High-precision real-time multi-path cooperative control method and system

Abstract

The application provides a high-precision real-time multi-path cooperative control method and system, comprising: establishing conversion relationships among a geodetic coordinate system, a geocentric coordinate system and a station-centered coordinate system; each device converts observation data of a target into global ECEF coordinates of the target through the conversion relationships according to its own geographic coordinates; an iterative weighted least square method is used to realize fusion of observation data of multiple cameras (integrated in a rotating cabin) to obtain the best estimation of the initial position of the target; the fused position is taken as an observation quantity and input into a Kalman filter based on a uniform motion model to dynamically adjust observation noise covariance and process noise covariance and to predict and track the motion state of the target, so that multi-camera cooperative control is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of target positioning and tracking, in particular to a high-precision real-time multi-path cooperative control method and system. BACKGROUND

[0002] In the technical field of target positioning and tracking, the traditional algorithm relying on a single sensor or intelligent body architecture is limited by narrow field of view range, weak anti-shielding ability and other defects, and has significant limitations in dynamic scenes. Although multi-sensor cooperative architecture is widely studied, there is still a lack of systematic modeling of multi-device or sensor coordinate conversion. Using only local coordinate data can cause spatial consistency conflicts due to different reference bases, and the tracking stability is insufficient in dynamic scenes. SUMMARY

[0003] Therefore, the purpose of the present application is to provide a high-precision real-time multi-path cooperative control method and system, which breaks through the technical bottlenecks of coordinate alignment, filtering adaptability and field of view linkage by means of multi-sensor data fusion and improved Kalman filtering, and improves the positioning and tracking accuracy of dynamic targets.

[0004] In a first aspect, the embodiments of the present application provide a high-precision real-time multi-path cooperative control method, which comprises:

[0005] Obtaining the LLA coordinates of each camera through GNSS, wherein the LLA coordinates include latitude, longitude and altitude;

[0006] Converting the LLA coordinates of each camera into the first coordinates in the global ECEF coordinate system through the earth ellipsoid model;

[0007] After calibrating the azimuth zero position and the pitch angle zero position of the camera, synchronously collecting the observation data of each camera, wherein the observation data includes distance, azimuth and pitch angle;

[0008] Obtaining the coordinates of the target in the local coordinate system of the camera through the trigonometric function relationship, and mapping the coordinates of the target in the local coordinate system of the camera to the global ECEF coordinate;

[0009] Fusing the global ECEF coordinates obtained by converting the coordinates in the local coordinate system of all cameras through the iterative weighted least squares algorithm to obtain the optimal global coordinates of the target;

[0010] Taking the optimal global coordinates of the target as the initial state position component, inputting into the Kalman filter constructed based on the uniform motion model to predict and track the motion state of the target, and obtaining the posterior optimal state estimation and the posterior optimal covariance at the current time;

[0011] calculate local polar coordinates of the target in a camera to be switched;

[0012] drive the camera to adjust an angle according to the local polar coordinates.

[0013] In a second aspect, an embodiment of the present application provides a high-precision real-time multi-path cooperative control system, which comprises:

[0014] an LLA coordinate acquisition module configured to acquire LLA coordinates of each camera through GNSS, wherein the LLA coordinates comprise latitude, longitude and altitude;

[0015] a conversion module configured to convert the LLA coordinates of each camera into first coordinates in a global ECEF coordinate system through an earth ellipsoid model;

[0016] an acquisition module configured to synchronously acquire observation data of each camera after calibration of azimuth zero position and pitch angle zero position of the camera is completed, wherein the observation data comprises distance, azimuth and pitch angle;

[0017] a mapping module configured to acquire coordinates of a target in a local coordinate system of a camera through a trigonometric function relationship, and map the coordinates of the target in the local coordinate system of the camera to a global ECEF coordinate;

[0018] a fusion module configured to fuse the global ECEF coordinates obtained by converting the coordinates in the local coordinate system of all the cameras through an iterative weighted least squares algorithm, to obtain optimal global coordinates of the target;

[0019] a prediction and tracking module configured to input the optimal global coordinates of the target as a position component of an initial state to a Kalman filter constructed based on a uniform motion model, to perform prediction and tracking of a motion state of the target, and to obtain a posterior optimal state estimation and a posterior optimal covariance at a current time;

[0020] a calculation module configured to calculate local polar coordinates of the target in a camera to be switched;

[0021] a driving module configured to drive the camera to adjust an angle according to the local polar coordinates.

[0022] In a third aspect, an embodiment of the present application provides an electronic device, which comprises a memory and a processor, the memory stores a computer program capable of running on the processor, and the processor implements the method as described above when executing the computer program.

[0023] In a fourth aspect, an embodiment of the present application provides a computer readable medium having non-volatile program codes executable by a processor, and the program codes cause the processor to execute the method as described above.

[0024] This invention provides a high-precision real-time multi-channel cooperative control method and system, comprising: acquiring the LLA coordinates of each camera via GNSS, wherein the LLA coordinates include latitude, longitude, and elevation; converting the LLA coordinates of each camera into first coordinates in the global ECEF coordinate system using an Earth ellipsoid model; synchronously acquiring observation data from each camera after calibrating the azimuth and elevation zero points of the cameras; wherein the observation data includes distance, azimuth, and elevation; obtaining the target's coordinates in the camera's local coordinate system using trigonometric functions, and mapping the target's coordinates in the camera and turntable's local coordinate systems to the global ECEF coordinates; and performing iterative weighted minimum... The little-squares algorithm fuses the global ECEF coordinates obtained from coordinate transformations in the local coordinate systems of all cameras to obtain the optimal global target coordinates. These optimal global target coordinates are then used as the initial state position components and input into a Kalman filter built on a uniform motion model for predicting and tracking the target's motion state, yielding the posterior optimal state estimate and posterior optimal covariance at the current moment. The algorithm also calculates the target's local polar coordinates in the camera to be switched and drives the camera to adjust its angle based on these local polar coordinates. Through multi-sensor data fusion and improved Kalman filtering techniques, the algorithm overcomes the technical bottlenecks of coordinate alignment, filter adaptability, and field-of-view linkage, thereby improving the accuracy of dynamic target localization and tracking.

[0025] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention are realized and obtained in accordance with the structures particularly pointed out in the description, claims and drawings.

[0026] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described below in detail with reference to the accompanying drawings. Attached Figure Description

[0027] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0028] Figure 1 This is a flowchart of a high-precision real-time multi-channel cooperative control method provided in Embodiment 1 of the present invention;

[0029] Figure 2 This is a schematic diagram illustrating the working principle of multiple cameras working together, provided in Embodiment 1 of the present invention.

[0030] Figure 3 This is a schematic diagram of a high-precision real-time multi-channel collaborative control system provided in Embodiment 2 of the present invention. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] To facilitate understanding of this embodiment, the embodiments of the present invention will be described in detail below.

[0033] Example 1:

[0034] Figure 1 The flowchart illustrates the high-precision real-time multi-channel cooperative control method provided in Embodiment 1 of the present invention.

[0035] Reference Figure 1 The method includes the following steps:

[0036] Step S101: Obtain the LLA coordinates of each camera via GNSS, where the LLA coordinates include latitude, longitude and elevation;

[0037] Step S102: Using the Earth ellipsoid model, convert the LLA coordinates of each camera (placed inside the turntable cabin) into the first coordinates in the global ECEF coordinate system.

[0038] Step S103: After the azimuth and elevation zero points of the cameras are calibrated, the observation data of each camera is collected synchronously; the observation data includes distance, azimuth and elevation.

[0039] Step S104: Obtain the target's coordinates in the camera's local coordinate system through trigonometric function relationships, and map the target's coordinates in the camera's local coordinate system to the global ECEF coordinate system;

[0040] Step S105: The global ECEF coordinates obtained by coordinate transformation in the local coordinate system of all cameras are fused through the iterative weighted least squares algorithm to obtain the optimal global target coordinates.

[0041] Step S106: The optimal global coordinates of the target are used as the position components of the initial state and input into the Kalman filter constructed based on the uniform motion model to predict and track the target motion state, so as to obtain the posterior optimal state estimate and posterior optimal covariance at the current time.

[0042] Step S107: Calculate the local polar coordinates of the target in the camera to be switched;

[0043] Step S108: Adjust the angle of the camera based on the local polar coordinates.

[0044] Specifically, addressing the challenges of coordinate system alignment, insufficient tracking accuracy in dynamic scenes, and limited field of view in existing multi-sensor collaborative systems, this application achieves spatial consistency fusion of multi-sensor observation data and high-precision tracking of dynamic targets through unified multi-camera coordinate modeling, improved Kalman filter tracking, and dual-turntable field-of-view expansion. This solves the problems of coordinate system alignment difficulties, insufficient tracking accuracy, and limited field of view in existing technologies. The algorithm demonstrates robustness and stability in dynamic tracking. The average error between the calculated and theoretical values ​​of the multi-camera coordinate transformation is lower than [a certain value]. The positional deviation predicted by the Kalman filter is ≤0.1m.

[0045] This application establishes a transformation relationship between ECEF, LLA, and the centroid coordinate system (ENU). Each device, based on its own geographic coordinates, calculates its observations of the target (distance, azimuth, and elevation) into the target's global ECEF coordinates using this transformation relationship. An iterative weighted least squares method is employed to fuse multi-camera observation data, obtaining the optimal estimate of the target's initial position. ;

[0046] The merged position As an observation, it is input into a Kalman filter constructed based on a uniform motion model, dynamically adjusting the observation noise covariance R and the process noise covariance Q to predict and track the target motion state; based on a dual-turntable angle conversion model, the turntable azimuth and pitch angles are calculated to achieve dual-turntable cooperative control.

[0047] This application includes four photoelectric cameras (integrated into the turntable cabin) and their calibration, camera coordinate transformation, iterative weighted least squares data fusion, improved Kalman filtering and other technologies.

[0048] The geodetic coordinates (LLA coordinates, i.e., latitude) of multiple cameras (set to four) are obtained through positioning. ,longitude and elevation ), and calibrate the azimuth zero point of the camera (adjust the azimuth measurement zero point (0 degrees) of the camera) to ensure that its physical pointing is exactly aligned with the due north direction; at the same time, define the due north as 0 degrees of the azimuth, the clockwise rotation direction as the increasing angle direction, and the measurement range covers 0 degrees to 360 degrees and the pitch zero point (taking the horizontal reference plane of the equipment installation position as the reference, adjust the pitch measurement zero point (0 degrees) of the camera) to ensure that its initial optical axis or rotation axis coincides with the horizontal reference plane; at the same time, define the horizontal reference plane as 0 degrees of the pitch, the upward deflection direction as the positive angle, and the downward deflection direction as the negative angle, and the measurement range covers -90 degrees to 90 degrees.

[0049] After the equipment deployment is completed, the observation data (including distance, azimuth, and pitch) from each camera are collected in real time, and then the coordinates of the target in the coordinate system of each camera are converted.

[0050] Refer to Figure 2 , and obtain the LLA coordinates of the camera through GNSS ( represented as the latitude of the camera , longitude and altitude ), and convert it to the coordinates in the global ECEF coordinate system through the earth ellipsoid model (CGCS2000) , as shown in formula (1):

[0051] (1)

[0052] Where, is the radius of curvature of the prime vertical; is the altitude of the equipment ; is the latitude of the equipment ; is the longitude of the equipment ; .

[0053] After the equipment positioning and zero point calibration are completed, start the multi-device synchronous clock, and collect the observation parameters of the target in real time: the distance of the target , the azimuth of its own rotation, and the pitch ; use the trigonometric function relationship to obtain the coordinates of the target in the local coordinate system of the camera: adopt the local coordinate system of "east - north - sky" (E, N, U) (the origin is the camera center, the E axis points east, the N axis points north, and the U axis points to the sky), and the calculation formula of the coordinates of the target in the local coordinate system refers to formula (2):

[0054] (2)

[0055] Then, the obtained local coordinates are uniformly mapped to global ECEF coordinates under the CGCS2000 datum. , refer to formula (3):

[0056] (3)

[0057] Furthermore, step S105 includes:

[0058] Step S201: Convert the observation data of each camera into corresponding global ECEF coordinate points;

[0059] Step S202: Calculate the weighted average of the corresponding global ECEF coordinate points to obtain the average value;

[0060] Step S203: Use the average value as the initial position estimate for the IWLS algorithm;

[0061] Step S204: Set the target position estimate for the t-th iteration and substitute it into the observation model to calculate the predicted observation value for each camera;

[0062] Step S205: Obtain the residuals and weight matrices based on the predicted observations corresponding to each camera;

[0063] Step S206: Calculate the Jacobian matrix of the observation model based on the current estimated point for each camera;

[0064] Step S207: Construct the normal equation for the position correction based on the Jacobian matrix, weight matrix, and residuals;

[0065] Step S208: After the iteration is completed, solve the normal equation for the position correction to obtain the correction amount;

[0066] Step S209: Update the target position estimate based on the correction amount;

[0067] Step S210: Repeat the above iterative process until the norm of the correction is less than the preset threshold or the number of iterations reaches the preset maximum number of iterations, and obtain the optimal target global coordinates.

[0068] Furthermore, step S205 includes the following steps:

[0069] Step S301: Calculate the difference between the predicted observation value and the actual observation value corresponding to each camera to obtain the residual;

[0070] Step S302: Calculate the weight of each camera based on the predicted observation value of each camera.

[0071] Step S303: Construct a weight matrix by assigning weights to each camera.

[0072] Specifically, the global ECEF coordinates obtained from each camera are transformed using the Iterative Weighted Least Squares (IWLS) algorithm. The optimal global coordinates of the target are obtained by fusing the data, and used as the initial position for subsequent Kalman filter tracking.

[0073] Let the ECEF global coordinates of the target after data fusion from multiple cameras be... (Optimal solution to be determined). First, weighted least squares (WLS) is used to independently convert the observation data of each camera into global ECEF coordinates and perform a weighted average. The result is used as the initial position estimate for the IWLS algorithm. Next, iterative optimization is performed. In each iteration, the estimated values ​​of the target parameters are updated through the following process:

[0074] Assume the target position estimate in the t-th iteration is Substitute it into the observation model In the model for transforming camera relative coordinates to global coordinates, the predicted observations for each camera are calculated. Then, the difference between the predicted observations and the actual observations (i.e., the initial input observations) is calculated to obtain the residuals. A fixed weight is calculated for the predicted observations of each camera, forming a weight matrix. For each camera k, at the current estimated point At this point, calculate the observation model. The Jacobian matrix is ​​given by formula (4):

[0075] (4)

[0076] Using Jacobi matrix Weight matrix and residual Construct a position correction amount The normal equation is given by formula (5):

[0077] (5)

[0078] After the iteration cycle ends, the correction amount is obtained by solving the above equation. Then update the target location estimate, referring to formula (6):

[0079] (6)

[0080] Repeat the above iterative process until the correction amount is reached. The norm is less than a preset threshold This process continues until the preset maximum number of iterations is reached, thus obtaining the optimal ECEF coordinate estimate. Then, it is converted into geodetic latitude, longitude, and altitude coordinates and output.

[0081] Furthermore, step S106 includes the following steps:

[0082] Step S401: When the velocity component is initialized to 0, the initial state vector is obtained based on the position component of the initial state.

[0083] Step S402: In the prediction stage under the uniform motion model, based on the posterior state vector and state transition matrix of the previous moment, predict the target prior state vector and state covariance at the current moment.

[0084] Step S403: In the update phase, the prior state vector and state covariance of the target are corrected to obtain the posterior optimal state estimate and posterior optimal covariance at the current time.

[0085] Furthermore, step S403 includes the following steps:

[0086] Step S501: Calculate the observation residuals based on the optimal global target coordinates, the target prior state vector, and the observation matrix;

[0087] Step S502: Adjust the adaptive factor according to the norm of the observation residuals and adjust the observation noise covariance matrix to obtain the adjusted observation noise covariance matrix.

[0088] Step S503: Calculate the Kalman gain based on the adjusted observation noise covariance matrix and state covariance.

[0089] Step S504: Correct the prior state vector of the target using Kalman gain to obtain the posterior optimal state estimate at the current time.

[0090] Step S505: Correct the state covariance using Kalman gain to obtain the posterior optimal covariance at the current time.

[0091] Specifically, an improved Kalman filter is designed to track the target's trajectory. In the Kalman filter model, considering the target's motion characteristics, a 6-dimensional state vector containing position and velocity is defined. .

[0092] Using the fused target position (ECEF coordinates) as the position component of the initial state, and initializing the velocity component to zero, we obtain the initial state vector. .

[0093] In the uniform velocity (CV) motion model, during the prediction phase, the posterior state vector from the previous moment is used as the basis. and state transition matrix Predict the prior state vector of the target at the current time k. State covariance Referring to formulas (7) and (8):

[0094] (7)

[0095] (8)

[0096] in, Here is the state transition matrix. The sampling period (determined by the device's sampling frequency). It is a 3-order identity matrix. is a 3rd order zero matrix; Q is the process noise covariance matrix.

[0097] Unlike traditional Kalman filters, this application introduces a residual adaptive mechanism to dynamically adjust the observation noise covariance R.

[0098] During the update phase, the prior states and prior covariance obtained in the "prediction phase" are corrected, and the posterior optimal state estimate for the current time step is finally output. and posterior optimal covariance .

[0099] First calculate the observation residuals , refer to formula (9):

[0100] (9)

[0101] in, The observation vector at time k (i.e., the ECEF coordinates obtained by fusion) represents the ECEF position after fusion of multiple cameras. This is the observation matrix.

[0102] Then, based on the magnitude of the observation residual norm, the adaptive factor is adjusted, thereby adjusting the observation noise covariance matrix, referring to formula (10):

[0103] (10)

[0104] in, Based on observation noise; To observe noise The adaptive adjustment factor.

[0105] Use the adjusted Calculate Kalman gain Referring to formulas (11) and (12):

[0106] (11)

[0107] (12)

[0108] in, R is the covariance; R is the observation noise covariance matrix; The observation matrix; Let be the state covariance.

[0109] Then, the prior state is corrected to obtain the optimal state estimate at the current moment, referring to formula (13):

[0110] (13)

[0111] Update the posterior optimal covariance, referring to formula (14):

[0112] (14)

[0113] Example 2:

[0114] Figure 3 This is a schematic diagram of a high-precision real-time multi-channel collaborative control system provided in Embodiment 2 of the present invention.

[0115] Reference Figure 3 The system includes:

[0116] The LLA coordinate acquisition module is used to acquire the LLA coordinates of each camera via GNSS. The LLA coordinates include latitude, longitude and elevation.

[0117] The conversion module is used to convert the LLA coordinates of each camera into the first coordinates in the global ECEF coordinate system using the Earth ellipsoid model;

[0118] The data acquisition module is used to synchronously acquire observation data from each camera after the azimuth and elevation zero points of the cameras have been calibrated; the observation data includes distance, azimuth, and elevation.

[0119] The mapping module is used to obtain the target's coordinates in the camera's local coordinate system through trigonometric functions, and then map the target's coordinates in the camera's local coordinate system to the global ECEF coordinate system.

[0120] The fusion module is used to fuse the global ECEF coordinates obtained from coordinate transformations in the local coordinate systems of all cameras using an iterative weighted least squares algorithm, thereby obtaining the optimal global coordinates of the target.

[0121] The prediction and tracking module is used to take the optimal global coordinates of the target as the position component of the initial state and input them into the Kalman filter built based on the uniform motion model to predict and track the target motion state, and obtain the posterior optimal state estimate and posterior optimal covariance at the current time.

[0122] The calculation module is used to calculate the local polar coordinates of the target in the camera to be switched.

[0123] The driving module is used to drive the camera to adjust its angle based on local polar coordinates.

[0124] Furthermore, the fusion module is specifically used for:

[0125] The observation data from each camera is converted into corresponding global ECEF coordinate points;

[0126] The corresponding global ECEF coordinates are weighted and averaged to obtain the average value.

[0127] The average value is used as the initial position estimate for the IWLS algorithm;

[0128] Set the target position estimate for the t-th iteration and substitute it into the observation model to calculate the predicted observation value for each camera;

[0129] Based on the predicted observations corresponding to each camera, obtain the residuals and weight matrices;

[0130] Calculate the Jacobian matrix of the observation model based on each camera at the current estimated point;

[0131] Based on the Jacobian matrix, weight matrix, and residuals, construct the normal equation for the position correction;

[0132] After the iteration is complete, solve the normal equation for the position correction to obtain the correction amount;

[0133] Update the target location estimate based on the correction amount;

[0134] Repeat the above iterative process until the norm of the correction is less than the preset threshold or the number of iterations reaches the preset maximum number of iterations, at which point the optimal target global coordinates are obtained.

[0135] This invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the high-precision real-time multi-channel cooperative control method provided in the above embodiments.

[0136] This invention also provides a computer-readable medium having processor-executable non-volatile program code, on which a computer program is stored, and which, when run by a processor, executes the steps of the high-precision real-time multi-channel cooperative control method described above.

[0137] The computer program product provided in this embodiment of the invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the methods described in the preceding method embodiments. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0138] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system and apparatus described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0139] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0140] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0141] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0142] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit it. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the technical scope disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A high-precision real-time multi-channel cooperative control method, characterized in that, The method includes: The LLA coordinates of each camera are obtained via GNSS, wherein the LLA coordinates include latitude, longitude and elevation; Using the Earth ellipsoid model, the LLA coordinates of each camera are converted into the first coordinates in the global ECEF coordinate system; After the azimuth and elevation zero points of the cameras are calibrated, observation data from each camera are collected synchronously; wherein, the observation data includes distance, azimuth, and elevation. The coordinates of the target in the local coordinate system of the camera are obtained by using trigonometric functions, and the coordinates of the target in the local coordinate system of the camera are mapped to the global ECEF coordinates. By using an iterative weighted least squares algorithm, the global ECEF coordinates obtained by coordinate transformation in the local coordinate system of all the cameras are fused to obtain the optimal global target coordinates. The optimal global coordinates of the target are used as the position components of the initial state and input into a Kalman filter constructed based on a uniform motion model to predict and track the target motion state, thereby obtaining the posterior optimal state estimate and posterior optimal covariance at the current moment. Calculate the local polar coordinates of the target in the camera to be switched; The camera angle is adjusted based on the local polar coordinates. By using an iterative weighted least squares algorithm, the global ECEF coordinates obtained from coordinate transformations in the local coordinate systems of all the cameras are fused to obtain the optimal global target coordinates, including: The observation data from each of the cameras are converted into corresponding global ECEF coordinate points; The corresponding global ECEF coordinate points are weighted and averaged to obtain the average value. The average value is used as the initial position estimate for the IWLS algorithm; Set the target position estimate for the t-th iteration and substitute it into the observation model to calculate the predicted observation value for each of the cameras; Based on the predicted observations corresponding to each of the cameras, obtain the residuals and weight matrices; Calculate the Jacobian matrix of the observation model for each of the cameras at the current estimated point; Based on the Jacobian matrix, the weight matrix, and the residual, construct the normal equation for the position correction; After the iteration is completed, the normal equation for the position correction is solved to obtain the correction amount; Update the target location estimate based on the correction amount; The iteration process is repeated until the norm of the correction is less than a preset threshold or the number of iterations reaches the preset maximum number of iterations, at which point the optimal target global coordinates are obtained.

2. The high-precision real-time multi-channel cooperative control method according to claim 1, characterized in that, Based on the predicted observations corresponding to each of the aforementioned cameras, the residuals and weight matrices are obtained, including: The residual is obtained by calculating the difference between the predicted observation and the actual observation for each of the cameras. Calculate the weight of each camera based on the predicted observation value corresponding to each camera; The weights corresponding to each of the cameras are used to form the weight matrix.

3. The high-precision real-time multi-channel cooperative control method according to claim 1, characterized in that, The optimal global coordinates of the target are used as the position components of the initial state and input into a Kalman filter constructed based on a uniform motion model to predict and track the target's motion state, obtaining the posterior optimal state estimate and posterior optimal covariance at the current moment, including: When the velocity component is initialized to 0, the initial state vector is obtained based on the position component of the initial state. In the prediction phase of the uniform motion model, based on the posterior state vector and state transition matrix of the previous moment, the target prior state vector and state covariance at the current moment are predicted. During the update phase, the prior state vector and the state covariance of the target are corrected to obtain the posterior optimal state estimate and the posterior optimal covariance at the current time.

4. The high-precision real-time multi-channel cooperative control method according to claim 3, characterized in that, During the update phase, the prior state vector and the state covariance of the target are corrected to obtain the posterior optimal state estimate and the posterior optimal covariance at the current time, including: The observation residual is calculated based on the optimal global coordinates of the target, the prior state vector of the target, and the observation matrix. The adaptive factor is adjusted according to the norm of the observation residuals, and the observation noise covariance matrix is ​​adjusted to obtain the adjusted observation noise covariance matrix. Calculate the Kalman gain based on the adjusted observation noise covariance matrix and the state covariance; The prior state vector of the target is corrected by the Kalman gain to obtain the posterior optimal state estimate at the current moment; The state covariance is corrected by the Kalman gain to obtain the posterior optimal covariance at the current time.

5. A high-precision real-time multi-channel cooperative control system, characterized in that, The system includes: The LLA coordinate acquisition module is used to acquire the LLA coordinates of each camera via GNSS, wherein the LLA coordinates include latitude, longitude and elevation; The conversion module is used to convert the LLA coordinates of each camera into the first coordinates in the global ECEF coordinate system using the Earth ellipsoid model; The acquisition module is used to synchronously acquire observation data from each of the cameras after the azimuth and elevation zero points of the cameras have been calibrated; wherein the observation data includes distance, azimuth, and elevation. The mapping module is used to obtain the coordinates of the target in the local coordinate system of the camera through trigonometric function relationships, and to map the coordinates of the target in the local coordinate system of the camera to the global ECEF coordinates; The fusion module is used to fuse the global ECEF coordinates obtained by coordinate transformation in the local coordinate systems of all the cameras through an iterative weighted least squares algorithm to obtain the optimal global target coordinates. The prediction and tracking module is used to take the optimal global coordinates of the target as the position component of the initial state and input them into the Kalman filter constructed based on the uniform motion model to predict and track the target motion state, so as to obtain the posterior optimal state estimate and posterior optimal covariance at the current time. A calculation module is used to calculate the local polar coordinates of the target in the camera to be switched. The driving module is used to drive the camera to adjust its angle according to the local polar coordinates; The fusion module is specifically used for: The observation data from each of the cameras are converted into corresponding global ECEF coordinate points; The corresponding global ECEF coordinate points are weighted and averaged to obtain the average value. The average value is used as the initial position estimate for the IWLS algorithm; Set the target position estimate for the t-th iteration and substitute it into the observation model to calculate the predicted observation value for each of the cameras; Based on the predicted observations corresponding to each of the cameras, obtain the residuals and weight matrices; Calculate the Jacobian matrix of the observation model for each of the cameras at the current estimated point; Based on the Jacobian matrix, the weight matrix, and the residual, construct the normal equation for the position correction; After the iteration is completed, the normal equation for the position correction is solved to obtain the correction amount; Update the target location estimate based on the correction amount; The iteration process is repeated until the norm of the correction is less than a preset threshold or the number of iterations reaches the preset maximum number of iterations, at which point the optimal target global coordinates are obtained.

6. The high-precision real-time multi-channel cooperative control system according to claim 5, characterized in that, The fusion module is specifically used for: The residual is obtained by calculating the difference between the predicted observation and the actual observation for each of the cameras. Calculate the weight of each camera based on the predicted observation value corresponding to each camera; The weights corresponding to each of the cameras are used to form the weight matrix.

7. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the method described in any one of claims 1 to 4.

8. A computer-readable medium having processor-executable non-volatile program code, characterized in that, The program code causes the processor to execute the method described in any one of claims 1 to 4.

Images