the data from the current observation.

[0061] The judgment module is used to determine that when the current observation is not the first observation, it performs model prediction on the modal states of the main tracking chain and the backup tracking chain to obtain the predicted states of each modality of the main tracking chain and the predicted states of each modality of the backup tracking chain.

[0062] The calculation module is used to calculate the likelihood probability of the current observation data belonging to each mode of the main tracking chain, the likelihood probability of the current observation data belonging to each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain based on the prediction state of each mode of the main tracking chain and the prediction state of each mode of the backup tracking chain. It also calculates the posterior probability of the current observation data belonging to each mode of the main tracking chain and the posterior probability of the current observation data belonging to each mode of the backup tracking chain.

[0063] The update module is used to update the state of each mode of the main tracking chain based on the posterior probability of the current observation data belonging to each mode of the main tracking chain, and to update the state of each mode of the backup tracking chain based on the posterior probability of the current observation data belonging to each mode of the backup tracking chain.

[0064] The output module is used to traverse the modal states of the updated main tracking chain and the backup tracking chain, and outputs the mean value corresponding to the mode with the highest popularity as the current observation tracking information.

[0065] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program performing the following steps:

[0066] Two tracking chains for radio direction finding are established and initialized separately to obtain the main tracking chain and the backup tracking chain. The main tracking chain corresponds to the main mode set, and the backup tracking chain corresponds to the backup mode set. Both the main mode set and the backup mode set include multiple modes, and the state of each mode includes heat and mean.

[0067] Obtain the data from the current observation;

[0068] When it is determined that the current observation is not the first observation, model prediction is performed on the modal states of the main tracking chain and the backup tracking chain to obtain the predicted states of each mode of the main tracking chain and the predicted states of each mode of the backup tracking chain.

[0069] Based on the predicted states of each modality in the main tracking chain and each modality in the backup tracking chain, calculate the likelihood probability of the current observed data belonging to each modality in the main tracking chain, the likelihood probability of the current observed data belonging to each modality in the backup tracking chain, the prior probability of each modality in the main tracking chain, and the prior probability of each modality in the backup tracking chain. Also calculate the posterior probability of the current observed data belonging to each modality in the main tracking chain and the posterior probability of the current observed data belonging to each modality in the backup tracking chain.

[0070] Based on the posterior probability of the current observation data belonging to each mode of the main tracking chain, update the state of each mode of the main tracking chain; based on the posterior probability of the current observation data belonging to each mode of the backup tracking chain, update the state of each mode of the backup tracking chain.

[0071] Iterate through the modal states of the updated main tracking chain and the backup tracking chain, and output the mean value corresponding to the mode with the highest popularity as the current observation tracking information.

[0072] A computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0073] Two tracking chains for radio direction finding are established and initialized separately to obtain the main tracking chain and the backup tracking chain. The main tracking chain corresponds to the main mode set, and the backup tracking chain corresponds to the backup mode set. Both the main mode set and the backup mode set include multiple modes, and the state of each mode includes heat and mean.

[0074] Obtain the data from the current observation;

[0075] When it is determined that the current observation is not the first observation, model prediction is performed on the modal states of the main tracking chain and the backup tracking chain to obtain the predicted states of each mode of the main tracking chain and the predicted states of each mode of the backup tracking chain.

[0076] Based on the predicted states of each modality in the main tracking chain and each modality in the backup tracking chain, calculate the likelihood probability of the current observed data belonging to each modality in the main tracking chain, the likelihood probability of the current observed data belonging to each modality in the backup tracking chain, the prior probability of each modality in the main tracking chain, and the prior probability of each modality in the backup tracking chain. Also calculate the posterior probability of the current observed data belonging to each modality in the main tracking chain and the posterior probability of the current observed data belonging to each modality in the backup tracking chain.

[0077] Based on the posterior probability of the current observation data belonging to each mode of the main tracking chain, update the state of each mode of the main tracking chain; based on the posterior probability of the current observation data belonging to each mode of the backup tracking chain, update the state of each mode of the backup tracking chain.

[0078] Iterate through the modal states of the updated main tracking chain and the backup tracking chain, and output the mean value corresponding to the mode with the highest popularity as the current observation tracking information.

[0079] The aforementioned Kalman tracking method, apparatus, equipment, and medium for radio direction finding address the technical problems of tracking failure caused by continuous outliers in UAV radio direction finding tracking and how to quickly adapt to maneuvers. Based on Kalman tracking, it performs modified adaptive angle tracking, introduces a multimodal tracking chain, eliminates outliers by adaptively switching between multimodal modes, and introduces a main tracking chain and a backup tracking chain. Through adaptive switching between the main tracking chain and the backup tracking chain, it achieves adaptation to rapid maneuvers, satisfying the adaptability of radio direction finding tracking under conditions with many outliers and rapid target maneuvers. Attached Figure Description

[0080] Figure 1 This is an application scenario diagram of a Kalman tracking method for radio direction finding in one embodiment;

[0081] Figure 2 This is a flowchart illustrating a Kalman tracking method using radio direction finding in one embodiment;

[0082] Figure 3 This is a schematic diagram of the framework of a Kalman tracking method for radio direction finding in one embodiment;

[0083] Figure 4 This is a typical data tracking effect diagram of radio direction finding in one embodiment;

[0084] Figure 5 This is a structural block diagram of a Kalman tracking device for radio direction finding in one embodiment;

[0085] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0086] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0087] It should be noted that all directional indicators (such as up, down, left, right, front, back, etc.) in the embodiments of this application are only used to explain the relative positional relationship and movement of each component in a certain specific posture (as shown in the figure). If the specific posture changes, the directional indicator will also change accordingly.

[0088] Furthermore, the use of terms such as "first" and "second" in this application is for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include at least one of those features. In the description of this application, "multiple sets" means at least two sets, such as two sets, three sets, etc., unless otherwise explicitly specified.

[0089] In this application, unless otherwise expressly specified and limited, the terms "connection," "fixed," etc., should be interpreted broadly. For example, "fixed" can mean a fixed connection, a detachable connection, or an integral part; it can mean a mechanical connection, an electrical connection, a physical connection, or a wireless communication connection; it can mean a direct connection or an indirect connection through an intermediate medium; it can mean the internal communication of two elements or the interaction between two elements, unless otherwise expressly limited. Those skilled in the art can understand the specific meaning of the above terms in this application according to the specific circumstances.

[0090] Furthermore, the technical solutions of the various embodiments of this application can be combined with each other, but only if they are based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed by this application.

[0091] The method provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. Terminal 102 may include, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. Server 104 may be a server corresponding to various portal websites or work system backends.

[0092] This application provides a Kalman tracking method for radio direction finding, such as Figure 2 As shown, in one embodiment, the method is applied to Figure 1 Taking the terminal in the example, the explanation includes:

[0093] Step 202: Establish two tracking chains for radio direction finding and initialize them respectively to obtain the main tracking chain and the backup tracking chain; the main tracking chain corresponds to the main mode set, and the backup tracking chain corresponds to the backup mode set. Both the main mode set and the backup mode set include multiple modes, and the state of each mode includes heat and mean.

[0094] In this step, two tracking chains are established for the UAV's radio direction finding, with the tracking chains based on azimuth θ and azimuth angular velocity. Using the state variable and the azimuth θ as the observation, the state is characterized by a Gaussian mixture model, which can adapt to the direction finding field value.

[0095] The modalities of the main tracing chain constitute the main modal set, and the modalities of the backup tracing chain constitute the backup modal set.

[0096] Initializing the tracking chain includes: parameter initialization of the tracking chain (including parameter initialization of the main tracking chain and parameter initialization of the backup tracking chain, the parameters of which are the same but have different values), state initialization of the tracking chain (state initialization of the main tracking chain and state initialization of the backup tracking chain), and divergence initialization of the main tracking chain and the backup tracking chain.

[0097] The parameters of the tracking chain include: maximum number of modes K (typically 5), forgetting factor γ, and observation noise. (Typical value is 5) 2 Orientation transformation variance of the main tracking chain (Typical value is 1) 2 ), Azimuth angular velocity transformation variance of the main tracking chain (Typical value is 0.5) 2 ), Location transformation variance of backup tracking chain (Typical value is 5) 2 ), azimuth angular velocity conversion variance of backup tracking chain (Typical value is 2) 2 And the initial state variance Σ0, where:

[0098]

[0099]

[0100] In the formula, 20 indicates that the time window is 20 times. Generally 180 2 , Determined based on the maximum range of angular velocity, such as 10. 2 .

[0101] The primary tracking chain has a smaller transfer variance and higher accuracy, but poorer adaptability to maneuvers. The backup tracking chain has a larger transfer variance and can adapt to rapid maneuvers, but its accuracy is slightly lower.

[0102] When initializing the state of each mode of the tracking chain, n k =0 indicates that all current modes are not activated.

[0103] Initialize the divergence between the main tracing chain and the backup tracing chain, setting the divergence Δ i =0.

[0104] Step 204: Obtain the data from the current observation.

[0105] In this step, Y i Let i be the data currently observed, and let i be the i-th observation in the sequence.

[0106] Step 206: When it is determined that the current observation is not the first observation, model prediction is performed on the modal states of the main tracking chain and the modal states of the backup tracking chain to obtain the predicted states of each modality of the main tracking chain and the predicted states of each modality of the backup tracking chain.

[0107] Specifically:

[0108] The modal states of the main tracing chain and the backup tracing chain are represented as follows:

[0109] {n k ,μ k ,Σ k}

[0110]

[0111] In the formula, n k To track the popularity of the k-th mode in the chain, μ is calculated by using a sliding window to count the number of observation samples belonging to that mode within a certain time window. k To track the mean of the k-th mode of the chain, Σ k To track the variance of the k-th mode of the tracking chain, θ represents the orientation of the tracking chain. To track the azimuth angular velocity of the chain;

[0112] The state transition equation is:

[0113]

[0114]

[0115]

[0116] In the formula, X i Here is the state update equation, where i is the i-th observation data, and η is the η-th observation data. i The state transition uncertainty follows a Gaussian distribution with mean 0 and variance Q. The variance of the positional shift is unknown. The variance of the transfer uncertainty of the azimuth angular velocity;

[0117] Model prediction is performed based on the modal states and state transition equations of the tracking chain. This process iterates through all modes of the tracking chain. If n k =0, skip the current mode; otherwise, the predicted states of each mode in the tracking chain are:

[0118]

[0119] In the formula, To track the predicted popularity of the k-th mode of the chain, To track the predicted mean of the k-th mode of the chain, To track the variance of the k-th mode in the chain, γ is the forgetting factor, typically taken as 0.95.

[0120] Step 208: Based on the predicted states of each mode of the main tracking chain and each mode of the backup tracking chain, calculate the likelihood probability of the current observation data belonging to each mode of the main tracking chain, the likelihood probability of the current observation data belonging to each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain. Also calculate the posterior probability of the current observation data belonging to each mode of the main tracking chain and the posterior probability of the current observation data belonging to each mode of the backup tracking chain.

[0121] Specifically:

[0122] Based on the predicted states of each mode in the main tracking chain and each mode in the backup tracking chain, calculate the likelihood probability of the current observation data belonging to each mode in the main tracking chain and the likelihood probability of the current observation data belonging to each mode in the backup tracking chain:

[0123] like but,

[0124] P(Y i |k)=0

[0125] Y i =θ i +ε i =BX i +ε i

[0126] B = [1 0]

[0127] otherwise,

[0128]

[0129]

[0130] In the formula, P(Y) i |k) represents the likelihood probability of the current observation data belonging to each mode of the tracking chain, calculated using a Gaussian distribution. Note that due to the 360° entanglement issue, the observations need to be unwrapped to within 180° of the distribution mean before being substituted into the Gaussian distribution probability formula. i The measurement equation (which is the expanded representation of the observed data), ε i The azimuth observation noise follows a mean of 0 and a variance of . The distribution is a Gaussian distribution, where N is the Gaussian probability distribution function. To observe the residual covariance;

[0131] Based on the predicted states of each modality in the main tracking chain and each modality in the backup tracking chain, calculate the prior probabilities of each modality in the main tracking chain and each modality in the backup tracking chain:

[0132]

[0133] In the formula, To track the prior probabilities of each mode in the chain, α is the Dirichlet distribution parameter, typically taken as 0.2.

[0134] Based on the likelihood probabilities of the current observation data belonging to each mode of the main tracking chain, the likelihood probabilities of the current observation data belonging to each mode of the backup tracking chain, the prior probabilities of each mode of the main tracking chain, and the prior probabilities of each mode of the backup tracking chain, calculate the posterior probabilities of the current observation data belonging to each mode of the tracking chain (the posterior probabilities of the current observation data belonging to each mode of the main tracking chain and the posterior probabilities of the current observation data belonging to each mode of the backup tracking chain are both calculated using the following formula):

[0135]

[0136]

[0137]

[0138] In the formula, P k The posterior probability of the current observed data belonging to each mode of the tracking chain. Let P(Y) be the prior probability of the new mode. i |new) represents the likelihood distribution of the current observation data belonging to the new mode, indicating a uniform distribution within a 360° range.

[0139] Step 210: Update the state of each mode of the main tracking chain according to the posterior probability of the current observation data belonging to each mode of the main tracking chain, and update the state of each mode of the backup tracking chain according to the posterior probability of the current observation data belonging to each mode of the backup tracking chain.

[0140] In this step, according to {P k ,Y i Update the state of each modality:

[0141]

[0142]

[0143]

[0144]

[0145]

[0146] In the formula, For the prediction error, W k Let I be the correction matrix, and let I be the identity matrix.

[0147] Step 212: Traverse the modal states of the updated main tracking chain and the backup tracking chain, and use the mean value corresponding to the mode with the highest popularity as the current observation tracking information and output it.

[0148] In this embodiment, the UAV has two tracking chains, namely the main tracking chain and the backup tracking chain. Each tracking chain has a state and corresponds to a mode set, and each mode set includes multiple modes.

[0149] The aforementioned Kalman tracking method for radio direction finding addresses the technical problems of tracking failure caused by continuous outliers in UAV radio direction finding tracking and how to quickly adapt to maneuvers. Based on Kalman tracking, this application performs modified adaptive angle tracking, introduces a multimodal tracking chain, and eliminates outliers by adaptively switching between multimodal modes. It introduces a main tracking chain and a backup tracking chain, and achieves adaptation to rapid maneuvers through adaptive switching between the main tracking chain and the backup tracking chain, thus satisfying the adaptability of radio direction finding tracking under conditions with many outliers and rapid target maneuvers.

[0150] It should be understood that, although Figure 2 The steps in the flowchart are shown sequentially as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified herein, there is no strict order in which these steps are executed, and they can be performed in other orders. Figure 2 At least some of the steps in the process may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least some of the sub-steps or stages of other steps.

[0151] like Figure 3 As shown, in one embodiment, after updating the states of each modality of the main tracking chain and the states of each modality of the backup tracking chain, the method further includes: updating the divergence between the main tracking chain and the backup tracking chain based on the updated states of each modality of the main tracking chain and the updated states of each modality of the backup tracking chain; when the divergence meets a preset first condition, copying the updated backup tracking chain to the updated main tracking chain to obtain a new main tracking chain; when the divergence does not meet the preset first condition, using the updated main tracking chain as the new main tracking chain; acquiring the data for the next observation, and performing model prediction based on the new main tracking chain and the updated backup tracking chain, and outputting the tracking information for the next observation.

[0152] in,

[0153] Δ i =γΔ i +(θ main -θ back )

[0154] In the formula, θ main The output tracking information of the main tracking chain, θ back To back up the output tracing information of the tracing chain.

[0155] The first condition can be set as: Δ i >2σ θ That is, an adaptive switching decision is made. If the first condition is met, the state of the backup tracking chain is copied to the main tracking chain. In other words, the modal states of the backup tracking chain are used to replace the modal states of the main tracking chain.

[0156] In one embodiment, it further includes: when determining that the current observation is the first observation, i.e., all... Then P new =1, establish and initialize a new mode; based on the initialized new mode, update the main mode set of the main tracking chain and the backup mode set of the backup tracking chain; obtain the data for the next observation, and based on the updated main mode set and the updated backup mode set, perform model prediction and output the tracking information for the next observation.

[0157] In one embodiment, the method further includes: calculating the posterior probability of the current observation data belonging to a new mode based on the likelihood probability of the current observation data belonging to each mode of the main tracking chain, the likelihood probability of the current observation data belonging to each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain; establishing and initializing a new mode when the posterior probability of the current observation data belonging to the new mode satisfies a preset second condition; updating the main mode set of the main tracking chain and the backup mode set of the backup tracking chain based on the initialized new mode; acquiring the data for the next observation, and performing model prediction and outputting the tracking information for the next observation based on the updated main mode set and the updated backup mode set.

[0158] Specifically, the posterior probability of the current observed data belonging to the new mode is calculated:

[0159]

[0160] In the formula, P new The posterior probability of the current observed data belonging to the new mode;

[0161] The second condition can be set as: P new >0.5;

[0162] Establishing a new mode: Search for the mode with the lowest popularity and its corresponding popularity value, denoted as .

[0163] like Then initialize the new mode k cold for:

[0164]

[0165] Here, the establishment of a new mode is determined by comparing it with the minimum heat value. The number of modes is fixed. The k modes with higher heat values ​​are obtained by comparison. The initial mean and initial variance of the new mode are calculated based on the observation values ​​of the new mode itself.

[0166] Otherwise, that is, if Maintain the original mode with the lowest heat and its corresponding heat value.

[0167] like Figure 4 As shown in the figure, the horizontal axis represents the time sequence, and the vertical axis represents the azimuth angle. This figure reveals that the original angle output is scattered and its distribution is rather disorganized; the backup tracking chain is highly mobile, maintaining relative consistency with the original output, and making minor corrections to the output based on the original data; the main tracking chain's output is relatively stable and concentrated compared to the original data. Compared with the actual results, the output of this main tracking chain is closest to the actual situation.

[0168] This application also provides a Kalman tracking device for radio direction finding, such as Figure 5 As shown, in one embodiment, it includes: an establishment module 502, an acquisition module 504, a judgment module 506, a calculation module 508, an update module 510, and an output module 512, wherein:

[0169] Module 502 is used to establish two tracking chains for radio direction finding and initialize them respectively to obtain a main tracking chain and a backup tracking chain. The main tracking chain corresponds to the main mode set, and the backup tracking chain corresponds to the backup mode set. Both the main mode set and the backup mode set include multiple modes, and the state of each mode includes heat and mean.

[0170] Module 504 is used to acquire the data of the current observation;

[0171] The judgment module 506 is used to determine that when the current observation is not the first observation, to perform model prediction on the modal states of the main tracking chain and the backup tracking chain, so as to obtain the predicted states of each modality of the main tracking chain and the predicted states of each modality of the backup tracking chain.

[0172] The calculation module 508 is used to calculate the likelihood probability of the current observation data belonging to each mode of the main tracking chain, the likelihood probability of the current observation data belonging to each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain based on the prediction state of each mode of the main tracking chain and the prediction state of each mode of the backup tracking chain, and to calculate the posterior probability of the current observation data belonging to each mode of the main tracking chain and the posterior probability of the current observation data belonging to each mode of the backup tracking chain.

[0173] The update module 510 is used to update the state of each mode of the main tracking chain according to the posterior probability of the current observation data belonging to each mode of the main tracking chain, and to update the state of each mode of the backup tracking chain according to the posterior probability of the current observation data belonging to each mode of the backup tracking chain.

[0174] Output module 512 is used to traverse the modal states of the updated main tracking chain and the modal states of the updated backup tracking chain, and outputs the mean value corresponding to the modality with the highest heat as the current observation tracking information.

[0175] Specific limitations regarding the Kalman tracking device for radio direction finding can be found in the limitations of the Kalman tracking method for radio direction finding mentioned above, and will not be repeated here. Each module in the above device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in hardware or independently of the processor in a computer device, or stored in software in the memory of a computer device, so that the processor can call and execute the operations corresponding to each module.

[0176] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, network interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The network interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a Kalman tracing method for radio direction finding. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the computer device casing, or an external keyboard, touchpad, or mouse.

[0177] Those skilled in the art will understand that Figure 4The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0178] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the method described above.

[0179] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0180] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0181] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0182] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A Kalman tracking method for radio direction finding, characterized by The method comprises the following steps: Two tracking chains of radio direction finding are established and initialized respectively to obtain a main tracking chain and a backup tracking chain; The main tracking chain corresponds to a main modal set, and the backup tracking chain corresponds to a backup modal set; the main modal set and the backup modal set each comprise a plurality of modes; the state of each mode comprises a hotness and a mean value; Data of a current observation is obtained; When the current observation is not the first observation, model prediction is performed on the state of each mode of the main tracking chain and the state of each mode of the backup tracking chain to obtain the predicted state of each mode of the main tracking chain and the predicted state of each mode of the backup tracking chain; According to the predicted state of each mode of the main tracking chain and the predicted state of each mode of the backup tracking chain, the likelihood probability of the current observation data belonging to each mode of the main tracking chain, the likelihood probability of the current observation data belonging to each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain are calculated, and the posterior probability of the current observation data belonging to each mode of the main tracking chain and the posterior probability of the current observation data belonging to each mode of the backup tracking chain are calculated; According to the posterior probability of the current observation data belonging to each mode of the main tracking chain, the state of each mode of the main tracking chain is updated, and according to the posterior probability of the current observation data belonging to each mode of the backup tracking chain, the state of each mode of the backup tracking chain is updated; The state of each mode of the updated main tracking chain and the state of each mode of the updated backup tracking chain are traversed, and the mean value corresponding to the mode with the maximum hotness is taken as the tracking information of the current observation and is outputted; When the current observation is not the first observation, model prediction is performed on the state of each mode of the main tracking chain and the state of each mode of the backup tracking chain to obtain the predicted state of each mode of the main tracking chain and the predicted state of each mode of the backup tracking chain, which comprises the following steps: The state of each mode of the main tracking chain and the state of each mode of the backup tracking chain are represented as: In the formula, For tracking chain number The popularity of each modality For tracking chain number The mean of each mode, For tracking chain number The variance of each modality To track the location of the chain, To track the azimuth angular velocity of the chain; The state transition equation is: In the formula, For the state update equation, For the first One observation data, The state transition uncertainty follows a pattern with a mean of 0 and a variance of . Gaussian distribution, The variance of the shift in orientation is uncertain. The variance of the transfer uncertainty of the azimuth angular velocity; According to the state of each mode of the tracking chain and the state transition equation, model prediction is performed, all modes of the tracking chain are traversed, if the current mode is skipped, otherwise the predicted state of each mode of the tracking chain is: wherein is the predicted temperature of the chain of tracks for the i-th modality, is the predicted temperature of the chain of tracks for the i-th modality, is the predicted mean of the chain of tracks for the i-th modality, is the predicted mean of the chain of tracks for the i-th modality, is the variance of the chain of tracks for the i-th modality, is the variance of the chain of tracks for the i-th modality, is the forgetting factor.

2. The Kalman tracking method of radio direction finding according to claim 1, characterized in that, After the state of each mode of the main tracking chain and the state of each mode of the backup tracking chain are updated, the following steps are further included: According to the state of each mode of the updated main tracking chain and the state of each mode of the updated backup tracking chain, the divergence degree of the main tracking chain and the backup tracking chain is updated; When the divergence degree meets a preset first condition, the updated backup tracking chain is copied to the updated main tracking chain to obtain a new main tracking chain; when the divergence degree does not meet the preset first condition, the updated main tracking chain is taken as the new main tracking chain; Data of a next observation is obtained, and model prediction is performed according to the new main tracking chain and the updated backup tracking chain to output the tracking information of the next observation.

3. The Kalman tracking method of radio direction finding according to claim 2, characterized in that, Further comprising the following steps: When the current observation is the first observation, a new mode is established and initialized; According to the initialized new mode, the main modal set of the main tracking chain and the backup modal set of the backup tracking chain are updated; Data of a next observation is obtained, and model prediction is performed according to the updated main modal set and the updated backup modal set to output the tracking information of the next observation.

4. The Kalman tracking method of radio direction finding according to claim 3, characterized in that, Further comprising the following steps: According to the likelihood probability of the current observation data belonging to each mode of the main tracking chain, the likelihood probability of the current observation data belonging to each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain, the posterior probability of the current observation data belonging to the new mode is calculated. the posterior probability of the current observation data belonging to the new mode meets a preset second condition, a new mode is established and initialized; according to the initialized new mode, a main mode set of the main tracking chain and a backup mode set of the backup tracking chain are updated; data of a next observation is obtained, and model prediction is performed according to the updated main mode set and the updated backup mode set, and tracking information of the next observation is output.

5. The Kalman tracking method of radio direction finding according to any one of claims 1 to 4, characterized in that, According to the predicted state of each mode of the main tracking chain and the predicted state of each mode of the backup tracking chain, the likelihood probability of the current observation data belonging to each mode of the main tracking chain, the likelihood probability of the current observation data belonging to each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain are calculated, including: According to the predicted state of each mode of the main tracking chain and the predicted state of each mode of the backup tracking chain, the likelihood probability of the current observation data belonging to each mode of the main tracking chain and the likelihood probability of the current observation data belonging to each mode of the backup tracking chain are calculated: If then, Otherwise, wherein is the likelihood probability of the current observation data belonging to each modality of the tracking chain, is the measurement equation, is the azimuth observation noise, is the Gaussian probability distribution function, is the observation residual covariance; According to the predicted state of each mode of the main tracking chain and the predicted state of each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain and the prior probability of each mode of the backup tracking chain are calculated: wherein is the prior probability of each modality of the tracking chain, is dirichlet distribution parameters.

6. The Kalman tracking method of radio direction finding according to claim 5, characterized in that, The posterior probability of the current observation data belonging to each mode of the main tracking chain and the posterior probability of the current observation data belonging to each mode of the backup tracking chain are calculated, including: According to the likelihood probability of the current observation data belonging to each mode of the main tracking chain, the likelihood probability of the current observation data belonging to each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain, the posterior probability of the current observation data belonging to each mode of the tracking chain is calculated. wherein is the posterior probability that the current observation belongs to the tracking chain of the modalities, is the prior probability of the new modality, is the likelihood distribution of the current observation belonging to the new modality.

7. A Kalman tracking arrangement for radio direction finding, characterized by The Kalman tracking method for radio direction finding according to any one of claims 1 to 6, comprising: A establishing module is configured to establish two tracking chains for radio direction finding and initialize them respectively to obtain a main tracking chain and a backup tracking chain; the main tracking chain corresponds to a main mode set, and the backup tracking chain corresponds to a backup mode set; the main mode set and the backup mode set each include a plurality of modes; and the state of each mode includes a temperature and a mean value. An obtaining module is configured to obtain current observation data. A judging module is configured to, when the current observation is not the first observation, perform model prediction on the state of each mode of the main tracking chain and the state of each mode of the backup tracking chain to obtain the predicted state of each mode of the main tracking chain and the predicted state of each mode of the backup tracking chain. A calculating module is configured to calculate, according to the predicted state of each mode of the main tracking chain and the predicted state of each mode of the backup tracking chain, the likelihood probability of the current observation data belonging to each mode of the main tracking chain, the likelihood probability of the current observation data belonging to each mode of the backup tracking chain, the prior probability of each mode of the main tracking chain, and the prior probability of each mode of the backup tracking chain, and calculate the posterior probability of the current observation data belonging to each mode of the main tracking chain and the posterior probability of the current observation data belonging to each mode of the backup tracking chain. An updating module is configured to update the state of each mode of the main tracking chain according to the posterior probability of the current observation data belonging to each mode of the main tracking chain, and update the state of each mode of the backup tracking chain according to the posterior probability of the current observation data belonging to each mode of the backup tracking chain. An output module is configured to traverse each modality state of the updated main tracking chain and each modality state of the updated backup tracking chain, and output the tracking information of the current observation as the mean value of the modality with the highest heat.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

9. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

Images