Dynamic achievement display system and method

Abstract

The invention discloses a dynamic achievement display system and method, and the system comprises a cloud service platform which is used for generating a dynamic scene content and a dynamic achievement content according to a user image, cultural data and behavior data; the wearable intelligent badge is used for receiving and displaying the content; the multi-mode positioning and triggering module is used for fusing multiple positioning signals such as GNSS, BLE and NFC to identify the position of a user and trigger a clock-in event so as to drive content updating; and the data synchronization and management module is used for carrying out association modeling on the culture entities, the user behaviors and the achievements, and managing data synchronization and adaptive display among multiple terminals. According to the method, the card punching behavior of the user is converted into a personalized dynamic visual achievement through the generative AI and multi-mode positioning technology, the problems of static content, weak user participation sense, low cross-end data synchronization efficiency and the like in the existing travel card punching experience are solved, and the immersion and culture exploration desire of the user are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of smart wearables and cultural tourism information technology, and in particular to a dynamic achievement display system and method. Background Technology

[0002] With the popularization of artificial intelligence and wearable device technologies, the cultural tourism industry is gradually adopting digital guides, AR interactions, and other forms to enhance the tourist experience. Traditional check-in methods, such as paper stamps or simple mobile phone scanning, can only achieve static recording functions, and the presentation is mostly in the form of icons, numbers, or static images, which has obvious shortcomings in experience: First, the content expression is monotonous and highly homogenized, and it is impossible to dynamically generate exclusive content based on the user's personalized characteristics and behavior, resulting in a lack of user role and insufficient sense of participation and resonance; Second, the achievement display lacks dynamism and visual appeal, making it difficult to form effective positive incentives and social dissemination; Third, location check-in mostly relies on a single technology (such as GPS or NFC only), which is prone to drift, false triggering, or coverage blind spots in complex scenarios, resulting in a disjointed experience; In addition, the lack of an efficient data collaboration architecture at the system level makes it difficult to synchronize the content status between mini-programs, mobile apps, and smart hardware devices in real time, resulting in problems such as refresh lag and fragmented experience; Finally, most wearable devices on the market focus on sports and health monitoring, and there is a lack of hardware carriers designed specifically for cultural immersion experiences that integrate multimodal perception and low-power dynamic display capabilities. Therefore, there is an urgent need for a systematic solution that can deeply integrate cultural content generation, high-precision context perception, and real-time dynamic display. Summary of the Invention

[0003] To address the problems of static achievement display in existing Chinese travel check-in experiences, lack of user engagement, low efficiency in multi-device data synchronization, and inaccurate and inconsistent location-based triggering, this invention proposes a dynamic achievement display system and method.

[0004] The specific technical solution is as follows: A dynamic achievement display system, comprising:

[0005] The cloud service platform receives user images, cultural data of scenic spots, and user behavior data, and generates dynamic scene content and dynamic achievement content related to user behavior based on the received content.

[0006] The wearable smart badge communicates with the cloud service platform to receive and display the dynamic scene content and the dynamic achievement content.

[0007] A multimodal positioning and triggering module is integrated into the wearable smart badge and / or cloud service platform. It integrates multiple positioning signals to identify the user's geographical location and triggers a check-in event when the user arrives at a preset scenic spot. The check-in event is used to drive the cloud service platform to generate or update dynamic achievement content.

[0008] The data synchronization and management module performs correlation modeling of cultural entities, user behaviors, and achievement items. Based on the modeling results, it manages the synchronization and adaptive display of dynamic scene content and dynamic achievement content between the cloud service platform and wearable smart badges. By constructing a collaborative system consisting of a cloud service platform, wearable smart badges, a multimodal positioning and triggering module, and a data synchronization and management module, a complete closed loop is achieved, enabling dynamic perception of user behavior, personalized content generation, precise triggering, and multi-terminal synchronized display. This fundamentally enhances the systematic nature and immersiveness of the cultural tourism experience.

[0009] Furthermore, the cloud service platform generates dynamic scene content, including:

[0010] A deep feature separation network based on the Transformer architecture is used to decouple identity features, dynamic features and style features from user images;

[0011] Semantic information of cultural data is extracted through a visual-language alignment model, and an attention mechanism is used to fuse cultural symbols with user characteristics.

[0012] A temporal modeling network is used to expand the fused single-frame images into a temporal dynamic scene sequence that includes basic actions and micro-narratives of interaction with the environment. By employing a Transformer-based feature separation network, a visual-language alignment model, and a temporal modeling network, it is possible to efficiently decouple user features, integrate cultural semantics, and generate a coherent dynamic scene sequence. This enables the user to transform from a "participant" to a "protagonist," greatly enhancing the personalization and narrative expressiveness of the content.

[0013] Furthermore, the time-series dynamic scene sequence is compressed using the H.266 / VVC standard and transmitted to the wearable smart badge via a Bluetooth 5.2 link. By employing the H.266 / VVC compression standard and the Bluetooth 5.2 low-power link, efficient and stable transmission of large-volume dynamic scene sequences to micro wearable devices can be achieved with limited bandwidth resources, solving the technical bottleneck of real-time display of dynamic content on the terminal.

[0014] Furthermore, the cloud service platform generates dynamic achievement content, including:

[0015] A reinforcement learning model trained using the Proximal Policy Optimization (PPO) algorithm is combined with a Beta-Binomial probability distribution model to dynamically generate achievement trigger rules. The Alpha and Beta parameters of the Beta-Binomial probability distribution model are dynamically adjusted based on the user's historical exploration data and current scene characteristics. This combination of a PPO-based reinforcement learning model and a Beta-Binomial probability distribution model for intelligent decision-making enables dynamic adjustment of achievement trigger rules based on user behavior and scene characteristics, achieving an achievement system that is adaptive, intelligent, and engaging.

[0016] Furthermore, the user's historical exploration data includes the user's check-in frequency and preference types over the past 30 days, and the current scene features include the scenic area level and cultural theme. Using the user's check-in frequency, preference types, scenic area level, and cultural theme over the past 30 days as the basis for adjusting model parameters enables the achievement triggering mechanism to more accurately reflect the user profile and scene context, achieving a truly personalized and contextualized achievement experience.

[0017] Furthermore, the multimodal positioning and triggering module fuses multiple positioning signals to identify the user's geographical location, including:

[0018] Coarse positioning is performed using the Global Navigation Satellite System (GNSS).

[0019] Once GNSS data indicates that the user has entered the preset geofence, Bluetooth beacon BLE scanning is initiated for precise positioning.

[0020] When BLE positioning determines that a user has entered the preset range of the check-in point, the Near Field Communication (NFC) module is activated to confirm the contact and trigger the check-in event. This three-tiered positioning triggering process, employing GNSS, BLE, and NFC, ensures sub-meter level precision positioning and millisecond-level final confirmation while maintaining wide-area coverage. This achieves high-precision and high-reliability triggering of check-in events even in complex environments.

[0021] Furthermore, the multimodal positioning and triggering module also evaluates the reliability of GNSS, BLE, and NFC data in real time through Bayesian filtering, and dynamically adjusts the weights of each positioning signal according to the signal-to-noise ratio, data update rate, and environmental occlusion level. By dynamically evaluating the reliability of each positioning signal and assigning weights using Bayesian filtering, the positioning results can be automatically optimized based on environmental signal-to-noise ratio, occlusion level, and other conditions, enabling the system to maintain high robustness and positioning accuracy even in complex scenarios with severe signal interference.

[0022] Furthermore, the wearable smart badge includes a circular low-power display and a multimodal sensing unit integrating GNSS, NFC, and BLE communication modules. This hardware design, combining a multimodal sensing unit integrating GNSS, NFC, and BLE with a circular low-power display, provides a dedicated and form-fitting hardware platform for immersive cultural experiences, enabling seamless and imperceptible interaction between users and cultural content.

[0023] A dynamic achievement display method, applicable to a dynamic achievement display system, includes:

[0024] The cloud service platform receives user images, cultural data, and user behavior data, and generates dynamic scene content and dynamic achievement content.

[0025] The multimodal positioning and triggering module integrates multiple positioning signals to identify the user's geographical location and triggers a check-in event when a preset scenic spot is reached, thereby driving the cloud service platform to generate or update dynamic achievement content.

[0026] The data synchronization and management module is used to model the association between cultural entities, user behavior and achievement items, and to manage the synchronization and adaptive display of dynamic scene content and dynamic achievement content between the cloud service platform and wearable smart badges.

[0027] Receive and display dynamic scene content and dynamic achievement content through wearable smart badges.

[0028] Furthermore, the method of integrating multiple positioning signals to identify the user's geographical location and trigger the check-in event includes:

[0029] A time window algorithm is used to suppress instantaneous drift in GNSS positioning data;

[0030] The Received Signal Strength Indicator (RSSI) dynamic signal strength model is used to probabilistically distinguish neighboring locations.

[0031] A decision hysteresis and multi-source consistency verification mechanism are employed to avoid false triggering. By using a time window algorithm, an RSSI dynamic signal strength model, and a multi-source consistency verification mechanism, the system can effectively suppress positioning drift, distinguish neighbor interference, and prevent false triggering, thus significantly improving the accuracy of the attendance event triggering rate.

[0032] The above technical solution has the following advantages or technical effects:

[0033] 1. This invention achieves a paradigm shift in cultural tourism check-in from discrete recording to immersive experience by constructing a collaborative system architecture that integrates "cloud-terminal-positioning-data". The system centrally processes complex calculations through a cloud service platform, focuses on low-power display and light interaction through wearable badges, ensures trigger accuracy through a multimodal positioning module, and connects multiple terminal states through a data synchronization module, thereby solving the systemic problem of fragmented links and inconsistent experience in existing technologies.

[0034] 2. This invention employs a user-centric dynamic scene generation technology based on generative AI, achieving a fundamental innovation in achievement content. It upgrades traditional static icons or text achievements into dynamic scene animations that integrate user personal image characteristics, cultural symbol elements, and chronological storylines, greatly enhancing the visual expressiveness, emotional resonance, and social dissemination value of achievements, and effectively improving users' sense of participation and achievement.

[0035] 3. This invention introduces an intelligent decision-making mechanism that integrates reinforcement learning and probabilistic models, achieving adaptive and intelligent achievement triggering rules. The system can dynamically adjust parameters based on the user's historical behavior profile and real-time scene characteristics, making achievement acquisition both deterministic and surprising. This solves the problem at the mechanism level that traditional achievement systems are either too monotonous or completely random, and can sustainably maintain the user's motivation to explore.

[0036] 4. This invention employs a multi-level collaborative positioning triggering strategy of "GNSS coarse positioning + BLE fine positioning + NFC touch confirmation," combined with a Bayesian filtering dynamic weighting algorithm, to achieve high-precision and high-reliability seamless check-in in complex environments. This technology effectively suppresses GPS drift, distinguishes high-density neighboring locations, and avoids false and missed triggers while ensuring large-area coverage. It solves the technical pain points of insufficient accuracy, poor environmental adaptability, and user experience interruption that exist when relying solely on a single positioning technology, ensuring the smoothness of the entire exploration process.

[0037] 5. This invention designs wearable smart badge hardware specifically for cultural exploration scenarios and combines lightweight rendering and synchronization strategies with cloud collaboration to achieve efficient display of dynamic content on low-power micro devices. It breaks the limitations of existing smart wearable devices with homogeneous functions and provides users with a dedicated cultural experience terminal that is focused, flexible to wear, and easy to interact with. This effectively reduces users' dependence on mobile phones and enhances the immersive experience of offline exploration. Attached Figure Description

[0038] Figure 1 This is a system structure block diagram of the present invention;

[0039] Figure 2 This is a flowchart of the method of the present invention. Detailed Implementation

[0040] To make the technical solution of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0041] Example 1

[0042] like Figure 1 As shown, a dynamic achievement display system includes:

[0043] The cloud service platform receives user images, cultural data of scenic spots, and user behavior data, and generates dynamic scene content and dynamic achievement content related to user behavior based on the received content.

[0044] Wearable smart badges communicate and connect with cloud service platforms to receive and display dynamic scene content and dynamic achievement content;

[0045] The multimodal positioning and triggering module is integrated into the wearable smart badge and / or cloud service platform. It integrates multiple positioning signals to identify the user's geographical location and triggers a check-in event when the user arrives at a preset scenic spot. The check-in event is used to drive the cloud service platform to generate or update dynamic achievement content.

[0046] The data synchronization and management module performs correlation modeling on cultural entities, user behaviors, and achievement items. Based on the modeling results, it manages the synchronization and adaptive display of dynamic scene content and dynamic achievement content between the cloud service platform and wearable smart badges.

[0047] The cloud service platform serves as the system's computing and decision-making hub, with its core being a user-centric dynamic scene generation system. This system is based on a multimodal large-scale model architecture (integrating visual, linguistic, and 3D stylization multi-task large-scale models), combined with edge computing and lightweight deployment technologies, breaking through the limitations of traditional static avatar generation to achieve end-to-end real-time generation from user features to dynamic scenes. The cloud service platform includes a user feature decoupling and stylization engine, a cultural symbol adaptive transfer module, and a dynamic content sequence generation platform.

[0048] During dynamic scene content generation, the cloud service platform processes user-uploaded images through user feature decoupling and a stylization engine. Utilizing an image generation and cultural semantic annotation engine, user images are fused with location-specific cultural content to synthesize user-characterized scenes in real time. This engine employs a deep feature separation network based on the Transformer architecture, accurately decoupling user-uploaded 2D photos into identity features (such as face and body shape), dynamic features (such as facial expressions and postures), and style features (such as clothing texture and lighting), thereby improving the accuracy of feature decoupling. The engine provides an editable stylization parameter space with preset style base vectors such as "Guochao" (Chinese trend), "Cyberpunk," and "Ink Painting." By combining Generative Adversarial Networks (GANs) and Neural Radiation Fields (NeRF) technology, it achieves stylized dynamic reconstruction of the user's image, supporting dynamic effects such as rotation and micro-expression-driven effects, significantly improving reconstruction speed.

[0049] Subsequently, the cultural symbol adaptive migration module begins operation. This module is based on a scenic area cultural symbol knowledge base (covering structured data such as architectural patterns, folk symbols, and regional colors) built using cross-modal knowledge distillation technology. It intelligently extracts the cultural semantics (such as architectural patterns and folk symbols) of the current check-in point using improved visual-language alignment models like CLIP. Finally, it dynamically adjusts the fusion weights of cultural symbols and user features using an attention mechanism to achieve natural integration. The dynamic content sequence generation platform integrates temporal modeling networks such as Temporal Fusion Transformer, expanding single-frame stylized images into 5-10 second temporal dynamic scene sequences. These sequences include basic actions and micro-narratives involving interaction with the environment, shortening the generation time of dynamic scenes and improving the coherence of temporal actions. The platform supports compression of the generated dynamic bitstream using the H.266 / VVC standard, ensuring real-time transmission to the badge terminal via low-bandwidth links such as Bluetooth 5.2, effectively reducing the demand for transmission bandwidth and the packet loss rate.

[0050] To achieve rapid synchronization with low bandwidth, the platform features a lightweight dynamic content conversion system. This system converts generated dynamic scene sequences into highly compressed formats such as GIF, and automatically matches the device resolution and transmission conditions (such as the bandwidth of a Bluetooth 5.2 link). Simultaneously, the task scheduling and synchronization module dynamically distributes relevant task clues, achievement update packages, and IP interaction content to the device based on the user's current exploration status, forming a complete data flow from cloud to edge.

[0051] When generating dynamic achievement content, the cloud service platform relies on the live achievement intelligent generation engine, which constructs a closed-loop system of "data collection - intelligent decision-making - dynamic rendering". Through user behavior analysis and generative AI technology, it achieves personalized live element input effects in the check-in scenario.

[0052] Its multi-source behavioral data perception layer collects user check-in behavior data, including stay duration, movement trajectory, and frequency of interaction with other tourists, through the inertial measurement unit (IMU), global navigation satellite system (GNSS) positioning module and near field communication (NFC) chip on the badge. It also builds user exploration profiles (such as "deep explorer" and "check-in expert" tags) locally on the device through a federated learning framework. All privacy data does not need to be uploaded to the cloud, thus providing stronger privacy compliance protection at the technical level.

[0053] The dynamic achievement decision center is the core, responsible for dynamically generating achievement triggering rules. It is a reinforcement learning model trained based on the proximal policy optimization (PPO) algorithm. The state space features of this policy model include: the normalized value of the user's check-in frequency in the past 30 days, the one-hot encoding of the user's preference type, the current scene feature vector (such as scenic spot level, cultural theme), and the user's historical achievement acquisition rate.

[0054] The decision-making center incorporates a Beta-Binomial probability distribution model as its core mechanism for ensuring randomness and fairness. The Alpha (α) and Beta (β) parameters of this model are dynamically adjusted based on user profiles and scenario characteristics, with the specific strategies as follows:

[0055] Based on user profiles: Parameters were set for novice users (α=3.0, β=2.0) to achieve a success rate of approximately 60%; for ordinary users (α=2.5, β=4.0) to achieve a success rate of approximately 38%; and for experienced users (α=2.0, β=6.0) to achieve a success rate of approximately 25% to maintain a challenging experience.

[0056] Based on scene characteristics: For 5A-level scenic spots, increase the β parameter by 1.0 to increase the difficulty; for 3A-level and below scenic spots, increase the α parameter by 0.5 to decrease the difficulty. For popular cultural themes, increase the β parameter by 0.5; for less popular themes, increase the α parameter by 0.5.

[0057] The reward function of this PPO model is specially designed, including: an immediate reward for achievements triggered by the expected value of the Beta distribution, a long-term reward signal derived from user satisfaction feedback, and a penalty for false triggers (calculated as -0.1 * number of false triggers). This design ensures that the achievement system is highly adaptable while controlling the randomness and fairness of acquisition through probability distribution, and reducing the false trigger rate even further. The sense of surprise brought by achievements is indirectly measured and optimized through proxy metrics, including: user dwell time after obtaining an achievement, achievement sharing rate, increase in the growth rate of subsequent continuous check-ins, increase in in-app interaction depth, achievement rarity score, and unpredictability score of acquisition time.

[0058] Its lightweight, dynamic rendering engine is primarily deployed on wearable smart badges, responsible for the final effect presentation. Addressing the low computational power of badges (typical power consumption <50mW), the engine employs a WebGL-based lightweight rendering pipeline, supporting real-time decoding and dynamic playback of GIF / APNG formats, with decoding latency controlled within 200 milliseconds, significantly reducing the rendering power consumption of dynamic elements. The engine combines 3D stylized element blending rendering technology to overlay dynamic elements onto the UI space scene, achieving rich 3D interactive effects such as "treasures emerging from the badge surface" and "NPCs jumping along the badge edge." The engine supports over-the-air (OTA) updates to the dynamic content generation model parameters, ensuring the system can quickly adapt to new achievement types for new checkpoints, thereby effectively improving user retention.

[0059] The wearable smart badge is the system's dedicated hardware carrier. It adopts a badge-style design, with its core consisting of a low-power display screen (including but not limited to a circular shape) and a multimodal sensing unit integrating GNSS (for outdoor positioning), NFC (for near-field communication), and BLE (for Bluetooth beacon scanning). The badge also embeds an edge computing unit (Edge AI Unit), solving the problems of low-power response and display updates under high-frequency check-in conditions.

[0060] The device adopts a wearable adaptive design, supporting multiple wearing methods such as badges, pendants, neckbands, and hats, offering low-interference and high-convenience options for cultural tourism or study tours. This flexible design breaks through the traditional limitation of smartwatches / bracelets being worn on the wrist, better adapting to the preferences and usage scenarios of different user groups.

[0061] The badge-based live achievement engine (corresponding to the device-side component in the documentation) is responsible for loading and switching lightweight animation content delivered from the cloud in real time, and dynamically linking it with the check-in results. Its collaborative positioning module, as part of the multimodal positioning and triggering module, integrates GNSS positioning (outdoor) and BLE / NFC sensing (near field), balancing check-in accuracy and environmental adaptability, and supporting fault-tolerant interaction under complex location conditions.

[0062] The badge employs a power consumption control mechanism, setting a standby / response dual state. The core decoding and synchronization process is only activated when the user approaches the check-in point or operates the button, effectively extending the battery life.

[0063] The badge features local data filtering, image caching, and lightweight WebGL-based rendering capabilities, enabling dynamic decoding and playback of GIF / APNG formats (decoding latency <200ms). This allows for the display of 3D interactive effects such as "treasure emerging" and "NPC jumping," significantly reducing reliance on real-time cloud transmission and overall power consumption. Dynamic content is cached within the edge computing unit, supporting the display of user-exclusive cultural achievements in animated (e.g., GIF) form, creating a visible, tangible, and shareable interactive experience in outdoor environments. The badge is operated via a single physical button or touch panel, making it convenient for outdoor use. As an optional implementation, the displayed content is limited to short animations, location feedback, or highly recognizable IP characters, excluding complex path navigation content, ensuring a balance between energy consumption and user experience.

[0064] The multimodal positioning and triggering module works collaboratively on the badge and the cloud, realizing "cross-modal location recognition + real-time triggering of multi-point check-in behavior + cultural task-driven interactive link". The module integrates high-precision GNSS, NFC near-field triggering, and BLE beacon data, employing an innovative multi-source positioning fusion algorithm to determine user location and behavior. Its core lies in the combination of hierarchical dynamic weighted fusion and semantic scene constraints, thereby achieving low-power, high-precision fusion.

[0065] Multi-stage position coarseness switching:

[0066] Phase 1 (Global Coarse Positioning): The high-precision GNSS module operates at a low frequency to quickly determine the user's initial position in macro space (such as the entire West Lake scenic area) and achieve low power consumption under large-area coverage.

[0067] The second stage (local fine positioning): When GNSS data indicates that the user has entered the preset geofence area, the system automatically wakes up the BLE beacon scan and achieves sub-meter level relative positioning by receiving the signal strength indicator (RSSI) and multi-point signal trilateration.

[0068] The third stage (precise contact point identification): When BLE positioning determines that the user has entered a 3-meter radius of the check-in point, the NFC module is activated briefly to complete the absolute contact point confirmation with a millisecond delay, ultimately achieving zero-error task triggering.

[0069] Dynamic weighted fusion algorithm: This module evaluates the reliability of GNSS, BLE, and NFC data in real time using Bayesian filtering, and dynamically adjusts the weights of each positioning signal based on the signal-to-noise ratio, data update rate, and environmental occlusion level. For example, in dense forests or areas obstructed by buildings, the weight of GNSS is automatically reduced while the weight of BLE is increased; in areas with sparse Bluetooth beacons, the opposite weighting strategy is implemented to ensure optimal positioning results are output in any environment.

[0070] Semantic Scenario Constraints: This module introduces the concept of a cultural task semantic map, where the spatial location of each task point not only has its geographical coordinates but also includes its logical relationships with surrounding paths, landmarks, and activity status. The positioning algorithm combines semantic features such as the user's movement trajectory pattern (e.g., walking or cycling), task chain progress, and reasonable time windows to logically verify the physical positioning results, thereby intelligently eliminating unreasonable jump points and drifts (e.g., preventing positioning errors such as users "crossing a lake"), greatly improving the intelligence and reliability of positioning.

[0071] Event-driven power optimization: This module employs a non-continuous scanning mode to optimize battery life to the extreme. The specific strategy is as follows: when the distance to the next cultural task point exceeds 50 meters, only low-frequency GNSS is maintained for tracking; BLE scanning is activated for precise positioning only when approaching the target location; and NFC is briefly activated for final confirmation only when BLE determines that the device is within 3 meters. This process is also combined with motion detection (IMU), where high-frequency scanning is paused when the user is stationary, further extending the wearable device's battery life.

[0072] Furthermore, the multimodal positioning and triggering module, in conjunction with a time window algorithm and an RSSI dynamic signal strength model, jointly addresses the "GNSS drift" problem at high-density locations. Specifically, it utilizes time consistency to suppress instantaneous GNSS drift, employs probabilistic RSSI modeling to distinguish neighboring locations, and finally uses decision hysteresis and multi-source consistency verification mechanisms to avoid false triggering. This ensures accurate identification of check-in points and achievement triggering, significantly improving the accuracy of location identification in high-density, complex environments.

[0073] Overall Judgment Framework:

[0074] Within a sliding time window of length T seconds (e.g., T = 6~10 s), the posterior is calculated for each candidate point k:

[0075]

[0076] in, For RSSI model, Represents GNSS residuals. Indicates kinematic consistency. Indicates task / time priors.

[0077] Triggering conditions (dual thresholds + hysteresis):

[0078] Entry Judgment: (hysteresis) );

[0079] The final achievement can be triggered by either a successful NFC touch or the peak RSSI mode at a very short distance.

[0080] RSSI dynamic signal strength model:

[0081] For each beacon i: ;

[0082] and Adaptive to the environment: Estimated using the rolling median within the window + MAD Using online least squares / EM pairs Fine-tuning;

[0083] Crowd / Obstruction Indicator: If the accelerometer shows low-speed congestion and the RSSI variance increases, increase... Lower the confidence level to avoid "false proximity";

[0084] Multi-beacon joint likelihood and robust regression:

[0085] Solving the weighted robust least squares / Huber regression at the geometric center of the POI yields the following results: With residual ;

[0086] Likelihood And give higher weight to the three strongest beacons;

[0087] Peak-shape criteria:

[0088] When approaching the correct checkpoint, RSSI will exhibit a peak followed by rapid decay, reaching its maximum value within the window. Test by the rising / falling slope:

[0089] (e.g., -50dBm), rise time <= 1s and half-decay time after peak <= 2s, if these conditions are met, it is considered "arrival at the pile" and can directly enter the trigger candidate, even if the GNSS has drift.

[0090] Time window algorithm:

[0091] Innovations in GNSS observations and BLE estimation: If the GNSS crosses the boundary m times consecutively (e.g., m>=3), it is determined to be GNSS drift, and the GNSS weight is temporarily downgraded to [a lower weight]. ;

[0092] Determining whether to stay or cross over:

[0093] Only when the duration of stay is >= (e.g., >= 2s) and velocity v <= (e.g., <= 0.6 m / s) POI posterior is allowed to exceed the limit. The "scan and trigger" behavior of GPS devices that pass by without stopping is suppressed;

[0094] Path / entry constraints: Define reachable corridors for each POI on the semantic map. If the trajectory deviates from the corridor by more than a certain threshold (e.g., 3m) within a time window, then the POI is considered a priori. Attenuation reduces accidental touches.

[0095] Multi-level confirmation triggered by achievements:

[0096] Level 1 (candidate): Posterior exceeds and Large enough;

[0097] Level 2 (in place): Meets the "peak-shape criterion" or NFC is successful;

[0098] Level 3 (Locked): Maintain >= (e.g., 1s) + delay without falling back, immediately triggering the achievement and recording it.

[0099] Quantifiable definitions and evaluation metrics in high-density, complex environments:

[0100] Environmental density and distinguishability indicators:

[0101] Point density: (pieces / 100m) 2 )

[0102] Minimum spacing

[0103] Effective separation degree: S>=2 is considered to be stably distinguishable, S∈[1,2) requires time window and NFC assistance, and S<1 requires denser beacon deployment / adjustment of points;

[0104] RF entropy: Normalized power distribution of visible beacons within a window ,definition

[0105] , A higher variance indicates more distinguishable information; however, if the variance is high but the entropy is low, multipath propagation is severe and the dominant signal is unclear.

[0106] Trigger accuracy metrics:

[0107] Point of Interest (POI) recognition accuracy (Top-1): Acc = Number of correct POI recognitions / Total number of attempts

[0108] False trigger rate: Number of false triggers per hour / Total number of triggers

[0109] Confusion matrix: the distribution of misidentifications among adjacent POIs

[0110] Distance resolution: in d min Acc@95% under conditions of =3 / 5 / 10m

[0111] Trigger delay: Median delay from entering a 3m radius to Level 3 triggering

[0112] Energy consumption metrics: mWh per trigger and estimated daily range;

[0113] Robustness layered use cases:

[0114] Dense forest cover / pavilions and corridors: a record The magnitude of the increase and the change in Acc

[0115] High-traffic periods (peak concurrent users > X people / minute): Comparison With false trigger rate

[0116] Rainy / humid weather: Adaptive adjustment of RSSI attenuation slope and peak value.

[0117] The data synchronization and management module forms the data foundation and collaborative framework of the system. This architecture is the first to combine "semantic knowledge graph, large model intelligent agent and edge rendering protocol" to construct a technical path of "adaptive expression of cultural content", which realizes the "revitalization" of digital culture from a technical level.

[0118] The core technical components of this module include:

[0119] Cultural Intelligent Agent System: This module constructs a cultural intelligent agent system based on nested knowledge graphs, serving as the core of data association. This system performs unified structured modeling of three types of data: scenic area cultural entities, user behavior, and check-in achievements, forming deep semantic relationships. For example, a "Su Shi" cultural entity will be connected with user behavior such as "reciting West Lake poems," the achievement item "Su Shi fan," and related check-in locations. This knowledge graph-based association method enables the system to mine complex relationships between cultural entities more efficiently and has a shorter response latency for matching user behavior and achievements.

[0120] Adaptive Front-End Rendering Protocol: To achieve a unified experience for cultural content across multiple devices, this module adopts an adaptive front-end rendering protocol. This protocol defines a set of specifications that separate content description from display logic, enabling dynamic adaptation and unified output of the same cultural content based on the screen size, computing power, and network status of the terminal device (such as a web browser, mobile app, or badge device). This allows a dynamic achievement to display a high-definition, complete animation on a mobile device, while automatically converting it to a low-resolution, low-color-depth GIF format for playback on badge devices. This significantly reduces the development and adaptation costs of applications across different platforms while ensuring consistent display across multiple devices.

[0121] Event-Driven Content Growth Chain: On the backend, this module employs an event-driven content growth chain to manage content state and synchronization. This chain treats each user's exploration action (such as checking in, staying, and interacting) as an event, driving the entire cultural content system to dynamically evolve and "grow." For example, when a user unlocks a new NPC, this event will synchronously update the user's personal encyclopedia, achievement list, and the viewing views of other teammates in real time. This event-driven state synchronization mechanism makes the encyclopedia unlocking faster and the synchronization error rate of NPC and other multi-device states lower, providing users with a growth-oriented experience where content continuously enriches as exploration progresses.

[0122] This system solves the problem of user immersion in non-mobile-dominated scenarios through a dual-interface, lightweight interaction mechanism, enabling users to have a low-barrier interactive experience with cultural content.

[0123] Once the user arrives at the designated location, the badge can be automatically checked in and the corresponding animation will switch automatically by touching the check-in area with NFC, without the need to operate the mobile phone.

[0124] Users can freely switch between "User Character Achievement Animation" and "IP Guide Prompt Scene" by clicking the physical buttons or touch panel on the device;

[0125] Users can also view or share exploration tasks and achievement scenes through the accompanying mini-program, creating a collaborative experience process across three terminals: mobile phone, cloud, and device.

[0126] Example 2

[0127] like Figure 2 As shown, a dynamic achievement display method includes the following steps:

[0128] S1: Content Generation Steps. The cloud service platform receives user images, cultural data of scenic spots, and user behavior data. Based on this data, it generates dynamic scene content featuring the user as the protagonist and dynamic achievement content associated with the user's behavior. Specifically, this includes using a deep feature separation network to decouple user features, fusing cultural symbols through a visual-language alignment model, and generating dynamic sequences using a temporal modeling network; simultaneously, it dynamically decides achievement trigger rules based on a reinforcement learning model and a Beta-Binomial probability distribution model.

[0129] S2: Location Triggering Step. The multimodal positioning and triggering module fuses GNSS, BLE, and NFC positioning signals to identify the user's geographical location. A time window algorithm is employed to suppress instantaneous GNSS drift, an RSSI dynamic signal strength model is used for probabilistic differentiation, and a decision hysteresis and multi-source consistency verification mechanism is used to avoid false triggering. When the user is determined to have reached a preset scenic spot, a check-in event is automatically triggered. This event drives the cloud service platform to generate or update dynamic achievement content.

[0130] S3: Data Synchronization and Adaptation Steps. Through the data synchronization and management module, cultural entities, user behaviors, and achievement items are modeled in association (e.g., a knowledge graph is constructed). Based on the modeling results, the synchronization of dynamic scene content and achievement content between the cloud and the badge is managed, and the content undergoes format conversion and adaptive encoding (e.g., conversion to low-resolution GIFs) to adapt to the display capabilities of wearable smart badges.

[0131] S4: Content Display Step. The wearable smart badge receives the dynamic scene content and achievement content processed in step S3, and renders and displays it through its circular low-power display, completing the closed loop from behavior to visualized achievement.

[0132] Example 3

[0133] As an extension of Embodiment 1 or 2, the presentation of the dynamic achievement content can be configured as a "virtual team growth" mode. Specifically, the badge initially displays the user's personal avatar leading the team, with several NPC phantoms following behind. Each time the user reaches a checkpoint and triggers a check-in via NFC sensing, the cloud generates a live animation of the corresponding cultural figure NPC and automatically sends it to the badge. The badge displays that the NPC has joined the team and moves dynamically with the main character, thus visualizing the check-in progress as the virtual team continuously growing, greatly enhancing the user's desire to collect and their motivation to explore.

[0134] Example 4

[0135] As a further extension of Embodiment 3, the system can support cross-scenic area collaboration. Specifically, NPCs unlocked by users in different scenic areas can be grouped together in the same team. In addition to the "NPC team," achievement displays can also be configured as dynamic scene animations featuring the user as the protagonist, such as "collecting food guides" or "poetry collections." By enriching the protagonist's performance with diverse themes, the system can satisfy the preferences of different users and further enhance the fun of exploration.

[0136] Example 5

[0137] As an extension of Embodiment 1 or 2, the system can be applied to study tour team scenarios. Specifically, a team mode is configured for students in the same study tour group, and their badge devices can synchronize check-in progress via BLE self-organizing network. When a member triggers a check-in, their badge animation updates, and the badges of other members in the team can update the team progress animation via an offline synchronization mechanism. In addition, badges can also be tapped with NFC to a large offline screen device to view the team's overall progress and ranking with one click. Parents can view the GPS location information of their children's badges in real time through a mobile app, enabling safety management and remote participation.

[0138] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A dynamic achievement display system, characterized in that, include: The cloud service platform receives user images, cultural data of scenic spots, and user behavior data, and generates dynamic scene content and dynamic achievement content related to user behavior based on the received content. The wearable smart badge communicates with the cloud service platform to receive and display the dynamic scene content and the dynamic achievement content. A multimodal positioning and triggering module is integrated into the wearable smart badge and / or cloud service platform. It integrates multiple positioning signals to identify the user's geographical location and triggers a check-in event when the user arrives at a preset scenic spot. The check-in event is used to drive the cloud service platform to generate or update dynamic achievement content. The data synchronization and management module performs correlation modeling on cultural entities, user behaviors, and achievement items. Based on the modeling results, it manages the synchronization and adaptive display of dynamic scene content and dynamic achievement content between the cloud service platform and wearable smart badges.

2. The dynamic achievement display system according to claim 1, characterized in that, The cloud service platform generates dynamic scene content, including: A deep feature separation network based on the Transformer architecture is used to decouple identity features, dynamic features and style features from user images; Semantic information of cultural data is extracted through a visual-language alignment model, and an attention mechanism is used to fuse cultural symbols with user characteristics. A temporal modeling network is used to expand the fused single-frame image into a temporal dynamic scene sequence that includes basic actions and micro-storylines of interaction with the environment.

3. The dynamic achievement display system according to claim 2, characterized in that, The time-series dynamic scene sequence is compressed using the H.266 / VVC standard and transmitted to the wearable smart badge via Bluetooth 5.2 link.

4. The dynamic achievement display system according to claim 1, characterized in that, The cloud service platform generates dynamic achievement content, including: The reinforcement learning model trained based on the near-end strategy optimization PPO algorithm is combined with the Beta-Binomial probability distribution model to dynamically generate achievement triggering rules. The Alpha and Beta parameters of the Beta-Binomial probability distribution model are dynamically adjusted according to the user's historical exploration data and the current scene characteristics.

5. A dynamic achievement display system according to claim 4, characterized in that, The user's historical exploration data includes the user's check-in frequency and preference types in the past 30 days, and the current scene features include the scenic area level and cultural theme.

6. The dynamic achievement display system according to claim 1, characterized in that, The multimodal positioning and triggering module fuses multiple positioning signals to identify the user's geographical location, including: Coarse positioning is performed using the Global Navigation Satellite System (GNSS). Once GNSS data indicates that the user has entered the preset geofence, Bluetooth beacon BLE scanning is initiated for precise positioning. When BLE positioning determines that the user has entered the preset range of the check-in point, the near field communication (NFC) module is activated to confirm the contact and trigger the check-in event.

7. A dynamic achievement display system according to claim 6, characterized in that, The multimodal positioning and triggering module also evaluates the reliability of GNSS, BLE and NFC data in real time through Bayesian filtering, and dynamically adjusts the weight of each positioning signal according to the signal-to-noise ratio, data update rate and environmental occlusion.

8. A dynamic achievement display system according to claim 1, characterized in that, The wearable smart badge includes a circular low-power display and a multimodal sensing unit that integrates GNSS, NFC and BLE communication modules.

9. A dynamic achievement display method, applicable to the dynamic achievement display system according to any one of claims 1 to 8, characterized in that, include: The cloud service platform receives user images, cultural data, and user behavior data, and generates dynamic scene content and dynamic achievement content. The multimodal positioning and triggering module integrates multiple positioning signals to identify the user's geographical location and triggers a check-in event when a preset scenic spot is reached, thereby driving the cloud service platform to generate or update dynamic achievement content. The data synchronization and management module is used to model the association between cultural entities, user behavior and achievement items, and to manage the synchronization and adaptive display of dynamic scene content and dynamic achievement content between the cloud service platform and wearable smart badges. Receive and display dynamic scene content and dynamic achievement content through wearable smart badges.

10. A dynamic achievement display method according to claim 9, characterized in that, The method of integrating multiple positioning signals to identify the user's geographical location and trigger a check-in event includes: A time window algorithm is used to suppress instantaneous drift in GNSS positioning data; The Received Signal Strength Indicator (RSSI) dynamic signal strength model is used to probabilistically distinguish neighboring locations. A decision delay and multi-source consistency verification mechanism is adopted to avoid false triggering.

Images