A security guard position calibration method based on space-time trajectory

Abstract

The application relates to the technical field of security personnel positioning, in particular to a security personnel position calibration method based on space-time trajectories. The method comprises the following steps: collecting terminal positions, motion characteristics and environment parameters to perform data alignment, and establishing initial space-time trajectories; identifying a patrol path boundary and a motion characteristic cycle, and intercepting trajectory data groups; extracting trajectory abnormal center features and correcting the motion characteristic cycle through local abnormal factor analysis; obtaining basic motion points and offset extreme nodes, and combining space and environment interference coordinate distances to establish a cooperative interval; obtaining a position deviation minimum value based on an approximation algorithm, and distributing fusion weight coefficients and calibration parameters to perform trajectory smoothing. Through dynamic calibration driven by multi-source data, the application effectively overcomes the influence of signal shielding and dynamic interference, and significantly improves positioning accuracy, trajectory coherence and system stability.

Description

Technical Field

[0001] This invention relates to the field of security monitoring and positioning technology, specifically to a method for calibrating the location of security personnel based on spatiotemporal trajectories. Background Technology

[0002] With the increasing intelligence of security systems, real-time location tracking and trajectory management of security personnel have become crucial for improving security efficiency. In the security operation system, location calibration, as a core link connecting front-end sensing and back-end dispatch, directly impacts the timeliness of emergency response and the overall coordination of security work. Therefore, establishing a scientific personnel location calibration mechanism plays a vital supporting role in accurately allocating human resources, optimizing patrol routes, and ensuring public safety.

[0003] Among them, the spatiotemporal trajectory-based position calibration method aims to construct a calibration model capable of eliminating positioning deviations by monitoring the movement trajectory of security personnel in complex environments, multi-source sensing data, and spatial constraint parameters from multiple dimensions. This technical direction mainly focuses on using multi-source data fusion methods to collaboratively analyze the real-time movement status and historical trajectory patterns of personnel, thereby achieving the technical goal of evolving security personnel positioning from single discrete coordinates to continuous and accurate trajectories through refined anomaly identification and trajectory optimization algorithms.

[0004] However, existing technologies largely rely on static calibration modes using single positioning techniques, making them ill-suited to signal obstruction and dynamic interference in complex environments both inside and outside buildings. Traditional methods often exhibit isolated characteristics during the data analysis phase, lacking spatiotemporal alignment and collaborative analysis between real-time location data, trajectory continuity, and the spatial boundaries of the activity area. This leads to inaccurate identification of abnormal locations and inconsistent calibration dimensions. Furthermore, the mining of trajectory movement patterns lacks scientific basis, particularly in the adaptive adjustment of multi-source data fusion strategies, failing to fully consider the guiding role of environmental characteristics, making it difficult to achieve accurate correlation between spatiotemporal trajectories and historical patterns. In addition, the calibration process lacks a global optimization mechanism for the entire trajectory, resulting in insufficient accuracy in correcting positioning drift and abnormal jumps. Moreover, the verification phase during trajectory optimization lacks a dynamic benchmark, making it impossible to adjust calibration parameters according to real-time scene characteristics. Consequently, the stability and reliability of security positioning systems fail to meet current requirements. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a method for calibrating the location of security personnel based on the correlation between spatiotemporal trajectories and historical patterns. The multi-source data fusion and intelligent algorithm significantly improve the accuracy of security personnel positioning, trajectory consistency, and system operation stability.

[0006] To achieve the above objectives, this invention provides a method for calibrating the location of security personnel based on the correlation between spatiotemporal trajectories and historical patterns, comprising: After collecting and aligning the terminal location data, motion state characteristics, and environmental perception parameters of security personnel, the system obtains the real-time spatial location characteristics of personnel based on the terminal location data and establishes the initial spatiotemporal trajectory of security personnel. Spatial data of the activity area is collected and processed for boundary analysis to identify the patrol path boundary and motion characteristic period, and the initial spatiotemporal trajectory data is processed based on the motion characteristic period to determine several spatiotemporal trajectory data groups. Local anomaly factor analysis is performed on the spatiotemporal trajectory data group to extract the trajectory anomaly center features corresponding to each spatiotemporal trajectory data group, and the motion feature period is corrected based on the trajectory anomaly center features. The initial spatiotemporal trajectory is processed based on the motion characteristic period to obtain the basic motion point nodes and motion offset extreme value nodes of the security personnel. Based on the environmental perception parameters in the same coordinate system, the spatial position coordinate distance and environmental interference coordinate distance to the expected predicted point node and the extreme value node of motion offset of the security personnel are obtained respectively. Based on the value range relationship of the spatial position coordinate distance and the environmental interference coordinate distance, a collaborative interval for point calibration and trajectory optimization is established. The approximation algorithm is used to obtain the minimum position deviation from multi-source sensing data, expected prediction point nodes, and extreme value nodes of motion offset. Based on the minimum position deviation and the corresponding real-time motion state features, fusion weight coefficients and matching calibration parameters are assigned to perform trajectory smoothing to complete the position calibration of security personnel.

[0007] Preferably, the step of collecting and aligning security personnel terminal location data, motion state characteristics, and environmental perception parameters, and then obtaining real-time spatial location characteristics of personnel based on the terminal location data includes: Establish the starting time for collecting security personnel location data; Based on the aforementioned start time, the terminal location data is periodically collected and processed to obtain latitude and longitude coordinates, horizontal position accuracy, vertical height, and signal strength, and to establish personnel positioning data characteristics. Based on the starting time, the motion state characteristics are periodically collected and processed to obtain acceleration, angular velocity and number of steps respectively, and to establish the motion state data characteristics of the person. The personnel positioning data features and personnel movement state data features are subjected to time synchronization calibration processing, and real-time spatial point features are calculated based on the personnel positioning data features. Among them, the environmental perception parameters are the geomagnetic intensity and air pressure data of the environment in which the person is located.

[0008] Furthermore, the spatial data of the collected activity area undergoes boundary analysis processing to identify the patrol path boundaries and motion characteristic cycles, including: Collect the boundary coordinates of the security zone and the route planning to establish spatial data of the activity area; The patrol path boundary is defined based on the spatial topological relationship corresponding to the boundary coordinates. The travel path planning is divided into movement characteristic cycles based on the changes in personnel patrol rate.

[0009] Furthermore, local anomaly factor analysis is performed on the spatiotemporal trajectory data sets to extract the trajectory anomaly center features corresponding to each spatiotemporal trajectory data set, including: Spatial trajectory density analysis is performed based on the spatiotemporal trajectory data set to obtain nodes with different position jump characteristics and nodes with different signal occlusion responses. Based on the different nodes with different position jump characteristics and different nodes with different signal blockage responses, the corresponding node feature values ​​are obtained and associated to obtain the different center values ​​of trajectory offset and different center values ​​of environmental interference respectively. Based on the trajectory offset anomaly center value and the environmental interference anomaly center value, drift point elimination processing is performed to obtain the trajectory anomaly center features corresponding to each spatiotemporal trajectory data group.

[0010] Furthermore, the correction process for the motion feature period based on the cluster center features includes: Based on the mean standard deviation of the trajectory offset dissimilar center value and the environmental interference dissimilar center value corresponding to the trajectory anomaly center feature, a trajectory verification threshold is established. Determine whether the number of abnormal centers corresponding to each spatiotemporal trajectory data group is 1, and determine whether the deviation of the trajectory offset dissimilar center value and the environmental interference dissimilar center value corresponding to the trajectory abnormal center feature is less than the trajectory verification threshold. When the number of abnormal centers in the same spatiotemporal trajectory data group is not 1, and the deviation of each abnormal center is not less than the trajectory verification threshold, the motion feature period is shortened by sliding window based on the deviation of the abnormal center and the trajectory verification threshold, and the spatiotemporal trajectory data is re-trimmed. When the number of anomaly centers corresponding to each spatiotemporal trajectory data group is 1, and the deviation of the anomaly center is less than the trajectory verification threshold, the correction process is completed.

[0011] Furthermore, based on the motion characteristic period, the initial spatiotemporal trajectory is truncated to obtain the basic motion point nodes and motion offset extreme value nodes of the security personnel, including: Based on the ratio of the real-time spatial location features to the center value of the environmental interference, the trajectory interception threshold is defined, and the initial spatiotemporal trajectory of the personnel is intercepted based on the motion feature period to obtain several spatiotemporal trajectory intervals. Based on the trajectory interception threshold, the spatiotemporal trajectory intervals are smoothed, and jump points in each spatiotemporal trajectory interval that exceed the trajectory interception threshold are deleted. The median coordinate values ​​corresponding to the several spatiotemporal trajectory intervals are collected to divide them into basic movement point nodes for security personnel, and the extreme values ​​of coordinate values ​​corresponding to the several spatiotemporal trajectory intervals are collected to divide them into movement offset extreme value nodes.

[0012] Furthermore, the establishment of the collaborative interval for point calibration and trajectory optimization includes: A unified spatiotemporal coordinate system is established with the horizontal position coordinates as the XY axis and the vertical height coordinates as the Z axis. Based on the unified spatiotemporal coordinate system, coordinate mapping processing is performed on the basic movement point nodes, the extreme value nodes of movement offset, and the boundary of the activity area of ​​the personnel. The Euclidean distances of the basic movement point nodes, the extreme value nodes of movement offset, and the boundary of the activity area are calculated respectively to obtain the point offset coordinate distance and the boundary constraint coordinate distance. Based on the point offset coordinate distance and boundary constraint coordinate distance corresponding to the motion characteristic cycle, the position deviation value range is divided into intervals, and based on the product of the environmental interference dissimilar center value corresponding to the trajectory anomaly center and the trajectory interception threshold, the relative deviation reasonable value range threshold is divided into intervals. When the point offset coordinate distance and the boundary constraint coordinate distance are both within the position deviation range, and the relative distance between the point offset coordinate distance and the boundary constraint coordinate distance is within the reasonable range threshold of the relative deviation, the point's allowed activity range and trajectory smoothing requirement range are divided based on the corresponding basic motion point node and the boundary of the activity area. The intersection of the permitted activity range and the trajectory smoothing requirement range at the specified points is processed to obtain the overlapping space and establish a collaborative range.

[0013] Furthermore, the step of obtaining the minimum positional deviation based on the approximation algorithm for multi-source sensing data, expected prediction point nodes, and motion offset extreme value nodes includes: The average value of the geometric center coordinates of the collaborative interval is divided into interval center position values, and the position deviation calculation benchmark is calculated based on the interval center position values ​​and the position requirement values ​​corresponding to the preset inspection path. Based on the acceleration and angular velocity corresponding to the characteristics of the personnel's motion state data, the real-time position of the terminal trajectory is calculated and obtained, and the coordinate values ​​corresponding to the expected prediction point nodes and the coordinate values ​​corresponding to the extreme value nodes of motion offset are collected respectively. Based on the real-time position calculated from the terminal trajectory, the coordinate values ​​corresponding to the expected predicted point nodes and the coordinate values ​​corresponding to the extreme value nodes of motion offset, the parameter dimensions are uniformly processed under a unified spatiotemporal coordinate system. Based on the Kalman filter approximation algorithm and the position deviation calculation benchmark, the coordinate deviation values ​​between the real-time position calculated from the terminal trajectory and the expected predicted point nodes and the extreme value nodes of motion offset are calculated respectively to establish a position deviation dataset. Based on the trajectory interception threshold, outlier deletion processing is performed on the position deviation dataset, and the minimum deviation data corresponding to the position deviation dataset is obtained as the minimum position deviation value.

[0014] Furthermore, the trajectory smoothing process, based on the minimum position deviation and the corresponding real-time motion state features, assigns fusion weight coefficients and matching calibration parameters, including: The difference between the real-time coordinates and the basic motion point nodes in the personnel positioning data features is used to classify the real-time positioning offset. The fusion weight coefficient is obtained by defining an inverse proportionality based on the minimum positional deviation. Based on the real-time positioning offset and the fusion weight coefficient, the point is accurately calibrated and allocated, and the trajectory is optimized by matching the calibration parameters. The fusion weight coefficient ranges from 0.1 to 0.9.

[0015] Furthermore, based on the real-time positioning offset and the fusion weight coefficient, precise point calibration and allocation processing is performed, and trajectory optimization is performed by matching calibration parameters, including: Based on the fusion weight coefficient and the real-time positioning offset, the basic calibration points and the matching calibration points are allocated according to the position requirement value corresponding to the preset inspection path. When the basic calibration point is greater than the real-time positioning offset, the real-time positioning offset is used as the basic calibration point and the matching calibration point is used as the auxiliary correction amount. If the sum of the basic calibration points and the auxiliary correction amount is not greater than the real-time positioning offset, and the difference between the basic calibration points and the auxiliary correction amount is not greater than the minimum position deviation, then the trajectory iterative verification process is performed based on the basic calibration points and the auxiliary correction amount; otherwise, the fusion weight coefficient is adjusted, and the point precise calibration allocation process is returned.

[0016] Compared with the closest existing technology, the present invention has the following advantages: By employing a three-dimensional analysis approach based on "points, lines, and surfaces" and aligning multi-source data, a high-precision spatiotemporal trajectory model was constructed. Furthermore, local anomaly factor analysis and an adaptive sliding window algorithm were introduced to intelligently extract environmental interference features and identify anomalies. This effectively overcomes the impact of signal obstruction and dynamic interference on positioning accuracy. The data-driven dynamic calibration method not only solves the problems of inconsistent and asynchronous dimensions among multi-source data but also reduces positioning drift and floor misjudgment rates through iterative verification algorithms, providing reliable technical support for personnel scheduling and emergency response in security systems. Attached Figure Description

[0017] Figure 1 This is a flowchart of a security personnel location calibration method based on the correlation between spatiotemporal trajectory and historical pattern provided by the present invention; Figure 2 This is a schematic diagram illustrating the principle framework of a security personnel location calibration method based on the correlation between spatiotemporal trajectory and historical pattern provided by the present invention. Figure 3 This is a logical flowchart of a security personnel location calibration method based on the correlation between spatiotemporal trajectory and historical pattern provided by the present invention. Figure 4 This invention provides a flowchart of the collaborative interval establishment and trajectory optimization smoothing process for a security personnel location calibration method based on the correlation between spatiotemporal trajectory and historical patterns. Detailed Implementation

[0018] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments 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. Example

[0020] This invention provides a method for calibrating the location of security personnel based on spatiotemporal trajectories, such as... Figure 1 As shown, it includes: S1. After collecting the terminal location data, motion state characteristics and environmental perception parameters of security personnel respectively and performing data alignment processing, the real-time spatial location characteristics of personnel are obtained based on the terminal location data, and the initial spatiotemporal trajectory of security personnel is established. S2. Collect spatial data of the activity area and perform boundary analysis to identify the patrol path boundary and motion characteristic cycle respectively, and extract the initial spatiotemporal trajectory data based on the motion characteristic cycle to determine several spatiotemporal trajectory data groups. S3. Perform local anomaly factor analysis on the spatiotemporal trajectory data group to extract the trajectory anomaly center features corresponding to each spatiotemporal trajectory data group, and correct the motion feature period based on the trajectory anomaly center features. S4. Based on the motion characteristic period, the initial spatiotemporal trajectory is truncated to obtain the basic motion point nodes and motion offset extreme value nodes of the security personnel. S5. Based on the environmental perception parameters in the same coordinate system, obtain the spatial position coordinate distance and environmental interference coordinate distance from the expected predicted point node and the extreme value node of the motion offset of the security personnel, respectively, and establish the collaborative interval of point calibration and trajectory optimization based on the value range relationship of the spatial position coordinate distance and the environmental interference coordinate distance. S6. Based on the approximation algorithm, obtain the minimum position deviation from the multi-source sensing data, expected prediction point nodes, and extreme value nodes of motion offset. Then, based on the minimum position deviation and the corresponding real-time motion state features, assign fusion weight coefficients and matching calibration parameters to perform trajectory smoothing processing, and complete the position calibration of security personnel.

[0021] In this embodiment, a method for calibrating the location of security personnel based on the association of spatiotemporal trajectory and historical patterns is proposed. Based on the NTP (Network Time Protocol) timing protocol, it ensures that the sampling clock deviation of all sensor nodes between the security personnel's handheld terminal and the back-end management server is controlled within 10 microseconds. Based on the start time, the security personnel's terminal is periodically sampled to establish personnel positioning data features, specifically including latitude and longitude coordinate vectors, horizontal precision factor (HDOP), vertical height scalar, signal strength (RSSI), and real-time motion status.

[0022] The sensor array worn by security personnel is synchronously collected and processed, and motion state features are extracted every 50 milliseconds. The collected parameters include the output value of the three-axis accelerometer, the angular velocity of the three-axis gyroscope, the geomagnetic intensity data of the electronic compass, and the barometer data as environmental perception parameters. Specifically, the periodic processing is unified based on cubic spline interpolation algorithm and Gaussian filtering, and all heterogeneous data streams are resampled to a unified 100-millisecond reference period.

[0023] Specifically, the security personnel positioning system involved in this embodiment adopts a multi-source data fusion architecture. Terminal location data is acquired in real time through the positioning module, including but not limited to satellite positioning signals such as BeiDou, GPS, and GLONASS, as well as auxiliary positioning signals such as Wi-Fi and Bluetooth. Motion state characteristics are collected through an inertial measurement unit, including sensor data from accelerometers, gyroscopes, and other sensors. Environmental perception parameters are acquired through environmental sensors, including data from geomagnetic intensity sensors and barometric pressure sensors. Since single positioning technologies are easily affected by factors such as building obstruction and multipath effects when security personnel are working in complex environments, leading to decreased positioning accuracy, this embodiment uses spatiotemporal alignment and collaborative analysis of multi-source data to fuse location data, motion state data, and environmental perception data. This achieves multi-dimensional calibration of positioning deviations and effectively eliminates the positioning deviation problem caused by single positioning technologies.

[0024] S1 specifically includes: S1-1. Establish the start time for collecting security personnel location data; S1-2. Based on the starting time, periodically collect and process the terminal location data to obtain latitude and longitude coordinates, horizontal position accuracy, vertical height and signal strength, and establish personnel positioning data characteristics. S1-3. Based on the starting time, periodically collect and process the motion state characteristics to obtain acceleration, angular velocity and number of steps respectively, and establish the motion state data characteristics of the person. S1-4. Perform time synchronization calibration on the personnel positioning data features and personnel movement state data features, and calculate the real-time spatial point features based on the personnel positioning data features. Among them, the environmental perception parameters are the geomagnetic intensity and air pressure data of the environment in which the person is located.

[0025] Specifically, after collecting and aligning the terminal location data, motion state characteristics, and environmental perception parameters of security personnel, the system obtains the real-time spatial location characteristics of personnel based on the terminal location data and establishes the initial spatiotemporal trajectory of the security personnel.

[0026] In this embodiment, the location data of security personnel terminals is acquired in real time through a positioning module, including but not limited to satellite positioning signals such as BeiDou, GPS, and GLONASS, as well as auxiliary positioning signals such as Wi-Fi and Bluetooth. Motion state characteristics are collected through an inertial measurement unit, including data from sensors such as accelerometers and gyroscopes. Environmental perception parameters are acquired through environmental sensors, including data from geomagnetic intensity sensors and barometric pressure sensors. The system first establishes a unified acquisition start time as a time reference. Based on this start time, the terminal location data is periodically acquired to obtain latitude and longitude coordinates, horizontal position accuracy, vertical height, and signal strength, thereby establishing personnel positioning data characteristics. Simultaneously, motion state characteristics are periodically acquired based on the same start time to obtain acceleration, angular velocity, and step count, thereby establishing personnel motion state data characteristics.

[0027] Because different sensor data sources have different acquisition frequencies and time bases, the system performs time synchronization calibration on the aforementioned personnel positioning data features and personnel movement status data features. All data streams are resampled to a unified reference period through interpolation and filtering algorithms, thereby eliminating data timing deviations caused by sensor asynchrony. Based on this, real-time spatial location features are calculated from the calibrated personnel positioning data features. These features contain the precise coordinate information of the personnel in three-dimensional space. Environmental perception parameters include the geomagnetic intensity and air pressure data of the personnel's environment, used to subsequently assist in identifying environmental interference and floor changes.

[0028] S2 specifically includes: S2-1. Collect the boundary coordinates of the security area and the route planning to establish spatial data of the activity area; S2-2, Based on the spatial topological relationship corresponding to the boundary coordinates, the patrol path boundary is divided; S2-3. The travel path planning is divided into motion characteristic cycles based on the changes in personnel patrol rate.

[0029] Specifically, spatial data of the activity area is collected and processed for boundary analysis to identify the patrol path boundary and motion characteristic cycle, and the initial spatiotemporal trajectory data is processed based on the motion characteristic cycle to determine several spatiotemporal trajectory data groups.

[0030] In this embodiment, the system collects the boundary coordinates and path planning of the security area to jointly establish activity area spatial data. This activity area spatial data covers all physical spaces accessible to security personnel, including building corridors, entrances and exits, stairwells, and outdoor patrol areas. Based on the spatial topology corresponding to the boundary coordinates, the system uses spatial topology analysis to divide the activity area into patrol path boundaries. These boundaries define the spatial activity range of personnel during normal operations and serve as a geographical constraint reference for subsequent trajectory calibration.

[0031] For path planning, the system divides movement characteristic periods based on changes in personnel patrol speed. Specifically, when a person's movement speed remains within a small fluctuation range over a continuous period, this continuous period is identified as a movement characteristic period. This period reflects the time interval during which the person's movement state remains relatively stable, such as walking in a straight line at a constant speed or staying at a fixed point. After identifying the movement characteristic period, the system uses this period as the extraction step size to segment the initial spatiotemporal trajectory data, dividing the continuous trajectory sequence into several discrete spatiotemporal trajectory data groups. Each data group corresponds to a segment of the person's movement trajectory within a movement characteristic period, providing basic data units for subsequent anomaly analysis.

[0032] S3 specifically includes: S3-1. Perform spatial trajectory density analysis processing based on the spatiotemporal trajectory data group to obtain nodes with different position jump characteristics and nodes with different signal occlusion responses respectively. S3-2. Based on the different nodes with different position jump characteristics and different nodes with different signal blockage response, obtain the corresponding node feature values ​​and perform correlation processing to obtain the different center values ​​of trajectory offset and different center values ​​of environmental interference respectively. S3-3. Based on the trajectory offset anomaly center value and the environmental interference anomaly center value, drift point elimination processing is performed respectively to obtain the trajectory anomaly center features corresponding to each spatiotemporal trajectory data group.

[0033] S3-4. Establish a trajectory verification threshold based on the mean standard deviation of the trajectory offset dissimilar center value and the environmental interference dissimilar center value corresponding to the trajectory anomaly center feature; S3-5. Determine whether the number of abnormal centers corresponding to each spatiotemporal trajectory data group is 1, and determine whether the deviation of the trajectory offset dissimilar center value and the environmental interference dissimilar center value corresponding to the trajectory abnormal center feature is less than the trajectory verification threshold. When the number of abnormal centers in the same spatiotemporal trajectory data group is not 1, and the deviation of each abnormal center is not less than the trajectory verification threshold, perform sliding window shortening processing on the motion feature period based on the deviation of the abnormal center and the trajectory verification threshold, and re-process the spatiotemporal trajectory data. S3-6. When the number of anomaly centers corresponding to each spatiotemporal trajectory data group is 1, and the deviation of the anomaly center is less than the trajectory verification threshold, the correction process is completed.

[0034] In this embodiment, a security personnel location calibration method based on the association between spatiotemporal trajectory and historical patterns utilizes the Local Anomaly Factor (LOF) algorithm. For different spatiotemporal trajectory data sets, it calculates the local density distribution between trajectory points in a high-dimensional space, identifying nodes with dissimilar location jump characteristics. These nodes represent instantaneous changes in positioning coordinates caused by multipath effects or signal obstruction. Simultaneously, it maps the environmental signal attenuation coefficient within the same time window to the feature space, extracting nodes with dissimilar signal obstruction responses. By performing spatiotemporal topological association on these two types of dissimilar nodes, the coupling weight between environmental obstruction and location jumps is identified, thereby obtaining the trajectory offset dissimilar center value and the environmental interference dissimilar center value.

[0035] Specifically, local anomaly factor analysis is performed on the spatiotemporal trajectory data group to extract the trajectory anomaly center features corresponding to each spatiotemporal trajectory data group, and the motion feature period is corrected based on the trajectory anomaly center features.

[0036] In this embodiment, the system processes each spatiotemporal trajectory data group based on the Local Anomaly Factor (LOF) algorithm. The LEF algorithm quantifies the outlier degree of each trajectory point by calculating the ratio of its local density in its neighborhood to the local density of other points in that neighborhood, effectively identifying location jump points caused by signal occlusion or multipath effects. Specifically, the system calculates the local reachability density of each trajectory point in the three-dimensional spatiotemporal feature space to obtain nodes with dissimilar location jump characteristics and nodes with dissimilar signal occlusion responses. Nodes with dissimilar location jump characteristics represent instantaneous changes in positioning coordinates caused by multipath effects or signal reflection, while nodes with dissimilar signal occlusion responses represent key locations where signal attenuation or loss occurs due to building obstruction.

[0037] Based on this, the system performs spatiotemporal correlation processing on the two types of dissimilar nodes, identifies the coupling mapping relationship between environmental interference and positioning jumps, and then obtains the trajectory offset dissimilar center value and environmental interference dissimilar center value corresponding to each spatiotemporal trajectory data group. The trajectory offset dissimilar center value reflects the degree of deviation and central trend of the positioning coordinates relative to the normal trajectory, while the environmental interference dissimilar center value reflects the degree of interference and central trend of environmental factors such as geomagnetic anomalies or sudden changes in air pressure on positioning accuracy. Subsequently, the system performs drift point elimination processing based on the trajectory offset dissimilar center value and the environmental interference dissimilar center value, respectively, to remove outlier trajectory points caused by sensor noise or occasional interference, thereby obtaining the trajectory anomaly center characteristics corresponding to each spatiotemporal trajectory data group.

[0038] After obtaining the anomaly center features of each trajectory, the system establishes a trajectory verification threshold based on the mean standard deviation of the trajectory offset anomaly center values ​​and the environmental interference anomaly center values ​​corresponding to the trajectory anomaly center features. This threshold is used to determine whether the current motion feature cycle division is reasonable. The system determines whether the number of anomaly centers corresponding to each spatiotemporal trajectory data group reaches a preset number (this preset number is 1, representing that there is at most one significant abnormal interference source in each motion cycle), and at the same time determines whether the deviation of the trajectory offset anomaly center values ​​and the environmental interference anomaly center values ​​corresponding to each anomaly center feature is less than the trajectory verification threshold. When the number of anomaly centers in the same spatiotemporal trajectory data group is not 1, and the deviation of each anomaly center is not less than the trajectory verification threshold, it indicates that the granularity of the current motion feature cycle division is too large, failing to effectively separate independent interference events. The system performs sliding window shortening processing on the motion feature cycle based on the deviation of the anomaly centers and the trajectory verification threshold, and re-processes the spatiotemporal trajectory data to enter the next round of anomaly analysis. When the number of anomaly centers in each spatiotemporal trajectory data group is 1, and the deviation of the anomaly centers is less than the trajectory verification threshold, it indicates that the current division of motion feature cycles can effectively characterize environmental interference features and complete the correction process.

[0039] S4 specifically includes: S4-1. Based on the ratio of the real-time spatial point features to the center value of the environmental interference, the trajectory interception threshold is divided, and the initial spatiotemporal trajectory of the personnel is intercepted based on the motion feature period to obtain several spatiotemporal trajectory intervals. S4-2. Based on the trajectory interception threshold, smooth the spatiotemporal trajectory interval and delete the jump position points in each spatiotemporal trajectory interval that are greater than the trajectory interception threshold; S4-3. Collect the median coordinate values ​​corresponding to the several spatiotemporal trajectory intervals to divide them into basic movement point nodes for security personnel, and collect the extreme values ​​of coordinate values ​​corresponding to several spatiotemporal trajectory intervals to divide them into movement offset extreme value nodes.

[0040] In this embodiment, a method for calibrating the location of security personnel based on the correlation between spatiotemporal trajectories and historical patterns is described, such as... Figure 2 As shown, the ratio of real-time spatial point features to the dissimilarity center value of environmental interference is calculated. This ratio characterizes the nonlinear relationship between the degree of positioning drift and the density of environmental interference, and is defined as the trajectory interception threshold. Using the motion feature period as the sliding step size, the initial trajectory is segmented. Within each spatiotemporal trajectory interval, outliers are removed according to the trajectory interception threshold, i.e., outliers whose coordinate change rate exceeds the threshold are deleted. Subsequently, the median of latitude, longitude, and altitude coordinates within each interval is extracted as the basic motion point node, reflecting the stable motion benchmark of personnel. Coordinate extreme points are also extracted as the maximum motion offset boundary under the influence of environmental interference, used for subsequent collaborative interval construction.

[0041] Specifically, the initial spatiotemporal trajectory is processed based on the motion characteristic period to obtain the basic motion point nodes and motion offset extreme value nodes of the security personnel.

[0042] In this embodiment, the system first calculates the trajectory interception threshold based on the ratio of real-time spatial point features to the dissimilarity center value of environmental interference. This threshold reflects the sensitivity of personnel to spatial position fluctuations under a unit environmental interference intensity and is used for subsequent identification and elimination of abnormal jump points. Subsequently, using the corrected motion feature period as the sampling step size, the initial spatiotemporal trajectory of the security personnel is segmented and intercepted, dividing the continuous trajectory sequence into several spatiotemporal trajectory intervals, with each interval corresponding to a personnel trajectory segment within a motion feature period.

[0043] Based on this, the system smooths each spatiotemporal trajectory interval according to the trajectory interception threshold, deleting abrupt change points in each spatiotemporal trajectory interval whose coordinate change rate exceeds the trajectory interception threshold. These abrupt change points are usually caused by spurious displacements due to instantaneous signal loss or multipath effects. After completing the outlier removal, the system collects the median coordinate values ​​corresponding to each spatiotemporal trajectory interval as the basic motion point node for security personnel. This node represents the stable spatial reference position of the personnel in each motion characteristic cycle. At the same time, the system collects the extreme values ​​of the coordinate values ​​corresponding to each spatiotemporal trajectory interval as the motion offset extreme value node. This node represents the maximum spatial offset boundary of the personnel when affected by environmental interference in each motion characteristic cycle.

[0044] S5 specifically includes: S5-1. Establish a unified spatiotemporal coordinate system with the horizontal position coordinates as the XY axis and the vertical height coordinates as the Z axis. Based on the unified spatiotemporal coordinate system, perform coordinate mapping processing on the basic movement point nodes, the extreme value nodes of movement offset, and the boundary of the activity area of ​​the personnel. Calculate the Euclidean distances of the basic movement point nodes, the extreme value nodes of movement offset, and the boundary of the activity area to obtain the point offset coordinate distance and the boundary constraint coordinate distance, respectively. S5-2. Based on the point offset coordinate distance and boundary constraint coordinate distance corresponding to the motion characteristic cycle, the position deviation value range is divided into intervals, and based on the product of the environmental interference dissimilar center value corresponding to the trajectory abnormal center and the trajectory interception threshold, the relative deviation reasonable value range threshold is divided into intervals. S5-3. When the point offset coordinate distance and the boundary constraint coordinate distance are both within the position deviation range, and the relative distance between the point offset coordinate distance and the boundary constraint coordinate distance is within the reasonable range threshold of the relative deviation, the point allowable activity range and the trajectory smoothing requirement range are divided based on the corresponding basic motion point node and the boundary of the activity area. S5-4. Perform intersection calculations on the allowed activity range of the points and the trajectory smoothing requirement range to obtain the overlapping space and establish a collaborative range.

[0045] In this embodiment, a method for calibrating the location of security personnel based on the correlation between spatiotemporal trajectories and historical patterns is described, such as... Figure 3 As shown, constructing a unified three-dimensional spatiotemporal coordinate system specifically includes: converting horizontal latitude and longitude into rectangular coordinates X and Y using Mercator projection, and using the altitude calibrated by a barometer as the Z-axis. All mapped feature points are scale-normalized. The system calculates the three-dimensional Euclidean distance between the real-time positioning point and the preset patrol path and activity area boundary to obtain the point offset coordinate distance and boundary constraint coordinate distance. The system performs logical judgment: when all the above distance indicators are within the preset deviation range, and the relative distance difference caused by signal fluctuations is not greater than a reasonable threshold, the current trajectory is determined to have high confidence. Based on the basic motion points and the structured map boundary, the system uses a convex hull algorithm to divide the allowed activity range and the smoothing requirement range of the points, and extracts the spatial overlap between the two to establish a collaborative range. This range defines the range of possible physical locations of personnel under the current environmental constraints.

[0046] Specifically, based on the environmental perception parameters in the same coordinate system, the spatial position coordinate distance and environmental interference coordinate distance of the expected prediction point node and the extreme value node of the motion offset of the security personnel are obtained respectively. Based on the value range relationship of the spatial position coordinate distance and the environmental interference coordinate distance, a collaborative interval for point calibration and trajectory optimization is established.

[0047] In this embodiment, the system establishes a unified spatiotemporal coordinate system with the horizontal position coordinates as the XY axis and the vertical height coordinates as the Z axis. Based on this unified spatiotemporal coordinate system, coordinate mapping processing is performed on the basic motion point nodes, motion offset extreme value nodes, and activity area boundaries of the personnel, mapping all feature points to the same three-dimensional vector space. Furthermore, the Euclidean distances between the basic motion point nodes, motion offset extreme value nodes, and the activity area boundaries are calculated to obtain the point offset coordinate distance and boundary constraint coordinate distance, respectively. The point offset coordinate distance reflects the degree of deviation of the basic motion point from the center of the activity area, while the boundary constraint coordinate distance reflects the degree of proximity of the motion offset extreme value to the activity area boundary.

[0048] The system divides the positional deviation range into intervals based on the coordinate distances of the point offset and boundary constraint corresponding to the motion characteristic cycle. This interval represents the reasonable range of values ​​for point offset and boundary constraint under normal operating conditions. Simultaneously, based on the product of the environmental interference dissimilarity center value corresponding to the trajectory anomaly center and the trajectory interception threshold, a reasonable relative deviation threshold is defined. This threshold is used to determine whether the relative relationship between the point offset coordinate distance and the boundary constraint coordinate distance is reasonable.

[0049] When both the offset coordinate distance and the boundary constraint coordinate distance are within the range of positional deviation values, and the relative distance between the offset coordinate distance and the boundary constraint coordinate distance is within the reasonable threshold of relative deviation, it indicates that the current trajectory point neither exceeds the boundary of the activity area nor is within a reasonable deviation range. Based on this, the system divides the allowable activity range and the trajectory smoothing requirement range according to the corresponding basic motion point nodes and the boundary of the activity area. The allowable activity range represents the range of actual physical locations that security personnel may appear in the current movement state, while the trajectory smoothing requirement range represents the calibration and adjustment range required to maintain trajectory continuity. The system performs an intersection operation on the allowable activity range and the trajectory smoothing requirement range to obtain the overlapping space between them. This overlapping space is the coordination range, representing the optimal spatial range for trajectory calibration under the current environmental constraints and movement state.

[0050] S6 specifically includes: S6-1. The average value of the geometric center coordinates of the collaborative interval is divided into interval center position values, and the position deviation calculation benchmark is calculated based on the interval center position values ​​and the position requirement values ​​corresponding to the preset inspection path. S6-2. Based on the acceleration and angular velocity corresponding to the characteristics of the personnel motion state data, calculate the real-time position of the terminal trajectory and collect the coordinate values ​​corresponding to the expected prediction point nodes and the coordinate values ​​corresponding to the extreme value nodes of motion offset respectively. S6-3. Based on the terminal trajectory, the coordinate values ​​corresponding to the real-time position, the expected predicted point node, and the extreme value node of motion offset are calculated and the parameter dimensions are unified in a unified spatiotemporal coordinate system. Based on the Kalman filter approximation algorithm and the position deviation calculation benchmark, the coordinate deviation values ​​between the terminal trajectory-calculated real-time position and the expected predicted point node and the extreme value node of motion offset are calculated respectively to establish a position deviation dataset. S6-4. Based on the trajectory interception threshold, perform outlier deletion processing on the position deviation dataset, and obtain the minimum deviation data corresponding to the position deviation dataset as the minimum position deviation value.

[0051] S6-5. Based on the difference between the real-time coordinates and the basic motion point nodes in the personnel positioning data features, the real-time positioning offset is divided. S6-6. Based on the minimum position deviation, define an inverse proportionality to obtain the fusion weight coefficient; S6-7. Based on the real-time positioning offset and the fusion weight coefficient, perform precise point calibration and allocation processing, and match calibration parameters to optimize the trajectory; The fusion weight coefficient ranges from 0.1 to 0.9.

[0052] In this embodiment, a security personnel location calibration method based on the correlation between spatiotemporal trajectory and historical patterns uses an inverse proportional function to obtain the fusion weight coefficient based on the minimum location deviation. The calculation formula for the fusion weight coefficient is as follows:

[0053] Where η is the fusion weight coefficient, γ is the proportional adjustment coefficient, and Δd min λ is the minimum position deviation, λ is the smoothing coefficient, the fusion weighting coefficient ranges from 0.1 to 0.9, and γ and λ are preset sensitivity adjustment operators. This coefficient determines the proportion of inertial navigation estimation (PDR) and GNSS positioning data in the final fused location.

[0054] Specifically, based on the approximation algorithm, the minimum position deviation is obtained from the multi-source sensing data, the expected prediction point nodes, and the extreme value nodes of motion offset. Then, based on the minimum position deviation and the corresponding real-time motion state features, the fusion weight coefficient and matching calibration parameters are assigned to perform trajectory smoothing processing to complete the position calibration of security personnel.

[0055] In this embodiment, the system first divides the area into center position values ​​based on the average value of the geometric center coordinates of the collaborative interval. These center position values ​​represent the spatial geometric center of the collaborative interval. Subsequently, based on these center position values ​​and the position requirement values ​​corresponding to the preset inspection path, a position deviation calculation benchmark is calculated. This benchmark serves as the reference standard for subsequent deviation calculations.

[0056] The system calculates the real-time position of the terminal trajectory based on the acceleration and angular velocity corresponding to the characteristics of personnel motion state data, using a trajectory estimation algorithm. This position, based on the principle of inertial navigation, utilizes the relative displacement calculated using the integral of acceleration and the integral of angular velocity, exhibiting high continuity over a short period. Simultaneously, the coordinate values ​​corresponding to the expected prediction point nodes and the extreme motion offset nodes are collected. Based on this, the coordinate values ​​of the terminal trajectory estimation real-time position, the expected prediction point nodes, and the extreme motion offset nodes are processed in a unified spatiotemporal coordinate system to unify the parameter dimensions, ensuring that each parameter has a consistent data structure and dimensions. Subsequently, based on the Kalman filter approximation algorithm and the position deviation calculation benchmark, the coordinate deviation values ​​between the terminal trajectory estimation real-time position and the expected prediction point nodes, and between the terminal trajectory estimation real-time position and the extreme motion offset nodes, are calculated to establish a position deviation dataset. The Kalman filter approximation algorithm, through two steps of state prediction and measurement update, optimally fuses the trajectory estimation position with external positioning observations, thereby quantifying the degree of deviation between each candidate position and the true position.

[0057] The system performs outlier removal on the position deviation dataset based on the trajectory interception threshold, removing outlier deviation data caused by sensor saturation or algorithm divergence, and extracts the minimum deviation data from the processed position deviation dataset as the minimum position deviation value.

[0058] After obtaining the minimum positional deviation, the system proceeds to the weight allocation and trajectory smoothing stage. First, based on the difference between the real-time coordinates and the baseline motion point nodes in the personnel positioning data, a real-time positioning offset is defined. This offset reflects the instantaneous deviation of the current positioning measurement value from the stable reference point. Second, a fusion weight coefficient is obtained based on an inversely proportional definition of the minimum positional deviation. This fusion weight coefficient uses an inversely proportional function; that is, the smaller the minimum positional deviation, the larger the fusion weight coefficient, indicating higher reliability of the trajectory estimation result; conversely, the larger the minimum positional deviation, the smaller the fusion weight coefficient, indicating relatively higher reliability of the external positioning observation. The value range of the fusion weight coefficient is within a preset interval, which is set according to the system's level of trust in trajectory estimation and external positioning.

[0059] The system performs precise calibration and allocation of points based on real-time positioning offset and fusion weight coefficients, and optimizes the trajectory by matching calibration parameters. Specifically, the system allocates basic calibration points and matching calibration points according to the location requirements corresponding to the preset patrol path. When the basic calibration point is greater than the real-time positioning offset, the real-time positioning offset is used as the upper limit of the basic calibration point, and the matching calibration point is used as an auxiliary correction, with the difference compensated by environmental perception data. After the initial allocation is completed, the system performs closed-loop verification: it checks whether the sum of the basic calibration point and the auxiliary correction is not greater than the real-time positioning offset, and whether the difference between the basic calibration point and the auxiliary correction is not greater than the minimum position deviation. If both conditions are met, it indicates that the current calibration allocation scheme conforms to physical kinematic constraints and is within the optimal deviation range. The system performs trajectory iterative verification based on the basic calibration point and the auxiliary correction, and outputs the calibrated points as the final positioning result. If either condition is not met, the system adjusts the fusion weight coefficients and returns to the precise calibration and allocation processing stage to re-allocate and verify the weights until the trajectory becomes smooth and stable.

[0060] S6-7 specifically includes: S6-7-1. Based on the fusion weight coefficient and the real-time positioning offset, the basic calibration points and the matching calibration points are allocated according to the position requirement value corresponding to the preset inspection path. When the basic calibration point is greater than the real-time positioning offset, the real-time positioning offset is used as the basic calibration point and the matching calibration point is used as the auxiliary correction amount. S6-7-2. Determine whether the sum of the basic calibration points and the auxiliary correction amount is not greater than the real-time positioning offset, and whether the difference between the basic calibration points and the auxiliary correction amount is not greater than the minimum position deviation. If so, perform trajectory iteration verification processing based on the basic calibration points and the auxiliary correction amount. Otherwise, adjust the fusion weight coefficient and return to the point precise calibration allocation processing.

[0061] In this embodiment, a method for calibrating the location of security personnel based on the correlation between spatiotemporal trajectories and historical patterns is described, such as... Figure 4 As shown, the precise calibration and allocation logic for the points includes: Based on the location accuracy requirements of the patrol path, a basic calibration amount is pre-allocated. If the calculated basic calibration amount exceeds the real-time observed positioning offset, all observable offsets are used as the basic calibration, and auxiliary corrections based on geomagnetic and barometric pressure data are introduced for fine-tuning compensation.

[0062] Initiate a closed-loop iterative process: Determine whether the calibrated points meet the physical kinematic constraints, i.e., the total calibration amount is not greater than the original offset, and the corrected residual is not greater than the calculated minimum deviation. If both constraints are met, execute trajectory smoothing output and officially update the real-time coordinates of the security personnel; if not, the system triggers a sliding window callback, fine-tunes the fusion weight coefficient η, and recalculates the calibration parameters.

[0063] Through the above steps, the system completes the entire process from multi-source data acquisition, activity area boundary analysis, motion cycle identification and correction, trajectory interception and key node extraction, collaborative interval establishment to weight allocation and trajectory smoothing, enabling security personnel to perform high-precision position calibration in complex environments.

[0064] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0065] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0066] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0067] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for calibrating the location of security personnel based on spatiotemporal trajectories, characterized in that, include: After collecting and aligning the terminal location data, motion state characteristics, and environmental perception parameters of security personnel, the system obtains the real-time spatial location characteristics of personnel based on the terminal location data and establishes the initial spatiotemporal trajectory of security personnel. Spatial data of the activity area is collected and processed for boundary analysis to identify the patrol path boundary and motion characteristic period, and the initial spatiotemporal trajectory data is processed based on the motion characteristic period to determine several spatiotemporal trajectory data groups. Local anomaly factor analysis is performed on the spatiotemporal trajectory data group to extract the trajectory anomaly center features corresponding to each spatiotemporal trajectory data group, and the motion feature period is corrected based on the trajectory anomaly center features. The initial spatiotemporal trajectory is processed based on the motion characteristic period to obtain the basic motion point nodes and motion offset extreme value nodes of the security personnel. Based on the environmental perception parameters in the same coordinate system, the spatial position coordinate distance and environmental interference coordinate distance to the expected predicted point node and the extreme value node of motion offset of the security personnel are obtained respectively. Based on the value range relationship of the spatial position coordinate distance and the environmental interference coordinate distance, a collaborative interval for point calibration and trajectory optimization is established. The approximation algorithm is used to obtain the minimum position deviation from multi-source sensing data, expected prediction point nodes, and extreme value nodes of motion offset. Based on the minimum position deviation and the corresponding real-time motion state features, fusion weight coefficients and matching calibration parameters are assigned to perform trajectory smoothing to complete the position calibration of security personnel.

2. The security personnel location calibration method based on spatiotemporal trajectory and historical pattern association as described in claim 1, characterized in that, The process of collecting and aligning security personnel terminal location data, motion state characteristics, and environmental perception parameters, and then obtaining real-time spatial location characteristics of personnel based on the terminal location data includes: Establish the starting time for collecting security personnel location data; Based on the aforementioned start time, the terminal location data is periodically collected and processed to obtain latitude and longitude coordinates, horizontal position accuracy, vertical height, and signal strength, and to establish personnel positioning data characteristics. Based on the starting time, the motion state characteristics are periodically collected and processed to obtain acceleration, angular velocity and number of steps respectively, and to establish the motion state data characteristics of the person. The personnel positioning data features and personnel movement state data features are subjected to time synchronization calibration processing, and real-time spatial point features are calculated based on the personnel positioning data features. Among them, the environmental perception parameters are the geomagnetic intensity and air pressure data of the environment in which the person is located.

3. The security personnel location calibration method based on spatiotemporal trajectory and historical pattern association as described in claim 2, characterized in that, The process of collecting spatial data of the activity area and performing boundary analysis to identify the patrol path boundaries and motion characteristic cycles includes: Collect the boundary coordinates of the security zone and the route planning to establish spatial data of the activity area; The patrol path boundary is defined based on the spatial topological relationship corresponding to the boundary coordinates. The travel path planning is divided into movement characteristic cycles based on the changes in personnel patrol rate.

4. The security personnel location calibration method based on spatiotemporal trajectory and historical pattern association as described in claim 3, characterized in that, Local anomaly factor analysis is performed on the spatiotemporal trajectory data sets to extract the trajectory anomaly center features corresponding to each spatiotemporal trajectory data set, including: Spatial trajectory density analysis is performed based on the spatiotemporal trajectory data set to obtain nodes with different position jump characteristics and nodes with different signal occlusion responses. Based on the different nodes with different position jump characteristics and different nodes with different signal blockage responses, the corresponding node feature values ​​are obtained and associated to obtain the different center values ​​of trajectory offset and different center values ​​of environmental interference respectively. Based on the trajectory offset anomaly center value and the environmental interference anomaly center value, drift point elimination processing is performed to obtain the trajectory anomaly center features corresponding to each spatiotemporal trajectory data group.

5. The security personnel location calibration method based on spatiotemporal trajectory and historical pattern association as described in claim 4, characterized in that, The correction process for the motion feature period based on the aforementioned trajectory anomaly center features includes: Based on the mean standard deviation of the trajectory offset dissimilar center value and the environmental interference dissimilar center value corresponding to the trajectory anomaly center feature, a trajectory verification threshold is established. Determine whether the number of abnormal centers corresponding to each spatiotemporal trajectory data group is 1, and determine whether the deviation of the trajectory offset dissimilar center value and the environmental interference dissimilar center value corresponding to the trajectory abnormal center feature is less than the trajectory verification threshold. When the number of abnormal centers in the same spatiotemporal trajectory data group is not 1, and the deviation of each abnormal center is not less than the trajectory verification threshold, the motion feature period is shortened by sliding window based on the deviation of the abnormal center and the trajectory verification threshold, and the spatiotemporal trajectory data is re-trimmed. When the number of anomaly centers corresponding to each spatiotemporal trajectory data group is 1, and the deviation of the anomaly center is less than the trajectory verification threshold, the correction process is completed.

6. The security personnel location calibration method based on spatiotemporal trajectory and historical pattern association as described in claim 5, characterized in that, The initial spatiotemporal trajectory is truncated based on the motion characteristic period to obtain the basic motion point nodes and motion offset extreme value nodes of the security personnel, including: Based on the ratio of the real-time spatial location features to the center value of the environmental interference, the trajectory interception threshold is defined, and the initial spatiotemporal trajectory of the personnel is intercepted based on the motion feature period to obtain several spatiotemporal trajectory intervals. Based on the trajectory interception threshold, the spatiotemporal trajectory intervals are smoothed, and jump points in each spatiotemporal trajectory interval that exceed the trajectory interception threshold are deleted. The median coordinate values ​​corresponding to the several spatiotemporal trajectory intervals are collected to divide them into basic movement point nodes for security personnel, and the extreme values ​​of coordinate values ​​corresponding to the several spatiotemporal trajectory intervals are collected to divide them into movement offset extreme value nodes.

7. The security personnel location calibration method based on spatiotemporal trajectory and historical pattern association as described in claim 6, characterized in that, The established collaborative interval for point calibration and trajectory optimization includes: A unified spatiotemporal coordinate system is established with the horizontal position coordinates as the XY axis and the vertical height coordinates as the Z axis. Based on the unified spatiotemporal coordinate system, coordinate mapping processing is performed on the basic movement point nodes, the extreme value nodes of movement offset, and the boundary of the activity area of ​​the personnel. The Euclidean distances of the basic movement point nodes, the extreme value nodes of movement offset, and the boundary of the activity area are calculated respectively to obtain the point offset coordinate distance and the boundary constraint coordinate distance. Based on the point offset coordinate distance and boundary constraint coordinate distance corresponding to the motion characteristic cycle, the position deviation value range is divided into intervals, and based on the product of the environmental interference dissimilar center value corresponding to the trajectory anomaly center and the trajectory interception threshold, the relative deviation reasonable value range threshold is divided into intervals. When the point offset coordinate distance and the boundary constraint coordinate distance are both within the position deviation range, and the relative distance between the point offset coordinate distance and the boundary constraint coordinate distance is within the reasonable range threshold of the relative deviation, the point's allowed activity range and trajectory smoothing requirement range are divided based on the corresponding basic motion point node and the boundary of the activity area. The intersection of the permitted activity range and the trajectory smoothing requirement range at the specified points is processed to obtain the overlapping space and establish a collaborative range.

8. The security personnel location calibration method based on spatiotemporal trajectory and historical pattern association as described in claim 7, characterized in that, The step of obtaining the minimum positional deviation based on the approximation algorithm for multi-source sensing data, expected prediction point nodes, and motion offset extreme value nodes includes: The average value of the geometric center coordinates of the collaborative interval is divided into interval center position values, and the position deviation calculation benchmark is calculated based on the interval center position values ​​and the position requirement values ​​corresponding to the preset inspection path. Based on the acceleration and angular velocity corresponding to the characteristics of the personnel's motion state data, the real-time position of the terminal trajectory is calculated and obtained, and the coordinate values ​​corresponding to the expected prediction point nodes and the coordinate values ​​corresponding to the extreme value nodes of motion offset are collected respectively. Based on the real-time position calculated from the terminal trajectory, the coordinate values ​​corresponding to the expected predicted point nodes and the coordinate values ​​corresponding to the extreme value nodes of motion offset, the parameter dimensions are uniformly processed under a unified spatiotemporal coordinate system. Based on the Kalman filter approximation algorithm and the position deviation calculation benchmark, the coordinate deviation values ​​between the real-time position calculated from the terminal trajectory and the expected predicted point nodes and the extreme value nodes of motion offset are calculated respectively to establish a position deviation dataset. Based on the trajectory interception threshold, outlier deletion processing is performed on the position deviation dataset, and the minimum deviation data corresponding to the position deviation dataset is obtained as the minimum position deviation value.

9. The security personnel location calibration method based on spatiotemporal trajectory and historical pattern association as described in claim 8, characterized in that, Trajectory smoothing is performed based on the minimum position deviation and the corresponding real-time motion state features, by assigning fusion weight coefficients and matching calibration parameters, including: The difference between the real-time coordinates and the basic motion point nodes in the personnel positioning data features is used to classify the real-time positioning offset. Based on the minimum positional deviation, an inverse proportional definition is used to obtain the fusion weight coefficient η, which is calculated as follows: Based on the real-time positioning offset and the fusion weight coefficient, the point is accurately calibrated and allocated, and the trajectory is optimized by matching the calibration parameters. Where η is the fusion weight coefficient, γ is the proportional adjustment coefficient, and Δd min λ is the minimum positional deviation, λ is the smoothing coefficient, and the fusion weight coefficient ranges from 0.1 to 0.

9.

10. The security personnel location calibration method based on spatiotemporal trajectory and historical pattern association as described in claim 9, characterized in that, Based on the real-time positioning offset and fusion weight coefficient, precise point calibration and allocation processing is performed, and trajectory optimization is performed by matching calibration parameters, including: Based on the fusion weight coefficient and the real-time positioning offset, the basic calibration points and the matching calibration points are allocated according to the position requirement value corresponding to the preset inspection path. When the basic calibration point is greater than the real-time positioning offset, the real-time positioning offset is used as the basic calibration point and the matching calibration point is used as the auxiliary correction amount. If the sum of the basic calibration points and the auxiliary correction amount is not greater than the real-time positioning offset, and the difference between the basic calibration points and the auxiliary correction amount is not greater than the minimum position deviation, then the trajectory iterative verification process is performed based on the basic calibration points and the auxiliary correction amount; otherwise, the fusion weight coefficient is adjusted, and the point precise calibration allocation process is returned.

Images