A hybrid behavioral value quantification assessment method and system integrating LBS and XR data

Abstract

This invention relates to the field of location service technology, specifically to a method and system for quantifying the value of hybrid behavior by integrating LBS and XR data. The method includes: collecting IMU data and GNSS location data of the target user; generating a PDR trajectory by extracting the user's physical motion characteristics and performing spatiotemporal consistency verification; calculating the user's grid-object value integral, including gaze duration, viewing angle, and optimal viewpoint matching degree; generating a dynamic scarcity coefficient by combining the inverse logarithmic function of historical access frequency, geographical remoteness, and terrain complexity; obtaining the accumulated physical displacement during the time period that passed the spatiotemporal consistency verification, and calculating the user's hybrid behavior value based on the location scarcity coefficient, physical displacement, and grid-object value integral; and generating an emotional valence map using the aggregated gaze duration and PDR wandering trajectory. This invention addresses the significant shortcomings of existing technologies in constructing high-value, fraud-resistant, and deeply interactive virtual-real fusion systems.

Description

Technical Field

[0001] This invention relates to the technical field of location services, specifically to a method and system for quantitatively evaluating the hybrid behavioral value of LBS and XR data. Background Technology

[0002] With the improvement of computing power of smart mobile terminals and the popularization of 5G networks, the integration of digital content and physical space has become an important gateway to the next generation of the Internet. However, the existing location-based services (LBS) and extended reality (XR) integration applications in the fields of cultural tourism, gaming, and urban digitalization face the following four key technical bottlenecks.

[0003] Firstly, there is the problem of ineffective verification of LBS location data and rampant cheating. In existing LBS applications, location verification mainly relies on the standard APIs provided by the mobile operating system. However, users can easily load simulated location applications through "developer options" or use software-defined radio to broadcast fake GPS signals, allowing a large number of users to simulate movement around the world "without leaving home." At the same time, existing systems often only verify the jump speed of GPS coordinates, but cannot identify low-speed spoofing scripts that simulate normal walking speed.

[0004] Secondly, the quantification of the value of virtual interactive behavior is singular and superficial. The interaction design of existing AR applications remains at the level of projecting virtual objects onto reality -> user taps the screen -> triggering feedback. However, this interaction mode simplifies the complex cognitive process into a mechanical muscle reaction. In order to obtain rewards, users often quickly tap the target on the screen, completely ignoring the visual details and cultural connotations carried by the target (such as inscriptions on cultural relics or the bracket structure of buildings). At the same time, the system lacks refined capture and quantification of user gaze, dwell time, and observation angle, making it impossible to distinguish between shallow interaction and deep immersion. This reduces XR technology to a visual gimmick rather than an effective medium for conveying information.

[0005] Third, there is an imbalance in the value mapping between physical behavior and digital rewards. Existing reward algorithms typically employ linear or static models (e.g., a fixed reward of 100 points for each checkpoint, or rewards calculated solely based on distance). However, this mechanism ignores the heterogeneity of geographical space. Reaching a convenient downtown shopping mall versus climbing a remote section of the Great Wall presents a significant difference in both physical and psychological effort, yet the existing system often yields the same numerical reward. Furthermore, due to the lack of dynamic adjustment based on "location scarcity," users tend to repeatedly accumulate points in low-cost areas (such as densely populated urban areas), resulting in high-cultural-value but geographically remote locations being neglected, leading to a severe imbalance in the digital distribution of cultural and tourism resources.

[0006] Fourth, the path for transforming individual experience data into public governance data is blocked. Currently, the data generated by LBS+XR applications is usually treated as closed commercial assets, used only for game operation or advertising. However, the data only records "who is where," lacking in-depth behavioral data such as "who is interested in what." At the same time, there is a lack of an algorithmic system that can clean, aggregate, de-identify, and transform the micro-behaviors of massive users (such as the length of time spent at a historical site or the confusion feedback of an AR guide sign) into a basis for urban spatial micro-renewal decisions, resulting in the waste of the public value of Phygital data. Summary of the Invention

[0007] The purpose of this invention is to provide a hybrid behavioral value quantification evaluation method and system that integrates LBS and XR data, in order to solve the problem that existing technical solutions have significant defects in building high-value, fraud-proof, and deeply interactive Phygital systems.

[0008] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0009] A hybrid behavioral value quantification assessment method integrating LBS and XR data includes the following steps:

[0010] S100: Collect IMU data and GNSS location data of the target user, generate PDR trajectory by extracting the user's physical motion characteristics, and perform spatiotemporal consistency verification by comparing the feature vectors of GNSS displacement and PDR displacement.

[0011] S200: Utilizes the 6DoF attitude estimation and raycasting algorithm of mobile devices to calculate the user grid value integral, which includes gaze duration, viewing angle, and optimal viewpoint matching degree.

[0012] S300, generates a dynamic scarcity coefficient based on the inverse logarithmic function of historical access frequency, geographical remoteness, and terrain complexity; obtains the physical displacement accumulated during the time period that has passed the spatiotemporal consistency verification, and calculates the user's mixed behavior value based on the location scarcity coefficient, physical displacement, and grid value integral.

[0013] S400: Using aggregated gaze duration and PDR hovering trajectory, an emotional valence map is generated.

[0014] Preferably, step S100 specifically includes:

[0015] S101. The system uses sensors to collect IMU data, including: acceleration vector. Angular velocity vector and magnetic field strength vector Simultaneously collect GNSS location data. ,in These represent latitude, longitude, altitude, and precision radius, respectively.

[0016] S102, Acceleration in the body coordinate system Acceleration converted to world coordinate system And based on acceleration Gait detection, stride length estimation, and heading estimation are performed to generate a PDR trajectory;

[0017] S103, in the sliding time window Within, obtain the GNSS displacement vector respectively. PDR displacement vector and acceleration energy density GPS anomaly detection is performed by comparing the feature vectors of GNSS displacement and PDR displacement.

[0018] Preferably, step S102, which involves gait detection, stride length estimation, and heading estimation to generate a PDR trajectory, specifically includes:

[0019] Fusion using complementary filtering algorithm and The attitude quaternion of the navigation device is calculated. Then, according to the formula: Acceleration in the body coordinate system Acceleration converted to world coordinate system ;in, It is the acceleration due to gravity. For quaternions The generated rotation matrix;

[0020] Obtain the vertical acceleration in the world coordinate system. A low-pass filter is then applied to obtain a smoothed signal. Gait detection is performed using a peak detection algorithm with adaptive thresholds.

[0021] ;

[0022] in, The mean of the sliding window. For dynamic thresholds;

[0023] Step size is estimated using an improved Weinberg model. : ;

[0024] in, These represent the maximum and minimum acceleration values ​​within a single-step period, respectively. Personalized constants for users, This is the terrain coefficient;

[0025] By combining magnetometer data with gyroscope integration, the heading angle is calculated using an extended Kalman filter. Generate the PDR trajectory at time t: .

[0026] Preferably, the GPS anomaly detection method described in step S103, which involves comparing the feature vectors of GNSS displacement and PDR displacement, specifically includes:

[0027] when ,but And step count At that time, it was determined to be GPS simulator spoofing; among them Indicates the energy density threshold;

[0028] when ,but At that time, it was determined to be indoor exercise, not a valid LBS trek; among them Indicates the movement threshold;

[0029] when However, the detected step frequency At that time, it was determined to be a spoofing attempt to modify GPS speed;

[0030] For the signal-to-noise ratio sequence of GNSS signals, calculate its variance. ,when If the condition is abnormal, it is considered an abnormality.

[0031] Preferably, step S200 specifically includes:

[0032] S201, Define the camera center point Camera forward vector Virtual cultural relic target points Virtual artifact surface normal vector By geometrically modeling the user's line of sight, we can obtain:

[0033] line-of-sight vector ,distance ;

[0034] Line of sight deviation angle: ;

[0035] Observe the azimuth angle: ;

[0036] S202, Define valid interaction states Only if all of the following conditions are met Otherwise, it is 0:

[0037] Distance constraints: ;in , Indicates the minimum and maximum distances;

[0038] Visual constraints: ;in Indicates the field of view;

[0039] Occlusion constraint: Using ARCore / ARKit's depth API or Raycast, detect whether there are real occlusions between the camera and the object. If the ray is blocked by a real object, it is considered invalid.

[0040] S203. Calculate the angle weight based on the observation angle: ;in, This is the best angle for viewing cultural relics. Tolerance / Degradation rate;

[0041] Based on gaze duration Calculate the duration saturation coefficient: ;in, This refers to the attenuation intensity parameter;

[0042] Therefore, the user's property value integral is calculated based on the angle weight and duration saturation coefficient: .

[0043] Preferably, step S300 specifically includes:

[0044] S301, Obtain each location point The server maintains a dynamic scarcity coefficient: ;

[0045] in, This represents the number of visitors to the location over the past week. Weight the distance of the location from the city center or transportation hub. Due to the complexity of the environment leading to that location, To access the weight of scarce items, For the remoteness item weight, To reach the complexity term weight, It is a natural constant;

[0046] S302, User Hybrid Behavior Value It is determined by both "verified trekking behavior" and "quantified investigation behavior": ;

[0047] in, This refers to the physical displacement mileage accumulated only within the period that passes the spatiotemporal consistency check. For XR interaction integrals based on gaze duration and angle, , Configure system parameters.

[0048] Preferably, step S400 specifically includes:

[0049] S401. Before uploading user trajectories, add Laplace noise to the coordinate points to ensure that it is impossible to reverse-locate to the home address of a specific individual, and only retain the statistically significant regional distribution; divide the city into... For hexagonal grids, select the corresponding H3 resolution and define the unit to which each reported event belongs as... ;

[0050] S402, Statistical analysis within a time window Each grid cell within Generate an emotional valence map: ;in, The user's contribution to the value of the unit. Valence of acrolateral facial expressions / emotions;

[0051] Set up pedestrian congestion and wayfinding warnings: Analyze the mixed behavior value of users in the PDR trajectory. When it is less than the value threshold, it is judged as "invalid wandering". The system automatically generates a suggestion report and sends it to the system administrator.

[0052] Preferably, the mixed behavior value quantification assessment system includes:

[0053] The multi-source sensing module, located on the mobile device, is used to collect GNSS, IMU, and camera data;

[0054] The spatiotemporal consistency verification engine, distributed on the edge and cloud sides, is used to perform spatiotemporal consistency verification by comparing the feature vectors of GNSS displacement and PDR displacement.

[0055] The hybrid value calculation engine, located in the cloud, is used to calculate the value of user hybrid behaviors based on location scarcity coefficients, physical displacement, and grid value integrals.

[0056] The public value feedback system, located in the cloud, is used for data aggregation and visualization.

[0057] A non-transitory computer-readable storage medium having computer instructions stored thereon that cause a computer to perform the methods described above.

[0058] A computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, wherein when the computer program is executed by a processor, the computer performs the method described above.

[0059] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0060] This invention achieves spatiotemporal consistency verification by integrating IMU physical motion characteristics with GNSS trajectories, completely eliminating cheating through "cloud tourism" using simulators, ensuring the authenticity of physical arrival, and maintaining commercial fairness. By calculating attention integrals through forced deep observation based on gaze duration and angle, it effectively recreates the cultural experience of "gazing," quantifies user cognitive input, and upgrades cultural dissemination from "reaching" to "touching." Through automatic adjustment of the value of less popular locations using an inverse logarithmic model for dynamic scarcity weighting, it automatically balances cultural and tourism traffic, activating marginal assets without manual operation and alleviating congestion in popular areas. By recording micro-level attention and emotions, it outputs governance suggestions to realize a public value flywheel, achieving the social application of data and providing objective quantitative evidence for urban micro-renewal (such as optimizing road signs and providing early warnings for repairs). Attached Figure Description

[0061] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings:

[0062] Figure 1 This is a framework diagram of the hybrid behavior value quantification assessment system of the present invention;

[0063] Figure 2 This is a flowchart of the spatiotemporal consistency verification process of the present invention;

[0064] Figure 3 This is a schematic diagram of the AR attention calculation geometry of the present invention;

[0065] Figure 4 This is a schematic diagram of the integral of the lattice value of the present invention;

[0066] Figure 5 This is the scarcity dynamic weighted curve of the present invention. Detailed Implementation

[0067] 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, and 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.

[0068] Please see Figures 1-5 The present invention provides the following technical solution:

[0069] Example 1: A hybrid behavioral value quantification assessment method integrating LBS and XR data, comprising the following steps:

[0070] S100: Collect IMU data and GNSS location data of the target user, generate PDR trajectory by extracting the user's physical motion characteristics, and perform spatiotemporal consistency verification by comparing the feature vectors of GNSS displacement and PDR displacement.

[0071] S200: Utilizes the 6DoF attitude estimation and raycasting algorithm of mobile devices to calculate the user grid value integral, which includes gaze duration, viewing angle, and optimal viewpoint matching degree.

[0072] S300, generates a dynamic scarcity coefficient based on the inverse logarithmic function of historical access frequency, geographical remoteness, and terrain complexity; obtains the physical displacement accumulated during the time period that has passed the spatiotemporal consistency verification, and calculates the user's mixed behavior value based on the location scarcity coefficient, physical displacement, and grid value integral.

[0073] S400: Using aggregated gaze duration and PDR hovering trajectory, an emotional valence map is generated.

[0074] Preferably, step S100 specifically includes: This module aims to solve the GPS spoofing problem. The core logic is to use the relative trajectory generated by pedestrian dead reckoning (PDR) to verify the absolute trajectory generated by GNSS;

[0075] S101. The system uses sensors to acquire IMU data at a frequency of 100Hz, including: acceleration vector. Angular velocity vector and magnetic field strength vector Simultaneously, GNSS location data is acquired at 1Hz. ,in These represent latitude, longitude, altitude, and precision radius, respectively.

[0076] S102, Acceleration in the body coordinate system Acceleration converted to world coordinate system And based on acceleration Gait detection, stride length estimation, and heading estimation are performed to generate a PDR trajectory;

[0077] S103, in the sliding time window Within, obtain the GNSS displacement vector respectively. PDR displacement vector and acceleration energy density GPS anomaly detection is performed by comparing the feature vectors of GNSS displacement and PDR displacement.

[0078] Preferably, step S102, which involves gait detection, stride length estimation, and heading estimation to generate a PDR trajectory, specifically includes:

[0079] Fusion using complementary filtering algorithm and The attitude quaternion of the navigation device is calculated. Then, according to the formula: Acceleration in the body coordinate system Acceleration converted to world coordinate system ;in, It is the acceleration due to gravity. For quaternions The generated rotation matrix;

[0080] Obtain the vertical acceleration in the world coordinate system. A low-pass filter (cutoff frequency 3Hz) is then applied to eliminate high-frequency noise and obtain a smooth signal. Gait detection is performed using a peak detection algorithm with adaptive thresholds.

[0081] ;

[0082] in, The mean of the sliding window. For dynamic thresholds;

[0083] Step size is estimated using an improved Weinberg model. : ;

[0084] in, These represent the maximum and minimum acceleration values ​​within a single-step period, respectively. Personalized constants for users (calibrated using historical data). This is the terrain coefficient (decreases when going uphill);

[0085] By combining magnetometer data with gyroscope integration, the heading angle is calculated using an extended Kalman filter. To address indoor and outdoor magnetic interference, a "quasi-static magnetic field detection" mechanism is introduced, updating the magnetic heading weights only when the magnetic field magnitude fluctuation is small, thus generating the PDR trajectory at time t.

[0086] .

[0087] Preferably, the GPS anomaly detection method described in step S103, which involves comparing the feature vectors of GNSS displacement and PDR displacement, specifically includes:

[0088] when (GPS shows significant movement), but (The accelerometer shows a stationary state or only slight vibration) and the step count is... At that time, it was determined to be GPS simulator spoofing; the principle is that movement in the physical world is always accompanied by changes in acceleration. Even when sitting in a car, there will be characteristic vibrations and acceleration / deceleration curves; among which... Indicates the energy density threshold;

[0089] when Very high (PDR shows movement), but When the GPS location remains essentially unchanged, it is considered indoor movement, not effective LBS trekking; the principle is that although physical movement occurs, no geographical displacement is generated, which does not meet the definition of "trekking"; among which Indicates the movement threshold;

[0090] when (Running speed), but the detected cadence When measuring walking cadence, it is determined to be a spoofing that modifies GPS speed;

[0091] For the signal-to-noise ratio sequence of GNSS signals, calculate its variance. In real-world environments, the SNR fluctuates due to multipath effects and occlusion, while the SNR of simulators is typically a constant value or a perfect normal distribution; when If the condition is abnormal, it is considered an abnormality.

[0092] Preferably, step S200 specifically includes: This module aims to solve the shallow problem of "click-based" interaction by calculating the "object value" by quantifying the user's visual attention. This solution utilizes device pose as an approximate proxy for gaze on ordinary mobile devices without an eye tracker.

[0093] S201, Define the camera center point Camera forward vector (Corresponding to the center ray of the screen), virtual cultural relic target point (Usually the geometric center or key feature point of the artifact model), virtual artifact surface normal vector (Used to determine whether the view is frontal), obtained through geometric modeling of the user's gaze:

[0094] line-of-sight vector ,distance ;

[0095] Line-of-sight deviation angle, the angle between the camera's optical axis and the line connecting the target: ;

[0096] Observe the azimuth angle, the angle between the inverse of the line-of-sight vector and the normal to the object's surface:

[0097] ;

[0098] S202, Define valid interaction states Only if all of the following conditions are met Otherwise, it is 0:

[0099] Distance constraints: (For example: 0.5 meters) (5 meters); too far and you can't see clearly, too close and you see through the lens; among them , Indicates the minimum and maximum distances;

[0100] Visual constraints: (e.g., 15 degrees); ensure the object is centered on the screen, not at the edge; where Indicates the field of view;

[0101] Occlusion constraint: Using ARCore / ARKit's depth API or Raycast, detect whether there are real occlusions between the camera and the object. If the ray is blocked by a real object, it is considered invalid.

[0102] S203. Calculate angle weights based on the observation angle to encourage users to observe from the best perspective: ;in, The optimal viewing angle for cultural relics (e.g., 0 degrees to the front of the inscription) is the angle from which they are viewed. The further away from the front, the faster the weight of the inscription diminishes. Tolerance / Degradation rate ( The smaller the size, the more "angle-dependent" it is.

[0103] Based on gaze duration Calculate the duration saturation coefficient to prevent infinite score farming; adjust according to gaze duration. As you increase, the returns per unit of time decrease (diminishing marginal utility): ;in, This is the attenuation intensity parameter (the larger the value, the faster the attenuation).

[0104] Therefore, the user's property value integral is calculated based on the angle weight and duration saturation coefficient: ;

[0105] To avoid hand tremors leading to Rapidly switch between 0 and 1, introduce a hysteresis comparator or set a short hold window (e.g., 0.5 seconds), and pause integration only if the device is out of sight for more than 0.5 seconds.

[0106] Preferably, step S300 specifically includes: this module solves the problem of balancing value incentives by automatically pricing the "exploration value" of a geographical location through an algorithm;

[0107] S301, Obtain each location point The server maintains a dynamic scarcity coefficient: ;

[0108] in, This represents the number of visitors to the location over the past week; the fewer the visitors, the higher the scarcity. The distance of the location from the city center or transportation hub is weighted; the more remote the location, the higher the weight. The environmental complexity of reaching this location (such as altitude changes and path tortuosity) is estimated based on the average energy consumption from historical PDR data. To access the weight of scarce items, For the remoteness item weight, To reach the complexity term weight, The natural constant (used to avoid) );

[0109] S302, User Hybrid Behavior Value It is determined by both "verified trekking behavior" and "quantified investigation behavior": ;

[0110] in, This refers to the physical displacement mileage accumulated only within the period that passes the spatiotemporal consistency check. For XR interaction integrals based on gaze duration and angle, , Configure system parameters to adjust the weighting of "action" and "knowledge".

[0111] Preferably, step S400 specifically includes: This module elaborates in detail how to transform the data generated by the above technology into urban governance assets.

[0112] Public Value Flywheel: User participation (voluntary) → Edge-side extraction and anonymization → Grid aggregation to form city-level indicators → Management-side optimization of wayfinding / facilities / content → Improved experience leading to more participation and higher-quality data → Forming a closed loop.

[0113] S401. Before uploading user trajectories, add Laplace noise to the coordinate points to ensure that it is impossible to reverse-locate to the home address of a specific individual, and only retain the statistically significant regional distribution; divide the city into... If the hexagonal grid (H3 index, an open-source hexagonal hierarchical indexed grid system from Uber) is selected, then the corresponding H3 resolution is chosen (actually based on the cell side length / area of ​​H3), and the unit to which each reported event belongs is defined as... ;

[0114] S402, Statistical analysis within a time window Each grid cell within Generate an emotional valence map to measure the spatial distribution of "cultural resonance intensity / experience quality": ;in, The contribution of the user's grid value within this unit (with time saturation already implemented). Valence of facial expressions / emotions is determined by combining facial expressions (such as smiling and frowning) captured by the front-facing camera with gaze duration. High-scoring areas represent "strong cultural resonance," while low-scoring areas with short dwell times represent "poor experience."

[0115] Set up pedestrian congestion and wayfinding warnings: Analyze the mixed behavior value of users in the PDR trajectory. When it is less than the value threshold, it is judged as "invalid wandering" (high number of steps, low displacement, low grid value). The system automatically generates a suggestion report and sends it to the system administrator.

[0116] Facility damage hotspots: If a certain virtual content... A sudden, precipitous drop from historical highs, coupled with a large number of users quickly turning off their cameras after turning them on, may indicate a change in the physical environment (such as being blocked by construction barriers, insufficient lighting, or unpleasant odors). The system will automatically trigger an anomaly warning.

[0117] Preferably, the mixed behavior value quantification assessment system includes:

[0118] The multi-source sensing module, located on the mobile device, is used to collect GNSS, IMU, and camera data;

[0119] The spatiotemporal consistency verification engine, distributed at the edge (initial screening) and cloud (final judgment), is used to perform spatiotemporal consistency verification by comparing the feature vectors of GNSS displacement and PDR displacement.

[0120] The hybrid value calculation engine, located in the cloud, is used to calculate the value of user hybrid behaviors based on location scarcity coefficients, physical displacement, and grid value integrals.

[0121] The public value feedback system, located in the cloud, is used for data aggregation and visualization.

[0122] Example 2: This example also provides a non-transitory computer-readable storage medium storing computer instructions that cause a computer to execute the hybrid behavior value quantification evaluation method that integrates LBS and XR data as described in Example 1.

[0123] Example 3: This example also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer executes the hybrid behavior value quantification evaluation method that integrates LBS and XR data as described in Example 1.

[0124] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A hybrid behavioral value quantification assessment method integrating LBS and XR data, characterized in that, Includes the following steps: S100: Collect IMU data and GNSS location data of the target user, generate PDR trajectory by extracting the user's physical motion characteristics, and perform spatiotemporal consistency verification by comparing the feature vectors of GNSS displacement and PDR displacement. S200: Utilizes the 6DoF attitude estimation and raycasting algorithm of mobile devices to calculate the user grid value integral, which includes gaze duration, viewing angle, and optimal viewpoint matching degree. S300, generates a dynamic scarcity coefficient based on the inverse logarithmic function of historical access frequency, geographical remoteness, and terrain complexity; obtains the physical displacement accumulated during the time period that has passed the spatiotemporal consistency verification, and calculates the user's mixed behavior value based on the location scarcity coefficient, physical displacement, and grid value integral. S400: Using aggregated gaze duration and PDR hovering trajectory, an emotional valence map is generated.

2. The method as described in claim 1, characterized in that, Step S100 specifically includes: S101. The system uses sensors to collect IMU data, including: acceleration vector. Angular velocity vector and magnetic field strength vector Simultaneously collect GNSS location data. ,in These represent latitude, longitude, altitude, and precision radius, respectively. S102, Acceleration in the body coordinate system Acceleration converted to world coordinate system And based on acceleration Gait detection, stride length estimation, and heading estimation are performed to generate a PDR trajectory; S103, in the sliding time window Within, obtain the GNSS displacement vector respectively. PDR displacement vector and acceleration energy density GPS anomaly detection is performed by comparing the feature vectors of GNSS displacement and PDR displacement.

3. The method as described in claim 2, characterized in that, Step S102, which involves gait detection, stride length estimation, and heading estimation to generate a PDR trajectory, specifically includes: Fusion using complementary filtering algorithm and The attitude quaternion of the navigation device is calculated. Then, according to the formula: Acceleration in the body coordinate system Acceleration converted to world coordinate system ;in, It is the acceleration due to gravity. For quaternions The generated rotation matrix; Obtain the vertical acceleration in the world coordinate system. A low-pass filter is then applied to obtain a smoothed signal. Gait detection is performed using a peak detection algorithm with adaptive thresholds. ； in, The mean of the sliding window. For dynamic thresholds; Step size is estimated using an improved Weinberg model. : ; in, These represent the maximum and minimum acceleration values ​​within a single-step period, respectively. Personalized constants for users, This is the terrain coefficient; By combining magnetometer data with gyroscope integration, the heading angle is calculated using an extended Kalman filter. Generate the PDR trajectory at time t: .

4. The method as described in claim 2, characterized in that, Step S103, which involves performing GPS anomaly detection by comparing the feature vectors of GNSS displacement and PDR displacement, specifically includes: when ,but And step count At that time, it was determined to be GPS simulator spoofing; among them Indicates the energy density threshold; when ,but At that time, it was determined to be indoor exercise, not a valid LBS trek; among them Indicates the movement threshold; when However, the detected step frequency At that time, it was determined to be a spoofing attempt to modify GPS speed; For the signal-to-noise ratio sequence of GNSS signals, calculate its variance. ,when When this occurs, it is determined to be an abnormal detection.

5. The method as described in claim 1, characterized in that, Step S200 specifically includes: S201, Define the camera center point Camera forward vector Virtual cultural relic target points Virtual artifact surface normal vector By geometrically modeling the user's line of sight, we can obtain: line-of-sight vector ,distance ; Line of sight deviation angle: ; Observe the azimuth angle: ; S202, Define valid interaction states Only if all of the following conditions are met Otherwise, it is 0: Distance constraints: ;in , Indicates the minimum and maximum distances; Visual constraints: ;in Indicates the field of view; Occlusion constraint: Using ARCore / ARKit's depth API or Raycast, detect whether there are real occlusions between the camera and the object. If the ray is blocked by a real object, it is considered invalid. S203. Calculate the angle weight based on the observation angle: ;in, This is the best angle for viewing cultural relics. Tolerance / Degradation rate; Based on gaze duration Calculate the duration saturation coefficient: ;in, This refers to the attenuation intensity parameter; Therefore, the user's property value integral is calculated based on the angle weight and duration saturation coefficient: .

6. The method as described in claim 1, characterized in that, Step S300 specifically includes: S301, Obtain each location point The server maintains a dynamic scarcity coefficient: ; in, This represents the number of visitors to the location over the past week. Weight the distance of the location from the city center or transportation hub. Due to the complexity of the environment leading to that location, To access the weight of scarce items, For the remoteness item weight, To reach the complexity term weight, It is a natural constant; S302, User Hybrid Behavior Value It is determined by both "verified trekking behavior" and "quantified investigation behavior": ; in, This refers to the physical displacement mileage accumulated only within the period that passes the spatiotemporal consistency check. For XR interaction integrals based on gaze duration and angle, , Configure system parameters.

7. The method as described in claim 1, characterized in that, Step S400 specifically includes: S401. Before uploading user trajectories, add Laplace noise to the coordinate points to ensure that it is impossible to reverse-locate to the home address of a specific individual, and only retain the statistically significant regional distribution; divide the city into... For hexagonal grids, select the corresponding H3 resolution and define the unit to which each reported event belongs as... ; S402, Statistical analysis within a time window Each grid cell within Generate an emotional valence map: ;in, The user's contribution to the value of the unit. Valence of acrolateral facial expressions / emotions; Set up pedestrian congestion and wayfinding warnings: Analyze the mixed behavior value of users in the PDR trajectory. When it is less than the value threshold, it is judged as "invalid wandering". The system automatically generates a suggestion report and sends it to the system administrator.

8. A hybrid behavior value quantification assessment system for implementing the hybrid behavior value quantification assessment method integrating LBS and XR data as described in any one of claims 1-7, comprising: The multi-source sensing module, located on the mobile device, is used to collect GNSS, IMU, and camera data; The spatiotemporal consistency verification engine, distributed on the edge and cloud sides, is used to perform spatiotemporal consistency verification by comparing the feature vectors of GNSS displacement and PDR displacement. The hybrid value calculation engine, located in the cloud, is used to calculate the value of user hybrid behaviors based on location scarcity coefficients, physical displacement, and grid value integrals. The public value feedback system, located in the cloud, is used for data aggregation and visualization.

9. A non-transitory computer-readable storage medium, characterized in that, It stores computer instructions that cause the computer to perform the method described in any one of claims 1-7.

10. A computer program product, characterized in that, The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer performs the method described in any one of claims 1-7.

Images