Smart bracelet real-time position tracking visualization method and system

Abstract

The invention relates to the technical field of position tracking, in particular to an intelligent bracelet real-time position tracking visualization method and system. The method comprises the following steps: acquiring multi-dimensional sensing monitoring parameters of a user based on a smart bracelet, performing real-time motion state evaluation of the user, and performing adaptive adjustment of a position acquisition frequency to generate an adaptive position acquisition frequency; performing real-time position positioning acquisition based on the adaptive position acquisition frequency, and performing frequency-by-frequency spatial position positioning calculation to obtain a high-precision compensation reconstruction trajectory; the method comprises the following steps: acquiring public electronic map information, performing scene key element identification, and performing key attribute cognitive marking to obtain a plurality of scene attribute memory anchor points; and performing relative position distribution analysis on the high-precision compensation reconstruction trajectory based on a plurality of scene attribute memory anchor points, and constructing an anchor point space trajectory visual mapping graph. According to the invention, high-precision real-time position tracking of the user is realized, and clear and visual multi-level map information rendering is presented.

Description

Technical Field

[0001] The present invention relates to the field of position tracking technology, and in particular to a method and system for visualizing the real-time position tracking of a smart bracelet. Background Art

[0002] Traditional smart bracelet location tracking methods typically rely on GPS positioning technology. While it can provide relatively accurate positioning in outdoor environments, positioning accuracy is significantly reduced indoors, in underground parking lots, and in areas with weak or unavailable GPS signals. At the same time, existing trajectory visualization methods are often too simple to accurately reflect the user's specific location, movement status, and environmental changes in real time, which creates significant limitations in application scenarios that require high-precision location information. Furthermore, existing smart bracelet location tracking systems generally use a fixed acquisition frequency and simplified trajectory reconstruction methods, making it difficult to adapt to the needs of different scenarios. This is especially true during long-term, complex movement, where data integrity, accuracy, and timeliness cannot be fully guaranteed.

[0003] With the rapid development of big data, artificial intelligence algorithms, and machine learning technologies, traditional positioning and trajectory visualization methods are gradually evolving towards higher precision, higher real-time performance, and greater intelligence. By combining inertial sensor data, external map information, and intelligent algorithms, smart wristbands not only achieve accurate real-time monitoring of multi-dimensional data but also dynamically adjust the frequency of location acquisition through adaptive algorithms, optimizing data processing and trajectory reconstruction, thereby significantly improving location tracking accuracy and system response speed. Many unresolved challenges remain in this area, such as how to design intelligent algorithms that adapt to various environmental changes while maintaining low power consumption and high accuracy; how to cope with complex indoor and outdoor environments; and how to achieve more intelligent and personalized trajectory visualization and user interaction experiences. To overcome these challenges, combining advanced machine learning techniques, dynamic frequency adjustment, and real-time scene recognition methods promises to provide a more efficient, accurate, and secure solution for real-time location tracking and visualization in smart wristbands. Summary of the Invention

[0004] In order to solve the above technical problems, the present invention proposes a smart bracelet real-time position tracking visualization method and system to solve at least one of the above technical problems.

[0005] To achieve the above objectives, the present invention provides a method for visualizing the real-time position tracking of a smart bracelet, comprising the following steps: Step S1: Based on the smart bracelet, the user's multi-dimensional sensor monitoring parameters are collected to evaluate the user's real-time motion state, and the position acquisition frequency is adaptively adjusted to generate an adaptive position acquisition frequency; Step S2: Perform real-time position acquisition based on the adaptive position acquisition frequency, and perform frequency-by-frequency spatial position calculation to obtain a high-precision compensated reconstructed trajectory; Step S3: Obtaining public electronic map information, identifying key elements of the scene, and performing key attribute recognition marking to obtain multiple scene attribute memory anchor points; Step S4: performing relative position distribution analysis on the high-precision compensated reconstructed trajectory based on multiple scene attribute memory anchor points, and performing dynamic trajectory visualization mapping to construct an anchor point space trajectory visualization mapping diagram; Step S5: Performing visual recognition of the geographic background layer based on the public electronic map information, and then rendering the semantic information of the anchor point spatial trajectory visualization map to construct a three-dimensional trajectory-geographic information fusion model; Step S6: Decisions on rendering resource allocation are made for the 3D trajectory-geographic information fusion model, and multi-level detail level modeling is performed to perform location tracking visualization operations.

[0006] In this specification, a smart bracelet real-time position tracking visualization system is provided, which is used to execute the smart bracelet real-time position tracking visualization method described above, including: The acquisition frequency adjustment module is used to collect multi-dimensional sensor monitoring parameters of the user based on the smart bracelet, conduct real-time motion status assessment of the user, and perform adaptive adjustment of the position acquisition frequency to generate an adaptive position acquisition frequency; The trajectory error compensation module is used to perform real-time position acquisition based on the adaptive position acquisition frequency, and perform frequency-by-frequency spatial position calculation to obtain a high-precision compensated reconstructed trajectory; The scene memory anchor module is used to obtain public electronic map information, identify key elements of the scene, and perform key attribute recognition marking to obtain multiple scene attribute memory anchors; The trajectory visualization module is used to analyze the relative position distribution of high-precision compensated reconstructed trajectories based on multiple scene attribute memory anchor points, perform dynamic trajectory visualization mapping, and construct an anchor point space trajectory visualization mapping diagram; The semantic information rendering module is used to perform visual recognition of the geographic background layer based on the public electronic map information, and then perform semantic information rendering on the anchor point spatial trajectory visualization map to construct a 3D trajectory-geographic information fusion model; The multi-level modeling module is used to make rendering resource allocation decisions for the 3D trajectory-geographic information fusion model and perform multi-level detail level modeling to perform location tracking visualization operations.

[0007] The beneficial effects of the present invention include: By collecting multi-dimensional data (such as acceleration, heart rate, and gait) through a smart bracelet, a user's exercise status can be assessed in real time. This approach automatically adjusts the frequency of location acquisition based on the user's exercise intensity, reducing unnecessary data collection and thus lowering energy consumption. The frequency of location acquisition can be dynamically adjusted for different exercise states (such as walking, running, and stationary), optimizing the data collection process and providing more accurate trajectory reconstruction. Using frequency-by-frequency spatial positioning calculation technology based on real-time positioning data collected at an adaptive frequency, a compensation algorithm can be used to correct positioning errors caused by factors such as unstable satellite signals, thereby ensuring high-precision location data. By optimizing positioning accuracy, users can obtain more accurate trajectory information in various environments, such as complex scenes such as urban high-rise buildings and indoor environments. By accessing publicly available electronic map information and identifying key elements (such as buildings, roads, and landmarks), the user's specific environment can be identified and key features within the scene can be marked. This not only enhances the system's perception capabilities but also provides important scene data support for subsequent trajectory reconstruction and visualization. These scene memory anchors serve as an important reference for subsequent trajectory reconstruction, helping to improve the accuracy of high-precision trajectories. By analyzing multiple scene attributes, memorizing anchor points, and reconstructing trajectories, the system generates dynamic trajectory visualizations in real time. This intuitive display helps users more clearly understand their movements over a specific time period. Not only does it provide accurate trajectory tracking, but it also allows users to visualize their movements in real time through visualization, helping them better plan their routes. The integration of trajectory data with background map information enhances trajectory comprehensibility. Users can clearly see how their trajectories fit within the geographic context (such as streets and buildings), providing a higher level of perception for practical applications. Semantic rendering not only displays the user's location but also provides additional context based on map information, such as highlighting specific area features, increasing the interpretability of trajectory data and improving the user experience. Through resource allocation decisions and modeling at different levels of detail, the system dynamically adjusts the level of detail displayed based on different devices and network environments, ensuring smooth operation and providing an optimal user experience even on low-performance devices. Rendering details are dynamically adjusted based on device performance and network bandwidth to avoid lag or delays caused by excessive rendering. Through multi-level modeling, users can enjoy smooth and clear trajectory visualization in various usage scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 This is a schematic diagram of the steps of a method for visualizing real-time position tracking of a smart wristband according to the present invention; Figure 2 Detailed implementation flow chart of step S1; Figure 3Detailed implementation flow chart of step S2; Figure 4 Schematic diagram of the detailed implementation steps of step S3. DETAILED DESCRIPTION

[0009] It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0010] This application example provides a method and system for visualizing the real-time location tracking of a smart wristband. The execution entities of the method and system include, but are not limited to, the following: mechanical equipment, data processing platforms, cloud server nodes, network upload devices, etc., which can be considered as general computing nodes of this application. The data processing platform includes, but is not limited to, at least one of an audio and image management system, an information management system, and a cloud data management system.

[0011] See also Figures 1 to 4 The present invention provides a method for visualizing the real-time position tracking of a smart bracelet, comprising the following steps: Step S1: Based on the smart bracelet, the user's multi-dimensional sensor monitoring parameters are collected to evaluate the user's real-time motion state, and the position acquisition frequency is adaptively adjusted to generate an adaptive position acquisition frequency; Step S2: Perform real-time position acquisition based on the adaptive position acquisition frequency, and perform frequency-by-frequency spatial position calculation to obtain a high-precision compensated reconstructed trajectory; Step S3: Obtaining public electronic map information, identifying key elements of the scene, and performing key attribute recognition marking to obtain multiple scene attribute memory anchor points; Step S4: performing relative position distribution analysis on the high-precision compensated reconstructed trajectory based on multiple scene attribute memory anchor points, and performing dynamic trajectory visualization mapping to construct an anchor point space trajectory visualization mapping diagram; Step S5: Performing visual recognition of the geographic background layer based on the public electronic map information, and then rendering the semantic information of the anchor point spatial trajectory visualization map to construct a three-dimensional trajectory-geographic information fusion model; Step S6: Decisions on rendering resource allocation are made for the 3D trajectory-geographic information fusion model, and multi-level detail level modeling is performed to perform location tracking visualization operations.

[0012] In the embodiment of the present invention, see Figure 1 , is a flowchart of the steps of a smart wristband real-time location tracking visualization method of the present invention. In this example, the steps of the smart wristband real-time location tracking visualization method include: Step S1: Based on the smart bracelet, the user's multi-dimensional sensor monitoring parameters are collected to evaluate the user's real-time motion state, and the position acquisition frequency is adaptively adjusted to generate an adaptive position acquisition frequency; In this embodiment, the sensors built into the smart bracelet typically include an accelerometer, a gyroscope, a heart rate sensor, a GPS module, and the like. These sensors can collect a variety of physiological and motion data in real time, such as the user's heart rate, gait, speed, acceleration, and direction of movement. Specifically, the bracelet can use the accelerometer and gyroscope to collect the user's motion information, such as walking, running, sitting, and standing. The GPS module provides the user's location information, and the heart rate sensor can reflect the user's exercise intensity and physiological state. This data forms multi-dimensional motion parameters of the user, which can help the system assess the user's real-time motion state. After acquiring this sensor data, the system performs real-time motion state assessment based on a preset motion model. For example, by analyzing the accelerometer and gyroscope data, the system can identify the user's current motion state (e.g., walking, running, cycling, etc.), while the heart rate data can help assess exercise intensity and physiological load. The system may use different motion state assessment models for different motion states. For example, when the user is in a "stationary" or "resting" state, the system will reduce the frequency of location collection to save battery; when the user is in a "fast running" or "cycling" state, the system will increase the frequency of collection to obtain more detailed location information.

[0013] Based on real-time motion state assessment, the system determines whether to adjust the frequency of location collection. This adjustment is adaptive, meaning the system dynamically adjusts the GPS collection frequency according to different motion states. For example, when the user is stationary or in a low-intensity motion state (such as walking or standing), the system can reduce the frequency of GPS data collection to extend battery life, as the user's location information fluctuates less. However, when the user is engaged in high-intensity motion (such as running or cycling), the system increases the collection frequency due to frequent location changes to ensure trajectory precision and accuracy. This process is achieved through real-time motion state monitoring and a frequency adjustment algorithm. The algorithm dynamically adjusts the collection interval (e.g., once per second, five times per second, etc.) based on changes in motion state, ensuring that the collection frequency matches the user's actual needs.

[0014] In practical applications, the system needs to incorporate experimental data to optimize the adaptive acquisition frequency adjustment algorithm. Experimental data typically includes heart rate, gait, speed changes, GPS data collection accuracy, and power consumption under different exercise conditions. For example, in one set of experiments, the user's gait changes can be recorded to determine the appropriate acquisition frequency for different conditions, such as walking, running, and cycling. Assuming that the user's location information changes little per minute when walking, the system can set a lower acquisition frequency (e.g., collecting location information every 2 seconds). However, during fast running, the system needs to increase the frequency (e.g., collecting location information every 0.5 seconds) to ensure tracking accuracy. When adaptively adjusting the location acquisition frequency, the system also needs to consider the battery life of the wristband. Based on experimental data, the system tests battery life at different acquisition frequencies to ensure the optimal balance for different usage scenarios. For example, during low-intensity exercise, the system may use a lower acquisition frequency (to conserve battery), while during high-intensity exercise, the acquisition frequency may be increased to ensure tracking accuracy. After implementing the above exercise state assessment and acquisition frequency adjustment, the system will generate an adaptive acquisition frequency. For example, the system can set a dynamic parameter called "position acquisition frequency," which adjusts as the motion state changes. This parameter works in conjunction with other modules within the system (such as real-time position calculation, trajectory drawing, and motion status display) to ensure that the system provides high-precision position tracking while minimizing device resource consumption.

[0015] Step S2: Perform real-time position acquisition based on the adaptive position acquisition frequency, and perform frequency-by-frequency spatial position calculation to obtain a high-precision compensated reconstructed trajectory; In this embodiment, the smart bracelet periodically triggers GPS or other positioning modules (such as Wi-Fi and Bluetooth) to collect data based on the aforementioned adaptive frequency strategy. This frequency can be every second, every two seconds, or even more frequently, depending on the user's current motion state. For example, when the user is running fast, the location collection frequency may be higher (e.g., every 0.5 seconds) to ensure trajectory smoothness and accuracy; while when walking slowly or stationary, the collection frequency may be lower (e.g., every 3 or 5 seconds). During this process, the bracelet's sensors automatically activate and record GPS coordinates, timestamps, direction, speed, and other data, generating a series of raw location data. For data collected at different frequencies, the system adjusts the time step for positioning calculations based on the change in collection frequency. At lower frequencies, the system uses interpolation methods to fill in the gaps between previous and subsequent location data to ensure trajectory continuity and smoothness. Interpolation methods such as linear interpolation and spline interpolation can help calculate the exact coordinates between two location collections, thereby ensuring a smooth trajectory transition.

[0016] The system also employs a Kalman filter or similar filtering techniques to reduce error accumulation. The Kalman filter integrates the previous position estimate with the currently acquired signal to optimize the final position estimate. By comprehensively analyzing physical quantities such as velocity and acceleration, the filtering algorithm reduces uncertainty caused by positioning errors and changes in motion state, ensuring effective matching between high-frequency and low-frequency acquisitions. Based on previously acquired high- and low-frequency position data, the system employs trajectory smoothing algorithms (such as weighted averaging and B-spline curve fitting) to optimize the original trajectory, remove noise, and correct irregular trajectory points. Furthermore, by considering the surrounding environment (such as road topology) and geographic context, the system can further correct trajectory deviations and eliminate errors caused by signal obstruction or low positioning accuracy. For example, in urban environments, where buildings can cause GPS positioning inaccuracies, the system uses known information about building locations to reconstruct the trajectory, making it more consistent with the actual road network.

[0017] Step S3: Obtaining public electronic map information, identifying key elements of the scene, and performing key attribute recognition marking to obtain multiple scene attribute memory anchor points; In this embodiment, vector maps and satellite imagery data for a specified area are batch-acquired from open electronic map platforms (such as OpenStreetMap and the Google Maps API). This data covers layers such as road networks, building outlines, green spaces, water bodies, transportation facilities, and landmark structures. The data acquisition range is within a 5-kilometer radius centered on the area where the wristband user is sampling, with a resolution of approximately 1 meter per pixel to ensure sufficient detail in the underlying geographic information. Key scene elements are identified for the layers in the map data, primarily using a deep convolutional neural network (such as ResNet50 and EfficientNet) combined with semantic segmentation techniques (such as U-Net and DeepLabv3+) to jointly deconstruct satellite imagery and vector maps. Key elements identified include typical geographic landmarks such as buildings (residential / commercial), main roads, intersections, public transportation nodes (bus stops / subway entrances), green parks, bridges, commercial centers, and medical points. To improve the model's generalization across diverse terrains, training utilizes datasets encompassing multiple cities and scenes (e.g., Cityscapes and MapillaryVistas). Data augmentation (random cropping, illumination perturbation, and rotation) is also introduced to adapt for stability recognition across multiple time periods. Following element recognition, key attribute recognition is further performed. Each identified element is assigned a number of semantic attribute labels, including spatial function (e.g., "transportation node," "commercial core"), pedestrian density level, structural stability, opening hours, and navigational directionality. These attributes are assigned by integrating geographic knowledge graphs (e.g., GeoNames and DBpedia) with local geocoding services (e.g., GeoCoding APIs). For dynamic elements (e.g., subway entrances and bus stops), the system also annotates their operating hours and frequency using open transportation data interfaces (e.g., GTFS). Based on the identified and labeled key elements, the system constructs a set of "scene attribute memory anchors." Each anchor contains a spatial location (latitude and longitude), a set of semantic attributes, an image feature description (extracted through visual feature encoding), and a correlation strength with the user's path (calculated based on visit frequency and path overlap). These anchor points are used for path semantic enhancement and behavioral scenario backtracking during location tracking by smart bracelets, providing auxiliary labels and contextual awareness when visualizing trajectory reconstruction. For example, when a user passes the same bus stop multiple times, the point will be memorized as a high-frequency scene anchor point and highlighted as an icon in the trajectory visualization diagram. In the actual test environment, this method was used to extract an average of approximately 150 to 200 stable anchor points per square kilometer in 5 different urban areas, with a coverage rate of over 93%, laying a key foundation for achieving high-semantic location perception in complex urban environments.

[0018] Step S4: performing relative position distribution analysis on the high-precision compensated reconstructed trajectory based on multiple scene attribute memory anchor points, and performing dynamic trajectory visualization mapping to construct an anchor point space trajectory visualization mapping diagram; In this embodiment, the semantic visualization capabilities of trajectory data are enhanced by performing spatial relative analysis of historical movement trajectories against a pre-built set of scene attribute memory anchor points, enabling the construction of intuitive user-oriented behavioral trajectory maps. First, the high-precision compensated and reconstructed trajectory data from steps S2 and S3 is obtained. This trajectory data is generated by combining GPS sampling (sampling frequency 1Hz) with IMU multi-dimensional parameters (such as accelerometer and gyroscope) for compensation, with an error control within ±2 meters. Trajectory points include timestamps, latitude and longitude, speed, cadence, and movement status (e.g., walking, cycling, or stationary). The system performs sliding window cluster analysis on each trajectory segment (with a window length of 30 seconds and 50% overlap) to extract stable motion segments. The compensated trajectory points are spatially matched against the scene attribute memory anchor point set. A KD-Tree spatial indexing algorithm is used to quickly locate anchor points near the trajectory points (with a query radius set to 50 meters). The spatial proximity and temporal overlap of each anchor point with the trajectory path are calculated to form an anchor point influence factor matrix. The system analyzes the relative position distribution of trajectory data based on these indicators, including: anchor point distribution density (the number of anchor points hit per kilometer), trajectory path deviation (the average deviation between the trajectory and the anchor centerline), and path similarity score (DTW dynamic time warping matching with historical paths) to determine the compatibility of the trajectory with the environmental structure.

[0019] After completing the relative location distribution analysis, the system performs dynamic trajectory visualization. The visualization module uses the WebGL geographic graphics engine, combined with vector mapping APIs (such as Mapbox GL or CesiumJS) to render the trajectories. The main trajectory path is drawn as a time-gradient line segment (with colors shifting from light blue to dark blue to represent time progression). Anchor points are displayed in a hierarchical manner using various icons: for example, transportation anchor points are represented by blue circles, commercial anchor points by orange hexagons, and medical anchor points by red crosses. To support interactive operation of dynamic trajectories, the system implements a timeline controller and path replay functionality, allowing users to drag the timeline to view the relationship between their spatial location and anchor points at a specific moment. The system also introduces an anchor point spatial density heatmap as an auxiliary layer, displaying the spatial concentration of user activity areas. This heatmap is generated using the kernel density estimation method. In field measurements, trajectory visualization analysis was conducted within a 5-square-kilometer area in Chaoyang District, Beijing. The average number of valid anchor points matched per trajectory was 37, resulting in a 91.3% anchor point coverage rate. The constructed "anchor point space trajectory visualization map" not only enhances the semantic recognition ability of motion paths, but also provides a visual basis for subsequent behavioral pattern recognition, abnormal trajectory detection and motion intervention recommendations.

[0020] Step S5: Performing visual recognition of the geographic background layer based on the public electronic map information, and then rendering the semantic information of the anchor point spatial trajectory visualization map to construct a three-dimensional trajectory-geographic information fusion model; In this embodiment, geographic information layer data for the target area is obtained by accessing publicly available electronic map resources (such as OpenStreetMap, Mapbox Map API, or AutoNavi Open Platform). This includes road networks, building outlines, terrain elevation, water distribution, green space boundaries, and points of interest (POIs). The original map data is typically loaded in a tiled structure, with spatial objects organized in GeoJSON, TopoJSON, or Vector Tiles. After acquiring the geographic background layer, the system employs a layer separation and recognition strategy to visually and semantically classify various spatial objects. First, an image semantic segmentation model (such as DeepLabV3+ or HRNet) is used to perform visual structured analysis of the map layer, labeling elements such as roads, buildings, water areas, and green spaces. This process supports GPU acceleration for real-time semantic recognition, with recognition accuracy maintained at an mIoU (mean Intersection-over-Union) level greater than 85%. The segmented layers are then reorganized by semantic category for subsequent trajectory fusion rendering. After completing the semantic geographic layer hierarchy, the system renders the semantic information of the "anchor point spatial trajectory visualization map" generated in the previous step. This process involves aligning the coordinates of the trajectory with the map's semantic layer (using the WGS-84 coordinate system projection) and mapping the trajectory to different colors, thicknesses, and animation effects based on the semantics of user behavior (such as "stay," "continuous movement," and "fast traversal"). The semantic relationships between anchor points and geographic features are also reflected in the map. For example, if a user remains stationary for an extended period near a "medical POI" anchor point, the system labels that trajectory segment as "medical treatment behavior" and renders it as a flashing red path. After the semantic trajectory is fused with the geographic layer, the system constructs a 3D trajectory-geographic information fusion model. This model, powered by a 3D geographic information engine (such as CesiumJS or Three.js with a geographic API extension), extends the 2D map to 3D space through building extrusion (building elevations are derived from the OSM height attribute or DEM digital elevation model) and terrain simulation (based on SRTM or ASTER elevation data). Trajectory data is displayed dynamically in 3D space, positioned at a fixed height above the ground (e.g., 1.5 meters to simulate the center of a human body). Dynamic anchor labels are used to visualize paths through urban structures and into building complexes. The resulting 3D fusion model allows users to analyze trajectory replays from multiple perspectives, including bird's-eye view, first-person perspective, timeline scrolling, and interactive anchor highlighting.

[0021] Step S6: Decisions on rendering resource allocation are made for the 3D trajectory-geographic information fusion model, and multi-level detail level modeling is performed to perform location tracking visualization operations.

[0022] In this embodiment, the system introduces a hierarchical rendering resource scheduling mechanism, hierarchically classifying the resources of rendering units involved in visualization operations (including buildings, terrain blocks, trajectory lines, anchor point annotations, etc.), dividing them into high-priority core areas (the user's current location, critical path segments, and areas of high activity) and low-priority peripheral areas (areas far from the field of view, static background areas), to support asynchronous loading and dynamic resource unloading. The system collects the user's terminal's current hardware configuration parameters (such as GPU model, video memory capacity, and browser WebGL performance score), and combines them with the user's real-time dynamic perspective (view orientation, zoom ratio, and camera height). It uses a dual mechanism based on view frustum culling (View Frustum Culling) and screen space error (SSE) to determine the accuracy of 3D models and allocate resources. For example, when a user is performing path tracing in first-person perspective mode, the forward path and building models are prioritized for loading as high-definition resources (e.g., LOD0 / LOD1 levels), while distant backgrounds or areas not in view are maintained at low-precision levels (e.g., LOD3 / LOD4), dynamically compressing video memory usage. This strategy effectively controls redundant resource loading. Tests show that on a standard embedded terminal (Android platform, 6GB of memory, Adreno GPU), the overall model loading speed is increased by approximately 36%, significantly reducing the frequency of lag.

[0023] The system establishes a level of detail (LOD) hierarchy for map scenes and trajectory rendering models. LOD modeling is based on the spatial complexity of objects, trajectory density, and relevance to user behavior. Trajectory lines are assigned dynamic weights based on their semantic levels (such as sharp turns, high-speed movement, and unusual stops). Trajectories are full-resolution dynamic curves at LOD0 (supporting color coding of motion speed and dynamic animation of direction arrows). At LOD1 and LOD2 levels, they degenerate into simplified polylines and reduce the number of real-time animation frames, thereby reducing the GPU rendering load. Geographic buildings and terrain blocks use an octree structure to divide spatial areas. Each node points to model resource files of different precision at different LODs. For example, LOD0 uses a complete building model (in GLTF format) containing textures and normal maps, while LOD3 uses a simplified version containing only the outline volume (Bounding Box). To maintain the consistency and real-time responsiveness of trajectory visualization during user interaction, the system incorporates a built-in rendering task priority queue management module. This dynamically controls the number of rendering tasks per frame, employing a strategy based on time slice scheduling and importance-driven rendering (Importance-Driven Rendering) based on a perception model. This ensures a minimum frame rate of 25 fps even in highly dynamic scenarios (such as rapid user movement and zooming). By combining a multi-level level of detail (LOD) model with a resource allocation mechanism, the intelligent location tracking visualization platform can operate flexibly across multiple terminals and networks, supporting functional modules such as high-precision motion trajectory visualization, behavior replay, and regional thermal analysis. In actual deployments, the average response time for path rendering and 3D interaction in a campus scenario (approximately 5 km²) was maintained within 0.8 seconds, meeting the real-time and immersive requirements of smart wristband tracking applications and providing a solid data and technical foundation for subsequent location recognition, anomaly detection, and spatial behavior modeling.

[0024] In this embodiment, refer to Figure 2 , is a flowchart of the detailed implementation steps of step S1. In this embodiment, the detailed implementation steps of step S1 include: Collecting multi-dimensional sensor monitoring parameters of the user based on the smart bracelet, wherein the multi-dimensional sensor monitoring parameters of the user include real-time heart rate monitoring of the user, user movement speed and user body posture vibration parameters; Perform periodic decomposition on the user's multi-dimensional sensor monitoring parameters and extract the sensor monitoring parameters of multiple periods; Perform real-time motion state evaluation based on multiple periodic sensor monitoring parameters to obtain the user's motion state in different periods; Performing a time-series motion state change evolution on the motion state to generate a user motion state change curve; The position acquisition frequency of the smart bracelet is adaptively adjusted based on the user's motion state change curve to generate an adaptive position acquisition frequency.

[0025] In this embodiment, based on the multi-dimensional sensor monitoring parameters collected by the smart bracelet, the system first obtains multiple key physiological and motion data. Specifically, the bracelet monitors the user's heart rate, movement speed and vibration characteristics of the body posture in real time through built-in optical heart rate sensors, accelerometers, gyroscopes and other sensors. Real-time heart rate monitoring mainly detects changes in blood flow through optical sensors to infer the user's heart rate. The movement speed is analyzed by combining the accelerometer and gyroscope to analyze the user's acceleration and angular velocity to derive the actual walking or movement speed. The vibration parameters of the body posture capture the movement frequency and amplitude of the user's body through the accelerometer to identify different types of movements, such as gait, jumping or violent movement changes. The real-time acquisition of these sensor data provides the basis for subsequent cycle analysis and motion state evaluation.

[0026] After data collection, it is next necessary to perform cycle decomposition. The purpose of cycle decomposition is to extract motion characteristics within each cycle for more detailed analysis. Typically, cycles can be divided into fixed time windows (such as every minute or every five minutes), or dynamically based on the user's movement rhythm. For each cycle, the collected data, including heart rate, speed, and posture vibration, is first preprocessed to remove noise and outliers. Next, statistical features within the cycle are extracted, such as the maximum, minimum, mean, and volatility of heart rate, the average and fluctuation range of speed, and the frequency characteristics of posture changes. These features reflect the user's movement status and physiological load at different time periods, forming the basis for subsequent movement status assessment and time series analysis.

[0027] Based on the periodically decomposed feature data, the smart bracelet performs real-time exercise status assessment. Using machine learning algorithms such as support vector machines (SVMs), decision trees, or deep learning models, the system can determine the user's exercise status in real time based on changes in heart rate, speed, and posture. For example, if the user's heart rate is high and their speed is fast, the system may determine that they are in a high-intensity exercise state; if their heart rate is low and their speed is slow, it may be judged as low-intensity exercise or resting. In this way, the system can identify the user's current exercise intensity and activity type in real time, providing data support for subsequent analysis of changes in exercise status.

[0028] After obtaining the user's real-time exercise status, the next step is to analyze time-series changes. The goal of this analysis is to track and predict changing trends in the user's exercise status. Using time-series analysis models (such as ARIMA models and long short-term memory (LSTM) networks), the system can model the user's exercise status over multiple periods and generate a curve of the user's exercise status. This curve reflects the user's exercise intensity, activity type, and their changing patterns over different time periods, helping the system understand the user's exercise patterns and recovery status. For example, if a user's exercise status changes frequently and their heart rate fluctuates significantly over a period of time, the system can identify intermittent intense exercise. On the other hand, if the heart rate remains relatively stable and the movement speed is low, it indicates low-intensity activity or a state of inactivity.

[0029] Based on the changing curve of the user's motion state, the smart bracelet can adaptively adjust its location collection frequency. Specifically, the system dynamically adjusts the location update frequency of the bracelet by analyzing the user's current motion intensity and state. When the user is stationary or exercising at low intensity, the bracelet can reduce the location collection frequency (for example, once per minute); when exercising at high intensity or when the state changes drastically, the bracelet will increase the collection frequency (such as once per second) to ensure the real-time and accuracy of the location. In this way, the smart bracelet can balance the accuracy of position tracking and battery life by adjusting the collection frequency. During this process, the system will adaptively adjust the frequency based on information such as fluctuations in the motion state change curve, acceleration, and heart rate to ensure the best balance between energy consumption and accuracy.

[0030] In this embodiment, refer to Figure 3 , is a flowchart of the detailed implementation steps of step S2. In this embodiment, the detailed implementation steps of step S2 include: Based on the adaptive location acquisition frequency, the smart bracelet is used to collect real-time location positioning and obtain the user's real-time location information stream; Identify abnormal location information of users' real-time location information stream and perform dynamic filtering optimization to obtain filtered optimized location information stream; Perform frequency-by-frequency spatial position positioning calculation on the filtered optimized position information stream to generate the spatial position coordinates of each acquisition frequency; Calculate the standard timestamp of each acquisition frequency based on the filtered optimized position information stream; Performing multi-time point position sequential evolution of the spatial position coordinates based on the standard timestamp to construct a real-time position coordinate tracking trajectory; The real-time position coordinate tracking trajectory is dynamically compensated and reconstructed to obtain a high-precision compensated reconstructed trajectory.

[0031] In this embodiment, the smart bracelet collects the user's location in real time based on an adaptive location acquisition frequency. The system first dynamically adjusts the location acquisition frequency based on the user's real-time motion state. For example, when the user is in a high-intensity motion state, the acquisition frequency increases to ensure real-time capture of the user's location; when the user is in a low-intensity or stationary state, the acquisition frequency decreases to reduce energy consumption. The system uses an adaptive algorithm to adjust the location acquisition frequency based on the motion state curve and fluctuations in data such as acceleration and heart rate. Typically, the acquisition frequency is dynamically adjusted based on a set threshold (for example, collecting location information once per second during high-intensity motion and once per minute during low-intensity motion). Through this process, the smart bracelet generates a real-time user location information stream—a stream of coordinate data that changes over time—ensuring real-time and accurate location tracking. After the real-time location information stream is generated, it is then subjected to abnormal location information identification and dynamic filtering optimization. Due to sensor errors, signal interference, or external environmental influences, the real-time location information stream may contain some abnormal data points, which can affect positioning accuracy. Therefore, an anomaly detection algorithm is first required to identify these abnormal location information. For example, statistical outlier detection methods (such as the 3σ rule) or clustering algorithms from machine learning can be used to identify outliers or data points with significant deviations. This anomalous location information may be caused by GPS signal loss, sensor noise, or sudden user movements. After identifying anomalies, the system optimizes the location information stream using dynamic filtering algorithms, such as Kalman filtering and particle filtering, to smooth the location information stream. Kalman filtering can predict and correct based on the system's dynamic model and measurement errors, effectively removing abnormal fluctuations while preserving the overall trend of the location information. The filtered location information stream becomes smoother, thereby improving the accuracy and reliability of location tracking.

[0032] The position information stream, optimized through dynamic filtering, is used for frequency-by-frequency spatial positioning. Specifically, at different acquisition frequencies, the smart bracelet calculates the spatial coordinates of each time point based on the location data and the acquisition frequency. For example, if the acquisition frequency is high (e.g., once per second), the system gradually calculates the spatial coordinates for each second based on the timestamp and the changes in the position information stream. If the acquisition frequency is low (e.g., once per minute), the spatial coordinates for each cycle are determined through interpolation or time extrapolation. The key to this step lies in how to properly interpolate the position data based on the uneven acquisition frequency, ensuring accurate spatial position calculation for each acquisition cycle. Common methods include linear interpolation and spline interpolation. After calculating the spatial coordinates for each acquisition cycle, a standard timestamp must be calculated for each acquisition frequency. The standard timestamp unifies the time base of data collected at different acquisition frequencies, ensuring accurate alignment of location data at different frequencies. This step typically uses interpolation to unify the timestamps corresponding to each acquisition point with those of other time points. For data collected at a high frequency, timestamps may be densely packed, while for data collected at a low frequency, timestamps may be sparser. Therefore, it is necessary to use interpolation algorithms (such as linear interpolation and timing alignment methods) to adjust the timestamps at different frequencies to a unified standard timestamp. These standard timestamps provide a time reference for subsequent trajectory evolution and trajectory compensation.

[0033] Based on standard timestamps, smart wristbands can sequentially evolve spatial coordinates at multiple points in time, constructing a complete real-time location tracking trajectory. By uniformly managing standard timestamps, the system can smoothly connect location data collected at different frequencies on the timeline to form a continuous trajectory. During this process, the smart wristband combines location information from multiple moments and applies trajectory prediction algorithms, such as Kalman filtering or Bayesian filtering, to infer the user's trajectory. The goal of trajectory evolution is to infer the user's movement path over different time periods using the known spatial coordinates of multiple time points. Trajectory evolution considers not only spatial coordinates but also the user's motion state (such as acceleration and velocity), resulting in more accurate trajectories. For real-time location tracking trajectories, the smart wristband performs dynamic trajectory compensation and reconstruction to produce a highly accurate compensated trajectory. In actual use, due to factors such as environmental interference and sensor errors, the tracking trajectory may deviate or be lost. Therefore, dynamic trajectory compensation and reconstruction can smooth and correct the trajectory based on the relationship between previous and subsequent positions and the user's motion model. Common compensation methods include trajectory compensation based on motion models and trajectory correction based on Kalman filtering. The system predicts the user's direction and path based on motion status, acceleration, and velocity information, and reconstructs the position trajectory. This allows the system to effectively correct incomplete or inaccurate position information, ensuring the resulting trajectory is as close to the true trajectory as possible.

[0034] In this embodiment, the specific steps of performing dynamic trajectory compensation and reconstruction on the real-time position coordinate tracking trajectory to obtain a high-precision compensated and reconstructed trajectory are as follows: Obtain inertial measurement unit sensor data based on smart bracelet; The Kalman filter estimation is performed based on the inertial measurement unit sensor data to obtain the user's relative displacement estimate; The user's real-time movement trajectory is estimated based on the user's relative displacement estimation value, and the trajectory deviation of the real-time position coordinate tracking trajectory is calculated point by point to generate multi-point trajectory deviation values; Based on the multi-point trajectory deviation values, deviation accumulation calculation is performed to generate auxiliary trajectory correction parameters; Perform road topology constraint analysis based on publicly available electronic map information and generate road topology constraint conditions for real-time trajectories; The real-time position coordinate tracking trajectory is dynamically compensated and reconstructed according to the road topology constraint conditions and the auxiliary trajectory correction parameters to obtain a high-precision compensated and reconstructed trajectory.

[0035] In this embodiment, the IMU incorporates multiple sensors, such as accelerometers, gyroscopes, and magnetometers. These sensors work together to capture information such as acceleration, angular velocity, and direction during a user's movement. In a dynamic environment, IMU sensors continuously record changes in the user's movements. Using the data collected by these sensors, the system can perceive the user's movements, rotations, and posture changes. This data is subject to noise and interference during the acquisition process, necessitating filtering. To obtain accurate position information from these sensors, the smart bracelet performs a preliminary Kalman filter on the collected raw data to reduce the impact of noise on displacement estimation and ensure the final motion information is as accurate as possible. During this process, the bracelet processes each frame of IMU data in real time, capturing the user's instantaneous displacement, velocity, and acceleration changes, laying the foundation for subsequent trajectory estimation. Based on the Kalman filtered IMU data, the system estimates the user's relative displacement. The Kalman filter algorithm is essentially a recursive optimization method that continuously refines the estimated value by integrating a dynamic model with sensor data to minimize error. When estimating the user's relative displacement, the system uses accelerometer data to calculate the user's acceleration change and derives the estimated displacement by integrating the velocity. Because IMU data can contain accumulated errors (such as integration errors), Kalman filtering dynamically adjusts the predicted value to correct for drift errors in the system. This method enables the wristband to accurately estimate the user's displacement relative to the initial position, forming a stable displacement estimation trajectory. In practice, this process is typically adjusted based on set filtering parameters (such as process noise and measurement noise) to adapt to different motion states and environmental conditions.

[0036] After obtaining the user's estimated relative displacement, the system further estimates the user's real-time trajectory based on these estimates. This step aims to depict the user's complete trajectory through continuous displacement updates. The smart bracelet uses the timestamp and calculated relative displacement values ​​to continuously update the user's current location and estimate their spatial coordinates at different points in time. However, due to the accumulation of inertial navigation errors, the estimated trajectory may deviate from the actual path. Therefore, the system performs point-by-point trajectory deviation calculations. By calculating the deviation between the user's current and previous positions, the system can assess the trajectory error at each point in time. Deviation calculations typically use metrics such as Euclidean distance or Manhattan distance, which quantify the user's positioning error at different points in time. These deviations provide the necessary basis for subsequent trajectory corrections, helping the system monitor and adjust trajectory accuracy in real time. Based on the multi-point trajectory deviations calculated point by point, the system accumulates the deviations to generate auxiliary trajectory correction parameters. Because inertial navigation errors accumulate over time, a single deviation value is insufficient for accurate trajectory correction. The trend of the accumulated deviations can reveal the pattern of error accumulation. By accumulating the deviations, the system can identify patterns in error variation and generate correction parameters. These correction parameters typically include adjustments to the trajectory's scale, direction, and positional deviation, providing critical reference data for subsequent dynamic compensation. For example, if the user's gait is relatively regular, the system can analyze the accumulation of trajectory errors to infer the specific direction and magnitude of trajectory correction, thereby generating accurate correction parameters.

[0037] After this, the smart bracelet uses publicly available electronic map information to perform a road topology constraint analysis and generate the road topology constraint conditions for the real-time trajectory. Road topology analysis relies on map data (such as road structure, road direction, and intersection configuration) and simulates the user's trajectory to ensure that the trajectory adheres to the road network's constraints. Electronic maps provide not only geographical information about roads but also information about road connections and traffic regulations, all of which can aid in trajectory correction. For example, if the system detects that the user's trajectory deviates from the actual road network, the system will "constrain" the trajectory to the legal road network based on the road topology analysis. This process helps eliminate trajectory anomalies caused by sensor errors or extreme environmental conditions, ensuring that the trajectory more closely matches the actual road network structure. Combining the road topology constraints and auxiliary trajectory correction parameters, the smart bracelet dynamically compensates and reconstructs the real-time position coordinate tracking trajectory, generating a highly accurate compensated trajectory. During this stage, the system combines the road topology constraints with the correction parameters and employs dynamic trajectory compensation algorithms (such as Kalman filtering or particle filtering) to optimize the real-time trajectory. These compensation algorithms, by combining road information with correction parameters, correct errors in the user's trajectory in real time, ensuring that the trajectory closely matches the actual movement path. During the compensation process, the system takes into account the topology of the road to avoid undesirable trajectories, such as crossing walls or straying from the road. Through multiple iterations, the resulting trajectory achieves a high balance between real-time performance and accuracy, meeting the smart bracelet's requirement for real-time tracking of the user's motion path.

[0038] In this embodiment, reference Figure 4 The above is a flowchart of the detailed implementation steps of step S3. In this embodiment, the detailed implementation steps of step S3 include: Obtain public electronic map information, perform position matching on the high-precision compensated reconstructed trajectory, and perform fuzzy perception of scene information to extract scene information of the user's path trajectory; Identifying the road structure and key scene elements of the scene information, and marking a plurality of key scene element points; Map attribute analysis is performed on multiple key scene element points, and key attribute recognition marking is performed to obtain multiple scene attribute memory anchor points.

[0039] In this embodiment, publicly available electronic map information is obtained to ensure that the system can accurately match and compensate trajectories based on the actual geographic environment. Electronic maps typically contain detailed geographic information about road networks, buildings, traffic signals, intersections, and more. In this step, the system extracts information about roads, landmarks, buildings, and traffic signs from open data sources (such as OpenStreetMap and Google Maps) to construct a model of the user's actual environment. This environmental model provides the underlying data for subsequent trajectory matching and scene recognition. After the map data is imported into the system, the system performs position matching based on the high-precision compensated trajectory collected by the user's smart bracelet and the map data. By comparing the user's movement path with the road network on the electronic map, the system can determine the user's exact location and direction of movement. At this point, map matching algorithms (such as nearest neighbor matching and dynamic time warping (DTW)) are used to correct and match the trajectory data to avoid trajectory drift caused by errors. Through this process, the system ensures that the user's movement path is accurately displayed on the map and can be adjusted in real time. The system also performs scene information fuzzy perception on the user's path. Because real-time trajectory data may contain certain noise or errors, simple path matching cannot meet the requirements of high-precision scene perception. Fuzzy perception algorithms play a crucial role in this regard, helping the system perceive and identify the environmental scenes traversed by the user even with inaccurate trajectory data. The core idea of ​​fuzzy perception is to extract key scene features by fuzzifying the temporal and spatial variations in trajectory data. For example, when a user's trajectory approaches a complex intersection or a landmark, the system can identify the likely scene location based on the trajectory's changing pattern and map information. This process utilizes methods such as fuzzy logic and Bayesian reasoning, using probabilistic estimation and fuzzy inference of different scene features to identify the environmental elements the user passed through. In experiments, through multiple comparisons of different trajectory data and environmental data, the system was able to accurately extract scene information traversed by the user within a certain error range, laying the foundation for subsequent scene analysis and attribute recognition.

[0040] Based on scene information extraction, the next step is to identify road structures and key scene elements within the extracted scene information. This process relies on the structured information of the map. The system compares the user's trajectory points with the road topology to identify key scene elements such as road types and intersection locations. For example, if the user's trajectory approaches a bridge or tunnel, the system automatically identifies these unique structures and marks them as key scene elements. Scene recognition algorithms typically employ topological analysis methods based on graph theory. By analyzing the road network graph, they determine whether the user's trajectory passes through specific geographical structures or transportation facilities (such as bridges, tunnels, and large buildings). Furthermore, the system further verifies the alignment of the trajectory path with the road network through contextual analysis of geographic information, thereby extracting precise scene elements. The key to this stage lies in the deep integration of the trajectory and the environment. By using the road topology, the system automatically identifies key scene elements that the user has passed through, ensuring accurate environmental perception of each segment of the route. The system then performs map attribute analysis on multiple key scene element points and labels them with key attribute recognition, generating multiple scene attribute memory anchors. Each scene element not only has its geographic coordinates but also includes a range of additional attribute information, such as road type, transportation facility type, building nature (e.g., residential, commercial, government), traffic light status, speed limit signs, and more. Based on this attribute information, the system uses attribute analysis algorithms (such as machine learning-based classification methods) to deeply identify key scene elements. For example, by analyzing road attributes in a digital map, the system can identify whether the road a user is traveling on is a main or secondary road, or whether the location is located in a densely trafficked area. This deep understanding of these attributes allows the system to assign attribute labels to each key scene element and generate memory anchors. These anchors represent important locations in the trajectory, enabling the system to quickly retrace and compare scenes in subsequent analysis, enhancing its understanding of the scene. The generation of scene attribute memory anchors is essentially a semantic integration of the trajectory with environmental information. Each anchor point is not only a geographic coordinate point but also carries contextual understanding of the location. For example, when a user passes through a transportation hub, the system can label that location as a "transportation hub anchor" and attach information such as traffic density and road type. This information will further help the system provide accurate scenario support in subsequent trajectory calculation and path planning.

[0041] In this embodiment, step S4 includes the following steps: Based on multiple scene attribute memory anchor points, the relative position distribution of the high-precision compensation reconstruction trajectory is analyzed to obtain the relative position of each anchor point; Performing omnidirectional distribution fitting of anchor points according to the relative positions to construct an anchor point distribution map; Calculate the path timestamp of each anchor point based on the high-precision compensated reconstructed trajectory; Dynamic trajectory visualization mapping is performed based on the path timestamps to construct an anchor point space trajectory visualization mapping diagram.

[0042] In this embodiment, the goal of performing relative position distribution analysis on high-precision compensated reconstructed trajectories based on multiple scene attribute memory anchor points is to accurately quantify the positional relationship of each scene anchor point relative to other anchor points. The core of this process is to analyze the trajectory using a spatial coordinate system, compare the position information of each anchor point with that of the other anchor points, and extract their relative position information. Specifically, the system calculates the distance, angle, and direction differences between each pair of scene anchor points to construct the relative position of each anchor point relative to the others. This analysis reveals the spatial distribution pattern between anchor points and provides a basic spatial framework for subsequent trajectory visualization. In practice, the system uses Euclidean distance, Manhattan distance, or other geometric distance calculation methods to quantify the spatial relationship between anchor points. In experiments, assuming multiple anchor points (such as intersections, traffic lights, buildings, etc.), the system calculates the relative position of each anchor point one by one and generates a data table or graphical model containing the relative position information of all anchor points. This process helps to establish the spatial relationship between anchor points and provides data support for subsequent trajectory fitting and visualization. The step of fitting the omnidirectional distribution of anchor points based on the relative positions aims to construct a comprehensive anchor point distribution map by analyzing the relative position data of each anchor point. This process statistically analyzes the relative position information of all anchor points to identify the spatial distribution pattern of the anchor points. Specifically, the system uses a fitting algorithm (such as least squares, polynomial fitting, or kriging interpolation) to model the spatial distribution of each anchor point. During the fitting process, the system calculates the distribution trend of the anchor points in two-dimensional or three-dimensional space and uses the fitted function or model to depict the spatial distribution map of the anchor points in the trajectory. For example, anchor points may be distributed at specific locations such as road intersections or near buildings. Based on this location information, the system generates a clear distribution map showing the layout of the anchor points throughout the trajectory. To enhance the accuracy of the fitting results, the system may also consider the error range in the experiment and adjust the fitting accuracy to ensure that the final distribution map conforms to the actual scenario. The anchor point distribution map allows the system to identify the most densely populated areas in the trajectory, further improving the accuracy of subsequent trajectory correction and optimization.

[0043] The system calculates the path timestamp for each anchor point based on the highly accurate compensated reconstructed trajectory. This process aims to assign a timestamp to each anchor point, indicating the specific time the user passed through it. To achieve this, the system matches the location data in the trajectory with the temporal information to calculate the time it took the user to reach each anchor point from the starting point. Specifically, the system associates each anchor point with a timestamp in the trajectory to obtain the transit time of each anchor point. For example, if a user's trajectory contains multiple time stamps, the system calculates the time the user arrived at each anchor point based on the location of the anchor points (such as intersections, traffic facilities, etc.) and the time interval between the two paths. This timestamp is crucial for trajectory visualization, as it helps the system reconstruct the user's real-time motion and accurately reconstruct the temporal changes in the motion path. This process can be achieved through methods such as linear interpolation or time series analysis, ensuring that the timestamps of each anchor point are sufficiently accurate and consistent with the temporal patterns of actual motion. Dynamic trajectory visualization is performed based on the path timestamps. The goal of constructing a spatial trajectory visualization map of anchor points is to visualize the user's real-time trajectory and the temporal information of the scene anchor points. This step enables the system to intuitively display the user's motion trajectory and the changes in path time. The system uses the route timestamps and the spatial distribution of the trajectory to construct a dynamic trajectory visualization. This map not only displays the user's movement path but also shows the position changes of each anchor point along the timeline. This map shows the user's spatial position at different time points, helping users understand the evolution of their movement trajectory over time. In practical applications, the system may use dynamic graphical display techniques such as heat maps, time-lapse animations, or 3D trajectory rendering to present the temporal and spatial relationships between anchor points and trajectories. For example, the system can use color gradients to represent changes over time or use animations to display the user's movement status over different time periods. This visualization not only helps users track their movement path in real time but also provides data support for subsequent path optimization, motion analysis, and other processes. Through these steps, the smart bracelet can build an intuitive and dynamic trajectory visualization system based on high-precision trajectory reconstruction, combined with multiple scene attribute anchor points and route time.

[0044] In this embodiment, step S5 includes the following steps: Calculate the real-time climb rate, slope change and elevation profile based on the high-precision compensated reconstructed trajectory to obtain the trajectory terrain characteristic parameters; Perform visual recognition of geographic background layers based on publicly available electronic map information and extract the geographic background layers; Based on the trajectory terrain characteristic parameters and geographic background layer, spatial trajectory correlation matching is performed to obtain trajectory geographic semantic information; Based on the trajectory geographic semantic information, the semantic information of the anchor point spatial trajectory visualization map is rendered to construct a three-dimensional trajectory-geographic information fusion model.

[0045] In this embodiment, the step of calculating real-time climb rate, slope change, and elevation profile based on the high-precision compensated reconstructed trajectory to obtain trajectory terrain characteristic parameters aims to combine the user's motion trajectory with terrain information to accurately assess the user's motion status in different geographical environments. This process relies on the longitude and latitude information and elevation data in the trajectory data. The climb rate and slope change are calculated by calculating the altitude change of each trajectory point in each frame. Specifically, the climb rate is typically calculated based on the altitude difference and horizontal distance between two consecutive trajectory points. The climb rate can be expressed as the ratio of the altitude difference divided by the horizontal distance, usually in units of percentage. In this process, the altitude of each trajectory point is recorded during the experiment. Combined with the time interval between adjacent trajectory points, the user's speed change over the terrain is calculated. In addition, slope change refers to the degree of change in the terrain undulation encountered by the user during movement. The slope is typically calculated based on the ratio of the elevation difference between adjacent trajectory points to the geographical horizontal distance to obtain the slope value. The system analyzes the trajectory elevation profile in real time and generates a curve based on this to show the terrain changes at different locations of the user. By analyzing the terrain characteristic parameters of a trajectory, the system accurately understands the user's geographic environment, providing data support for subsequent route planning and motion status analysis. The system visually identifies and extracts a geographic background layer based on publicly available electronic map information. This process combines the electronic map's background data (such as terrain, roads, buildings, and natural landscapes) with the trajectory data to provide users with richer context. The geographic background layer includes various types of geographic information, such as elevation data, topographic maps, road networks, building distribution, and green spaces. The system accesses publicly available map data sources (such as OpenStreetMap, Google Maps, and SRTM) to extract these geographic elements and present them as a map background layer. In practice, the system queries the map for the corresponding geographic information based on the coordinates of each point in the trajectory data. Using a geographic information extraction algorithm, the system extracts the geographic background information associated with the trajectory from the map and visually overlays it on the trajectory's visualization layer. For example, if a user passes through a mountainous area or forest, the system can automatically display a terrain relief layer, providing more intuitive context and helping the user understand their surroundings.

[0046] Based on the trajectory's terrain characteristic parameters and the geographic background layer, the system performs spatial trajectory correlation matching to obtain the trajectory's geographic semantics. This process extracts geographically meaningful features by matching the geographic data in the trajectory with the geographic background information in the electronic map. Specifically, the system first matches the spatial coordinates of the user's movement trajectory, associating each point in the trajectory with a specific location on the map. Then, based on the trajectory's terrain characteristic parameters (such as climb rate and slope), combined with the geographic information in the background layer, the system performs a comprehensive analysis to obtain the geographic semantics of the trajectory path. For example, if the user moves on an uphill section of road, the system can infer that this section of the path is a "ramp" and assign it a geographic semantic label such as "climbing section." If the user passes through a forest, the system can label this section of the trajectory as "forest area." Through in-depth matching of trajectory and map data, the system can provide geographically meaningful semantic information for each trajectory point or segment. This geographic semantic information not only supports subsequent trajectory analysis but also enhances the user experience, allowing the trajectory visualization to not only display the location but also convey rich content about the environment and terrain. Based on the trajectory's geographic semantics, the system renders the anchor point's spatial trajectory visualization map with semantic information, constructing a 3D trajectory-geographic information fusion model. This process combines geographic semantics with the user's movement trajectory, displaying the user's movement path and the geographic environment through 3D visualization technology. Specifically, the system renders the trajectory points according to 3D spatial coordinates based on the geographic areas traversed by the user and the temporal changes in the trajectory. Different visual effects are applied to each area based on the geographic semantics. For example, the system might assign different colors or gradients to uphill sections to visually demonstrate the gradient; for forested areas, the system might use a green marker to emphasize the area's natural beauty. During the visualization process, the user's trajectory is not only displayed on a flat map but also converted into a 3D graphic, showcasing elevation changes and terrain undulations. By fusing trajectory and geographic semantics, the system can present a more vivid and accurate trajectory visualization. This 3D model not only helps users understand their movement but also provides decision support for subsequent movement planning and path optimization.

[0047] In this embodiment, step S6 includes the following steps: Calculate the trajectory complexity of the 3D trajectory-geographic information fusion model and obtain a visual trajectory complexity index; Perform real-time network status and device performance detection on the smart bracelet to generate the real-time status of the smart bracelet; Make rendering resource allocation decisions based on the visualization trajectory complexity index according to the real-time status of the smart bracelet and generate adaptive rendering strategy parameters; Based on the adaptive rendering strategy parameters, multi-level detail level modeling is performed to construct a four-layer visualization rendering engine consisting of map base layer, trajectory line layer, motion status identification layer and interactive control layer to perform position tracking visualization tasks.

[0048] In this embodiment, trajectory complexity is not only affected by factors such as the user's trajectory length and speed, but also by multiple factors such as terrain undulations, road conditions, and trajectory curvature. In this step, the system analyzes the curve changes of the high-precision compensated trajectory in three-dimensional space, for example, calculating the curvature, climbing rate, and slope change of each trajectory segment. Specifically, trajectory complexity can be quantified using the following methods: First, the curvature change of each trajectory segment is calculated; the greater the curvature, the more complex the trajectory; second, the severity of the altitude change within the trajectory is assessed; the steeper the slope, the higher the trajectory complexity; finally, the frequency of trajectory changes in space and the number of path turns are considered; frequent turns and frequent slope changes increase trajectory complexity. These complexity parameters are typically quantified using mathematical models such as curvature calculation and gradient analysis. In the experiment, a trajectory segment is assumed to pass through multiple different types of roads (e.g., straight roads, curves, and slopes). By analyzing each segment, a complexity index for each segment can be obtained. Finally, the system comprehensively calculates these complexity data to generate an overall visual trajectory complexity index to describe the complexity of the entire trajectory.

[0049] The wristband monitors factors such as network signal strength, processor load, memory usage, and battery charge in real time. Network status monitoring typically includes key metrics such as signal strength, data transfer rate, and latency; device performance monitoring focuses on CPU utilization, memory usage, and device temperature. The system periodically collects and analyzes this information using built-in diagnostic tools. For example, if the device's processor load is too high, the system will reduce unnecessary rendering tasks to reduce the pressure on computing resources. If the network is unstable, the system will reduce the frequency of data transmission or reduce the amount of real-time updates to ensure the stability of real-time location tracking. In experiments, different network conditions and device performance environments can be simulated to test the system's response strategies in various situations, ensuring that the device can operate normally under different workloads and network environments. Based on the real-time status of the smart wristband, the system will make rendering resource allocation decisions based on the visualization trajectory complexity index and generate adaptive rendering strategy parameters. The visualization rendering details and data update frequency are intelligently adjusted based on the wristband's current status (such as device load, network bandwidth, and battery charge). For example, when device performance is poor or the network is poor, the system will reduce the level of detail of the visualization, reducing the rendering workload for each frame of graphics to avoid device overload or latency caused by excessive rendering demands. When the device is operating efficiently, the system will enhance the rendering details to provide higher-quality graphics. Furthermore, the system dynamically allocates rendering resources based on the trajectory complexity index. For simple trajectories, the system uses a lower rendering level, while for complex trajectories, it selects a higher rendering quality. This process balances device performance and user experience by dynamically adjusting rendering detail and frequency. During the experiment, the effectiveness of the system's adaptive rendering strategy was tested in multiple environments simulating different performance conditions and network bandwidths to ensure a stable user experience in various environments. Based on the adaptive rendering strategy parameters, the system performs multi-level detail modeling and constructs a four-layer visualization rendering engine: a map base layer, a trajectory line layer, a motion status identification layer, and an interactive control layer. This process, through a layered design, allocates rendering tasks to different rendering layers, ensuring that the rendering load is properly distributed across different components and avoiding performance degradation caused by excessive rendering. Specifically: Map base layer: This is the most basic layer, containing geographic background information such as roads, landmarks, and buildings. This layer has relatively low rendering complexity and primarily displays static map content. Based on an adaptive rendering strategy, when device performance is poor, the base map's level of detail can be reduced, eliminating unnecessary graphic detail and thus alleviating rendering pressure.

[0050] Track Lines Layer: This layer primarily displays the user's movement trajectory. Based on the trajectory's complexity index, the system dynamically adjusts the trajectory's drawing details. In simple trajectory scenarios, the track lines can be simplified; in complex paths or areas with many curves, the track's line details are increased to ensure clarity.

[0051] Motion Status Identification Layer: This layer displays the user's current motion status, such as walking, running, and paused. The content and presentation of this layer may vary depending on the wristband's real-time status. For example, under high load, the system may reduce or simplify the update frequency of the status icon to ensure rendering efficiency.

[0052] Interaction Control Layer: This layer provides users with real-time interactive features, such as zooming in and out, and rotating the map. This layer is highly complex, requiring real-time responses to user input. When device performance is low, the system will appropriately slow down the interactive response to ensure a smooth user experience.

[0053] In the design of the entire rendering engine, the level of detail of each layer can be dynamically adjusted according to the current state of the device and the complexity of the trajectory. Through this multi-level detail modeling, the system can always provide a smooth and efficient rendering experience under different operating conditions. For example, if the device is in a low-performance state, the system will reduce the complexity of the interactive control layer and shorten the time of each rendering; when the device performance is better, the rendering details of each layer can be enhanced to provide a more refined image performance. Through the collaborative work of these four layers of visual rendering engines, the system can adaptively adjust the rendering quality according to the current device performance and network conditions, ensuring that users can get a balanced performance and user experience when using smart bracelets for location tracking.

[0054] In this embodiment, a smart wristband real-time position tracking visualization system is provided, which is used to execute the smart wristband real-time position tracking visualization method described above, including: The acquisition frequency adjustment module is used to collect multi-dimensional sensor monitoring parameters of the user based on the smart bracelet, conduct real-time motion status assessment of the user, and perform adaptive adjustment of the position acquisition frequency to generate an adaptive position acquisition frequency; The trajectory error compensation module is used to perform real-time position acquisition based on the adaptive position acquisition frequency, and perform frequency-by-frequency spatial position calculation to obtain a high-precision compensated reconstructed trajectory; The scene memory anchor module is used to obtain public electronic map information, identify key elements of the scene, and perform key attribute recognition marking to obtain multiple scene attribute memory anchors; The trajectory visualization module is used to analyze the relative position distribution of high-precision compensated reconstructed trajectories based on multiple scene attribute memory anchor points, perform dynamic trajectory visualization mapping, and construct an anchor point space trajectory visualization mapping diagram; The semantic information rendering module is used to perform visual recognition of the geographic background layer based on the public electronic map information, and then perform semantic information rendering on the anchor point spatial trajectory visualization map to construct a 3D trajectory-geographic information fusion model; The multi-level modeling module is used to make rendering resource allocation decisions for the 3D trajectory-geographic information fusion model and perform multi-level detail level modeling to perform location tracking visualization operations.

[0055] The present invention is therefore intended to be illustrative and non-restrictive in all respects, with the scope of the invention being defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the application documents are intended to be embraced therein.

[0056] The foregoing description is intended only to provide specific embodiments of the present invention, which are intended to enable those skilled in the art to understand and implement the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not intended to be limited to the embodiments shown herein, but is to be construed in the widest manner consistent with the principles and novel features disclosed herein.

Claims

1. A method for visualizing real-time position tracking of a smart bracelet, characterized in that: The following steps are involved: Step S1: Based on the smart bracelet, the user's multi-dimensional sensor monitoring parameters are collected to evaluate the user's real-time motion state, and the position acquisition frequency is adaptively adjusted to generate an adaptive position acquisition frequency; Step S2: Perform real-time position acquisition based on the adaptive position acquisition frequency, and perform frequency-by-frequency spatial position calculation to obtain a high-precision compensated reconstructed trajectory; Step S3: Obtaining public electronic map information, identifying key elements of the scene, and performing key attribute recognition marking to obtain multiple scene attribute memory anchor points; Step S4: performing relative position distribution analysis on the high-precision compensated reconstructed trajectory based on multiple scene attribute memory anchor points, and performing dynamic trajectory visualization mapping to construct an anchor point space trajectory visualization mapping diagram; Step S5: Performing visual recognition of the geographic background layer based on the public electronic map information, and then rendering the semantic information of the anchor point spatial trajectory visualization map to construct a three-dimensional trajectory-geographic information fusion model; Step S6: Decisions on rendering resource allocation are made for the 3D trajectory-geographic information fusion model, and multi-level detail level modeling is performed to perform location tracking visualization operations.

2. The method for visualizing the real-time position tracking of a smart bracelet according to claim 1, characterized in that: The specific steps of step S1 are: Collecting multi-dimensional sensor monitoring parameters of the user based on the smart bracelet, wherein the multi-dimensional sensor monitoring parameters of the user include real-time heart rate monitoring of the user, user movement speed and user body posture vibration parameters; Perform periodic decomposition on the user's multi-dimensional sensor monitoring parameters and extract the sensor monitoring parameters of multiple periods; Perform real-time motion state evaluation based on multiple periodic sensor monitoring parameters to obtain the user's motion state in different periods; Performing a time-series motion state change evolution on the motion state to generate a user motion state change curve; The position acquisition frequency of the smart bracelet is adaptively adjusted based on the user's motion state change curve to generate an adaptive position acquisition frequency.

3. The method for visualizing the real-time position tracking of a smart bracelet according to claim 1, characterized in that: The specific steps of step S2 are: Based on the adaptive location acquisition frequency, the smart bracelet is used to collect real-time location positioning and obtain the user's real-time location information stream; Identify abnormal location information of users' real-time location information stream and perform dynamic filtering optimization to obtain filtered optimized location information stream; Perform frequency-by-frequency spatial position positioning calculation on the filtered optimized position information stream to generate the spatial position coordinates of each acquisition frequency; Calculate the standard timestamp of each acquisition frequency based on the filtered optimized position information stream; Performing multi-time point position sequential evolution of the spatial position coordinates based on the standard timestamp to construct a real-time position coordinate tracking trajectory; The real-time position coordinate tracking trajectory is dynamically compensated and reconstructed to obtain a high-precision compensated reconstructed trajectory.

4. The method for visualizing the real-time position tracking of a smart bracelet according to claim 3, characterized in that: The specific steps of performing dynamic trajectory compensation and reconstruction on the real-time position coordinate tracking trajectory to obtain a high-precision compensated and reconstructed trajectory are as follows: Obtain inertial measurement unit sensor data based on smart bracelet; The Kalman filter estimation is performed based on the inertial measurement unit sensor data to obtain the user's relative displacement estimate; The user's real-time movement trajectory is estimated based on the user's relative displacement estimation value, and the trajectory deviation of the real-time position coordinate tracking trajectory is calculated point by point to generate multi-point trajectory deviation values; Based on the multi-point trajectory deviation values, deviation accumulation calculation is performed to generate auxiliary trajectory correction parameters; Perform road topology constraint analysis based on publicly available electronic map information and generate road topology constraint conditions for real-time trajectories; The real-time position coordinate tracking trajectory is dynamically compensated and reconstructed according to the road topology constraint conditions and the auxiliary trajectory correction parameters to obtain a high-precision compensated and reconstructed trajectory.

5. The method for visualizing the real-time position tracking of a smart bracelet according to claim 1, characterized in that: The specific steps of step S3 are: Obtain public electronic map information, perform position matching on the high-precision compensated reconstructed trajectory, and perform fuzzy perception of scene information to extract scene information of the user's path trajectory; Identifying the road structure and key scene elements of the scene information, and marking a plurality of key scene element points; Map attribute analysis is performed on multiple key scene element points, and key attribute recognition marking is performed to obtain multiple scene attribute memory anchor points.

6. The method for visualizing the real-time position tracking of a smart bracelet according to claim 1, characterized in that: The specific steps of step S4 are: Based on multiple scene attribute memory anchor points, the relative position distribution of the high-precision compensation reconstruction trajectory is analyzed to obtain the relative position of each anchor point; Performing omnidirectional distribution fitting of anchor points according to the relative positions to construct an anchor point distribution map; Calculate the path timestamp of each anchor point based on the high-precision compensated reconstructed trajectory; Dynamic trajectory visualization mapping is performed based on the path timestamps to construct an anchor point space trajectory visualization mapping diagram.

7. The method for visualizing the real-time position tracking of a smart bracelet according to claim 1, characterized in that: The specific steps of step S5 are: Calculate the real-time climb rate, slope change and elevation profile based on the high-precision compensated reconstructed trajectory to obtain the trajectory terrain characteristic parameters; Perform visual recognition of geographic background layers based on publicly available electronic map information and extract the geographic background layers; Based on the trajectory terrain characteristic parameters and geographic background layer, spatial trajectory correlation matching is performed to obtain trajectory geographic semantic information; Based on the trajectory geographic semantic information, the semantic information of the anchor point spatial trajectory visualization map is rendered to construct a three-dimensional trajectory-geographic information fusion model.

8. The method for visualizing the real-time position tracking of a smart bracelet according to claim 1, characterized in that: The specific steps of step S6 are: Calculate the trajectory complexity of the 3D trajectory-geographic information fusion model and obtain a visual trajectory complexity index; Perform real-time network status and device performance detection on the smart bracelet to generate the real-time status of the smart bracelet; Make rendering resource allocation decisions based on the visualization trajectory complexity index according to the real-time status of the smart bracelet and generate adaptive rendering strategy parameters; Based on the adaptive rendering strategy parameters, multi-level detail level modeling is performed to construct a four-layer visualization rendering engine consisting of map base layer, trajectory line layer, motion status identification layer and interactive control layer to perform position tracking visualization tasks.

9. A smart wristband real-time location tracking and visualization system, characterized in that: The method for performing the real-time position tracking visualization method of the smart bracelet according to claim 1 comprises: The acquisition frequency adjustment module is used to collect multi-dimensional sensor monitoring parameters of the user based on the smart bracelet, conduct real-time motion status assessment of the user, and perform adaptive adjustment of the position acquisition frequency to generate an adaptive position acquisition frequency; The trajectory error compensation module is used to perform real-time position acquisition based on the adaptive position acquisition frequency, and perform frequency-by-frequency spatial position calculation to obtain a high-precision compensated reconstructed trajectory; The scene memory anchor module is used to obtain public electronic map information, identify key elements of the scene, and perform key attribute recognition marking to obtain multiple scene attribute memory anchors; The trajectory visualization module is used to analyze the relative position distribution of high-precision compensated reconstructed trajectories based on multiple scene attribute memory anchor points, perform dynamic trajectory visualization mapping, and construct an anchor point space trajectory visualization mapping diagram; The semantic information rendering module is used to perform visual recognition of the geographic background layer based on the public electronic map information, and then perform semantic information rendering on the anchor point spatial trajectory visualization map to construct a 3D trajectory-geographic information fusion model; The multi-level modeling module is used to make rendering resource allocation decisions for the 3D trajectory-geographic information fusion model and perform multi-level detail level modeling to perform location tracking visualization operations.

Images