Parking space lock Internet of Things system for parking timing charging

Abstract

The invention relates to the technical field of Internet of Things, in particular to a parking space lock Internet of Things system for parking timing charging, which comprises an environment sensing module, a dynamic decision module and a parking lock linkage control module, a multi-dimensional environment feature vector is generated through cross-modal fusion, and the data credibility is verified through space-time relevance; the dynamic decision-making module predicts a basic rate floating interval by using a long-short-term memory network time sequence model based on credible data input, dynamically adjusts the model weight in combination with the accumulated occupation duration of the facility and generates an abnormal level mark; when the abnormal level exceeds a threshold value and the occupied time exceeds an allowable maximum value, the ground lock linkage control module triggers a dual-channel forced lifting ground lock; through multi-source data collaborative verification, dynamic rate optimization and a multi-level risk determination mechanism, accurate adaptation of parking rates, rapid response to abnormal occupation and efficient regulation and control of parking space resources are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to an Internet of Things system for parking locks and timed parking charges. Background Art

[0002] Current parking metering systems are mostly based on single sensors or manual inspections, resulting in insufficient data dimensionality and delayed dynamic response. Traditional ground lock systems typically rely solely on mechanical control of parking space occupancy status, lacking comprehensive awareness of regional pedestrian density, facility operating status, and emergencies. This results in rigid rate adjustments and untimely handling of abnormal occupancy. For example, during peak hours when parking space turnover is low, pricing strategies cannot be dynamically optimized, while vehicles parked beyond their designated time limit rely solely on fixed thresholds to trigger ground locks, which can easily lead to disputes and waste resources.

[0003] Furthermore, existing systems lack a mechanism to verify the credibility of sensor data. The temporal and spatial correlations between emergencies (such as temporary traffic control) and regular data are not effectively utilized, potentially leading to misjudgments. Achieving the coordinated optimization of multi-source heterogeneous data fusion, dynamic rate decision-making, and ground lock linkage has become a key challenge in improving parking management efficiency. Summary of the Invention

[0004] The purpose of the present invention is to provide an Internet of Things system for parking locks and metered parking charges to solve the problems raised in the above-mentioned background technology. Specific technologies include how to improve the accuracy of parking management decisions through multi-dimensional data fusion and credibility verification; how to dynamically adjust rates based on real-time environmental characteristics and accurately identify abnormal occupancy behavior; and how to achieve multi-level risk assessment and reliable triggering of forced operations in ground lock control.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a parking lock IoT system for time-based parking fee collection, comprising an environment perception module, a dynamic decision module, and a ground lock linkage control module, wherein: Dynamic data is collected through the distributed sensor network of the environmental perception module (regional millimeter-wave radar, facility-level sensors, and event-level visual units). The cross-modal fusion algorithm of the fusion unit is used to generate a multidimensional feature vector that includes space utilization, facility status, and event urgency. The verification unit further analyzes the spatiotemporal causal relationship between emergencies and regional pedestrian flows through density clustering, and dynamically evaluates data credibility (such as switching historical data to supplement low-credibility data) to ensure reliable decision-making input.

[0006] The dynamic decision-making module uses the time series prediction algorithm of the long short-term memory network to predict the floating range of the basic rate. It adjusts the model weights (such as increasing the weight of the spatial density dimension) through an adaptive weighted algorithm based on the cumulative occupancy time in the facility-level status data, and simultaneously generates abnormal level marks; the dual feedback mechanism realizes the gradient evaluation of rate ladder increase and abnormal occupancy, avoiding erroneous operations caused by single threshold triggering.

[0007] When the ground lock linkage control module detects that the abnormality level exceeds the threshold and the cumulative occupancy time in the facility-level status data exceeds the maximum allowable value, it triggers dual-channel redundant control through multi-level risk judgment, first sending a soft unlocking command, and if there is no response, it starts the hard wire to forcibly lift the ground lock. At the same time, a structured evidence package is generated and stored in the cloud evidence storage system; the hardware interlocking circuit ensures switching to the local emergency strategy under abnormal conditions, thereby improving the robustness of the system.

[0008] Compared with the prior art, the present invention has the following beneficial effects: Through the fusion of multi-source heterogeneous data and spatiotemporal correlation verification, the comprehensiveness and credibility of environmental perception data are significantly improved, providing reliable input for dynamic decision-making; based on the collaborative mechanism of time series prediction and adaptive weight adjustment of long-short-term memory networks, the accurate generation of rate floating ranges and graded marking of abnormal occupancy are achieved, optimizing the efficiency of parking resource allocation; multi-level risk assessment is combined with dual-channel redundant control to ensure the high reliability of forced operation of ground locks, while the structured evidence storage mechanism provides a traceable basis for dispute resolution; dynamic compensation, filtering and wave elimination technologies are used to reduce the impact of short-term interference, and local-cloud collaborative storage is used to enhance data security, thereby improving the overall intelligence level and emergency response capabilities of the parking management system. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a schematic diagram of the overall module of the present invention; Figure 2 Schematic diagram of the environment perception module unit of the present invention.

[0010] In the figure: 100, environmental perception module; 101, acquisition unit; 102, fusion unit; 103, verification unit; 200, dynamic decision module; 300, ground lock linkage control module. DETAILED DESCRIPTION

[0011] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0012] Next, see Figure 1 The present invention provides a technical solution: a parking space lock Internet of Things system for parking time charging, including an environment perception module 100, a dynamic decision module 200 and a ground lock linkage control module 300.

[0013] See also Figure 2 The acquisition unit 101 in the environmental perception module 100 is configured to collect regional dynamic data, facility-level status data and event-level emergency data in real time. A distributed sensor network architecture is adopted to deploy three types of data acquisition equipment: regional millimeter-wave radar, facility-level Internet of Things sensors (including pressure, displacement, and current detection) and event-level visual recognition units. The regional millimeter-wave radar collects macro indicators such as target movement trajectory and regional crowd density at a frequency of 30 times per second as regional dynamic data. The facility-level Internet of Things sensor obtains the operating status, electromechanical parameters and cumulative occupancy time of a single facility through industrial bus polling as facility-level status data. The event-level visual recognition unit is triggered in real time by the edge computing node to capture the occurrence time, location coordinates and image features of emergencies such as illegal parking and abnormal intrusion, as well as event-level emergency data. All data are processed by the time synchronization server to ensure that data from different sources are strictly aligned on the timeline.

[0014] The fusion unit 102 in the environmental perception module 100 adopts a cross-modal fusion algorithm based on evidence theory to map regional-level dynamic data, facility-level status data, and event-level emergency data into a unified high-dimensional vector space. The specific implementation is divided into three layers, in which the bottom layer uses a filtering algorithm to eliminate sensor noise, the middle layer constructs a spatiotemporal relationship map to establish associations between different data sources (for example, mapping event-level emergency data into a regional heat map), and the output layer generates a multi-dimensional environmental feature vector (for example, 128 dimensions) through a weighted mechanism, which includes multiple standardized indicators such as space utilization rate, facility operating status, and event urgency. Each dimension is standardized.

[0015] The verification unit 103 in the environmental perception module 100 verifies the credibility of the regional-level dynamic data based on the spatiotemporal correlation of the event-level emergency data, and constructs a verification model based on the spatiotemporal characteristics of the event-level emergency data. Specifically, the spatial distribution pattern of emergencies in the event-level emergency data is analyzed through a density clustering algorithm to identify abnormal data clusters; secondly, the causal relationship between the time interval between the occurrence of emergencies and the change in regional pedestrian density in the regional-level dynamic data is calculated, and a credibility warning is triggered when the correlation is lower than the set safety threshold; the verification process adopts a dynamic probabilistic network model to evaluate the credibility score of the regional data in real time within a set time window (for example, 5 minutes), and automatically switches to historical data for the same period as a supplement when the score does not meet the standard, to ensure that the data input into the decision module is true and reliable.

[0016] The dynamic decision-making module 200 receives a credibility-verified multi-dimensional environmental feature vector as input, and generates a basic rate floating range through a time series prediction model. Specifically, a time series prediction algorithm based on a long short-term memory network is used as the time series prediction model, and the credibility-verified multi-dimensional environmental feature vector is used as the model input; the model first performs a correlation analysis on historical rate data and the multi-dimensional environmental feature vector, screens out key dimensions that affect rate fluctuations (such as space utilization rate, event urgency, etc.), and then extracts time series features through a sliding time window mechanism; the model output layer uses quantile regression technology to generate a basic rate floating range (such as 80%-120% of the standard rate) in the future time (for example, within 15 minutes), and the prediction result is updated every 30 seconds; in order to improve the prediction accuracy, it will be dynamically compensated in combination with external variables such as real-time weather data and holiday factors, and Kalman filtering will be used to eliminate short-term fluctuations in the prediction results.

[0017] The dynamic decision-making module 200 dynamically adjusts the weight parameters of the time series prediction model based on the cumulative occupancy time in the facility-level status data and simultaneously generates an abnormal occupancy level mark. A dual feedback mechanism is established through the cumulative occupancy time parameters in the facility-level status data. On the one hand, an adaptive weighting algorithm is adopted. When the cumulative occupancy time of a single facility exceeds a preset time threshold (such as 30 minutes), the weight coefficient of the spatial density dimension in the time series prediction model is gradually increased, and can be adjusted to 1.5 times the initial value. On the other hand, an occupancy time gradient assessment model is constructed to divide continuous occupancy into three levels: normal (0-15 minutes), warning (15-30 minutes), and abnormal (more than 30 minutes), and a corresponding abnormal occupancy level mark is generated. The abnormal occupancy level mark triggers a three-level response strategy, including a step-by-step rate increase (a 10% rate increase for each level of abnormality increase) and the sending of a priority disposal instruction to the ground lock linkage module. All parameter adjustment processes are smoothed through the PID algorithm in cybernetics to avoid sudden decision-making.

[0018] The ground lock linkage control module 300 is communicatively connected to the dynamic decision module 200 and the collection unit 101 in the environmental perception module 100. When the abnormal occupancy level mark exceeds the risk threshold and the cumulative occupancy time in the facility-level status data exceeds the maximum allowable time corresponding to the basic rate floating range, the ground lock forced operation protocol is triggered, specifically including: A real-time data exchange channel is established via industrial Ethernet with the dynamic decision module 200 and the acquisition unit 101 in the environmental perception module 100, and a dual-channel redundant design is used to ensure communication reliability. A built-in multi-level risk determination logic (e.g., three levels) first monitors abnormal occupancy level marks in real time and initiates a primary warning when the mark exceeds a preset risk threshold (e.g., the abnormal level persists for more than 10 minutes). Secondly, the accumulated occupancy time in the facility-level status data is compared with the maximum allowed time in the basic rate floating range issued by the dynamic decision module 200 (e.g., the 90-minute upper limit corresponding to the basic rate floating range). When both conditions are met, the ground lock forced operation protocol is triggered. The forced ground lock operation protocol first controls the target ground lock into a physically locked state, achieving mechanical locking through a mechatronic drive. Simultaneously, a structured operation evidence package is generated, containing key data such as the base rate fluctuation range at the time of triggering (e.g., 150% of the standard rate), cumulative occupancy time in facility-level status data (accurate to seconds with timestamps), abnormal event records, and facility status snapshots. The evidence package uses digital signature technology to ensure it cannot be tampered with, and is synchronously stored in a local security chip and a cloud-based blockchain evidence storage system via a dual-channel transmission mechanism. The execution process of the forced operation protocol for ground locks is divided into three stages. First, a soft unlocking command is sent to the target ground lock via power carrier communication, accompanied by an audible and visual warning. If no feedback on the facility status change is received within 30 seconds, the hard-wired control loop is activated to forcibly lift the ground lock mechanical structure. Simultaneously, a structured operation evidence package is generated, containing key data such as the base rate fluctuation range at the time of triggering, the cumulative occupancy time in the facility-level status data, abnormal event records, and facility status snapshots. To ensure safety, a hardware interlock circuit is integrated, automatically switching to a local emergency strategy in the event of an abnormal facility current or network interruption, executing the preset protection action based on the last valid parameters received. The entire linkage process complies with the fail-safe principles of industrial control systems, ensuring that the ground lock remains locked in the event of system anomalies.

[0019] The parking meter and parking lock IoT system provided in this embodiment utilizes the environmental perception module 100 to achieve high-precision collection and credibility verification of multi-source heterogeneous data. Combined with the long-short-term memory network-based time series prediction model and dual feedback mechanism of the dynamic decision-making module 200, it dynamically generates rate fluctuation ranges and accurately identifies abnormal occupancy behavior. Furthermore, the ground lock linkage control module 300 employs multi-level risk assessment and dual-channel redundant operation to reliably trigger forced locking of the ground lock and generate tamper-proof evidence in the event of abnormal overtime occupancy. Its technical benefits include: improving environmental perception reliability through data fusion and spatiotemporal correlation analysis; achieving coordinated optimization of rates and parking occupancy based on dynamic model weight adjustment; and enhancing the timeliness and compliance of abnormal handling through redundant control and blockchain evidence storage. Ultimately, this system achieves efficient scheduling of parking resources, precise rate adaptation, and intelligent closed-loop management of illegal occupancy, significantly improving the responsiveness, resource utilization, and operational safety of urban parking systems.

[0020] The above shows and describes the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and improvements may be made to the present invention without departing from the spirit and scope of the present invention. Such changes and improvements fall within the scope of the present invention. The scope of protection claimed in the present invention is defined by the appended claims and their equivalents.

Claims

1. The parking lock IoT system for parking time and fee collection is characterized by: It comprises an environment perception module (100), a dynamic decision module (200) and a ground lock linkage control module (300), wherein: The environmental perception module (100) is configured to collect regional-level dynamic data, facility-level status data, and event-level emergency data in real time, generate a multi-dimensional environmental feature vector through heterogeneous data fusion, and verify the credibility of the regional-level dynamic data based on the spatiotemporal correlation of the event-level emergency data; The dynamic decision module (200) receives the multi-dimensional environmental feature vector that has been verified for credibility as input, generates a basic rate floating interval through a time series prediction model, and dynamically adjusts the weight parameters of the time series prediction model based on the accumulated occupancy time in the facility-level status data, and simultaneously generates an abnormal occupancy level mark; The ground lock linkage control module (300) is communicatively connected to the dynamic decision module (200) and the environment perception module (100), and triggers the ground lock forced operation protocol when the abnormal occupancy level mark exceeds the risk threshold and the accumulated occupancy time in the facility-level status data exceeds the maximum allowed time corresponding to the basic rate floating interval.

2. The parking lock Internet of Things system for parking time and fee collection according to claim 1 is characterized in that: The environmental perception module (100) includes a collection unit (101), wherein the collection unit (101) uses a distributed sensor network architecture to deploy a regional millimeter wave radar, a facility-level Internet of Things sensor, and an event-level visual recognition unit; wherein the regional millimeter wave radar collects target movement trajectory and regional crowd density data at a preset frequency as regional dynamic data; The facility-level IoT sensor obtains the operating status, electromechanical parameters and cumulative occupancy time of a single facility through industrial bus polling as facility-level status data; the event-level visual recognition unit is triggered in real time by the edge computing node to capture the occurrence time, location coordinates and image features of the emergency as event-level emergency data.

3. The parking lock Internet of Things system for parking time and fee collection according to claim 1 is characterized in that: The environmental perception module (100) includes a fusion unit (102), which adopts a cross-modal fusion algorithm based on evidence theory, eliminates sensor noise through a filtering algorithm, constructs a spatiotemporal relationship map to establish data association, and generates a multi-dimensional environmental feature vector containing space utilization rate, facility operation status and event urgency indicators through a weighting mechanism.

4. The parking lock Internet of Things system for parking time and fee collection according to claim 1 is characterized in that: The environmental perception module (100) includes a verification unit (103), which analyzes the spatial distribution pattern of emergencies through a density clustering algorithm, calculates the causal relationship between the time interval between emergencies and the change in regional pedestrian density in regional dynamic data, triggers a credibility warning when the correlation is lower than a set safety threshold, and uses a dynamic probability network model to evaluate the credibility score of regional data in real time within a set time window. When the credibility score does not meet the standard, it switches to historical data of the same period as a supplement.

5. The parking lock Internet of Things system for parking time and fee collection according to claim 1 is characterized in that: The dynamic decision module (200) adopts the time series prediction algorithm of the long short-term memory network as the time series prediction model, extracts time series features through a sliding time window mechanism, combines external variables for dynamic compensation, and uses quantile regression technology to generate the basic rate floating range for the future set time period. The prediction results are periodically updated and short-term fluctuations are eliminated through a filtering algorithm.

6. The parking lock Internet of Things system for parking time and fee collection according to claim 5 is characterized in that: The time series forecasting model screens key dimensions through correlation analysis and outputs a basic rate floating range including a standard rate floating range.

7. The parking lock Internet of Things system for parking time and fee collection according to claim 1 is characterized in that: The dynamic decision module (200) establishes a dual feedback mechanism. When the cumulative occupancy time of a single facility exceeds a preset time threshold, the weight coefficient of the spatial density dimension in the time series prediction model is gradually increased through an adaptive weighting algorithm. At the same time, an occupancy time gradient evaluation model is constructed to generate an abnormal occupancy level mark containing multiple levels of abnormality.

8. The parking lock Internet of Things system for parking time and fee collection according to claim 7 is characterized in that: The abnormal occupancy level mark triggers a rate step-up strategy. The rate increases according to a preset ratio every time the abnormal level increases, and a smooth transition of parameter adjustment is achieved through a control algorithm.

9. The parking lock Internet of Things system for parking time and fee collection according to claim 1 is characterized in that: The ground lock linkage control module (300) adopts a multi-level risk determination logic, and when the abnormal occupancy level mark continues to exceed the preset risk threshold, and the accumulated occupancy time in the facility-level status data exceeds the maximum allowed time in the basic rate floating range, the ground lock forced operation protocol is triggered through a dual-channel redundant design.

10. The parking lock Internet of Things system for parking time and fee collection according to claim 9 is characterized in that: The ground lock forced operation protocol includes sending a soft unlocking command, starting the hard-wired control loop to forcibly lift the ground lock mechanical structure after no response within the preset waiting time, and simultaneously generating a structured operation evidence package containing the basic rate floating range, cumulative occupancy time and facility status snapshot, using data verification technology to store it in local and cloud evidence systems, and switching to a local emergency strategy under abnormal conditions through a hardware interlocking circuit.

Images