Intelligent parking system and device based on digital twinning and artificial intelligence

Abstract

The application relates to the technical field of digital twinning and artificial intelligence, and provides a smart parking system and device based on digital twinning and artificial intelligence. Specifically, construction data is extracted, a basic digital twinning model is established, device coordinates are read, the device coordinates are mapped to the basic digital twinning model, a correction mapping with the basic digital twinning model is established according to collected data, time sequence collection information of an image collection device is read, parking features and fixed features are identified from the time sequence collection information, feature identification results are updated in the basic digital twinning model based on the correction mapping results, a digital twinning model is generated, parking space state identification is performed on the digital twinning model, path planning results are generated based on received vehicle parking requests, and smart parking management is performed based on the path planning results. Therefore, the combination of digital twinning and artificial intelligence technology can improve the efficiency and control accuracy of smart parking management.

Description

Technical Field

[0001] This application relates to the fields of digital twin and artificial intelligence technologies, and in particular to smart parking systems and devices based on digital twins and artificial intelligence. Background Technology

[0002] With rapid urbanization and a continuous increase in car ownership, parking difficulties have become a major problem in urban transportation. Traditional parking management methods suffer from low parking space utilization, opaque parking fees, and long search times, failing to meet the efficient, convenient, and intelligent parking requirements of modern cities. Therefore, intelligent parking systems and devices based on digital twin and artificial intelligence technologies have emerged as a new solution to the urban parking problem.

[0003] Digital twin technology combines physical parking lots with virtual ones, enabling comprehensive management and optimization of parking lots by monitoring various parameters in real time, such as parking space occupancy and vehicle traffic flow. Simultaneously, artificial intelligence technology can analyze and predict historical parking data, providing strong support for parking lot planning and management. Smart parking systems and devices based on digital twins and artificial intelligence can improve parking lot utilization and the parking experience for drivers, alleviating urban traffic congestion.

[0004] In summary, existing smart parking management methods suffer from low efficiency and low control precision. Summary of the Invention

[0005] This application provides a smart parking system and device based on digital twins and artificial intelligence, aiming to solve the technical problems of traditional parking management methods being inefficient, unable to meet the demand for high efficiency, and having limited parking data processing capabilities, which threatens the control accuracy of the system terminal for parking management.

[0006] In view of the above problems, this application provides a smart parking system and device based on digital twins and artificial intelligence.

[0007] The first aspect disclosed in this application provides a smart parking method based on digital twins and artificial intelligence. The method includes: extracting construction data of a target parking lot and establishing a basic digital twin model using the construction data; reading the device coordinates of an image acquisition device and mapping the device coordinates to the basic digital twin model, and establishing a correction mapping between the image acquisition device and the basic digital twin model based on the acquisition data of the image acquisition device; reading the time-series acquisition information of the image acquisition device, identifying parking features and fixed features in the time-series acquisition information, and updating the feature identification results in the basic digital twin model based on the correction mapping results to generate a digital twin model; identifying parking space status in the digital twin model and generating a path planning result based on received vehicle parking requests; and performing smart parking management based on the path planning results.

[0008] Another aspect of this application discloses a smart parking system based on digital twins and artificial intelligence. The system is used in the aforementioned method and includes: an extraction module for extracting construction data of a target parking lot and establishing a basic digital twin model using the construction data; a correction mapping module for reading the device coordinates of an image acquisition device and mapping the device coordinates to the basic digital twin model, establishing a correction mapping between the image acquisition device and the basic digital twin model based on the acquired data; a digital twin model generation module for reading the time-series acquisition information of the image acquisition device, performing parking feature and fixed feature recognition on the time-series acquisition information, and updating the feature recognition results in the basic digital twin model based on the correction mapping results to generate a digital twin model; a first path planning module for recognizing parking space status in the digital twin model and generating path planning results based on received vehicle parking requests; and a parking management module for performing smart parking management based on the path planning results.

[0009] Another aspect of this application discloses a smart parking device based on digital twins and artificial intelligence. The device is used in the aforementioned system and includes: a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are connected via the bus. When the computer program is executed by the processor, it implements the steps of the smart parking system based on digital twins and artificial intelligence as described in any of the above methods.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] The aforementioned intelligent parking system and device based on digital twins and artificial intelligence works by reading the coordinates of an image acquisition device and mapping them onto a basic digital twin model. Then, data from the image acquisition device is used to establish a correction mapping with the basic digital twin model, thereby improving the model's accuracy. Subsequently, the system reads the time-series data acquired by the image acquisition device, identifies parking and stationary features, and updates these feature recognition results in the basic digital twin model, generating a more accurate digital twin model. The system further identifies the parking space status within this digital twin model and generates corresponding path planning results upon receiving a vehicle parking request. Finally, based on these path planning results, the system implements intelligent parking management, thereby optimizing the parking lot's operational efficiency. In summary, this method integrates digital twin and artificial intelligence technologies to achieve efficient and highly precise parking management.

[0012] The above description is merely an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart illustrating a smart parking system based on digital twins and artificial intelligence in one embodiment.

[0015] Figure 2 This is a schematic diagram illustrating the feature recognition results obtained by a smart parking system based on digital twins and artificial intelligence in one embodiment.

[0016] Figure 3 This is a system architecture diagram of a smart parking system based on digital twins and artificial intelligence in one embodiment.

[0017] Figure 4 This is a device diagram of a smart parking system based on digital twins and artificial intelligence in one embodiment.

[0018] Figure labeling: Extraction module 1, Correction mapping module 2, Digital twin model generation module 3, First path planning module 4, Parking management module 5. Detailed Implementation

[0019] This application provides a smart parking system based on digital twins and artificial intelligence, which solves the technical problems of low efficiency and low control precision in the prior art.

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0021] It should be noted that the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or modules that are not explicitly listed or that are inherent to such process, method, product, or device.

[0022] Example 1

[0023] like Figure 1 As shown, this application provides a smart parking method based on digital twins and artificial intelligence, the method comprising:

[0024] Extract the construction data of the target parking lot and establish a basic digital twin model based on the construction data;

[0025] With rapid societal development and accelerated urbanization, automobiles have become an essential mode of transportation. However, traditional parking lot management methods are struggling to meet the ever-increasing parking demand. Therefore, improving parking lot utilization efficiency and management has become one of the most pressing issues to address.

[0026] In this embodiment, building the future of smart parking begins with a precise digital twin model. The construction data refers to various data from the construction and planning process of the target parking lot, including detailed information such as the parking lot's structural design, geographical location, parking space layout, entrance and exit locations, and facility distribution. The basic digital twin model is a virtual model built based on the construction data of the target parking lot. This model uses digital technology to replicate the physical structure and environment of the parking lot in virtual space, achieving a one-to-one correspondence between the real parking lot and the virtual model. The basic digital twin model provides a foundational framework for the smart parking system, used for subsequent intelligent management operations such as data analysis, feature recognition, and path planning. The system terminal extracts the construction data of the target parking lot and performs data cleaning, integration, and transformation on the collected raw data. Utilizing 3D modeling, environmental rendering, and digital representation of facilities and equipment, it begins to build a digital twin model based on the processed data. This allows the system terminal to create a virtual digital model of the target parking lot that is consistent with the actual situation. This process achieves a digital representation of the parking lot environment, providing fundamental support for subsequent smart parking management. In summary, this step transforms real-world parking lots into a virtual digital world for more efficient and precise management and planning.

[0027] Read the device coordinates of the image acquisition device and map the device coordinates to the basic digital twin model. Establish a correction mapping between the image acquisition device and the basic digital twin model based on the acquisition data of the image acquisition device.

[0028] In one embodiment, deep fusion of the image acquisition device and the digital twin model is also required. The device coordinates refer to the physical coordinates of the image acquisition device, which pinpoint its exact location within the parking lot. The correction mapping refers to the process of correcting and adjusting the basic digital twin model using the data collected by the image acquisition device. By reading the device coordinates of the image acquisition device and mapping them onto the basic digital twin model, a connection between the model and the actual device can be established. After reading the device coordinates of the image acquisition device, the system terminal processes the acquired image data to extract key feature points, such as parking space markers, vehicles, and obstacles. Then, it finds the parts corresponding to these feature points in the basic digital twin model and maps them to the basic digital twin model, thus achieving a connection between the actual device and the virtual model. To ensure the accuracy of this connection, the basic digital twin model is corrected using data collected in real-time by the image acquisition device and based on the mapping relationship, making the model closer to reality and reflecting the close relationship between the real parking lot and the digital world. In summary, this correction mapping method improves the accuracy of the model and further promotes efficient and high-precision parking management.

[0029] Read the time-series acquisition information of the image acquisition device, identify parking features and stationary features in the time-series acquisition information, and update the feature recognition results in the basic digital twin model based on the correction mapping results to generate a digital twin model;

[0030] In one embodiment, after the calibration mapping, the temporal acquisition information of the image acquisition device needs to be identified to generate a digital twin model. The temporal acquisition information refers to a series of image or video data continuously acquired by the image acquisition device, which contains information about the time sequence. The parking features refer to features related to vehicle parking in the temporal acquisition information of the image acquisition device, including the vehicle's parking location, the occupancy status of parking spaces, and vehicle entry and exit behavior. The fixed features refer to relatively static and unchanging features in the parking lot environment, including the parking lot's entrance and exit, lane markings, parking space layout, walls, pillars, and other infrastructure. The feature recognition result refers to the recognition result of parking features and fixed features obtained after processing the temporal acquisition information, including various attributes, labels, or coordinate information, used to describe and locate key elements such as vehicles, parking spaces, and infrastructure in the parking lot. The digital twin model refers to a virtual model generated after calibration mapping based on the temporal acquisition information and feature recognition results of the image acquisition device. This model is an accurate replica of the real parking lot in the digital world, containing various features and attributes of the parking lot, as well as the relationships between them. The system terminal continuously and in real-time monitors the parking lot by reading the time-series data acquired by the image acquisition device. This data is further analyzed to identify parking and stationary features of the parking lot. After identification, these features are matched with features in the basic digital twin model. This step ensures that the identified features are correctly associated with their corresponding features in the model. Based on the previously established calibration mapping relationship, the coordinates of the identified features are transformed into the coordinate system of the basic digital twin model. Transformations such as translation, rotation, and scaling are used to ensure the accurate position of the features in the model, thereby ensuring data accuracy and consistency. Then, based on the calibration mapping results, the processed new feature data is integrated into the basic digital twin model, involving updating the object states, attributes, or relationships in the model to reflect the latest situation of the parking lot. For example, updating the real-time location of vehicles and the occupancy status of parking spaces in the model. The updated digital twin model is then verified to ensure that the updated features match the actual situation and that no inconsistencies or errors have been introduced. Through continuous updates to the feature recognition results, the digital twin model is gradually generated and improved. This model not only includes the static layout of the parking lot but also incorporates real-time dynamic information, such as vehicle movement and parking space occupancy. Based on this real-time updated digital twin model, the system terminal can extract historical data, real-time data, and other information related to parking lot characteristics from the updated model. Then, based on the prediction and decision-making objectives, relevant features are selected for analysis. For example, to predict parking demand for a specific time period, it is necessary to analyze historical parking data, time periods, weather, and other factors.A predictive model is built using time series analysis through steps including defining the prediction target, collecting time series data, data preprocessing, observing time series features, selecting a time series model (ARIMA, SARIMA, etc.), estimating model parameters, and model validation and diagnostics. This model is trained based on the selected features and used to predict future conditions. Using current and past data as input, the model generates future predictions. These results include parking demand and parking space occupancy at a future point in time. This process enables real-time updates to the digital twin model, keeping it synchronized with the real parking lot, and predicting future parking conditions through the digital twin model. In summary, this approach allows for the construction of a dynamic and accurate digital twin model, providing real-time and accurate data support for smart parking systems, thereby achieving efficient and highly precise parking management.

[0031] Furthermore, such as Figure 2 As shown, this application provides a method for obtaining feature recognition results, which further includes:

[0032] Image contour features are extracted from the time-series acquired information, and feature analysis is performed based on the image contour feature extraction results to identify parking features and stationary features.

[0033] Preferably, to obtain a more accurate digital twin model, further identification and extraction of the time-series acquired information is required. Image contour feature extraction refers to the process of extracting the outer edge shape features of objects from the time-series acquired information of the image acquisition device, including vehicles, parking spaces, walls, pillars, etc. Feature identification refers to the process of identifying and marking parking features and fixed features based on the image contour feature extraction results. After the system terminal performs in-depth processing of the time-series acquired information provided by the image acquisition device, it first performs image contour feature extraction. This step involves extracting the contours of various objects from the original image, which clearly outline the shape and boundaries of key elements such as vehicles and parking spaces. After obtaining these contour features, these features are analyzed in detail. Through analysis, it is possible to accurately distinguish which contours correspond to vehicles and which correspond to parking spaces or other fixed facilities. Then, feature identification of parking and fixed features is performed. When performing parking feature recognition, the time-series acquired information is first preprocessed to remove noise, enhance image quality, and extract key features. In the preprocessed image, image processing technology is used to extract key information related to parking features. The extracted feature information is transmitted to the PLC (Programmable Logic Controller) system. The PLC system scans the input signals from the image acquisition device and performs various logical operations, comparisons, timer counting, and other processing and analysis tasks on the input signals according to a pre-written program by industry professionals. Through the execution of the pre-written program, the PLC can identify, classify, and analyze parking features and generate corresponding output results. Based on the PLC's control and processing, the system can identify parking features in the time-series acquisition information. Furthermore, based on the output results, the PLC can generate control signals and output them to the actuators, facilitating subsequent parking lot management. For example, these output signals can be used to control parking lot equipment, indicator lights, turnstiles, etc., enabling real-time monitoring and corresponding control operations. When performing fixed feature recognition, after extracting fixed features from the pre-processed image using image processing technology, these features need to be described. Feature descriptive factors, such as SIFT, SURF, and ORB, are used to calculate the shape, size, color, and other attributes of the features, transforming the features into a comparable mathematical expression. Then, the feature templates pre-stored in the PLC system are traversed, and similarity calculations are performed to find the template with the most similar extracted features. In this embodiment, to eliminate false matches, industry professionals pre-set a similarity threshold. Only when the similarity between features exceeds this threshold is a matching feature considered to have been found. This reduces false identifications and improves identification accuracy. Finally, based on these analysis and identification results, feature labeling is performed on parking and stationary features, i.e., these features are labeled to clarify their identity and attributes.This series of processes enables the system terminal to accurately extract the required information from the original images, providing a solid data foundation for subsequent smart parking management.

[0034] If the feature identifier is a fixed feature, then fixed occupancy identification is performed, and a fixed occupancy feature identification result is generated;

[0035] If the feature is identified as a parking feature, the multi-angle sensor is activated to perform multi-angle data acquisition, and a vehicle occupancy feature recognition result is established based on the multi-angle data acquisition results.

[0036] Preferably, the fixed occupancy identification refers to the process of identifying and determining whether fixed facilities or areas within a parking lot are occupied. For example, for parking spaces, fixed occupancy identification can determine whether a vehicle is parked in the parking space, thereby determining the occupancy status of the parking space. The fixed occupancy feature identification result refers to the identification result obtained after performing fixed occupancy identification on fixed features, used to indicate whether the fixed facility is occupied and the specific circumstances of the occupancy, including the occupancy status and occupancy time. The vehicle occupancy feature identification result refers to the result of analyzing vehicle parking conditions based on multi-angle data collection results, including features such as the vehicle's position, shape, and color. When the feature is identified as a fixed feature, the system terminal will perform fixed occupancy identification to determine the usage status of fixed facilities within the parking lot. This step helps to understand the static layout of the parking lot and the occupancy status of objects. When the feature is identified as a parking feature, in order to more accurately grasp the vehicle parking situation, the system terminal will activate multi-angle sensors to collect data. These sensors collect data from multiple angles, enabling a more comprehensive understanding of the vehicle's position and status. Based on this data, a vehicle occupancy feature identification result can be established, thereby accurately determining how vehicles occupy parking space. In summary, whether targeting fixed or parking characteristics, this series of operations aims to more accurately grasp the usage of parking lots, optimize the overall operational efficiency of parking lots, and improve the efficiency and control precision of parking management.

[0037] The feature recognition result is obtained by using the fixed occupancy feature recognition result and the vehicle occupancy feature recognition result.

[0038] Preferably, the system terminal obtains more comprehensive and accurate feature recognition results by combining the fixed occupancy feature recognition results and the vehicle occupancy feature recognition results. This result not only includes the usage status of fixed facilities within the parking lot but also accurately describes the parking status of vehicles. In summary, these feature recognition results provide reliable data support for the smart parking system, enabling it to manage parking lots more efficiently and improve overall operational efficiency.

[0039] Furthermore, this application provides a method for identifying parking space status using the digital twin model, which further includes:

[0040] The basic digital twin model is used to establish the relationship between basic parking spaces;

[0041] Optionally, parking space status identification also needs to be performed on the digital twin model. The basic parking space association refers to establishing fundamental relationships and connections between parking spaces in the digital twin model, including relationships such as adjacency, entrance / exit, parking space type, and parking space occupancy. By establishing basic parking space associations through the basic digital twin model, the system terminal essentially maps the relationships between parking spaces in a virtual digital world. This process helps understand the spatial relationships, usage status, and other specific attributes of parking spaces. Thus, when the system terminal needs to query, manage, or plan parking spaces, it can utilize this association information to make decisions more efficiently and accurately. In summary, by establishing basic parking space associations, the management and understanding of real-world parking lots can be strengthened at the digital level, thereby improving the efficiency and control precision of parking management.

[0042] The digital twin model is used to identify the parking space status within the model based on a state recognition network, and an occupancy recognition result is generated.

[0043] Based on the occupancy identification results, the association between the basic parking spaces is invoked, and the occupancy impact value is generated;

[0044] Vehicle parking requests are matched based on occupancy impact values.

[0045] Optionally, the state recognition network refers to a deep learning network specifically designed for recognizing and classifying parking space status in a digital twin model. The occupancy impact value refers to a numerical value generated based on the parking space occupancy recognition results and the correlation between basic parking spaces, used to measure the impact of parking space occupancy on surrounding parking spaces or the overall parking lot. The system terminal inputs data from the digital twin model into the state recognition network. This data may include information such as the shape, location, and color of the parking space, as well as possible image data of the surrounding environment. The state recognition network extracts features from the input data, identifying features effective for parking space status recognition. The extracted features are then fed into a classifier for status classification, categorizing parking space status into "occupied" or "vacant." After classification, the network generates parking space occupancy recognition results and outputs these results for subsequent steps. By recognizing parking space status in the digital twin model as described above, it is possible to determine which parking spaces are occupied and which are vacant. Then, using the obtained occupancy recognition results and the correlation between basic parking spaces, the occupancy impact value is calculated. This value helps understand the impact of a specific parking space's occupancy on surrounding spaces and the entire parking lot. Finally, the system terminal processes vehicle parking requests based on this occupancy impact value, performing request matching to find the most suitable parking space for each vehicle. This considers both the current status of the parking space and the mutual influence between parking spaces to achieve the optimal parking strategy. In summary, this process ensures more efficient and accurate fulfillment of vehicle parking needs, improves the overall operational efficiency of the parking lot, and ultimately enhances the efficiency and precision of parking management.

[0046] Furthermore, this application provides a method for generating path planning results based on the ranking and optimization results, which also includes:

[0047] The vehicle parking request is parsed to generate a parsing result, which includes vehicle information, user information, and destination information.

[0048] The occupancy impact value is calculated by matching the vehicle information and the user information to obtain the matching calculation result;

[0049] Optionally, the vehicle parking request also needs to be parsed for subsequent route planning results generated based on the ranking and optimization results. The vehicle information refers to various information related to the vehicle requesting parking, including license plate number, vehicle model, size, and color. The user information refers to information about the user initiating the parking request, including username, contact information, user type, and user driving information. The matching calculation refers to calculating and matching the vehicle and user information with existing occupancy impact values ​​to obtain a specific occupancy impact value for the vehicle and user. When a vehicle parking request is received, the system terminal first parses the request, generating a parsing result. This parsing result includes vehicle information, user information, and destination information. Then, using this vehicle and user information, and based on the parking space status identification results, currently available parking spaces are determined. These spaces should be unoccupied and meet vehicle size requirements. The occupancy impact values ​​between parking spaces are then matched with the vehicle and user information. This matching process considers the user's preferences for parking spaces, such as proximity to the destination and ease of access. Simultaneously, vehicle characteristics, such as vehicle size and whether charging is required, are also considered. After matching calculations, a matching result for the vehicle and user is obtained. This result helps understand the potential impact of the vehicle and user in the parking lot, and which parking space best suits the user's needs. In summary, this process is a comprehensive matching process that considers vehicle, user, and parking space information to ensure that the system terminal provides the most suitable parking solution for each vehicle, while optimizing the overall parking lot's operational efficiency, thereby improving parking management efficiency.

[0050] The matching results are used to perform distance correlation calculations based on the destination information;

[0051] The matching calculation results are sorted and optimized based on the distance correlation calculation results;

[0052] Path planning results are generated based on the ranking and optimization results.

[0053] Optionally, the distance association calculation refers to calculating the distance between each matched parking space and the destination based on the destination information in the vehicle parking request, and associating these distances with the parking space matching calculation results. After parsing the vehicle parking request and obtaining the matching calculation results, the system terminal further considers the destination information. Based on the destination information, distance association calculations are performed on these matched parking spaces, that is, by obtaining the destination coordinates, calculating the distance from each matched parking space to the destination, and associating the distance with the matching results to calculate the actual distance from each possible parking space to the destination. Then, using the results of these distance association calculations, all matched parking spaces are ranked and optimized. When performing ranking and optimization, firstly, the ranking criteria need to be determined. For example, the distance between the parking space and the destination, the availability of the parking space, the location of the parking space, etc. These criteria will be used to evaluate the quality of each matched parking space. Then, data related to the ranking criteria is extracted from the matching calculation results. For example, the coordinates of each parking space are extracted for distance calculation, and the availability status of the parking space is extracted for evaluating whether it is suitable for parking, etc. According to the set ranking criteria, an evaluation value is calculated for each matched parking space. For example, a distance assessment value can be calculated based on the distance from the parking space to the destination, with a higher assessment value for closer distances. Combined with parking space availability, the assessment value for available parking spaces will be even higher. Then, merge sort is used to rank the matching results based on the calculated assessment values. The goal of sorting is to prioritize better parking spaces for subsequent selection. After the initial sorting, further optimization operations can be performed, such as adjusting the sorting results according to user needs. For example, if the user is price-sensitive, lower-cost parking spaces can be ranked higher. Finally, based on the optimized sorting results, a route planning result is generated for the user. This result guides the user on how to reach their destination most quickly and conveniently. In summary, this process comprehensively considers factors such as parking space suitability and distance from the destination, aiming to provide users with a more convenient and user-friendly parking and walking route planning experience, thereby improving the control precision of parking management.

[0054] Furthermore, methods for achieving optimal sorting also include:

[0055] Configure the normalization coefficients for the distance and matching values;

[0056] The distance correlation calculation result and the matching calculation result are normalized by the normalization coefficient to complete the sorting optimization.

[0057] Optionally, in the ranking and optimization process that comprehensively considers distance and matching values, normalization coefficients for distance and matching values ​​are first configured. These coefficients are used to balance the weight of distance and matching values ​​in the evaluation, ensuring they have appropriate influence in the ranking process. Through normalization, distance correlation calculation results and matching calculation results with different units and value ranges can be transformed into a unified scale, allowing them to be directly compared and comprehensively evaluated. The configuration of normalization coefficients can be adjusted according to actual needs to adapt to different scenarios and user preferences. Finally, based on the results after normalization, the ranking and optimization are completed, resulting in an ordered list that comprehensively considers distance and matching factors, providing users with better parking options.

[0058] Furthermore, this application provides a method for real-time path planning, which also includes:

[0059] Obtain route congestion information for the target parking lot;

[0060] Real-time path planning is performed using the sorting optimization results and the path congestion information to obtain the path planning results.

[0061] Optionally, in finding the optimal parking solution, in addition to considering the distance and matching degree between the parking space and the destination, it is also necessary to obtain real-time route congestion information for the target parking lot. This means that the system terminal needs to understand the road congestion situation within the parking lot to prevent users from entering congested areas. Combining the previously obtained ranking and optimization results, distance, matching degree, and route congestion can be comprehensively considered to perform real-time route planning. Such planning ensures that users not only park in a suitable location but also reach it in the shortest possible time. Ultimately, the system terminal obtains a route planning result that comprehensively considers multiple factors, providing users with a more convenient and efficient parking solution, thereby improving the efficiency of parking management.

[0062] The digital twin model is used to identify parking space status and generate route planning results based on the received vehicle parking requests.

[0063] In one embodiment, using a digital twin model to identify parking space status enables real-time monitoring of every parking space in a parking lot, clearly identifying which spaces are vacant and which are occupied. Based on this information, when a parking request is received, the system generates a route planning result by combining vehicle and user information from the request with route congestion information. This process comprehensively considers factors such as parking space availability, distance, and congestion conditions, aiming to provide users with an optimal parking solution, ensuring that users can quickly and conveniently find an available parking space and arrive smoothly. In summary, by utilizing digital twin model technology, more intelligent and efficient parking services can be provided, thereby improving the efficiency and control accuracy of parking management.

[0064] Intelligent parking management is implemented based on the route planning results.

[0065] In one embodiment, intelligent parking management based on path planning results enables more intelligent and efficient parking lot operations. By analyzing and applying the path planning results, the occupancy status of parking spaces and the parking routes of vehicles can be monitored in real time, allowing for precise allocation and management of parking resources. This enables the system terminal to maximize the utilization of parking lot capacity, reduce parking space idle time, and improve vehicle parking efficiency. Simultaneously, intelligent parking management also provides convenient service experiences, such as real-time parking space inquiry, reservation, and navigation functions, enhancing user satisfaction. In summary, intelligent parking management based on path planning results helps optimize parking lot operational efficiency, improve the parking experience, achieve more intelligent and convenient parking management, and ultimately improve the efficiency and control precision of parking management.

[0066] Furthermore, this application provides a method for invoking an intelligent parking robot and generating a parking space location plan based on parking space status identification, the method further including:

[0067] Determine whether the vehicle parking request is an automatic parking request;

[0068] When the vehicle parking request is an automatic parking request, the intelligent parking robot is invoked, and a parking space positioning plan is generated based on the parking space status, and the automatic position of the intelligent parking robot is determined simultaneously.

[0069] Once the intelligent parking robot aligns with the target vehicle at the designated automatic location, it completes intelligent parking management based on the parking space positioning plan.

[0070] Preferably, before finalizing smart parking management, an intelligent parking robot needs to be invoked to generate a parking space location plan. This parking space location plan involves identifying a specific, suitable parking space for the intelligent parking robot and planning the path from its current location to that space. This process involves identifying and analyzing the parking space status within the parking lot to determine which spaces are vacant and suitable for parking. The automatic positioning refers to the intelligent parking robot's ability to determine its own location information in real time and synchronize it with the parking space location plan while performing automatic parking tasks. This location information is a crucial reference for the robot's navigation and vehicle parking. When a vehicle sends a parking request, the system terminal first determines whether the request is an automatic parking request. If it is confirmed to be an automatic parking request, the system invokes the intelligent parking robot to assist in completing the parking. The intelligent parking robot processes and analyzes the current parking space status, then determines available parking spaces and plans parking space locations, thereby generating a parking space location plan to ensure the vehicle can park in the most suitable location. Simultaneously, the intelligent parking robot also determines its own automatic position to ensure accurate movement and vehicle parking. Once the intelligent parking robot arrives at the designated location and engages with the target vehicle, it completes the entire smart parking management process based on the previously generated parking space positioning plan, enabling the vehicle to park safely and quickly. In summary, this process, facilitated by the intelligent parking robot, achieves smart and efficient parking management, significantly improving the convenience and efficiency of parking.

[0071] In summary, the embodiments of this application have at least the following technical effects:

[0072] This application's embodiments successfully construct an efficient and precise parking management solution by integrating advanced technologies such as digital twins and artificial intelligence. First, the system uses digital twin technology to construct a digital twin model of the parking lot, accurately replicating the real-world parking environment. This enables comprehensive perception and real-time monitoring of the parking environment, providing precise data support for smart parking management and significantly improving management efficiency. Second, the system fully utilizes artificial intelligence technology to achieve real-time monitoring and identification of parking space status. Through deep learning algorithms, the system can quickly and accurately determine parking space occupancy, providing managers with real-time and comprehensive parking information. Simultaneously, by combining received vehicle parking requests, the system can generate optimized route planning results based on real-time parking information and user needs, greatly improving parking convenience. These technological effects collectively achieve efficient and highly precise parking management, solving the problems of low parking management efficiency and low control precision.

[0073] Example 2

[0074] Based on the same inventive concept as the smart parking method based on digital twins and artificial intelligence in the foregoing embodiments, such as Figure 3 As shown, this application provides a smart parking system based on digital twins and artificial intelligence. The system and method embodiments in this application are based on the same inventive concept. The system includes:

[0075] Extraction Module 1: The extraction module 1 is used to extract the construction data of the target parking lot and establish a basic digital twin model based on the construction data;

[0076] Correction mapping module 2: The correction mapping module 2 is used to read the device coordinates of the image acquisition device and map the device coordinates to the basic digital twin model, and establish a correction mapping with the basic digital twin model based on the acquisition data of the image acquisition device;

[0077] Digital twin model generation module 3: The digital twin model generation module 3 is used to read the time-series acquisition information of the image acquisition device, perform parking feature and fixed feature recognition on the time-series acquisition information, and update the feature recognition results in the basic digital twin model based on the correction mapping results to generate a digital twin model;

[0078] First path planning module 4: The first path planning module 4 is used to identify the parking space status of the digital twin model and generate path planning results based on the received vehicle parking requests;

[0079] Parking Management Module 5: The parking management module 5 is used for intelligent parking management based on the path planning results.

[0080] Furthermore, the digital twin model generation module 3 is used to perform the following method:

[0081] Image contour features are extracted from the time-series acquired information, and feature analysis is performed based on the image contour feature extraction results to identify parking features and stationary features.

[0082] If the feature identifier is a fixed feature, then fixed occupancy identification is performed, and a fixed occupancy feature identification result is generated;

[0083] If the feature is identified as a parking feature, the multi-angle sensor is activated to perform multi-angle data acquisition, and a vehicle occupancy feature recognition result is established based on the multi-angle data acquisition results.

[0084] The feature recognition result is obtained by using the fixed occupancy feature recognition result and the vehicle occupancy feature recognition result.

[0085] Furthermore, the digital twin model generation module 3 is used to perform the following method:

[0086] The basic digital twin model is used to establish the relationship between basic parking spaces;

[0087] The digital twin model is used to identify the parking space status within the model based on a state recognition network, and an occupancy recognition result is generated.

[0088] Based on the occupancy identification results, the association between the basic parking spaces is invoked, and the occupancy impact value is generated;

[0089] Vehicle parking requests are matched based on occupancy impact values.

[0090] Furthermore, the digital twin model generation module 3 is used to perform the following method:

[0091] The vehicle parking request is parsed to generate a parsing result, which includes vehicle information, user information, and destination information.

[0092] The occupancy impact value is calculated by matching the vehicle information and the user information to obtain the matching calculation result;

[0093] The matching results are used to perform distance correlation calculations based on the destination information;

[0094] The matching calculation results are sorted and optimized based on the distance correlation calculation results;

[0095] Path planning results are generated based on the ranking and optimization results.

[0096] Furthermore, the digital twin model generation module 3 is used to perform the following method:

[0097] Configure the normalization coefficients for the distance and matching values;

[0098] The distance correlation calculation result and the matching calculation result are normalized by the normalization coefficient to complete the sorting optimization.

[0099] Furthermore, the digital twin model generation module 3 is used to perform the following method:

[0100] Obtain route congestion information for the target parking lot;

[0101] Real-time path planning is performed using the sorting optimization results and the path congestion information to obtain the path planning results.

[0102] Furthermore, the parking management module 5 is used to perform the following methods:

[0103] Determine whether the vehicle parking request is an automatic parking request;

[0104] When the vehicle parking request is an automatic parking request, the intelligent parking robot is invoked, and a parking space positioning plan is generated based on the parking space status, and the automatic position of the intelligent parking robot is determined simultaneously.

[0105] Once the intelligent parking robot aligns with the target vehicle at the designated automatic location, it completes intelligent parking management based on the parking space positioning plan.

[0106] Example 3

[0107] Smart parking devices based on digital twins and artificial intelligence, such as Figure 4 As shown, it includes a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are connected via the bus. The computer program, when executed by the processor, implements the steps of the intelligent parking system based on digital twins and artificial intelligence as described in any of the above method embodiments.

[0108] In addition, this application also provides an electronic device, including a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor. The transceiver, the memory, and the processor are respectively connected via the bus. When the computer program is executed by the processor, it implements the various processes of the above-described method embodiment for controlling output data and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0109] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, specific embodiments have been described above. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims can be performed in a different order than that shown in the embodiments and still achieve the desired results. Additionally, the processes depicted in the drawings do not necessarily require a specific and sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0110] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0111] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and modifications fall within the scope of this application and its equivalents, this application intends to include such modifications and modifications.

Claims

1. A smart parking system based on digital twins and artificial intelligence, characterized in that: The system includes: Extraction module: Extracts the construction data of the target parking lot and establishes a basic digital twin model based on the construction data; Correction mapping module: reads the device coordinates of the image acquisition device and maps the device coordinates to the basic digital twin model, and establishes a correction mapping between the image acquisition device and the basic digital twin model based on the acquisition data of the image acquisition device; Digital twin model generation module: reads the time-series acquisition information of the image acquisition device, performs parking feature and stationary feature recognition on the time-series acquisition information, and updates the feature recognition results in the basic digital twin model based on the correction mapping results to generate a digital twin model; First path planning module: Recognizes parking space status of the digital twin model and generates path planning results based on the received vehicle parking requests; Parking management module: Performs intelligent parking management based on the route planning results; The system also includes: Feature extraction module: extracts image contour features from the time-series acquired information, performs feature analysis based on the image contour feature extraction results, and performs feature identification of parking features and stationary features; Fixed Feature Recognition Module: If the feature identifier is a fixed feature, then perform fixed occupancy recognition and generate a fixed occupancy feature recognition result; Parking feature recognition module: If the feature is identified as a parking feature, the multi-angle sensor is activated to perform multi-angle data acquisition, and a vehicle occupancy feature recognition result is established based on the multi-angle data acquisition results; Feature recognition result acquisition module: Obtains the feature recognition result through the fixed occupancy feature recognition result and the vehicle occupancy feature recognition result.

2. The system as described in claim 1, characterized in that, The process of identifying parking space status using the digital twin model further includes: Basic parking space association module: Establishes basic parking space associations through the aforementioned basic digital twin model; Occupancy identification result module: Based on the state recognition network, the module identifies the parking space status within the digital twin model and generates occupancy identification results; Occupation Impact Value Module: Based on the occupancy identification results, it calls the association between the basic parking spaces and generates the occupancy impact value; Request matching module: Matches vehicle parking requests based on occupancy impact values.

3. The system as described in claim 2, characterized in that, The system also includes: Parking request parsing module: Parses the vehicle parking request and generates parsing results, which include vehicle information, user information, and destination information; Matching calculation module: Performs matching calculation of occupancy impact value based on the vehicle information and the user information to obtain matching calculation results; Distance association calculation module: performs distance association calculation on the matching calculation results based on destination information; The sorting and optimization module sorts and optimizes the matching calculation results based on the distance association calculation results. The second path planning module generates path planning results based on the ranking and optimization results.

4. The system as described in claim 3, characterized in that, The system also includes: Coefficient configuration module: Configures the normalization coefficients for distance and matching values; Normalization module: Normalizes the distance association calculation result and the matching calculation result using the normalization coefficient to complete the sorting optimization.

5. The system as described in claim 3, characterized in that, The system also includes: Congestion information acquisition module; acquires congestion information along the route to the target parking lot; The third path planning module performs real-time path planning based on the sorting optimization results and the path congestion information to obtain the path planning results.

6. The system as described in claim 1, characterized in that, The system also includes: Request determination module: Determines whether the vehicle parking request is an automatic parking request; Location planning module: When the vehicle parking request is an automatic parking request, the intelligent parking robot is invoked, and a parking space location plan is generated based on the parking space status, and the automatic position of the intelligent parking robot is determined simultaneously. Smart parking management module: After the intelligent parking robot cooperates with the target vehicle at the automatic location, it completes smart parking management according to the parking space positioning plan.

7. A smart parking device based on digital twins and artificial intelligence, comprising a bus, a transceiver, a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the transceiver, the memory, and the processor are connected via the bus, characterized in that, When the computer program is executed by the processor, it implements the steps of the smart parking system based on digital twins and artificial intelligence as described in any one of claims 1-6.

Images