A parking garage adaptive linkage system with anti-following error breaking function and control method

Abstract

The application relates to a parking garage adaptive linkage system with a car-following mistaken breaking prevention function and a control method. The method comprises the following steps: establishing a vehicle multi-source identity recognition model, acquiring vehicle license plates, appearances, RFID and ETC information, and generating a unique vehicle identity; establishing a vehicle continuous trajectory tracking model, acquiring vehicle position, speed, direction and vehicle distance and other running parameters; constructing a car-following risk assessment model, calculating the car-following risk level according to the vehicle identity consistency, vehicle distance, passing time interval and trajectory similarity; establishing a parking scene adaptive analysis model, determining the threshold value of dynamic risk according to the traffic flow and parking space state, using edge computing nodes to complete data fusion, risk prediction and control decision, and linkage controlling the barrier gate, traffic signal lamp and information display device, uploading the evaluation result to the parking management platform, and realizing event management, statistical analysis and model optimization. The application has the effects of identifying the car-following mistaken breaking behavior and reducing the mistaken parking vehicle rate.

Description

Technical Field

[0001] This application relates to the field of smart parking and intelligent traffic control, and in particular to an adaptive linkage system and control method for parking garages with a function to prevent accidental entry by following vehicles. Background Technology

[0002] With the rapid development of smart city and intelligent transportation technologies, parking garage management is gradually moving towards unmanned operation, automatic payment, and intelligent control. Currently, urban public parking lots, commercial complex parking garages, residential parking lots, and transportation hub parking lots commonly use a combination of license plate recognition, barrier gate control, video surveillance, and parking management platforms to manage vehicle access. By using automatic license plate recognition technology to obtain vehicle identity information and control the automatic opening of barrier gates, vehicle throughput efficiency can be effectively improved and manual management costs reduced, thus becoming an important component of modern smart parking systems.

[0003] However, in actual parking garage operation, the phenomenon of vehicles mistakenly following another vehicle into the garage is quite common. This occurs when a following vehicle takes advantage of the gate being opened by a legitimate vehicle to enter or exit the garage, causing the system to only recognize the information of the preceding vehicle and fail to effectively verify the identity of the following vehicle. This type of mistaken following not only easily leads to lost parking fees but may also result in unauthorized vehicles entering the parking area, affecting parking lot operation and management and vehicle safety. Furthermore, in complex scenarios such as peak hours, low-light conditions at night, inclement weather, and densely packed queues, the accuracy of traditional parking management systems in recognizing following behavior further decreases, posing significant safety hazards.

[0004] Existing anti-following solutions for parking garages mostly use inductive loop detectors, infrared beam detectors, vehicle detectors, anti-collision radar, or single video recognition devices to detect vehicle passage. These systems can only determine the presence of a vehicle in the gate area, but cannot continuously verify the consistency of vehicle identification. When the gate opens, if a following vehicle passes close to the one in front, existing systems often cannot accurately distinguish between the two vehicles, easily leading to false passage. Furthermore, traditional systems mostly rely on fixed thresholds for vehicle detection, lacking comprehensive analysis of factors such as traffic flow, queue length, parking space occupancy, weather conditions, and time periods. They cannot dynamically adjust anti-following strategies based on the real-time operating status of the parking garage, ensuring both safety and vehicle throughput efficiency.

[0005] On the other hand, license plate recognition equipment, video surveillance equipment, radar detection equipment, and barrier gate control equipment in existing parking management systems typically operate independently. There is a lack of multi-source data fusion and collaborative control mechanisms between these devices, making it impossible to achieve continuous vehicle identity verification, real-time vehicle trajectory tracking, and prediction of following risks. Furthermore, the large amount of video data uploaded to the cloud for analysis is susceptible to network latency and bandwidth limitations, making it difficult to meet the requirements for real-time control of parking garage entrances and exits, resulting in slow system response speeds and affecting normal vehicle passage.

[0006] In response to the aforementioned technologies, the inventors believe that there is a need to provide an adaptive linkage system and control method for parking garages with anti-following and anti-mistaken entry functions. By integrating technologies such as multi-source vehicle identification, continuous trajectory tracking, anti-following risk assessment, scene adaptive analysis, and edge computing linkage control, continuous vehicle identification and real-time prediction of following risks can be achieved. The control strategy can be dynamically adjusted according to the parking garage's operating status, thereby improving the accuracy of vehicle identification and anti-following capabilities, reducing the false entry rate, and enhancing the intelligent, safe, and efficient management level of parking garage entrances and exits. Summary of the Invention

[0007] To address the technical problems in existing parking garage vehicle entrance and exit management, such as difficulty in real-time identification of vehicles following each other and insufficient continuous vehicle identity verification capabilities, low degree of fusion of multi-source sensing information, lack of adaptive adjustment of control strategies under dynamic changes in parking scenarios, and low level of intelligent parking management, this application provides an adaptive linkage system and control method for parking garages with anti-following-and-entry function.

[0008] The technical solution provided in this application for an adaptive linkage system and control method for parking garages with anti-accidental entry function is as follows: Firstly, an adaptive linkage control method for parking garages with anti-accidental entry prevention function includes the following steps: S1. Establish a multi-source vehicle identification model for parking garages. Obtain vehicle license plate information, vehicle appearance information, vehicle radio frequency identification information, and electronic non-stop toll collection identification information of vehicles at the entrance and exit of the parking garage, and establish a vehicle identity feature model; The vehicle identity features include license plate number, vehicle type, vehicle color, vehicle appearance features, RFID identity information, and ETC identity information; The vehicle identity is fused through a multi-source identity recognition module to generate a unique vehicle identity identifier. S2. Establish a vehicle continuous trajectory tracking model The vehicle trajectory tracking module is used to continuously acquire the vehicle's running trajectory in the parking garage entrance and exit area; The vehicle trajectory tracking module includes a video tracking unit, millimeter-wave radar, lidar, trajectory prediction unit, and vehicle re-identification unit; Acquire the vehicle's position, speed, direction of travel, acceleration, and distance between vehicles, and establish a continuous trajectory model for the vehicle. S3. Establish a vehicle anti-following risk assessment model Vehicle identity consistency is calculated based on vehicle identity characteristics and continuous vehicle trajectory. The vehicle following risk value is calculated by combining vehicle spacing, time interval, speed changes, and trajectory similarity. The anti-following recognition module includes a vehicle spacing calculation unit, a time interval calculation unit, a speed analysis unit, a trajectory similarity analysis unit, and a risk calculation unit; The risk calculation unit outputs the vehicle following risk level. S4. Establish an adaptive analysis model for parking scenarios. Obtain real-time operating status parameters of the parking garage; The operational status parameters include vehicle flow, queue length, number of remaining parking spaces, weather information, and time period information. The complexity of the parking scenario is calculated using the scenario adaptive analysis module; The threshold for determining vehicle following risk and the linkage control parameters are dynamically adjusted based on the complexity of the parking scenario.

[0009] S5. Perform real-time risk analysis based on edge computing nodes; Send vehicle identity data, vehicle trajectory data, and parking scenario operation data to edge computing nodes; The edge computing nodes perform data fusion processing, identity consistency verification, vehicle trajectory prediction, anti-following risk calculation, and control decisions. When the risk value of following another vehicle exceeds a preset threshold, a vehicle abnormality event is generated. S6, Execute adaptive linkage control for parking garage. Based on abnormal vehicle events, the parking garage linkage control module executes corresponding control strategies. The control strategy includes: Increase vehicle identity verification level, keep the barrier gate closed, switch traffic lights to a no-passing state, activate voice alarm, activate LED information prompts, and upload abnormal alarm information; S7. Generate parking garage anti-following management results; Parking management logs are generated based on vehicle identification results, risk assessment results, and linkage control results; The data is then uploaded to the parking management platform for event storage, statistical analysis, remote monitoring, and model optimization.

[0010] By adopting the above technical solution, the functions of multi-source identity fusion recognition at vehicle entrances and exits of parking garages, continuous trajectory tracking of vehicles in the whole process, and intelligent judgment of the illegal intrusion of closely-following vehicles are realized. By fusing multi-dimensional identity information such as license plates, vehicle appearances, RFID and ETC, the accuracy of vehicle identity confirmation is improved; through the real-time vehicle trajectory collection and risk assessment model, dynamic analysis of the distance between front and rear vehicles, time interval and motion state is realized, and abnormal closely-following behavior can be accurately identified; through adaptive analysis of parking scenarios and edge computing, dynamic adjustment of risk thresholds, real-time decision-making and rapid linkage control are realized; through the cooperative work of the barrier gate, alarm and information prompt equipment, timely blocking and early warning of abnormal vehicles are realized, thereby reducing the risk of unauthorized vehicles intruding by mistake, and improving the safety, reliability and intelligence level of parking garage management.

[0011] Optionally, the multi-source identity recognition module comprises a license plate recognition camera, a vehicle appearance recognition camera, an RFID recognizer, an ETC recognizer and a vehicle identity fusion unit; The vehicle identity fusion unit establishes a unique vehicle identity feature model according to the vehicle license plate number, vehicle color, vehicle type, vehicle appearance features, RFID identity information and ETC identity information.

[0012] By adopting the above technical solution, the functions of multi-dimensional vehicle identity information collection and fusion recognition in the parking garage are realized. The basic vehicle identity information is obtained through the license plate recognition camera, the vehicle color, vehicle model and appearance features are collected through the vehicle appearance recognition camera, the electronic vehicle identity information is obtained through the RFID recognizer and the ETC recognizer, and the vehicle identity fusion unit is used to perform association matching on data from different sources to establish a unique vehicle identity feature model. The technical solution can avoid the problem of recognition failure caused by single license plate recognition being affected by occlusion, contamination or environment, improve the accuracy and reliability of vehicle identity recognition, and provide an accurate data basis for subsequent continuous vehicle tracking, risk judgment of closely-following prevention and linkage control of the parking garage.

[0013] Optionally, the vehicle trajectory tracking module adopts video recognition, millimeter-wave radar and lidar for multi-source fusion positioning; a continuous vehicle trajectory is established according to the continuously collected vehicle position, speed, direction and acceleration; and continuous consistency check of vehicle identity is realized through a vehicle re-identification unit.

[0014] By adopting the above technical solutions, the real-time perception and continuous tracking of vehicle operation status in parking garages are achieved. By integrating detection data from video recognition, millimeter-wave radar, and lidar, dynamic parameters such as vehicle spatial position, speed, direction of travel, and acceleration can be acquired simultaneously, improving the accuracy and stability of vehicle positioning in complex environments. By establishing a continuous vehicle trajectory model, the movement status of vehicles from entering the detection area to passing through the barrier gate is continuously monitored. The vehicle re-identification unit re-matches vehicle appearance features and identity information to achieve continuous consistency verification of vehicle identity, avoiding identity confusion caused by vehicle obstruction, parallel passage, or following other vehicles. This provides reliable data support for subsequent risk identification and linkage control of following vehicles.

[0015] Optionally, the risk calculation unit calculates the vehicle following risk value based on vehicle spacing, vehicle passage time interval, vehicle speed difference, and trajectory overlap rate; The expression for calculating the risk value is: R = αD + βT + γΔV + δS Where R is the vehicle following risk value; D is the vehicle spacing; T is the vehicle passage time interval; ΔV is the vehicle speed difference; S is the vehicle trajectory overlap rate; and α, β, γ, and δ are the weighting coefficients.

[0016] By adopting the above technical solution, the functions of quantitative analysis of vehicle following behavior and intelligent judgment of risk level are realized. By establishing a risk calculation model by comprehensively considering parameters such as vehicle spacing, passage time interval, speed difference and trajectory overlap rate, the correlation between vehicles in front and behind can be accurately assessed, reducing misjudgment caused by a single detection parameter and improving the accuracy and real-time performance of vehicle following accident identification.

[0017] Optionally, the edge computing node includes a data access unit, a video analysis unit, an identity fusion computing unit, a risk prediction unit, and a control decision unit; The data access unit is used to receive data uploaded by various detection devices in the parking garage; The video analysis unit is used to perform vehicle target detection and vehicle feature extraction; The identity fusion computing unit is used for vehicle identity consistency analysis; The risk prediction unit is used to calculate the vehicle following risk value; The control decision unit is used to output linkage control commands.

[0018] By adopting the above technical solution, the parking garage achieves real-time vehicle risk analysis and edge-based intelligent decision-making functions. The edge computing node enables rapid access, fusion processing, feature analysis, and risk prediction of multi-source detection data, reducing data transmission latency and improving the speed of abnormal following behavior identification. The control decision unit outputs linkage control commands in real time, enabling rapid response of the barrier gate, alarm, and prompting equipment.

[0019] Optionally, the linkage control module includes a barrier gate control unit, a traffic signal light control unit, a voice alarm unit, an LED display control unit, an alarm sending unit, and a backup control unit; Depending on the risk level of the vehicle following, the control measures are implemented as follows: normal passage, deceleration reminder, prohibition of passage, alarm, and manual confirmation.

[0020] By adopting the above technical solution, the parking garage achieves graded response and intelligent linkage control functions for abnormal vehicles. Based on the risk level of vehicle following, the system can automatically match control strategies such as normal passage, warning, prohibition of passage, and manual confirmation, and link with the barrier gate, traffic lights, voice alarms, and display equipment to promptly block and warn against unauthorized following behavior, thereby improving the security management capabilities of the parking garage entrance and exit.

[0021] Optionally, the parking management platform is used to receive vehicle identity data, vehicle trajectory data, vehicle following risk data, and linkage control data in the parking garage; Establish a database of historical events; The model is trained on historical vehicle traffic data using machine learning to predict future trends in vehicle following risk in parking garages, thereby optimizing the vehicle following risk assessment model.

[0022] By adopting the above technical solutions, the functions of centralized management of parking garage operation data and continuous optimization of risk models are realized. By establishing a historical event database, vehicle identity, trajectory, risk and control data are stored and analyzed. Machine learning models are used to mine historical traffic patterns, realize the prediction of following risk trends and optimization of model parameters, and improve the accuracy of risk identification, adaptability and intelligence level of the system in the long-term operation process.

[0023] Secondly, an adaptive linkage system for parking garages with anti-accidental entry function includes: a multi-source identity recognition module, used to obtain vehicle identity information and establish a unique vehicle identity identifier; The vehicle trajectory tracking module is used to continuously acquire vehicle movement trajectories and perform continuous identity verification. The anti-following recognition module is used to calculate the risk value of following a vehicle and output the risk level; The scene adaptive analysis module is used to obtain the parking lot's operating status and dynamically adjust the risk assessment parameters; Edge computing nodes are used to perform data fusion, risk calculation, and control decisions; The linkage control module is used to control the barrier gate, traffic lights, voice alarm and LED display according to the risk level; The parking management platform is used to realize event management, data storage, statistical analysis, visual monitoring, and model training. The modules are interconnected to enable adaptive linkage control to prevent vehicles from accidentally entering the parking garage.

[0024] By adopting the above technical solutions, the modular collaborative control function of the parking garage vehicle anti-following and accidental entry system is realized. Through the collaborative work of the multi-source identity recognition module, vehicle trajectory tracking module, and anti-following recognition module, accurate vehicle identification, full-process trajectory monitoring, and real-time assessment of following risks are achieved. The scenario adaptive analysis module dynamically adjusts risk judgment parameters according to the parking lot's operating status, improving the system's adaptability to different traffic environments. Edge computing nodes enable rapid data fusion processing and real-time control decision-making, reducing response latency. Through the linkage control module and collaboration with the parking management platform, abnormal vehicles are promptly blocked, risk event alarms are triggered, operational data is stored, and models are continuously optimized, thereby improving the safety, reliability, and intelligence level of parking garage entrance and exit management.

[0025] Thirdly, an electronic device includes a processor, a memory, and a computer program stored in the memory; When the computer program is executed by the processor, the processor executes the parking garage adaptive linkage control method with anti-following-vehicle-entry function as described in any one of claims 1 to 8.

[0026] By adopting the above technical solution, the software-based deployment and automatic execution of the parking garage anti-following and accidental entry control method are realized. Through the processor calling the computer program in the memory, steps such as vehicle identification, trajectory tracking, risk assessment, scene analysis, and linkage control can be automatically executed, achieving intelligent detection and rapid response to abnormal vehicle passage behavior in the parking garage. Simultaneously, this electronic device features high data processing efficiency, fast control response speed, and strong operational stability, making it easily applicable to different types of parking garage management scenarios and improving the level of intelligence in parking garage entrance and exit management.

[0027] In summary, this application includes at least one of the following beneficial technical effects: This application establishes a multi-source vehicle identification model that integrates license plate information, vehicle appearance information, RFID identification information, and ETC identification information to achieve multi-dimensional vehicle identification and unique identifier generation. Compared with the traditional single license plate identification method, it improves the accuracy and reliability of vehicle identification and reduces identification errors caused by vehicle obstruction, environmental changes, or missing information. This application establishes a vehicle continuous trajectory tracking model and combines multi-source detection methods such as video recognition, millimeter-wave radar and lidar to obtain dynamic parameters such as vehicle position, speed, direction, acceleration and vehicle spacing in real time. It also achieves continuous identity verification through vehicle re-identification, which can effectively avoid problems such as vehicle following, parallel driving and identity confusion. This application constructs a vehicle anti-following risk assessment model, calculates the following risk value based on parameters such as vehicle spacing, passage time interval, speed change and trajectory similarity, realizes real-time quantitative analysis and risk level judgment of following-by mistaken behavior, improves the accuracy of abnormal vehicle identification, and reduces the risk of unauthorized vehicles entering parking garages. This application sets up a scenario adaptive analysis module to calculate the complexity of parking scenarios based on operational status parameters such as vehicle flow, queue length, parking space occupancy rate, weather conditions, and time periods, and dynamically adjusts the risk judgment threshold, enabling the system to adapt to complex operating environments such as peak congestion, low light, and severe weather, thereby improving the adaptability of the anti-following control strategy. This application uses edge computing nodes to perform real-time fusion analysis of vehicle identity data, trajectory data, and risk data, thereby achieving rapid data processing and risk prediction, reducing data transmission latency, and improving the detection and control response speed of abnormal events at parking garage entrances and exits. This application uses a linkage control module to execute a graded control strategy based on the risk level of vehicle following, which can automatically control the barrier gate, traffic lights, voice alarm device and LED display device to realize the release of normal vehicles, the reminder of suspected following vehicles and the blocking of high-risk vehicles, thereby improving the operational safety of parking garage entrances and exits. This application establishes a historical event database through a parking management platform and uses machine learning models to train and analyze vehicle traffic data, thereby enabling the prediction of following risk trends and continuous optimization of risk models. This allows the system to have self-learning and self-adaptive capabilities during long-term operation, improving the level of intelligent parking management. Attached Figure Description

[0028] Figure 1 This application presents an embodiment of an adaptive linkage system for parking garages with a function to prevent accidental entry by following vehicles.

[0029] Figure 2 This application provides a schematic diagram of a parking garage vehicle detection and multi-source identity recognition structure.

[0030] Figure 3 This application provides a schematic diagram of a vehicle continuous trajectory tracking and anti-following identification process.

[0031] Figure 4 This application provides a schematic diagram of a risk assessment model for preventing following vehicles.

[0032] Figure 5 This application provides a schematic diagram of the structure of an adaptive analysis model for parking scenarios.

[0033] Figure 6 This application provides an embodiment of a parking garage linkage control flowchart.

[0034] Figure 7 This application provides a schematic diagram of an edge computing real-time processing and cloud platform collaborative structure.

[0035] Figure 8 This application provides a schematic diagram of the deployment of an adaptive linkage system for a parking garage with a function to prevent accidental entry by following vehicles.

[0036] Figure 9 This is a schematic diagram of the deployment of an adaptive linkage system at a parking garage with a function to prevent accidental entry by following vehicles, according to an embodiment of this application.

[0037] Figure 10 This is a schematic diagram of the deployment of an edge computing node in an adaptive linkage system for parking garages with anti-accidental entry function according to an embodiment of this application.

[0038] Figure 11 This is a schematic diagram of the deployment of the adaptive linkage control module of an adaptive linkage system for parking garages with anti-accidental entry function according to an embodiment of this application.

[0039] Figure 12 This is a schematic diagram of the deployment of a parking management platform for an adaptive linkage system for parking garages with anti-accidental entry function, according to an embodiment of this application.

[0040] Attached diagram labels: 1. Parking garage entrance / exit; 101. Entrance lane; 102. Exit lane; 103. Barrier gate; 104. Anti-smashing radar; 105. Traffic light; 106. Voice prompt device; 107. LED information display screen; 108. Vehicle detection area; 2. Multi-source identity recognition module; 201. License plate recognition camera; 202. Vehicle exterior recognition camera; 203. RFID reader; 204. ETC reader; 205. Vehicle identity fusion unit; 3. Vehicle trajectory tracking module; 301. Video tracking unit; 302. Millimeter-wave radar; 303. LiDAR; 304. Trajectory prediction unit; 305. Vehicle re-identification (Re-ID) unit; 4. Anti-following recognition module; 401. Vehicle distance calculation unit; 402. Time interval calculation unit; 403. Speed ​​analysis unit; 404. Trajectory similarity analysis unit; 405. Following risk calculation unit; 5. Scene Adaptive Analysis Module; 501. Traffic Flow Statistics Unit; 502. Queue Length Analysis Unit; 503. Parking Space Status Analysis Unit; 504. Weather Environment Analysis Unit; 505. Scene Complexity Calculation Unit; 506. Dynamic Threshold Adjustment Unit; 6. Edge Computing Node; 601. Data Access Unit; 602. Video Analysis Unit; 603. Identity Fusion Calculation Unit; 604. Risk Prediction Unit; 605. Control Decision Unit; 7. Adaptive Linkage Control Module; 701. Barrier Gate Control Unit; 702. Traffic Light Control Unit; 703. Voice Alarm Unit; 704. LED Display Control Unit; 705. Alarm Information Sending Unit; 706. Backup Control Unit; 8. Parking Management Platform; 801. Data Storage Unit; 802. Risk Analysis Unit; 803. Event Management Unit; 804. Visual Monitoring Unit; 805. Remote Operation and Maintenance Unit; 806. Model Training Unit. Detailed Implementation

[0041] The following is in conjunction with the appendix Figure 1-8 This application will be described in further detail.

[0042] This application discloses an adaptive linkage system and control method for parking garages with anti-accidental entry prevention function. For example... Figures 1 to 8As shown in the illustration, this application provides an adaptive linkage system for a parking garage with anti-following and anti-accidental entry functions. The system includes a multi-source identity recognition module 2, a vehicle trajectory tracking module 3, an anti-following recognition module 4, a scene adaptive analysis module 5, an edge computing node 6, an adaptive linkage control module 7, and a parking management platform 8, all located at the parking garage entrance / exit 1. The parking garage entrance / exit 1 includes an entrance lane 101, an exit lane 102, a barrier gate 103, an anti-collision radar 104, traffic lights 105, a voice prompt device 106, an LED information display screen 107, and a vehicle detection area 108. During vehicle entry or exit from the parking garage, the aforementioned equipment performs vehicle detection, identity recognition, risk analysis, and linkage control.

[0043] Example 1 Example 1 illustrates the process of multi-source vehicle identification and continuous verification, such as... Figure 2 As shown, when a vehicle enters the vehicle detection area 108 at the parking garage entrance / exit 1, the multi-source identification module 2 starts working. The multi-source identification module 2 includes a license plate recognition camera 201, a vehicle exterior recognition camera 202, an RFID reader 203, an ETC reader 204, and a vehicle identity fusion unit 205. The license plate recognition camera 201 is used to collect vehicle license plate information; the vehicle exterior recognition camera 202 is used to acquire appearance information such as vehicle color, model, body outline, and vehicle feature points; the RFID reader 203 is used to read vehicle electronic tag information; and the ETC reader 204 is used to acquire vehicle electronic toll collection identity information. The data collected by each identification device is sent to the vehicle identity fusion unit 205, and a unique vehicle identity is established through multi-source information fusion: ID=f(L,C,T,R,E). Wherein: L represents vehicle license plate information; C represents vehicle color information; T represents vehicle type information; R represents RFID identity information; E represents ETC identity information. Through the above fusion method, even if the vehicle license plate is damaged, obscured, or has some missing information, the vehicle can still be identified by combining the vehicle appearance and electronic identity information, thus improving the reliability of vehicle identification.

[0044] Example 2: Vehicle Continuous Trajectory Tracking Process Example 2 is the process of continuous vehicle trajectory tracking, such as Figure 3As shown, after the vehicle completes identity recognition, the vehicle trajectory tracking module 3 continuously monitors the vehicle's operating status before and after passing through the barrier gate 103. The vehicle trajectory tracking module 3 includes a video tracking unit 301, a millimeter-wave radar 302, a lidar 303, a trajectory prediction unit 304, and a vehicle re-identification unit 305. The video tracking unit 301 obtains the vehicle's real-time position through a target detection algorithm; the millimeter-wave radar 302 is used to detect the vehicle's distance, speed, and direction of movement; the lidar 303 is used to obtain information on changes in the vehicle's spatial position. The data collected by multiple detection devices are fused to obtain the vehicle's continuous motion trajectory, Trajectory={P1,P2,...,Pn}, where P represents the vehicle's position coordinates at different time points; the trajectory prediction unit 304 predicts the vehicle's next position based on its historical motion status; and the vehicle re-identification unit 305 re-matches the vehicle's appearance features to achieve continuous identity verification of the vehicle from entering the detection area to passing through the barrier gate.

[0045] Example 3 Example 3 describes the process for identifying the risk of accidentally entering a vehicle while following it, such as... Figure 4 As shown, the anti-following recognition module 4 analyzes following behavior based on vehicle identity information and trajectory information. The anti-following recognition module 4 includes a vehicle distance calculation unit 401, a time interval calculation unit 402, a speed analysis unit 403, a trajectory similarity analysis unit 404, and a risk calculation unit 405. When two vehicles are detected to be simultaneously located in the gate area, the vehicle distance calculation unit 401 calculates the distance D between the vehicles, the time interval calculation unit 402 calculates the time difference T between the two vehicles passing through the detection area, the speed analysis unit 403 calculates the speed difference ΔV between the two vehicles, and the trajectory similarity analysis unit 404 calculates the correlation degree S between the movement trajectories of the two vehicles. Risk calculation unit 405 establishes a following risk model based on the above parameters: R = αD + βT + γΔV + δS Where: R is the vehicle following risk value; α, β, γ, δ are the corresponding parameter weights. Based on the comparison between the risk value R and the preset threshold: when R is lower than the first threshold, it is judged as normal passage; when R is between the first threshold and the second threshold, it is judged as suspected following; when R is higher than the second threshold, it is judged as high-risk following behavior.

[0046] Example 4 Example 4 is the adaptive analysis process for parking scenarios, such as... Figure 5As shown, to avoid misjudgments caused by fixed thresholds, this application uses a scenario adaptive analysis module 5 to dynamically analyze the parking operation environment. The scenario adaptive analysis module 5 includes a traffic flow statistics unit 501, a queue length analysis unit 502, a parking space status analysis unit 503, a weather environment analysis unit 504, a scenario complexity calculation unit 505, and a dynamic threshold adjustment unit 506. Among them, the traffic flow statistics unit 501 is used to obtain the number of vehicles entering and leaving per unit time, the queue length analysis unit 502 is used to obtain the vehicle queue length, the parking space status analysis unit 503 is used to obtain the number of remaining parking spaces in the parking garage, the weather environment analysis unit 504 is used to obtain environmental factors such as rain, snow, and sunlight, and the scenario complexity calculation unit 505 calculates C=f(N,L,P,W) based on the above parameters, where N represents the vehicle flow, L represents the queue length, P represents the parking space utilization rate, and W represents environmental factors. Based on the calculated scenario complexity C, the dynamic threshold adjustment unit 506 adjusts the following risk judgment threshold, so that the system can improve traffic efficiency during peak hours and improve the safety level in complex environments.

[0047] Example 5 Example 5 illustrates the real-time analysis process of edge computing, such as... Figure 7 As shown, this application uses an edge computing node 6 for real-time data processing. The edge computing node 6 includes a data access unit 601, a video analysis unit 602, an identity fusion computing unit 603, a risk prediction unit 604, and a control decision unit 605. The data access unit 601 receives data from camera equipment, radar equipment, and identity recognition equipment. The video analysis unit 602 performs vehicle target detection, vehicle feature extraction, and trajectory analysis. The identity fusion computing unit 603 matches vehicle identity information. The risk prediction unit 604 calculates the following risk level based on vehicle status data. The control decision unit 605 generates control commands based on the risk level. Through edge computing, the vehicle risk analysis process is completed on-site in the parking garage, reducing network latency caused by data upload and improving control real-time performance.

[0048] Example 6 Example 6 illustrates an adaptive linkage control process, such as... Figure 6 As shown, when the control decision unit 605 determines that the vehicle is at risk of accidentally following another vehicle, it sends a control command to the adaptive linkage control module 7. The adaptive linkage control module 7 includes a barrier gate control unit 701, a traffic signal light control unit 702, a voice alarm unit 703, an LED display control unit 704, an alarm information sending unit 705, and a backup control unit 706. The specific control process is as follows: When normal vehicle passage is detected, the barrier gate control unit 701 controls the barrier gate 103 to open; when suspected following behavior is detected, the traffic signal light 105 is switched to a warning state, and the driver is reminded to keep a distance through the voice prompt 106; when high-risk following behavior is detected, the barrier gate control unit 701 remains closed, while the voice alarm unit 703 and the LED display control unit 704 are activated, and the abnormal event is sent to the parking management platform 8 through the alarm information sending unit 705.

[0049] Example 7 Example 7 describes the data management process of the parking management platform. The parking management platform 8 receives vehicle identity information, trajectory data, risk assessment results, and control records. The parking management platform 8 includes a data storage unit 801, a risk analysis unit 802, an event management unit 803, a visualization monitoring unit 804, a remote operation and maintenance unit 805, and a model training unit 806. Specifically: the data storage unit 801 stores vehicle passage records; the risk analysis unit 802 statistically analyzes following events; the event management unit 803 generates abnormal event records; the visualization monitoring unit 804 displays vehicle operating status; the remote operation and maintenance unit 805 performs equipment maintenance; and the model training unit 806 optimizes the risk prediction model based on historical vehicle passage data.

[0050] The working principle of an adaptive linkage system for parking garages with anti-following and unauthorized entry function in this application embodiment is as follows: After a vehicle enters the parking garage entrance / exit 1, the multi-source identity recognition module 2 collects the vehicle's identity information and establishes a unique vehicle identity identifier; subsequently, the vehicle trajectory tracking module 3 continuously collects the vehicle's movement status to achieve full-process vehicle tracking; the anti-following recognition module 4 calculates the following risk level based on the vehicle's identity consistency and movement parameters; the scene adaptive analysis module 5 dynamically adjusts the risk judgment strategy in conjunction with the real-time operating environment of the parking garage; the edge computing node 6 completes real-time risk analysis and outputs control decisions; the adaptive linkage control module 7 controls the barrier gate, alarm equipment, and display equipment according to the risk level; finally, the parking management platform 8 completes data storage, event analysis, and model optimization. Through the above structure, continuous vehicle identity confirmation and real-time judgment of following risk can be achieved during the process of a vehicle passing through the parking garage entrance / exit, effectively preventing unauthorized vehicles from following the legitimate vehicle after it opens the barrier gate, thus improving the security and intelligence level of parking garage vehicle management.

[0051] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. An adaptive linkage control method for parking garages with anti-accidental entry function, characterized in that, Includes the following steps: S1. Establish a multi-source vehicle identification model for parking garages. Obtain vehicle license plate information, vehicle appearance information, vehicle radio frequency identification information, and electronic non-stop toll collection identification information of vehicles at the entrance and exit of the parking garage, and establish a vehicle identity feature model; The vehicle identity features include license plate number, vehicle type, vehicle color, vehicle appearance features, RFID identity information, and ETC identity information; The vehicle identity is fused through the multi-source identity recognition module (2) to generate a unique vehicle identity identifier; S2. Establish a vehicle continuous trajectory tracking model The vehicle trajectory tracking module (3) is used to continuously acquire the vehicle's running trajectory in the parking garage entrance and exit area; The vehicle trajectory tracking module (3) includes a video tracking unit (301), a millimeter-wave radar (302), a lidar (303), a trajectory prediction unit (304), and a vehicle re-identification unit (305). Acquire the vehicle's position, speed, direction of travel, acceleration, and distance between vehicles, and establish a continuous trajectory model for the vehicle. S3. Establish a vehicle anti-following risk assessment model Vehicle identity consistency is calculated based on vehicle identity characteristics and continuous vehicle trajectory. The vehicle following risk value is calculated by combining vehicle spacing, time interval, speed changes, and trajectory similarity. The anti-following recognition module (4) includes a vehicle distance calculation unit (401), a time interval calculation unit (402), a speed analysis unit (403), a trajectory similarity analysis unit (404), and a risk calculation unit (405). The risk calculation unit outputs the vehicle following risk level. S4. Establish an adaptive analysis model for parking scenarios. Obtain real-time operating status parameters of the parking garage; The operational status parameters include vehicle flow, queue length, number of remaining parking spaces, weather information, and time period information. The complexity of the parking scenario is calculated using the scenario adaptive analysis module (5); The vehicle following risk assessment threshold and linkage control parameters are dynamically adjusted according to the complexity of the parking scenario. S5. Perform real-time risk analysis based on edge computing nodes; Send vehicle identity data, vehicle trajectory data and parking scenario operation data to the edge computing node (6); The edge computing nodes perform data fusion processing, identity consistency verification, vehicle trajectory prediction, anti-following risk calculation, and control decisions. When the risk value of following another vehicle exceeds a preset threshold, a vehicle abnormality event is generated. S6, Execute adaptive linkage control for parking garage. The parking garage linkage control module (7) executes the corresponding control strategy based on the vehicle abnormal event; The control strategies include: increasing the vehicle identity verification level, keeping the barrier gate closed, switching the traffic lights to a no-passing state, activating the voice alarm, activating the LED information prompt, and uploading abnormal alarm information. Parking management logs are generated based on vehicle identification results, risk assessment results, and linkage control results; And upload it to the parking management platform (8) to realize event storage, statistical analysis, remote monitoring and model optimization.

2. The adaptive linkage control method for parking garages with anti-accidental entry function according to claim 1, characterized in that: The multi-source identity recognition module (2) includes a license plate recognition camera (201), a vehicle appearance recognition camera (202), an RFID reader (203), an ETC reader (204), and a vehicle identity fusion unit (205). The vehicle identity fusion unit (205) establishes a unique vehicle identity feature model based on the vehicle license plate number, vehicle color, vehicle type, vehicle appearance features, RFID identity information and ETC identity information.

3. The adaptive linkage control method for parking garages with anti-accidental entry function according to claim 1, characterized in that: The vehicle trajectory tracking module (3) uses video recognition, millimeter-wave radar and lidar for multi-source fusion positioning; A continuous vehicle trajectory is established based on continuously collected vehicle position, speed, direction, and acceleration. The vehicle identity continuity and consistency verification is achieved through the vehicle re-identification unit (305).

4. The adaptive linkage control method for parking garages with anti-accidental entry function according to claim 1, characterized in that: The risk calculation unit (405) calculates the vehicle following risk value based on the vehicle spacing, vehicle passage time interval, vehicle speed difference and trajectory overlap rate. The expression for calculating the risk value is: R = αD + βT + γΔV + δS Where R is the vehicle following risk value; D is the vehicle spacing; T is the vehicle passage time interval; ΔV is the vehicle speed difference; S is the vehicle trajectory overlap rate; and α, β, γ, and δ are the weighting coefficients.

5. The adaptive linkage control method for parking garages with anti-accidental entry function according to claim 1, characterized in that: The scene adaptive analysis module (5) includes a traffic flow statistics unit (501), a queue length analysis unit (502), a parking space status analysis unit (503), a weather environment analysis unit (504), a scene complexity calculation unit (505), and a dynamic threshold adjustment unit (506). The complexity of the parking scenario is calculated based on traffic flow, queue length, parking space occupancy rate, weather conditions, and time period, and the threshold for judging the risk of following other vehicles is dynamically adjusted.

6. The adaptive linkage control method for parking garages with anti-following-vehicle-accidental-entry function according to claim 1, characterized in that: The edge computing node (6) includes a data access unit (601), a video analysis unit (602), an identity fusion computing unit (603), a risk prediction unit (604), and a control decision unit (605). The data access unit (601) is used to receive data uploaded by each detection device in the parking garage; The video analysis unit (602) is used to perform vehicle target detection and vehicle feature extraction; The identity fusion computing unit (603) is used for vehicle identity consistency analysis; The risk prediction unit (604) is used to calculate the vehicle following risk value; The control decision unit (605) is used to output linkage control commands.

7. The adaptive linkage control method for parking garages with anti-accidental entry function according to claim 1, characterized in that: The linkage control module (7) includes a barrier gate control unit (701), a traffic signal light control unit (702), a voice alarm unit (703), an LED display control unit (704), an alarm sending unit (705), and a backup control unit (706). Depending on the risk level of the vehicle following, the control measures are implemented as follows: normal passage, deceleration reminder, prohibition of passage, alarm, and manual confirmation.

8. The adaptive linkage control method for parking garages with anti-accidental entry function according to claim 1, characterized in that: The parking management platform (8) is used to receive vehicle identity data, vehicle trajectory data, vehicle following risk data and linkage control data in the parking garage; Establish a database of historical events; The model is trained on historical vehicle traffic data using machine learning to predict future trends in vehicle following risk in parking garages, thereby optimizing the vehicle following risk assessment model.

9. A parking garage adaptive linkage system with anti-accidental entry function, characterized in that, include: The multi-source identity recognition module (2) is used to obtain vehicle identity information and establish a unique vehicle identity identifier; The vehicle trajectory tracking module (3) is used to continuously acquire the vehicle's running trajectory and perform continuous identity verification; The anti-following identification module (4) is used to calculate the risk value of following a vehicle and output the risk level; The scene adaptive analysis module (5) is used to obtain the parking lot operation status and dynamically adjust the risk judgment parameters; Edge computing nodes (6) are used to complete data fusion, risk calculation and control decisions; The linkage control module (7) is used to control the barrier gate, traffic lights, voice alarm and LED display according to the risk level; The parking management platform (8) is used to realize event management, data storage, statistical analysis, visual monitoring and model training; The modules are interconnected to enable adaptive linkage control to prevent vehicles from accidentally entering the parking garage.

10. An electronic device, characterized in that, Includes a processor, a memory, and a computer program stored in the memory; When the computer program is executed by the processor, the processor executes the parking garage adaptive linkage control method with anti-following-vehicle-entry function as described in any one of claims 1 to 8.

Images