A parking lot comprehensive situation display system

Abstract

The present invention provides a comprehensive situation display system for a vehicle yard, which includes a multi-source data input module, an intelligent fusion and situation deduction processing module, and a multi-dimensional dynamic output module. The multi-source data input module collects multi-source heterogeneous data streams such as vehicle terminals, video surveillance, and Internet of Things sensors; the intelligent fusion and situation deduction processing module uses MDP, LSTM, and CNN algorithms for structured processing and intelligent deduction; the multi-dimensional dynamic output module displays the results through three-dimensional digital twins, dynamic dashboards, and multi-terminal linkage, and supports permission grading. Through the data fusion mechanism under the unified space-time framework, combined with timestamp calibration and edge caching technology, the present invention solves the problems of data islands, analysis lag, single visualization, and weak decision-making, and realizes the real-time, intelligent, and visual vehicle state assessment, task prediction, environmental risk identification, resource guarantee suggestions, and combat readiness evaluation, significantly improving the vehicle yard management efficiency and decision-making scientificity.

Description

Technical Field

[0001] This invention relates to the field of comprehensive situational awareness display, and in particular to a comprehensive situational awareness display system for parking lots. Background Technology

[0002] With the development of intelligent and digital technologies, parking lot management needs are shifting from traditional static monitoring to dynamic perception, intelligent analysis, and collaborative decision-making. However, existing parking lot management systems suffer from the following key problems: Data silo problem: The system adopts a distributed architecture, with isolated data sources, making it difficult to integrate multi-source heterogeneous information (such as vehicle terminal status, video surveillance images, IoT sensor data, etc.). For example, data from the vehicle positioning system and the environmental monitoring system cannot be correlated, making comprehensive situational analysis impossible.

[0003] Analysis lag issues: The lack of a unified data fusion mechanism and intelligent inference model leads to lagging situational analysis. Although the existing system can display vehicle status and allocation information, it lacks correlation analysis between failure trends and environmental risks; although it can display task execution progress, it cannot predict task delays or resource conflicts.

[0004] The problem with limited visualization: Visualization methods still primarily rely on two-dimensional charts, lacking advanced display methods such as three-dimensional digital twins, making it difficult to intuitively reflect the overall situation of the vehicle depot. For example, it cannot simultaneously display the spatiotemporal relationship between vehicle location, environmental risks, and mission status.

[0005] The problem of weak decision support is the lack of comprehensive assessment and prediction capabilities for vehicle status, mission processes, environmental risks, resource support, and combat readiness. Although the existing system can provide spare parts and materials information, it lacks intelligent prediction of consumption trends; although it has basic statistical analysis functions, it lacks comprehensive assessment of external factors (such as weather effects).

[0006] These technical issues lead to inefficient parking lot management, making it difficult to detect and address potential risks in a timely manner, thus affecting the safety and efficiency of parking lot operations. Summary of the Invention

[0007] The present invention aims to solve at least one of the technical problems existing in the prior art, and provides a comprehensive vehicle parking situation display system.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: In a first aspect, the present invention provides a comprehensive parking lot situation display system, comprising a multi-source data input module, an intelligent fusion and situation inference processing module, and a multi-dimensional dynamic output module connected in sequence, wherein: The multi-source data input module is used to collect and access multi-source heterogeneous data streams from vehicle terminals, video surveillance, IoT sensors, electronic fences, dispatch systems, and environmental monitoring equipment; The intelligent fusion and situational inference processing module is used to perform structured processing, deep correlation analysis and intelligent inference on the multi-source heterogeneous data stream, and generate a result dataset; The multidimensional dynamic output module is used to present the result dataset in the form of a three-dimensional digital twin, a dynamic dashboard, and multi-terminal linkage, and supports hierarchical permission and cross-system linkage.

[0009] In some possible embodiments, the multi-source data input module supports timestamp calibration and edge caching mechanisms, and performs timeline calibration on asynchronous data using a sliding window synchronization algorithm. Its synchronization objective function is: ,in, For the first timestamps for similar devices Used as the base time.

[0010] In some possible embodiments, the intelligent fusion and situational analysis processing module includes the following sub-modules: The semantic knowledge graph construction submodule: Based on the entity-relation-attribute triple model, it constructs a four-element relation graph of "vehicle-task-person-environment". This provides structured data support for subsequent sub-modules; Vehicle Status Assessment Submodule: This module calls upon vehicle attribute data from the aforementioned quaternary relationship graph, combines it with real-time status information from the onboard terminal, and calculates the status evaluation value using the Vehicle Lifecycle Status (LCS) quantification model. : Where Q is the quality index, M is the maintenance completion rate, F is the failure frequency, R is the availability rate, and the weights are... Adaptively adjusts based on vehicle type and age; Task flow deduction submodule: Models task paths based on Markov Decision Process (MDP) and defines reward functions. for: And calculate the task success rate. : Where T is the transition probability, The discount factor (range 0.8 to 0.95) allows for a minimum task path deduction step of 1 minute, which can predict the task execution status and conflict risk within the next 24 hours.

[0011] In some possible embodiments, the intelligent fusion and situational inference processing module further includes the following deeply coupled sub-modules: Environmental safety perception submodule: fused visible light and infrared images to generate a 3D heat map It also uses a convolutional neural network (CNN) to detect abnormal hotspots and determine the fire hazard level. The calculation is as follows: ,in The intensity of temperature anomalies in the image patch. It uses the Sigmoid activation function and integrates acoustic event detection components to analyze noise signals using Short Time Fourier Transform (STFT). It can identify collision, horn blast, or unusual sound events. Resource Guarantee Prediction Submodule: Based on the future task requirements output by the Task Flow Deduction Submodule. Long Short-Term Memory (LSTM) network is used to process the spare parts consumption sequence. Modeling is performed to predict future demand. : ,in, For spare parts consumption sequence, Here are the model parameters, where This is a time series forecasting model built on LSTM. The LSTM model training data covers 3 months of historical consumption records, and the prediction window is the next 7 days. On the test set, the mean absolute percentage error (MAPE) is ≤8%. Combined with a hierarchical inventory threshold strategy (dynamically configuring safety stock levels according to spare parts type and depot level), when the predicted value exceeds the inventory threshold, replenishment suggestions and resource allocation plans are automatically generated. The prediction results are used to update the weights of resource-task association edges in the four-element relationship graph to achieve dynamic optimization of resource allocation. Combat readiness assessment submodule: integrates the aforementioned status evaluation values The fire hazard level mentioned above The future demand mentioned Weather impact coefficients obtained from external meteorological APIs Construct a comprehensive evaluation system based on five rates and calculate the revised combat readiness index. Its expression is as follows: ,in, For the first The original values ​​of the five indicators. For the first The minimum value of each of the five indicators recorded in historical operating data is used as the lower bound benchmark for normalization calculation. For the first The maximum values ​​of the five indicators recorded in historical operating data are used as the upper bound benchmark for normalization calculations. The weights of the corresponding indicators, As a weather-related attenuation factor, The weather impact coefficient is obtained in real time from an external meteorological service interface. The corrected combat readiness index is fed back to the quaternary relationship graph, which enables the quaternary relationship graph to dynamically update the correlation between combat readiness, vehicles / missions / resources and drive subsequent decision optimization.

[0012] In some possible embodiments, the multidimensional dynamic output module includes the following cooperatively operating sub-modules: The 3D digital twin visualization submodule integrates the resulting dataset, dynamically renders the 3D scene of the vehicle yard, and realizes multi-dimensional interactive browsing of timeline data synchronization, task chain tracking and geospatial positioning based on the sliding window algorithm, reflecting the vehicle status, environmental risks and combat readiness in real time. Dynamic dashboard module: Based on the three-color alarm mechanism and combat readiness trend analysis, it displays the status and changing trends of key indicators in real time; Multi-terminal collaborative push submodule: Pushes situation summaries, alarm events and prediction suggestions to the command screen, mobile APP and duty terminal through standardized API interface, and supports hierarchical permission; Cross-system linkage interface submodule: Reserves interface channels with intelligent fire protection, security control, and dispatch command systems to achieve event-driven cross-system linkage response.

[0013] In some possible embodiments, the intelligent fusion and situational inference processing module is deployed in an edge-cloud collaborative architecture, with edge nodes responsible for real-time data cleaning and local alarms, and the cloud platform responsible for map construction and model training, supporting offline operation.

[0014] In some possible embodiments, the vehicle state assessment submodule introduces a fuzzy membership function to classify the state: ,in, For actual performance parameters, Set as the standard performance value for the vehicle. To allow for different tolerance ranges based on vehicle type and age, new vehicles Smaller values, older vehicles The requirements are appropriately relaxed to ensure that the condition assessment accurately reflects the actual vehicle condition; As one of the input parameters of the task flow deduction submodule, and Together, they constitute the vehicle health assessment system.

[0015] In a second aspect, the present invention provides an electronic device, comprising: One or more processors; The storage unit is used to store one or more programs, which, when executed by the one or more processors, enable the one or more processors to implement the comprehensive vehicle situation display system described above.

[0016] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, can realize the comprehensive vehicle situation display system described above.

[0017] The parking lot comprehensive situation display system of this invention has the following beneficial effects: This invention employs a unified spatiotemporal calibration mechanism for multi-source heterogeneous data, combined with a sliding window synchronization algorithm and edge caching technology. It can achieve unified timeline calibration and reliable local caching for all types of heterogeneous data, including vehicle terminals, video surveillance, and IoT sensors. This fundamentally solves the data silo problem commonly found in traditional parking lot management systems, ensuring accurate alignment and efficient collaborative flow of multi-source data, significantly improving the integrity, real-time performance, and reusability of all parking lot data. Advanced intelligent algorithms such as Markov Decision Process (MDP), Long Short-Term Memory (LSTM), and Convolutional Neural Network (CNN) are deeply embedded into the entire situational simulation process. This enables accurate simulation of task flows for the next 24 hours, spare parts consumption prediction errors ≤8%, and fire hazard identification accuracy ≥98%. This drives a fundamental shift in parking lot comprehensive situational awareness from passive recording and monitoring to proactive perception and early warning, and from static data display to dynamic intelligent simulation. A 3D digital twin layered rendering technology is used to build a full-domain visualization scene of the parking lot, maintaining a stable rendering frame rate of over 60fps, while also supporting historical situational timelines. Retrospective analysis, end-to-end task chain tracking, and precise geospatial positioning provide a clear and intuitive view of vehicle status, environmental risks, and combat readiness capabilities, fully meeting the needs of efficient command and dispatch and real-time management in parking lots. Leveraging knowledge distillation technology, the CNN model is compressed to less than 5MB. Combined with an edge-cloud collaborative deployment architecture, embedded GPU inference latency is reduced to less than 100ms. It also supports offline operation during network outages and resume download after network recovery, perfectly adapting to the low-computing-power, high-real-time application scenarios of parking lot edge computing. A comprehensive evaluation system based on five rates is constructed, allowing for dynamic correction of the combat readiness index. It supports adaptive configuration of indicator weights according to different parking lot types such as civilian, emergency, and support functions, and achieves cross-system linkage response with intelligent fire protection, security control, and dispatch command systems, comprehensively improving the intelligence level, collaborative response capabilities, and scientific decision-making of parking lot management. The system adopts a modular and loosely coupled architecture design, with each module capable of independent upgrades and flexible configuration. It can quickly adapt to the needs of parking lot scenarios of different scales and purposes, possessing strong versatility, practicality, and scalability. Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the structure of an example electronic device for the comprehensive parking lot situation display system of the present invention; Figure 2 This is a schematic diagram of the structure of the comprehensive parking lot situation display system of the present invention. Detailed Implementation

[0019] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] Figure 1 This is a schematic diagram of an example electronic device for implementing a comprehensive parking lot situation display system according to the present invention. Figure 1 As shown, the electronic device 100 includes one or more processors 110, one or more storage devices 120, one or more input devices 130, one or more output devices 140, etc., and these components are interconnected through a bus system 150 and / or other forms of connection mechanisms. In this embodiment, the electronic device 100 can be deployed in a vehicle control center, an edge computing node, or a mobile duty terminal, adapting to the computing power and power consumption requirements of different deployment scenarios. It should be noted that... Figure 1 The components and structures of the electronic devices shown are merely exemplary and not limiting; other components and structures may be used as needed.

[0021] The processor 110 may be a central processing unit (CPU), or may be composed of multiple processing cores, or other forms of processing unit with data processing capabilities and / or instruction execution capabilities. The processor 110 may also integrate embedded GPUs, NPUs and other acceleration units to support low-latency inference of AI models such as CNN and LSTM, meet the requirements of real-time situational awareness, and control other components in the electronic device 100 to perform the desired functions.

[0022] Storage device 120 may include one or more computer program products, which may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. The volatile memory may include, for example, random access memory (RAM) and / or cache memory. The non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. Storage device 120 is also configured with a local data cache partition for temporarily storing real-time heterogeneous data collected by the multi-source input module, avoiding data loss and supporting basic situational awareness display in offline mode. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor may execute the program instructions to implement the client functions (implemented by the processor) in the embodiments of this disclosure described below, and / or other desired functions. Various applications and various data may also be stored in the computer-readable storage medium, such as various data used and / or generated by the applications.

[0023] The input device 130 can be a device used by a user to input commands, and may include one or more of a keyboard, mouse, microphone, and touch screen. The input device 130 can also be connected to dedicated equipment such as RFID card readers, fingerprint readers, and voice command collectors for parking lots to realize identity verification and quick command input.

[0024] Output device 140 can output various information (such as images or sounds) to external sources (e.g., users) and may include one or more of a display, speaker, etc. Output device 140 includes dedicated output devices such as high-definition command screens, vehicle-mounted terminal displays, and audible and visual alarms, and pushes situation information and alarm signals in a tiered manner.

[0025] Figure 2 This is a schematic diagram of the parking lot integrated situation display system of the present invention. Figure 2As shown, a comprehensive situation display system for a parking lot includes a multi-source data input module 201, an intelligent fusion and situation inference processing module 202, and a multi-dimensional dynamic output module 203 connected in sequence. The three modules adopt a loosely coupled and highly cohesive design concept. Data communication between the modules is achieved through a standardized message queue, and single-module upgrades and replacements are supported without affecting the overall system operation. The multi-source data input module 201 is used to collect and access multi-source heterogeneous data streams from vehicle terminals (GNSS positioning, CAN bus status, OBD fault codes, sampling frequency 1Hz), video surveillance (eagle-eye panoramic camera 4K video stream, infrared thermal imaging data, frame rate 30fps), IoT sensors (temperature and humidity sensors, smoke sensors, geomagnetic sensors, etc., sampling frequency 5Hz), electronic fences, scheduling systems (task work orders, duty logs, API interface docking), and environmental monitoring equipment (meteorological data and acoustic data); the intelligent fusion and situation inference processing module 202 is used to perform structured processing, deep correlation analysis and intelligent inference on the multi-source heterogeneous data streams, and generate result datasets; the multi-dimensional dynamic output module 203 is used to present the result datasets in the form of three-dimensional digital twins, dynamic dashboards and multi-terminal linkage, and supports hierarchical permission and cross-system linkage.

[0026] Specifically, the results dataset includes, but is not limited to, vehicle status assessment results, task process prediction results, environmental risk identification results, resource support recommendations, and combat readiness capability evaluation results. The results dataset is divided into three categories according to data timeliness: real-time snapshots, historical retrospectives, and future predictions. Among them, real-time snapshots are used to support dynamic visualization and real-time monitoring of the three-dimensional scene of the vehicle yard, historical retrospectives support event review and situation retrospective analysis based on edge cache data, and future predictions are generated through intelligent inference models to provide forward-looking decision-making basis for resource scheduling and task planning.

[0027] In some embodiments, the multi-source data input module supports timestamp calibration and edge caching mechanisms, and performs timeline calibration on asynchronous data using a sliding window synchronization algorithm. Its synchronization objective function is: ,in, For the first timestamps for similar devices Using the base time, the window size of the sliding window can be dynamically adjusted according to the data transmission delay of the parking lot. The default window duration is 500ms. An edge caching mechanism is deployed at the data access end, and the cache capacity is dynamically allocated according to the size of the parking lot (e.g., 500MB for small parking lots, 1GB for medium parking lots, and 2GB for large parking lots). It supports breakpoint resume and high-concurrency data processing to ensure data integrity.

[0028] In some embodiments, the intelligent fusion and situational simulation processing module includes the following sub-modules: The semantic knowledge graph construction submodule: Based on the entity-relation-attribute triple model, it constructs a four-element relation graph of "vehicle-task-person-environment". This provides structured data support for subsequent sub-modules; The vehicle condition assessment submodule calls upon vehicle attribute data from the quaternary relation graph, combines it with real-time condition information from the onboard terminal, and uses the Vehicle Lifecycle State (LCS) quantification model to calculate the condition evaluation value. : Where Q is the quality index, M is the maintenance completion rate, F is the failure frequency, R is the availability rate, and the weights are... Adaptively adjusted based on vehicle type and age (e.g., new passenger cars). old cars ); Task flow deduction submodule: Models task paths based on Markov Decision Process (MDP) and defines reward functions. for: And calculate the task success rate. : Where T is the state transition probability. The discount factor (range 0.8 to 0.95) allows for a minimum task path deduction step of 1 minute, which can predict the task execution status and conflict risk within the next 24 hours.

[0029] Specifically, the quaternary relation graph is implemented using the Neo4j graph database, supporting 100,000 relation writes per second. Node attributes include vehicle model, task type, environmental parameters, etc. The size of the quaternary relation graph is dynamically expanded according to the size of the parking lot (100,000 nodes for small parking lots and 1 million nodes for large parking lots).

[0030] Specifically, The normalization task priority is obtained by mapping to the closed interval [0,1] using the Min-Max extremum normalization algorithm. Complete the standardized conversion from the original cumulative return value to the business quantitative priority; Simultaneously, it serves as the core basis for prioritizing task scheduling and drives the dynamic update of task-vehicle relationships in the four-element relationship graph: when the priority of a task on a certain vehicle is lower than a preset threshold (such as 60%), the weight of the corresponding task-vehicle relationship edge in the graph is automatically reduced (relative reduction of 30%), and the system is triggered to reschedule globally; the training model is iterated every 5 minutes to adapt to the dynamic changes in parking lot conditions, task arrangement, and environmental factors in real time.

[0031] In some embodiments, the intelligent fusion and situational analysis processing module further includes the following deeply coupled sub-modules: Environmental safety perception submodule: fused visible light and infrared images to generate a 3D heat map It also uses a convolutional neural network (CNN) to detect abnormal hotspots and determine the fire hazard level. The calculation is as follows: ,in The intensity of temperature anomalies in the image patch. It uses the Sigmoid activation function and integrates acoustic event detection components to analyze noise signals using Short Time Fourier Transform (STFT). It can identify collision, horn blast, or unusual sound events. Resource Assurance Prediction Submodule: Based on future task requirements output by the Task Flow Deduction Submodule Long Short-Term Memory (LSTM) network is used to process the spare parts consumption sequence. Modeling is performed to predict future demand. : ,in, For spare parts consumption sequence, Here are the model parameters, where This is a time series forecasting model built on LSTM. The LSTM model training data covers 3 months of historical consumption records, and the forecast window is the next 7 days. On the test set, the mean absolute percentage error (MAPE) is ≤8%. It is combined with a hierarchical inventory threshold strategy (dynamically configuring safety stock levels according to spare parts type and depot level). When the predicted value exceeds the inventory threshold, replenishment suggestions and resource allocation plans are automatically generated. The prediction results are used to update the weights of resource-task association edges in the quaternary relationship graph to achieve dynamic optimization of resource allocation. Combat readiness assessment submodule: Comprehensive status evaluation value Fire hazard level Future demand Weather impact coefficients obtained from external meteorological APIs Construct a comprehensive evaluation system based on five rates and calculate the revised combat readiness index. Its expression is as follows: ,in, For the first The original values ​​of the five indicators. For the first The minimum value of each of the five indicators recorded in historical operating data is used as the lower bound benchmark for normalization calculation. For the first The maximum values ​​of the five indicators recorded in historical operating data are used as the upper bound benchmark for normalization calculations. The weights of the corresponding indicators, This is the weather impact attenuation factor (with a value range of [0.6, 1.0], reflecting the degree to which severe weather inhibits overall combat readiness). The weather impact coefficient (e.g., 0 for sunny days, 1 for extreme weather such as heavy rain or snow) is obtained in real time from the external meteorological service interface. The corrected combat readiness index is fed back to the quaternary relationship graph, which enables the quaternary relationship graph to dynamically update the relationship between combat readiness, vehicles / missions / resources and drive subsequent decision optimization.

[0032] Specifically, the environmental safety perception submodule is used to detect fire hazards, abnormal acoustic events, and the risk of crowd gathering in the parking lot. The 3D heat map is generated based on temperature field data collected by infrared cameras, with an update frequency of no less than 1 frame / second. The CNN model in the environmental safety perception submodule adopts the MobileNetV2 architecture and is compressed to 4.8MB through knowledge distillation technology. It can run in real time on embedded GPUs with an average inference latency of ≤85ms and a 95th percentile of ≤100ms. Even after compression, the model maintains an anomaly detection accuracy of ≥98%, making it suitable for low-computing-power hardware environments at the edge nodes of the parking lot. The detection threshold of the acoustic event detection component can be customized, and the accuracy of abnormal sound recognition is ≥95%.

[0033] Specifically, the "five rates" indicators in the combat readiness assessment sub-module are vehicle availability rate, mission completion rate, hidden danger rectification rate, resource support rate, and environmental compliance rate; attenuation factor The combat readiness index is dynamically assigned based on the severity of the weather, ranging from 0.6 to 1.0; it is expressed as a percentage, with a maximum score of 100. When the system is activated, it will automatically trigger a Level 1 combat readiness alarm; and the weights can be dynamically configured according to the type of vehicle depot (such as civilian vehicle depot, emergency support vehicle depot, etc.). and threshold range For example, the civilian vehicle depot focuses on resource guarantee rate and environmental compliance rate, while the emergency vehicle depot focuses on vehicle availability rate and task completion rate. The weight configuration supports one-click switching, which facilitates quick adaptation to different operating scenarios.

[0034] In some embodiments, the multidimensional dynamic output module includes the following cooperating sub-modules: The 3D digital twin visualization submodule integrates the result dataset, dynamically renders the 3D scene of the parking lot, and realizes multi-dimensional interactive browsing with timeline data synchronization, task chain tracking and geospatial positioning based on the sliding window algorithm, reflecting the vehicle status, environmental risks and combat readiness in real time. Dynamic dashboard module: Based on the three-color alarm mechanism and combat readiness trend analysis, it displays the status and changing trends of key indicators in real time; Multi-terminal collaborative push submodule: Pushes situation summaries, alarm events and prediction suggestions to the command screen, mobile APP and duty terminal through standardized API interface, and supports hierarchical permission; Cross-system linkage interface submodule: Reserves interface channels with intelligent fire protection, security control, and dispatch command systems to achieve event-driven cross-system linkage response.

[0035] Specifically, the 3D digital twin visualization submodule uses the WebGL engine to construct the 3D scene of the parking lot and employs layered rendering technology: static backgrounds are presented through high-resolution texture maps, dynamic vehicles and equipment are rendered using GPU instantiation to improve batch rendering efficiency, alarm signs are displayed as overlaid semi-transparent hemispheres, the frame rate is stable at over 60fps, and it supports real-time updates and dynamic mapping of 100,000 data points per second. The timeline backtracking function is based on the edge caching mechanism and supports continuous playback of historical situations for up to 1 hour, meeting the needs of fault review and scheduling analysis.

[0036] Specifically, the dynamic dashboard module uses a three-color alarm mechanism: green indicates normal, yellow indicates warning, and red indicates alarm. Alarm information is sorted by priority, and alarm thresholds are dynamically calculated based on historical data. , ,in, It represents the moving average of a certain indicator over the past 7 days; the combat readiness trend is calculated using a sliding average window (width = 30 minutes) to ensure a smooth trend and to display the change trend of the combat readiness index in real time; it also supports the comparison display of data from the past 7 days and the past 30 days, making it easy to identify periodic patterns and abnormal fluctuations.

[0037] Specifically, when the multi-terminal collaborative push submodule pushes information through a standardized API interface, the push latency is ≤200ms. The information granularity and operation permissions displayed by terminals with different permissions are different. For example, ordinary users can only display basic status (vehicle status, environmental heat map), while administrators can access the complete analysis report (including the detailed calculation process of the "five rates", including the dynamic weight configuration of indicators such as vehicle availability rate and task completion rate), and support synchronous updates from multiple terminals.

[0038] Specifically, the cross-system linkage interface submodule uses common protocols such as MQTT and HTTP. The linkage triggering condition for the cross-system linkage interface submodule, for example, is when... The fire protection system will be automatically activated when... The scheduling priority is dynamically adjusted in real time, and all linkage operations are implemented through an event-driven mechanism, supporting log recording and 72-hour backtracking analysis.

[0039] In some embodiments, the intelligent fusion and situational inference processing module is deployed in an edge-cloud collaborative architecture. The edge node (NVIDIA Jetson AGX Xavier) is responsible for real-time data cleaning and local alarms (latency ≤100ms), while the cloud platform (Alibaba Cloud ECS) is responsible for map construction and model training (incremental updates every 24 hours) and supports offline mode operation (data is cached to local storage, which can support 24-hour offline analysis).

[0040] Specifically, the edge nodes and cloud platform adopt a breakpoint resume mechanism, and offline data is automatically synchronized after the network is restored; in offline mode, local data collection, basic assessment and alarm output can be completed independently to ensure uninterrupted parking lot management.

[0041] In some embodiments, the vehicle state assessment submodule introduces a fuzzy membership function to classify the state: ,in, For actual performance parameters, Set as the standard performance value for the vehicle. To allow for different tolerance ranges based on vehicle type and age, new vehicles Smaller values ​​(e.g., 0.05 for passenger cars), older vehicles Appropriately relax (e.g., for commercial vehicles that are 5 years old) =0.14), ensuring that the condition assessment accurately reflects the actual vehicle condition; As one of the input parameters of the task flow deduction submodule, and Together, they constitute the vehicle health assessment system.

[0042] The parking lot comprehensive situation display system of this invention has the following advantages: This invention employs a unified spatiotemporal calibration mechanism for multi-source heterogeneous data, combined with a sliding window synchronization algorithm and edge caching technology. It can achieve unified timeline calibration and reliable local caching for all types of heterogeneous data, including vehicle terminals, video surveillance, and IoT sensors. This fundamentally solves the data silo problem commonly found in traditional parking lot management systems, ensuring accurate alignment and efficient collaborative flow of multi-source data, significantly improving the integrity, real-time performance, and reusability of all parking lot data. Advanced intelligent algorithms such as Markov Decision Process (MDP), Long Short-Term Memory (LSTM), and Convolutional Neural Network (CNN) are deeply embedded into the entire situational simulation process. This enables accurate simulation of task flows for the next 24 hours, spare parts consumption prediction errors ≤8%, and fire hazard identification accuracy ≥98%. This drives a fundamental shift in parking lot comprehensive situational awareness from passive recording and monitoring to proactive perception and early warning, and from static data display to dynamic intelligent simulation. A 3D digital twin layered rendering technology is used to build a full-domain visualization scene of the parking lot, maintaining a stable rendering frame rate of over 60fps, while also supporting historical situational timelines. Retrospective analysis, end-to-end task chain tracking, and precise geospatial positioning provide a clear and intuitive view of vehicle status, environmental risks, and combat readiness capabilities, fully meeting the needs of efficient command and dispatch and real-time management in parking lots. Leveraging knowledge distillation technology, the CNN model is compressed to less than 5MB. Combined with an edge-cloud collaborative deployment architecture, embedded GPU inference latency is reduced to less than 100ms. It also supports offline operation during network outages and resume download after network recovery, perfectly adapting to the low-computing-power, high-real-time application scenarios of parking lot edge computing. A comprehensive evaluation system based on five rates is constructed, allowing for dynamic correction of the combat readiness index. It supports adaptive configuration of indicator weights according to different parking lot types such as civilian, emergency, and support functions, and achieves cross-system linkage response with intelligent fire protection, security control, and dispatch command systems, comprehensively improving the intelligence level, collaborative response capabilities, and scientific decision-making of parking lot management. The system adopts a modular and loosely coupled architecture design, with each module capable of independent upgrades and flexible configuration. It can quickly adapt to the needs of parking lot scenarios of different scales and purposes, possessing strong versatility, practicality, and scalability.

[0043] In another aspect, this invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, enables the implementation of the method described above. The storage medium supports hot-swapping and data encryption, and the stored program and data comply with parking lot data security management regulations to prevent information leakage.

[0044] The computer-readable medium may be included in the apparatus, device, or system disclosed herein, or it may exist independently.

[0045] The computer-readable storage medium can be any tangible medium that contains or stores a program, and can be an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device. More specific examples include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, optical fibers, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0046] The computer-readable storage medium may also include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code, specific examples of which include, but are not limited to, electromagnetic signals, optical signals, or any suitable combination thereof.

[0047] Parameter description: All parameter values ​​mentioned in this manual are not subjective assumptions, but rather a comprehensive result of three factors: the security / efficiency requirements of the application scenario, the requirements of industry standards and specifications, and the empirical thresholds of industry practice. In actual applications, the parameters will be fine-tuned according to the relevant scenarios. Parameter fine-tuning can be completed through the system's backend visual configuration interface without modifying the code, reducing the difficulty of operation and maintenance.

[0048] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A comprehensive parking lot situation display system, characterized in that, It includes a multi-source data input module, an intelligent fusion and situational analysis processing module, and a multi-dimensional dynamic output module connected in sequence, wherein: The multi-source data input module is used to collect and access multi-source heterogeneous data streams from vehicle terminals, video surveillance, IoT sensors, electronic fences, dispatching systems, and environmental monitoring equipment. The intelligent fusion and situational inference processing module is used to perform structured processing, deep correlation analysis and intelligent inference on the multi-source heterogeneous data stream, and generate a result dataset; The multidimensional dynamic output module is used to present the result dataset in the form of a three-dimensional digital twin, a dynamic dashboard, and multi-terminal linkage, and supports hierarchical permission and cross-system linkage.

2. The comprehensive parking lot situation display system according to claim 1, characterized in that, The multi-source data input module supports timestamp calibration and edge caching mechanisms. It performs timeline calibration on asynchronous data using a sliding window synchronization algorithm, and its synchronization objective function is: ,in, For the first timestamps for similar devices Used as the base time.

3. The comprehensive parking lot situation display system according to claim 1, characterized in that, The intelligent fusion and situational simulation processing module includes the following sub-modules: The semantic knowledge graph construction submodule: Based on the entity-relation-attribute triple model, it constructs a four-element relation graph of "vehicle-task-person-environment". This provides structured data support for subsequent sub-modules; Vehicle Status Assessment Submodule: This module calls upon vehicle attribute data from the aforementioned quaternary relationship graph, combines it with real-time status information from the onboard terminal, and calculates the status evaluation value using the Vehicle Lifecycle Status (LCS) quantification model. : Where Q is the quality index, M is the maintenance completion rate, F is the failure frequency, R is the availability rate, and the weights are... Adaptively adjusts based on vehicle type and age; Task flow deduction submodule: Models task paths based on Markov Decision Process (MDP) and defines reward functions. for: And calculate the task success rate. : Where T is the transition probability, The discount factor (range 0.8 to 0.95) allows for a minimum task path deduction step of 1 minute, which can predict the task execution status and conflict risk within the next 24 hours.

4. The comprehensive parking lot situation display system according to claim 3, characterized in that, The intelligent fusion and situational simulation processing module also includes the following deeply coupled sub-modules: Environmental safety perception submodule: fused visible light and infrared images to generate a 3D heat map It also uses a convolutional neural network (CNN) to detect abnormal hotspots and determine the fire hazard level. The calculation is as follows: ,in The intensity of temperature anomalies in the image patch. It uses the Sigmoid activation function and integrates acoustic event detection components to analyze noise signals using Short Time Fourier Transform (STFT). It can identify collision, horn blast, or unusual sound events. Resource Assurance Prediction Submodule: Based on the future task requirements output by the Task Flow Deduction Submodule. Long Short-Term Memory (LSTM) network is used to process spare parts consumption sequences. Modeling is performed to predict future demand. : ,in, For spare parts consumption sequence, Here are the model parameters, where This is a time series forecasting model built on LSTM. The LSTM model training data covers 3 months of historical consumption records, and the prediction window is the next 7 days. On the test set, the mean absolute percentage error (MAPE) is ≤8%. Combined with a hierarchical inventory threshold strategy (dynamically configuring safety stock levels according to spare parts type and depot level), when the predicted value exceeds the inventory threshold, replenishment suggestions and resource allocation plans are automatically generated. The prediction results are used to update the weights of resource-task association edges in the four-element relationship graph to achieve dynamic optimization of resource allocation. Combat readiness assessment submodule: integrates the aforementioned status evaluation values The fire hazard level mentioned above The future demand mentioned Weather impact coefficients obtained from external meteorological APIs Construct a comprehensive evaluation system based on "five rates" and calculate the revised combat readiness index. Its expression is as follows: ,in, For the first The original values ​​of the five indicators. For the first The minimum value of each of the five indicators recorded in historical operating data is used as the lower bound benchmark for normalization calculation. For the first The maximum values ​​of the five indicators recorded in historical operating data are used as the upper bound benchmark for normalization calculations. The weights of the corresponding indicators, As a weather-related attenuation factor, The weather impact coefficient is obtained in real time from an external meteorological service interface. The corrected combat readiness index is fed back to the quaternary relationship graph, which enables the quaternary relationship graph to dynamically update the correlation between combat readiness, vehicles / missions / resources and drive subsequent decision optimization.

5. The comprehensive parking lot situation display system according to claim 1, characterized in that, The multidimensional dynamic output module includes the following collaboratively operating sub-modules: The 3D digital twin visualization submodule integrates the resulting dataset, dynamically renders the 3D scene of the parking lot, and realizes multi-dimensional interactive browsing of timeline data synchronization, task chain tracking and geospatial positioning based on the sliding window algorithm, reflecting the vehicle status, environmental risks and combat readiness in real time. Dynamic dashboard module: Based on the three-color alarm mechanism and combat readiness trend analysis, it displays the status and changing trends of key indicators in real time; Multi-terminal collaborative push submodule: Pushes situation summaries, alarm events and prediction suggestions to the command screen, mobile APP and duty terminal through standardized API interface, and supports hierarchical permission; Cross-system linkage interface submodule: Reserves interface channels with intelligent fire protection, security control, and dispatch command systems to achieve event-driven cross-system linkage response.

6. The comprehensive parking lot situation display system according to claim 1, characterized in that, The intelligent fusion and situational inference processing module is deployed in an edge-cloud collaborative architecture. Edge nodes are responsible for real-time data cleaning and local alarms, while the cloud platform undertakes map construction and model training, and supports offline operation.

7. The comprehensive parking lot situation display system according to claim 3, characterized in that, The vehicle state assessment submodule introduces a fuzzy membership function to classify the state: ,in, For actual performance parameters, Set as the standard performance value for the vehicle. To allow for different tolerance ranges based on vehicle type and age, new vehicles Smaller values, older vehicles The requirements are appropriately relaxed to ensure that the condition assessment accurately reflects the actual vehicle condition; As one of the input parameters of the task flow deduction submodule, and Together, they constitute the vehicle health assessment system.

8. An electronic device, characterized in that, include: One or more processors; A storage unit is used to store one or more programs, which, when executed by one or more processors, enable the one or more processors to implement the parking lot integrated situation display system according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it can realize the comprehensive parking lot situation display system according to any one of claims 1 to 7.

Images