Video linkage control method and device, computer equipment and storage medium

Abstract

The invention relates to a video linkage control method and device, computer equipment and a storage medium. The method comprises the following steps: collecting multi-source positioning data from a Beidou receiver, an inertial measurement unit and an odometer in parallel, and evaluating the availability of Beidou satellite positioning signals; fusing the multi-source positioning data, and outputting real-time pose information of the to-be-monitored target; wherein when the Beidou satellite positioning signal is evaluated to be in an available state, online dynamic calibration and compensation are carried out on measurement parameters of the odometer by utilizing real-time pose information; based on the real-time pose information, predicting a future motion track of the to-be-monitored target; according to preset monitoring area gridding information, determining at least one target monitoring sub-area passed by the future motion trail; and generating a video monitoring scheduling instruction based on the target monitoring sub-region and the corresponding video monitoring device. By adopting the method, continuous high-precision positioning can be realized in a complex environment, and the video resources are intelligently scheduled according to the predicted trajectory.

Description

Technical Field

[0001] This application relates to the field of navigation and positioning technology, and in particular to a video linkage control method, device, computer equipment, computer-readable storage medium, and computer program product. Background Technology

[0002] With the development of intelligent video surveillance technology, intelligent scheduling based on target localization and trajectory prediction has become crucial for improving monitoring efficiency. Existing systems largely rely on single satellite navigation signals for positioning. In complex environments such as building obstructions and underground spaces, these signals are easily interfered with or interrupted, leading to inaccurate positioning and hindering automated tracking and response in video surveillance. While multi-sensor fusion technologies, such as combining inertial navigation and odometers, can compensate for the limitations of satellite signals to some extent, existing solutions still have limitations in deep fusion of sensor data and real-time online compensation for odometer errors. Furthermore, current video linkage control is mostly based on fixed rules or simple area triggering, lacking the ability to perform forward-looking and adaptive scheduling based on accurate target trajectory prediction, resulting in wasted monitoring resources or target loss.

[0003] Therefore, there is an urgent need for a video linkage control method, device, computer equipment, computer-readable storage medium, and computer program product that can achieve continuous high-precision positioning in complex environments and intelligently schedule video resources based on predicted trajectories to improve the robustness and accuracy of the monitoring system. Summary of the Invention

[0004] Therefore, it is necessary to provide a video linkage control method, device, computer equipment, computer-readable storage medium, and computer program product that can achieve continuous high-precision positioning in complex environments and intelligently schedule video resources based on predicted trajectories to improve the robustness and accuracy of the monitoring system, in order to address the above-mentioned technical problems.

[0005] Firstly, this application provides a video linkage control method, including:

[0006] The system collects multi-source positioning data from BeiDou receivers, inertial measurement units, and odometry in parallel, and evaluates the availability of BeiDou satellite positioning signals in real time based on the number of visible satellites and the signal-to-noise ratio.

[0007] A deep fusion navigation model based on extended Kalman filtering is constructed to fuse the multi-source positioning data, and the real-time pose information of the target to be monitored is output using the deep fusion navigation model; wherein, when the BeiDou satellite positioning signal is evaluated as usable, the measurement parameters of the odometer are dynamically calibrated and compensated online using the real-time pose information.

[0008] Based on the real-time pose information, the future motion trajectory of the target to be monitored is predicted;

[0009] Based on the preset monitoring area grid information, at least one target monitoring sub-area is determined through which the future movement trajectory will pass. The monitoring area grid information represents the coverage association relationship between the sub-area and each video monitoring device.

[0010] Based on the target monitoring sub-area and the corresponding video monitoring equipment, a video monitoring scheduling instruction is generated. The video monitoring scheduling instruction is used to trigger at least one operation among pre-activation, viewing angle adjustment or focus adjustment of the corresponding video monitoring equipment.

[0011] In one embodiment, the process of generating the preset monitoring area grid information includes:

[0012] Obtain the deployment location of each video surveillance device within the monitored area;

[0013] Based on the deployment location, the Voronoi diagram algorithm is used to divide the space and form a gridded information of the monitoring area;

[0014] According to the gridded information of the monitoring area, each sub-area is associated with a video surveillance device, and any location within the sub-area is closest to the associated video surveillance device.

[0015] In one embodiment, the assessment process for determining whether the BeiDou satellite positioning signal is in a usable state includes:

[0016] Determine whether the number of visible satellites in real time has reached a preset threshold.

[0017] Determine whether the average signal-to-carrier-to-noise ratio of all currently visible satellites has reached the preset quality threshold;

[0018] When the number of visible satellites reaches the preset number threshold and the average signal carrier-to-noise ratio reaches the preset quality threshold, the BeiDou satellite positioning signal is determined to be in the available state.

[0019] In one embodiment, the step of using the real-time pose information to perform online dynamic calibration and compensation of the odometer's measurement parameters includes:

[0020] Establish an error parameter model for the odometer;

[0021] When the BeiDou satellite positioning signal is in the available state, the parameters in the error parameter model are identified and compensated in real time online using the real-time pose information as a reference and a filtering algorithm.

[0022] In one embodiment, predicting the future trajectory of the target to be monitored includes:

[0023] Based on the real-time pose information, the current motion state of the target to be monitored is extracted;

[0024] Based on a preset motion model and a predicted time step, the position sequence of the target to be monitored in future time periods is calculated.

[0025] In one embodiment, the method further includes:

[0026] After executing the video surveillance scheduling instruction, obtain system operation feedback data;

[0027] Based on the feedback data, the parameters of the deep fusion navigation model and / or trajectory prediction model are optimized and adjusted in a closed loop.

[0028] Secondly, this application also provides a video linkage control device, comprising:

[0029] The acquisition module is used to acquire multi-source positioning data from BeiDou receivers, inertial measurement units, and odometers in parallel, and to evaluate the availability of BeiDou satellite positioning signals in real time based on the number of visible satellites and the signal-to-noise ratio.

[0030] The processing module is used to construct a deep fusion navigation model based on extended Kalman filtering, fuse the multi-source positioning data, and output the real-time pose information of the target to be monitored using the deep fusion navigation model; wherein, when the BeiDou satellite positioning signal is evaluated as usable, the real-time pose information is used to perform online dynamic calibration and compensation of the measurement parameters of the odometer.

[0031] The prediction module is used to predict the future motion trajectory of the target to be monitored based on the real-time pose information.

[0032] The processing module is also used to determine at least one target monitoring sub-area through which the future movement trajectory will pass based on the preset monitoring area grid information, wherein the monitoring area grid information represents the coverage association relationship between the sub-area and each video monitoring device;

[0033] The control module is used to generate video monitoring scheduling instructions based on the target monitoring sub-area and the corresponding video monitoring equipment. The video monitoring scheduling instructions are used to trigger at least one operation among pre-activation, viewing angle adjustment or focus adjustment of the corresponding video monitoring equipment.

[0034] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0035] The system collects multi-source positioning data from BeiDou receivers, inertial measurement units, and odometry in parallel, and evaluates the availability of BeiDou satellite positioning signals in real time based on the number of visible satellites and the signal-to-noise ratio.

[0036] A deep fusion navigation model based on extended Kalman filtering is constructed to fuse the multi-source positioning data, and the real-time pose information of the target to be monitored is output using the deep fusion navigation model; wherein, when the BeiDou satellite positioning signal is evaluated as usable, the measurement parameters of the odometer are dynamically calibrated and compensated online using the real-time pose information.

[0037] Based on the real-time pose information, the future motion trajectory of the target to be monitored is predicted;

[0038] Based on the preset monitoring area grid information, at least one target monitoring sub-area is determined through which the future movement trajectory will pass. The monitoring area grid information represents the coverage association relationship between the sub-area and each video monitoring device.

[0039] Based on the target monitoring sub-area and the corresponding video monitoring equipment, a video monitoring scheduling instruction is generated. The video monitoring scheduling instruction is used to trigger at least one operation among pre-activation, viewing angle adjustment or focus adjustment of the corresponding video monitoring equipment.

[0040] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0041] The system collects multi-source positioning data from BeiDou receivers, inertial measurement units, and odometry in parallel, and evaluates the availability of BeiDou satellite positioning signals in real time based on the number of visible satellites and the signal-to-noise ratio.

[0042] A deep fusion navigation model based on extended Kalman filtering is constructed to fuse the multi-source positioning data, and the real-time pose information of the target to be monitored is output using the deep fusion navigation model; wherein, when the BeiDou satellite positioning signal is evaluated as usable, the measurement parameters of the odometer are dynamically calibrated and compensated online using the real-time pose information.

[0043] Based on the real-time pose information, the future motion trajectory of the target to be monitored is predicted;

[0044] Based on the preset monitoring area grid information, at least one target monitoring sub-area is determined through which the future movement trajectory will pass. The monitoring area grid information represents the coverage association relationship between the sub-area and each video monitoring device.

[0045] Based on the target monitoring sub-area and the corresponding video monitoring equipment, a video monitoring scheduling instruction is generated. The video monitoring scheduling instruction is used to trigger at least one operation among pre-activation, viewing angle adjustment or focus adjustment of the corresponding video monitoring equipment.

[0046] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0047] The system collects multi-source positioning data from BeiDou receivers, inertial measurement units, and odometry in parallel, and evaluates the availability of BeiDou satellite positioning signals in real time based on the number of visible satellites and the signal-to-noise ratio.

[0048] A deep fusion navigation model based on extended Kalman filtering is constructed to fuse the multi-source positioning data, and the real-time pose information of the target to be monitored is output using the deep fusion navigation model; wherein, when the BeiDou satellite positioning signal is evaluated as usable, the measurement parameters of the odometer are dynamically calibrated and compensated online using the real-time pose information.

[0049] Based on the real-time pose information, the future motion trajectory of the target to be monitored is predicted;

[0050] Based on the preset monitoring area grid information, at least one target monitoring sub-area is determined through which the future movement trajectory will pass. The monitoring area grid information represents the coverage association relationship between the sub-area and each video monitoring device.

[0051] Based on the target monitoring sub-area and the corresponding video monitoring equipment, a video monitoring scheduling instruction is generated. The video monitoring scheduling instruction is used to trigger at least one operation among pre-activation, viewing angle adjustment or focus adjustment of the corresponding video monitoring equipment.

[0052] The aforementioned video-linked control method, device, computer equipment, computer-readable storage medium, and computer program products achieve high-precision real-time target pose information in complex environments with unstable satellite signals by parallel acquisition of multi-source sensor data and deep fusion of extended Kalman filtering. Through online dynamic calibration and compensation of odometer parameters based on satellite signal quality, the cumulative drift of positioning errors is effectively suppressed, improving the long-term stability and positioning accuracy of the system. By combining high-precision positioning information with trajectory prediction algorithms, accurate prediction of future motion trajectories is achieved. Finally, through preset grid mapping of monitoring areas and intelligent scheduling logic, the predicted trajectory is automatically converted into forward-looking control commands for specific video surveillance equipment, achieving precise matching and efficient coordination between monitoring resources and target motion. This significantly improves the tracking success rate, response speed, and automation level of the video surveillance system in dynamic and complex environments. Attached Figure Description

[0053] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0054] Figure 1 This is an application environment diagram of the video linkage control method in one embodiment;

[0055] Figure 2 This is a flowchart illustrating a video linkage control method in one embodiment;

[0056] Figure 3 This is a flowchart illustrating the video linkage control method in another embodiment;

[0057] Figure 4 This is a flowchart illustrating the video linkage control method in the most detailed embodiment;

[0058] Figure 5 This is a structural block diagram of a video linkage control device in one embodiment;

[0059] Figure 6 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0060] 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.

[0061] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0062] The video linkage control method provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on another network server.

[0063] Server 104 collects multi-source positioning data from BeiDou receiver, inertial measurement unit, and odometer in parallel through terminal 102. Server 104 evaluates the availability of BeiDou satellite positioning signals in real time based on the number of visible satellites and signal-to-noise ratio. It constructs a deep fusion navigation model based on extended Kalman filtering, fuses the multi-source positioning data, and outputs the real-time pose information of the target to be monitored using the deep fusion navigation model. When the BeiDou satellite positioning signal is evaluated as available, the measurement parameters of the odometer are dynamically calibrated and compensated online using the real-time pose information. Based on the real-time pose information, the future trajectory of the target to be monitored is predicted. According to the preset monitoring area grid information, at least one target monitoring sub-area is determined through which the future trajectory will pass. The monitoring area grid information represents the coverage association between the sub-area and each video monitoring device. Based on the target monitoring sub-area and the corresponding video monitoring device, a video monitoring scheduling command is generated. The video monitoring scheduling command is used to trigger at least one operation of pre-activation, viewing angle adjustment, or focus adjustment of the corresponding video monitoring device.

[0064] The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle systems, and projection devices. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted displays. Head-mounted displays can be virtual reality (VR) devices, augmented reality (AR) devices, and smart glasses. The server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0065] In one exemplary embodiment, such as Figure 2 As shown, a video linkage control method is provided, which is applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps S202 to S210. Wherein:

[0066] Step S202: Multi-source positioning data from BeiDou receiver, inertial measurement unit and odometer are collected in parallel, and the availability of BeiDou satellite positioning signal is evaluated in real time based on the number of visible satellites and the signal-to-noise ratio.

[0067] The multi-source positioning data includes:

[0068] BeiDou satellite positioning data: originates from BeiDou receivers and provides absolute geographical coordinates, but it is easily blocked or interfered with in environments such as urban canyons and tunnels, and the signal may be unstable or lost.

[0069] Inertial Measurement Unit (IMU): Sourced from the inertial measurement unit, it provides high-frequency, continuous acceleration and angular velocity information of the carrier over a short period of time through accelerometers and gyroscopes. It can independently calculate relative displacement and attitude changes, but it has cumulative errors.

[0070] Odometer data: It comes from vehicle wheel speed sensors or visual odometers. It estimates the relative travel distance by measuring the number of wheel rotations or the displacement of image features. It has good short-term accuracy, but it is also affected by wheel slippage and uneven ground, which can cause errors.

[0071] Specifically, upon system startup, the first step is to initialize the multi-source sensors, including powering on the BeiDou receiver, inertial measurement unit, and odometer. The formula for determining the availability of BeiDou satellite positioning signals is as follows:

[0072] ;

[0073] In the formula, This refers to the number of visible satellites. It is the carrier-to-noise ratio of the i-th satellite. is the indicator function, and is the set threshold.

[0074] Step S204: Construct a deep fusion navigation model based on extended Kalman filter, fuse multi-source positioning data, and use the deep fusion navigation model to output the real-time pose information of the target to be monitored; wherein, when the BeiDou satellite positioning signal is assessed as usable, the real-time pose information is used to perform online dynamic calibration and compensation of the odometer measurement parameters.

[0075] Specifically, a deep fusion navigation model based on extended Kalman filtering is constructed. The system state parameters typically include the position, velocity, and attitude of the target to be monitored, as well as sensor error parameters (such as the scaling factor and bias of the odometer). In each processing cycle, the model first uses high-frequency data from the inertial measurement unit and the odometer to predict the target's motion state through a state equation. Subsequently, satellite observation data provided by the BeiDou receiver is used as measurement input to optimally correct the predicted state, thereby outputting the fused real-time pose information (i.e., position and attitude).

[0076] When the BeiDou satellite positioning signal is assessed as usable, the odometry measurement parameters are dynamically calibrated and compensated online using real-time pose information. When the satellite signal quality is good, the fused pose output by the model has high reliability and can be regarded as the true reference value. By comparing this high-precision reference pose with the pose derived from the current odometry's original measurement, the error parameters in the odometry measurement model (mainly the scaling factor and bias) can be deduced. Subsequently, these parameter estimates are updated in real time using a filtering algorithm and immediately used to correct the subsequent original odometry measurement data. This mechanism enables online identification and adaptive compensation of systematic errors in the odometry, effectively curbing the cumulative drift caused by long-term operation. It ensures that during periods of temporary satellite signal interruption, the system can still maintain high positioning accuracy by relying on inertial navigation and the calibrated odometry, thereby guaranteeing the continuity and reliability of positioning throughout the entire time.

[0077] The formulas for predicting and updating the state of BeiDou satellite positioning signals are as follows:

[0078] ;

[0079] In the formula, The predicted state at time k is... Here is the state transition matrix. This represents the state of k-1 at the previous time step. To control the input matrix, To control the input factor.

[0080] ;

[0081] In the formula, For the updated state estimate, For observation data from sensors, It is the observation matrix.

[0082] Step S206: Based on real-time pose information, predict the future motion trajectory of the target to be monitored.

[0083] Specifically, the trajectory prediction formula is as follows:

[0084] ;

[0085] In the formula, To predict the future location of the target to be monitored. The estimated position at the current moment. The current speed of the target to be monitored. For the acceleration of the target to be monitored, For time step.

[0086] Step S208: Based on the preset monitoring area grid information, determine at least one target monitoring sub-area that the future movement trajectory will pass through. The monitoring area grid information represents the coverage association relationship between the sub-area and each video monitoring device.

[0087] Specifically, the preset monitoring area grid information is generated based on the physical deployment location of all video surveillance devices, dividing the entire monitoring area into several logical sub-areas, and clearly specifying which specific video surveillance device (or group of devices) is responsible for primary monitoring or optimal coverage of each sub-area, ensuring that for any point in the area, the system can uniquely or preferentially determine the corresponding monitoring responsible device.

[0088] Determining at least one target monitoring sub-region that the future trajectory will traverse is an intelligent decision-making process combining spatial mapping and prediction. The system compares and analyzes a predicted continuous trajectory (a series of future location points) with the aforementioned gridded mapping rules. Through calculation, it identifies which logical sub-regions the trajectory will cross in space; these identified sub-regions are the target monitoring sub-regions.

[0089] Ultimately, the abstract spatial trajectory (the target will pass through area A and area B) can be automatically and accurately converted into specific equipment scheduling instructions (therefore, it is necessary to activate and adjust camera 1 responsible for area A and camera 2 responsible for area B in advance).

[0090] Step S210: Based on the target monitoring sub-area and the corresponding video monitoring equipment, a video monitoring scheduling instruction is generated. The video monitoring scheduling instruction is used to trigger at least one operation among pre-activation, viewing angle adjustment or focus adjustment of the corresponding video monitoring equipment.

[0091] The video surveillance scheduling function is as follows:

[0092] ;

[0093] In the formula, These are video surveillance equipment scheduling commands used for pre-activation, viewing angle adjustment, or focus adjustment of the corresponding video surveillance equipment. The predicted position of the target to be monitored at time k is given. The predicted behavior of the target to be monitored.

[0094] In the aforementioned video-linked control method, parallel acquisition of multi-source sensor data and deep fusion with extended Kalman filtering enable continuous acquisition of high-precision real-time target pose information in complex environments with unstable satellite signals. Online dynamic calibration and compensation of odometer parameters based on satellite signal quality effectively suppresses the cumulative drift of positioning errors, improving the long-term stability and positioning accuracy of the system. Combining high-precision positioning information with trajectory prediction algorithms achieves accurate prediction of future motion trajectories. Finally, through preset gridded mapping of monitoring areas and intelligent scheduling logic, the predicted trajectory is automatically converted into forward-looking control commands for specific video surveillance equipment, achieving precise matching and efficient coordination between monitoring resources and target motion. This significantly improves the tracking success rate, response speed, and automation level of the video surveillance system in dynamic and complex environments.

[0095] In one exemplary embodiment, such as Figure 3 As shown, the process of generating the preset gridded information of the monitoring area includes:

[0096] Obtain the deployment location of each video surveillance device within the monitored area;

[0097] Based on the deployment location, the Voronoi diagram algorithm is used to divide the space and form a gridded information of the monitoring area;

[0098] Based on the gridded information of the monitoring area, each sub-area is associated with a video surveillance device, and any location within the sub-area is closest to the associated video surveillance device.

[0099] Specifically, firstly, the deployment coordinates of all video surveillance devices within the monitoring area in a unified geographic coordinate system are obtained as the generators for spatial partitioning. Secondly, based on the aforementioned set of deployment locations, the Voronoi diagram algorithm from computational geometry is used to automatically partition the continuous monitoring area. This algorithm calculates based on the nearest neighbor principle, generating a set of non-overlapping convex polygonal sub-regions that cover the entire monitoring area; each sub-region is called a Voronoi cell.

[0100] The resulting gridded monitoring area defines a clear mapping relationship: each Voronoi unit is uniquely associated with one and only one video surveillance device; furthermore, for any point within that unit, its Euclidean distance to the associated device is strictly less than its distance to any other device. This characteristic mathematically guarantees the optimality and uniqueness of the spatial partitioning for a given set of devices, providing a deterministic spatial logic foundation for subsequent intelligent device scheduling based on target locations.

[0101] In this embodiment, the optimal monitoring responsibility area is automatically generated by an algorithm, replacing manual experience-based division and ensuring the accuracy and objectivity of the scheduling logic. It establishes a deterministic optimal mapping relationship between location and device, avoiding conflicts between monitoring blind spots and resource scheduling from the source.

[0102] In one exemplary embodiment, the assessment process for determining whether a BeiDou satellite positioning signal is in a usable state includes:

[0103] Determine whether the number of visible satellites in real time has reached a preset threshold.

[0104] Determine whether the average signal-to-carrier-to-noise ratio of all currently visible satellites has reached the preset quality threshold;

[0105] When the number of visible satellites reaches a preset threshold and the average signal carrier-to-noise ratio reaches a preset quality threshold, the BeiDou satellite positioning signal is determined to be available.

[0106] Specifically, firstly, the availability of satellite geometry is assessed to determine whether the number of visible satellites required for receiver processing at the current moment is not less than a preset threshold. This threshold is typically set based on the positioning model (e.g., at least four satellites are required for 3D positioning) and system redundancy requirements. Secondly, the reliability of signal reception quality is assessed by calculating the arithmetic mean of the carrier-to-noise ratio (C / N0) of all currently visible satellite signals and determining whether it reaches a preset quality threshold. This threshold is typically set based on receiver sensitivity and anti-interference threshold.

[0107] Finally, a comprehensive judgment is made using logic and criteria: the system determines that the BeiDou satellite positioning signal is currently available only when both the aforementioned conditions regarding the number of visible satellites and the average signal-to-noise ratio are met. This judgment result will serve as a prerequisite enabling condition for key operations such as weighted strategy selection in subsequent navigation fusion algorithms and online odometer calibration.

[0108] In this embodiment, the evaluation method achieves refined and adaptive discrimination of the reliability of BeiDou positioning signals by fusing the dual criteria of satellite quantity and signal quality. It effectively distinguishes scenarios where signals are available but unusable, providing a high-confidence mode-switching basis for subsequent multi-source fusion navigation algorithms. This avoids introducing significant errors by blindly trusting satellite data when signal geometry is poor or the signal-to-noise ratio is too low, thereby improving the robustness and overall positioning accuracy of the fusion system under complex electromagnetic and environmental interference.

[0109] In one exemplary embodiment, such as Figure 3 As shown, online dynamic calibration and compensation of the odometer's measurement parameters are performed using real-time pose information, including:

[0110] Step S302: Establish the error parameter model of the odometer;

[0111] Step S304: When the BeiDou satellite positioning signal is available, the parameters in the error parameter model are identified and compensated in real time online using a filtering algorithm based on the real-time pose information.

[0112] The odometer measurement formula is as follows:

[0113] ;

[0114] In the formula, For the measurement gain of the odometer, The scaling factor for the odometer. The vehicle speed measured by the odometer. For the odometer bias, This is the noise term.

[0115] Specifically, establishing an error parameter model for the odometer is a prerequisite for calibration. There is a nonlinear mapping error between the odometer's raw measurements (such as pulse counting) and the actual motion, mainly including scaling factor error (reflecting the deviation in the proportional relationship between the measured value and the actual displacement) and bias error (reflecting the output offset at zero input). This model parameterizes these errors, making them estimable state variables.

[0116] The real-time pose information serves as the reference source for calibration. When the BeiDou satellite positioning signal is comprehensively evaluated as usable, the real-time pose (position, attitude) output by the deep fusion navigation model has high absolute accuracy and reliability, and can be used as the ground truth or high-precision reference benchmark required for calibration.

[0117] The calibration execution mechanism is described by using a filtering algorithm to identify and compensate parameters in the error parameter model in real time. The odometer's error parameters are incorporated into the state vector of state estimation algorithms such as Extended Kalman Filtering. During periods when BeiDou signal is available, the algorithm continuously compares the pose calculated based on the original odometer measurements and the current estimates of the error parameters with the reference pose from the fusion model. Using this difference (i.e., the innovation), the filtering algorithm updates the optimal estimates of the error parameters (scaling factor, bias) online and recursively. Subsequently, the updated parameters are immediately used to correct the real-time measurement output of the odometer, thereby achieving dynamic compensation for its errors.

[0118] In this embodiment, using high-precision fusion positioning results as a benchmark, the error parameters of the odometer (such as the scaling factor and offset) are estimated and corrected online in real time, effectively suppressing the cumulative drift of the odometer. This significantly improves the autonomous navigation accuracy and endurance during periods of limited or interrupted satellite signals, enhances the long-term stability and environmental adaptability of positioning, and thus ensures the continuous and reliable location information in all scenarios, providing stable and accurate trajectory input for intelligent video linkage.

[0119] In one exemplary embodiment, predicting the future trajectory of the target to be monitored includes:

[0120] Based on real-time pose information, extract the current motion state of the target to be monitored;

[0121] Based on a preset motion model and predicted time step, the position sequence of the target to be monitored in future time periods is estimated.

[0122] Specifically, the system does not directly use the raw pose information stream, but instead calculates and extracts key dynamic state variables for motion extrapolation in real time, mainly including the current three-dimensional velocity vector and the current acceleration vector. This step transforms the positioning data into initial conditions that conform to the laws of kinematics.

[0123] The system inputs the extracted current motion state as initial values ​​into a pre-defined parametric motion dynamics model. This model mathematically describes the evolution of the state over time, and its complexity can be selected as needed, such as a uniform velocity model, a uniform acceleration (CA) model, or a cooperative turning (CT) model. Combined with a set prediction time step (Δt) or prediction time series, it iterative forward integration is performed through the model's discrete state transition equations to systematically deduce the set of predicted positions of the target at multiple consecutive future moments, thus forming an ordered sequence of future spatiotemporal positions.

[0124] In this embodiment, high-precision real-time positioning information is transformed into trajectory prediction with time foresight, providing a crucial state prediction window for the video surveillance system. This enables the system to calculate and schedule monitoring resources in advance, achieving seamless pre-alignment and relay tracking of cameras. It fundamentally eliminates tracking interruptions caused by device response delays or blind spots, significantly improving the continuity, automation level, and tracking success rate of monitoring in complex dynamic environments.

[0125] In one exemplary embodiment, the method further includes:

[0126] After executing video surveillance scheduling instructions, obtain system operation feedback data;

[0127] Based on feedback data, the parameters of the deep fusion navigation model and / or trajectory prediction model are optimized and adjusted in a closed loop.

[0128] Specifically, after a video surveillance dispatch command is executed, the system actively collects two types of key operational feedback data: first, feedback information from the video surveillance equipment itself, such as the actual arrival accuracy of preset positions, focus status, and image quality, which is used to evaluate the execution effect of the dispatch command; second, the actual motion data of the monitored target. During the period when the target is successfully tracked, its more accurate actual trajectory can be obtained through video analysis, radar, or other auxiliary sensors, which is used to compare with the predicted trajectory inside the system.

[0129] The system uses collected feedback data, particularly the deviation between predicted and actual trajectories, and the differences between scheduling instructions and execution results, as input for performance evaluation. Based on this, adaptive filtering, machine learning, or optimization algorithms (such as gradient-based adjustment) are employed to perform online, recursive parameter fine-tuning or weight updates on the front-end deep fusion navigation model (e.g., parameters such as process noise covariance and observation noise covariance) and / or trajectory prediction model (e.g., motion model selection parameters and maneuver uncertainty parameters).

[0130] The normalized innovation formula for closed-loop optimization adjustment is as follows:

[0131] ;

[0132] In the formula, NIS represents the normalized squared innovation. For innovation vectors, To innovate the covariance matrix.

[0133] In this embodiment, by analyzing the command execution effect and actual motion data, the core model parameters such as navigation filtering and trajectory prediction are fine-tuned online, enabling the system to automatically adapt to environmental changes and differences in target behavior. The technical effect is a significant improvement in the system's long-term positioning accuracy, prediction reliability, and scheduling matching degree, enhancing its overall adaptability and robustness.

[0134] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0135] The most detailed embodiment of this application is as follows:

[0136] like Figure 4 As shown, after the system powers on, it first initializes the BeiDou satellite positioning receiver, inertial measurement unit, and odometer to acquire the initial position, velocity, and attitude information of the target to be monitored. Then, based on the geographical location of each video surveillance device within the monitoring area, the Voronoi diagram algorithm is used to perform spatial gridding.

[0137] In practical implementation, suppose there are n video surveillance devices within the monitoring area, with their location coordinates {p1, p2, ..., pn}. The system automatically uses these coordinate points as generators of the Voronoi diagram, dividing the monitoring area into n non-overlapping convex polygonal sub-regions {V1, V2, ..., Vn}, and establishing a mapping relationship between the sub-regions and the devices. For any point q, if d(q, pn) = ... i ) <d(q,p j If for all j ≠ i, then point q belongs to subregion V. i , where d(·) represents the Euclidean distance.

[0138] The system collects positioning data in parallel from the BeiDou receiver, inertial measurement unit, and odometer. For BeiDou satellite signals, the system monitors the number of visible satellites and the signal-to-noise ratio in real time, and determines signal availability based on preset thresholds.

[0139] During implementation, the system collects data from each sensor at a fixed frequency (e.g., 10Hz). When the number of visible satellites is greater than or equal to a preset threshold N_min (e.g., N_min=4), and the average carrier-to-noise ratio of all visible satellites is greater than or equal to a preset quality threshold CN0_min (e.g., CN0_min=35dB-Hz), the BeiDou satellite positioning signal is determined to be available; otherwise, it is determined to be unavailable.

[0140] The system constructs a deep fusion navigation model based on extended Kalman filtering. The state vector includes position, velocity, attitude, sensor error parameters, etc. Multi-source data fusion is achieved through two steps: state prediction and measurement update.

[0141] In practical implementation, the system's state vector is designed as follows:

[0142] X=[x,y,z,v_x,v_y,v_z,φ,θ,ψ,b_ax,b_ay,b_az,b_gx,b_gy,b_gz,k,b]^T.

[0143] Where (x,y,z) is the three-dimensional position, (v_x,v_y,v_z) is the three-dimensional velocity, (φ,θ,ψ) is the three-axis attitude angle, (b_ax,b_ay,b_az) is the accelerometer zero bias, (b_gx,b_gy,b_gz) is the gyroscope zero bias, k is the odometer scaling factor, and b is the odometer bias.

[0144] When the BeiDou satellite positioning signal is assessed as available, the system activates the dynamic calibration procedure for the odometer parameters. Using the high-precision pose information fused at this time as a reference, the odometer's scaling factor and bias parameters are estimated online through a filtering algorithm.

[0145] During implementation, the system records the correspondence between the speed information output by the fusion model and the original odometer measurements during the available BeiDou signal periods. The odometer error model parameters are updated in real time using methods such as least squares or Kalman filtering. The updated parameters are immediately used to correct subsequent odometer measurements, achieving error compensation.

[0146] Based on the fused real-time pose information, the system extracts the target's current motion state, including position, velocity, and acceleration. Combined with a pre-defined motion model, it predicts the target's trajectory over a future period.

[0147] In practical implementation, the system uses a uniform acceleration motion model for trajectory prediction:

[0148] p(t+Δt)=p(t)+v(t)Δt+0.5a(t)Δt 2 ;

[0149] Where p(t) is the current position, v(t) is the current velocity, a(t) is the current acceleration, and Δt is the prediction time step. The system can predict the position sequence at multiple future moments, forming a predicted trajectory.

[0150] Based on the predicted trajectory, the system queries the pre-defined gridded information of the monitoring area to determine one or more target monitoring sub-areas that the trajectory will pass through. Based on the mapping relationship between the sub-areas and the video surveillance equipment, corresponding scheduling instructions are generated.

[0151] During implementation, the scheduling command includes the device ID, operation type, and parameter settings. Operation types include pre-activation, viewing angle adjustment, and focus adjustment. Parameter settings are calculated based on the predicted target position and direction of movement to ensure that the corresponding device is in optimal monitoring condition when the target enters the monitoring area.

[0152] The system evaluates filter consistency using a normalized innovation square formula and performs closed-loop optimization of navigation model parameters based on the evaluation results. Simultaneously, the system collects video scheduling execution performance and actual monitoring data to optimize the trajectory prediction model and scheduling strategy.

[0153] In practice, the system maintains a sliding time window and collects performance metrics within that window. When a metric deviates from a preset range, a parameter adjustment program is automatically triggered to continuously optimize system performance.

[0154] In extreme environments where satellite navigation signals are continuously unavailable, the system switches to a dead reckoning mode based on inertial navigation and odometry, and adjusts the video scheduling strategy to a more conservative wide-area monitoring mode to maintain basic monitoring functions.

[0155] The above embodiments illustrate the basic implementation scheme of the present invention. In practical applications, parameters and algorithm details can be adjusted according to specific needs, and these adjustments should all be considered to be within the protection scope of the present invention.

[0156] Based on the same inventive concept, this application also provides a video linkage control device for implementing the video linkage control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more video linkage control device embodiments provided below can be found in the limitations of the video linkage control method described above, and will not be repeated here.

[0157] In one exemplary embodiment, such as Figure 5 As shown, a video linkage control device is provided, comprising:

[0158] The acquisition module 502 is used to acquire multi-source positioning data from the BeiDou receiver, inertial measurement unit and odometer in parallel, and to evaluate the availability of BeiDou satellite positioning signals in real time based on the number of visible satellites and the signal-to-noise ratio.

[0159] The processing module 504 is used to construct a deep fusion navigation model based on extended Kalman filtering, fuse multi-source positioning data, and output the real-time pose information of the target to be monitored using the deep fusion navigation model; wherein, when the BeiDou satellite positioning signal is assessed as usable, the real-time pose information is used to perform online dynamic calibration and compensation of the odometer measurement parameters.

[0160] Prediction module 506 is used to predict the future motion trajectory of the target to be monitored based on real-time pose information;

[0161] The processing module 504 is also used to determine at least one target monitoring sub-area through which the future movement trajectory will pass based on the preset monitoring area grid information, wherein the monitoring area grid information represents the coverage association relationship between the sub-area and each video monitoring device;

[0162] The control module 508 is used to generate video monitoring scheduling instructions based on the target monitoring sub-area and the corresponding video monitoring equipment. The video monitoring scheduling instructions are used to trigger at least one operation among pre-activation, viewing angle adjustment or focus adjustment of the corresponding video monitoring equipment.

[0163] In an exemplary embodiment, the processing module 504 is further configured to obtain the deployment location of each video surveillance device within the monitoring area; based on the deployment location, the Voronoi diagram algorithm is used to divide the space to form gridded information of the monitoring area; wherein, according to the gridded information of the monitoring area, each sub-area is associated with a video surveillance device, and any location within the sub-area is closest to the associated video surveillance device.

[0164] In an exemplary embodiment, the processing module 504 is further configured to determine whether the number of visible satellites in real time has reached a preset number threshold; determine whether the average signal-to-noise ratio of all currently visible satellites has reached a preset quality threshold; and determine that the BeiDou satellite positioning signal is in an available state when the number of visible satellites reaches the preset number threshold and the average signal-to-noise ratio reaches the preset quality threshold.

[0165] In an exemplary embodiment, the processing module 504 is further configured to establish an error parameter model for the odometer; when the BeiDou satellite positioning signal is available, the parameters in the error parameter model are identified and compensated in real time online using a filtering algorithm based on real-time pose information.

[0166] In an exemplary embodiment, the prediction module 506 is further configured to extract the current motion state of the target to be monitored based on real-time pose information; and to estimate the position sequence of the target to be monitored in a future period based on a preset motion model and prediction time step.

[0167] In an exemplary embodiment, the adjustment module is configured to obtain system operation feedback data after executing video surveillance scheduling instructions; and to perform closed-loop optimization adjustment of the parameters of the deep fusion navigation model and / or trajectory prediction model based on the feedback data.

[0168] Each module in the aforementioned video linkage control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0169] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 6As shown, this computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media. The database stores multi-source positioning data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a video-linked control method.

[0170] Those skilled in the art will understand that Figure 6 The 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.

[0171] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0172] 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 above-described method.

[0173] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps of the method described above.

[0174] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0175] 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, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0176] 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 application.

[0177] 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 application should be determined by the appended claims.

Claims

1. A video linkage control method, characterized in that, The method includes: The system collects multi-source positioning data from BeiDou receivers, inertial measurement units, and odometry in parallel, and evaluates the availability of BeiDou satellite positioning signals in real time based on the number of visible satellites and the signal-to-noise ratio. A deep fusion navigation model based on extended Kalman filtering is constructed to fuse the multi-source positioning data, and the real-time pose information of the target to be monitored is output using the deep fusion navigation model; wherein, when the BeiDou satellite positioning signal is evaluated as usable, the measurement parameters of the odometer are dynamically calibrated and compensated online using the real-time pose information. Based on the real-time pose information, the future motion trajectory of the target to be monitored is predicted; Based on the preset monitoring area grid information, at least one target monitoring sub-area is determined through which the future movement trajectory will pass. The monitoring area grid information represents the coverage association relationship between the sub-area and each video monitoring device. Based on the target monitoring sub-area and the corresponding video monitoring equipment, a video monitoring scheduling instruction is generated. The video monitoring scheduling instruction is used to trigger at least one operation among pre-activation, viewing angle adjustment or focus adjustment of the corresponding video monitoring equipment.

2. The method according to claim 1, characterized in that, The process of generating the preset gridded information of the monitoring area includes: Obtain the deployment location of each video surveillance device within the monitored area; Based on the deployment location, the Voronoi diagram algorithm is used to divide the space and form a gridded information of the monitoring area; According to the gridded information of the monitoring area, each sub-area is associated with a video surveillance device, and any location within the sub-area is closest to the associated video surveillance device.

3. The method according to claim 1, characterized in that, The assessment process for determining whether the BeiDou satellite positioning signal is in a usable state includes: Determine whether the number of visible satellites in real time has reached a preset threshold. Determine whether the average signal-to-carrier-to-noise ratio of all currently visible satellites has reached the preset quality threshold; When the number of visible satellites reaches the preset number threshold and the average signal carrier-to-noise ratio reaches the preset quality threshold, the BeiDou satellite positioning signal is determined to be in the available state.

4. The method according to claim 1, characterized in that, The step of using the real-time pose information to perform online dynamic calibration and compensation of the odometer's measurement parameters includes: Establish an error parameter model for the odometer; When the BeiDou satellite positioning signal is in the available state, the parameters in the error parameter model are identified and compensated in real time online using the real-time pose information as a reference and a filtering algorithm.

5. The method according to claim 1, characterized in that, The prediction of the future trajectory of the target to be monitored includes: Based on the real-time pose information, the current motion state of the target to be monitored is extracted; Based on a preset motion model and a predicted time step, the position sequence of the target to be monitored in future time periods is calculated.

6. The method according to claim 1, characterized in that, The method further includes: After executing the video surveillance scheduling instruction, obtain system operation feedback data; Based on the feedback data, the parameters of the deep fusion navigation model and / or trajectory prediction model are optimized and adjusted in a closed loop.

7. A video linkage control device, characterized in that, The device includes: The acquisition module is used to acquire multi-source positioning data from BeiDou receivers, inertial measurement units, and odometers in parallel, and to evaluate the availability of BeiDou satellite positioning signals in real time based on the number of visible satellites and the signal-to-noise ratio. The processing module is used to construct a deep fusion navigation model based on extended Kalman filtering, fuse the multi-source positioning data, and output the real-time pose information of the target to be monitored using the deep fusion navigation model; wherein, when the BeiDou satellite positioning signal is evaluated as usable, the real-time pose information is used to perform online dynamic calibration and compensation of the measurement parameters of the odometer. The prediction module is used to predict the future motion trajectory of the target to be monitored based on the real-time pose information. The processing module is also used to determine at least one target monitoring sub-area through which the future movement trajectory will pass based on the preset monitoring area grid information, wherein the monitoring area grid information represents the coverage association relationship between the sub-area and each video monitoring device; The control module is used to generate video monitoring scheduling instructions based on the target monitoring sub-area and the corresponding video monitoring equipment. The video monitoring scheduling instructions are used to trigger at least one operation among pre-activation, viewing angle adjustment or focus adjustment of the corresponding video monitoring equipment.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

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

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images