Traffic signal light fault emergency management system and method based on internet of things large model

Abstract

The application provides a traffic signal lamp fault emergency management system and method based on an Internet of Things large model, and relates to the field of urban traffic signal lamp supervision. The method is executed by an emergency supervision management platform of the traffic signal lamp fault emergency management system based on the Internet of Things large model, and comprises the following steps: determining a state evaluation result of a signal lamp based on first monitoring data and second monitoring data of the signal lamp; determining a target signal lamp based on the state evaluation result; cutting off the power supply of a fault signal lamp and issuing a detour suggestion to target personnel; determining a fault prediction situation of an abnormal signal lamp based on the first monitoring data, the second monitoring data and the state evaluation result of the abnormal signal lamp; and determining a maintenance parameter based on the fault prediction situation and sending the maintenance parameter to maintenance personnel. The application can accurately diagnose the state of the signal lamp, cut off the power supply and issue a detour suggestion when a fault occurs, predict a fault and determine a maintenance parameter when an abnormality occurs, realize rapid response and active intervention, and improve operation and maintenance efficiency.

Description

Technical Field

[0001] This specification relates to the field of urban traffic signal monitoring, and in particular to a traffic signal fault emergency management system and method based on a large Internet of Things (IoT) model. Background Technology

[0002] In urban traffic, traffic lights are a crucial component of traffic control. A malfunction can cause traffic congestion, blockages, and even accidents. Traditional traffic light troubleshooting typically involves regular manual inspections or spot checks, which are insufficient for timely responses to sudden malfunctions.

[0003] Therefore, there is an urgent need to provide a traffic signal light fault emergency management system and method based on the Internet of Things (IoT) big data model, which can detect the operating status of traffic lights in urban traffic in real time, determine whether there is a signal fault, report relevant fault information to the traffic management department, and at the same time shut down potentially faulty traffic lights and intelligently adjust the associated traffic lights at surrounding intersections to prevent traffic congestion. Summary of the Invention

[0004] To address the challenges of real-time and accurate monitoring of traffic signal status in urban areas, timely detection and handling of sudden malfunctions, and intelligent coordination of surrounding traffic to avoid congestion and accidents, this specification provides a traffic signal malfunction emergency management system and method based on a large-scale Internet of Things (IoT) model.

[0005] The invention includes a traffic signal light malfunction emergency management system based on an Internet of Things (IoT) big data model. The system includes an emergency monitoring and management platform configured to execute a traffic signal light malfunction emergency management method based on an IoT big data model.

[0006] The invention includes a traffic signal light malfunction emergency management method based on an Internet of Things (IoT) big data model. The method is executed by an emergency monitoring and management platform of a traffic signal light malfunction emergency management system based on an IoT big data model. The method includes: determining a status assessment result for the traffic signal light based on first and second monitoring data, wherein the first and second monitoring data are collected through different sensing devices; identifying target traffic lights based on the status assessment result, including faulty traffic lights and abnormal traffic lights; cutting off the power supply to the faulty traffic light and issuing detour suggestions to the target personnel; determining a predicted fault condition for the abnormal traffic light based on the first and second monitoring data and the status assessment result; and determining maintenance parameters based on the predicted fault condition and sending them to maintenance personnel.

[0007] The beneficial effects of the above invention include, but are not limited to: (1) Through the hierarchical verification evaluation process, accurate and reliable traffic light status can be output; on the other hand, by using external and independent sensor data to cross-verify the internal data of the traffic light itself, the accuracy of fault judgment is greatly improved, and false alarms and missed alarms caused by damage to a single sensor or data drift can be effectively avoided, ensuring the reliability of the final status evaluation result. (2) The first and second abnormal thresholds are dynamically adjusted according to the traffic status of the intersection, making the traffic light fault emergency management system based on the Internet of Things large model more intelligent and able to more accurately identify potential faults at high-risk intersections. (3) By remotely fine-tuning parameters to fine-tune abnormal traffic lights, such as increasing the drive current and resetting the communication module, the working status can be temporarily reset or the effective working life can be extended to buy time for maintenance personnel. (4) Within the time window of equipment maintenance, by increasing the acquisition frequency, more transient abnormal data and performance fluctuation characteristics can be effectively captured, providing a more complete and timely fault diagnosis and analysis. The data support; at the same time, this adaptive adjustment mechanism avoids the waste of resources caused by continuous high-frequency collection, enabling the edge computing unit to maintain efficient operation while ensuring the monitoring effect, and achieving a balance between accurate monitoring and resource utilization; (5) By integrating the first and second monitoring data collected by different sensing devices, the accurate diagnosis and classification of the traffic light operation status are realized: for traffic lights that have already failed, the power supply can be cut off and detour suggestions can be issued immediately, effectively ensuring road safety; for abnormal traffic lights with potential risks, fault prediction is made based on multi-dimensional data, and accurate maintenance parameters are generated to guide maintenance work. This hierarchical handling mechanism not only ensures rapid response to serious faults, but also realizes early warning and proactive intervention for potential faults, thereby significantly improving the level of intelligence and emergency response efficiency of urban traffic management, and ultimately building a safe, efficient and reliable urban traffic light operation and maintenance system; (6) By introducing the first machine learning model, the traffic light fault emergency management system based on the Internet of Things big model can predict the specific problems that may occur in the traffic lights, fundamentally changing the traditional passive maintenance mode after discovering the fault, and transforming it into a preventive maintenance mode of proactive intervention after predicting potential faults, significantly improving operation and maintenance efficiency. Attached Figure Description

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0009] Figure 1 This is an exemplary schematic diagram of a traffic signal light malfunction emergency management system based on an Internet of Things (IoT) big data model, as shown in some embodiments of this specification. Figure 2This is an exemplary flowchart of a traffic signal light malfunction emergency management method based on an Internet of Things (IoT) big data model, as shown in some embodiments of this specification. Figure 3 This is an exemplary schematic diagram of a first machine learning model shown according to some embodiments of this specification; Figure 4 This is an exemplary schematic diagram illustrating the control of associated signals according to some embodiments of this specification. Detailed Implementation

[0010] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0011] Figure 1 This is an exemplary schematic diagram of a traffic signal malfunction emergency management system based on an IoT big data model, according to some embodiments of this specification. It should be noted that the following embodiments are for illustrative purposes only and do not constitute a limitation thereof.

[0012] In some embodiments, the large-scale IoT model refers to an IoT model architecture that enables the efficient operation of large amounts of data within the IoT system. Simultaneously, AI models can be applied to the IoT model architecture to enhance data sensing and processing capabilities.

[0013] In some embodiments, such as Figure 1 As shown, the traffic signal light malfunction emergency management system 100 based on the Internet of Things (IoT) big data model may include an emergency monitoring user platform 110, an emergency monitoring service platform 120, an emergency monitoring management platform 130, an emergency monitoring sensor network platform 140, and an emergency monitoring perception and control platform 150.

[0014] In some embodiments, one or more platforms in the traffic signal malfunction emergency management system 100 based on an IoT big data model can exchange information and / or data via a network. In some embodiments, the network can be any one or more of a wired network or a wireless network.

[0015] The Emergency Monitoring User Platform 110 refers to a platform where monitoring users supervise the operation of the entire Internet of Things (IoT) system. Monitoring users can be personnel from traffic safety management departments, etc. In some embodiments, the Emergency Monitoring User Platform may include at least one user interaction device, such as a mobile phone or computer. In some embodiments, monitoring users can use the Emergency Monitoring User Platform 110 to monitor the status assessment results of traffic lights in the urban traffic system in real time, achieving comprehensive status awareness of traffic lights throughout the urban traffic system.

[0016] Emergency monitoring service platform 120 refers to a platform used for receiving and transmitting data and / or information. In some embodiments, emergency monitoring service platform 120 may be configured as a communication network or gateway, etc. In some embodiments, emergency monitoring service platform 120 can perform bidirectional data interaction with emergency monitoring user platform 110 and emergency monitoring management platform 130. For example, emergency monitoring service platform 120 can receive fault prediction information of abnormal traffic lights from emergency monitoring management platform 130 and send the fault prediction information of abnormal traffic lights to emergency monitoring user platform 110 so that the monitoring user can take corresponding emergency measures.

[0017] The emergency monitoring and management platform 130 is a comprehensive management platform that manages and coordinates the connections and collaboration between multiple platforms. The emergency monitoring and management platform can be configured as a server or a processor. In some embodiments, the emergency monitoring and management platform 130 is communicatively connected to the emergency monitoring and sensing control platform 150 via the emergency monitoring sensor network platform 140.

[0018] In some embodiments, the emergency monitoring and management platform 130 is configured to implement a traffic signal malfunction emergency management method based on an IoT big data model.

[0019] The emergency monitoring sensor network platform 140 is used for comprehensive management of sensor information and serves as a communication transmission platform for bidirectional data interaction between the emergency monitoring management platform 130 and the emergency monitoring perception and control platform 150. In some embodiments, the emergency monitoring sensor network platform 140 can be configured as a communication network or gateway. The emergency monitoring sensor network platform 140 is responsible for uploading real-time sensor data collected by the emergency monitoring perception and control platform 150 to the emergency monitoring management platform 130, and for issuing control commands generated by the emergency monitoring management platform 130 to the corresponding emergency monitoring perception and control platform 150 for execution.

[0020] The emergency monitoring and control platform 150 is a platform for generating monitoring information and executing control information. In some embodiments, the emergency monitoring and control platform 150 may include various monitoring, sensing, and interactive devices deployed at intersections in urban traffic systems, such as a first sensor array, roadside radar, V2X communication module, smart camera, edge computing unit, AI agent, etc.

[0021] In some embodiments, the emergency monitoring and control platform 150 may be pre-authorized by the emergency monitoring management platform 130 to analyze and judge monitoring information through an AI agent and generate control instructions.

[0022] For more information about the above platforms, please refer to [link / reference]. Figures 2-4 And related explanations.

[0023] The traffic signal light malfunction emergency management system 100, based on the Internet of Things (IoT) big data model, enables communication between its various functional platforms, forming a closed loop of information operation. Under the unified management of the emergency supervision and management platform, it coordinates and operates regularly, realizing the informatization and intelligentization of traffic signal light malfunction supervision.

[0024] It should be noted that the above description of the traffic signal light fault emergency management system 100 based on the Internet of Things big model is for ease of description only and should not limit this specification to the scope of the embodiments described.

[0025] Figure 2 This is an exemplary flowchart illustrating an emergency management method for traffic signal malfunctions based on a large-scale Internet of Things (IoT) model, according to some embodiments of this specification. Figure 2 As shown, process 200 includes the following steps. In some embodiments, process 200 may be executed by an emergency monitoring and management platform.

[0026] Step 210: Based on the first monitoring data and the second monitoring data of the traffic lights, determine the status assessment result of the traffic lights.

[0027] Traffic lights are traffic control devices that use changes in specific colors of light or graphic symbols to convey instructions such as proceed, stop, and warning to road users. Examples include motor vehicle traffic lights and pedestrian traffic lights.

[0028] In some embodiments, the first and second monitoring data of the traffic light can be collected by different sensing devices.

[0029] The first monitoring data refers to the monitoring data collected by the first sensor array. The first sensor array refers to a combination of sensors deployed on a traffic light. In some embodiments, the first sensor array includes a brightness sensor, a color sensor, a current sensor, a voltage sensor, a temperature sensor, an attitude sensor, and an ambient light sensor, etc.

[0030] In some embodiments, the first monitoring data includes brightness data, chromaticity data, electrical data, temperature data, attitude data, and ambient light data. Brightness data refers to data collected by brightness sensors installed on red, yellow, and green light groups, reflecting the real-time brightness of each color of light. A light group refers to a light-emitting module composed of multiple bulbs. Chromaticity data refers to data collected by a chromaticity sensor, used to determine the spectral deviation of the light color. Electrical data refers to data collected by current and voltage sensors, used to monitor the power supply status of each light group. Temperature data refers to data collected by a temperature sensor, used to monitor the internal temperature of the light box and the operating temperature of each light group. Attitude data refers to data collected by an attitude sensor, used to monitor the tilt or vibration state of the traffic light pole. Ambient light data refers to data collected by an ambient light sensor, used to monitor the ambient light intensity at the intersection.

[0031] The second monitoring data includes monitoring data collected by a second sensor. The second sensor refers to a sensor installed in the intersection area where the traffic lights are located. In some embodiments, the second sensor includes a high-definition camera, millimeter-wave radar, lidar, etc.

[0032] In some embodiments, the second monitoring data includes visual data and traffic flow data. Visual data refers to the actual status information of the traffic lights collected by a high-definition camera, such as the color and flashing pattern of the traffic lights. Traffic flow data refers to traffic parameters such as real-time vehicle queue length, vehicle speed, and traffic volume at the intersection where the traffic lights are located. In some embodiments, traffic flow data can be acquired by millimeter-wave radar or lidar.

[0033] In some embodiments, the second monitoring data may further include communication data.

[0034] Communication data refers to the data obtained by the Vehicle to Everything (V2X) communication module at the intersection where the traffic lights are located. This includes the data generated when the V2X communication module communicates with passing vehicles at the corresponding intersection (such as buses, taxis, and other vehicles equipped with On-Board Diagnostics (OBD) data).

[0035] A condition assessment result refers to a conclusive evaluation or judgment obtained after analyzing the condition of a traffic light. In some embodiments, the condition assessment result includes whether there is a fault and the degree of abnormality. The degree of abnormality refers to the degree of deviation from the normal state.

[0036] In some embodiments, the status assessment result includes three scenarios: the traffic light is in a normal state, the traffic light is in a fault state, and the traffic light is in an abnormal state. A normal state means the traffic light is functioning properly and requires no intervention; a fault state means the traffic light is no longer working properly and requires immediate intervention and repair. Examples include the traffic light being completely off, an open circuit in the cable, or an inability to switch colors as instructed. An abnormal state refers to a situation where one or more of the traffic light's first or second monitoring parameters deviate from their historical normal baseline, but have not yet resulted in a complete loss of function.

[0037] In some embodiments, the emergency monitoring and management platform can compare the first monitoring data and the second monitoring data of the traffic light with the parameters of the corresponding historical first monitoring data and historical second monitoring data in the normal state, abnormal state and fault state of the traffic light in historical data, and determine the status assessment result of the traffic light.

[0038] In some embodiments, the emergency monitoring and management platform may also determine the initial assessment result of the traffic lights based on the first monitoring data; verify the initial assessment result based on the second monitoring data; and use the verified initial assessment result as the status assessment result.

[0039] The initial assessment result refers to the conclusion obtained by the emergency monitoring and management platform after conducting a preliminary analysis and judgment on the status of traffic lights based on the first monitoring data.

[0040] In some embodiments, the emergency monitoring and management platform can determine the initial assessment result based on the first monitoring data through various methods. For example, the emergency monitoring and management platform can define the safe range or abnormal triggering conditions of various data (brightness data, chromaticity data, electrical data, temperature data, attitude data, and ambient light data, etc.) in the first monitoring data based on historical data and relevant professional knowledge, store the safe range or abnormal triggering conditions in the ECU, and control the ECU to determine the initial assessment result according to the safe range or abnormal triggering conditions.

[0041] Abnormal triggering conditions may include: a sudden drop in current of a certain group of lights in the traffic lights exceeding a preset threshold or dropping to zero; or the brightness of the traffic lights remaining 30% or more below the normal value for a preset duration. The preset threshold and preset duration can be preset by technicians based on experience.

[0042] In some embodiments, the emergency monitoring and management platform can control the ECU to compare multiple data points in the first monitoring data of the traffic light to determine whether they are within safe ranges, thus determining the initial assessment result. For example, if any one of the multiple data points in the first monitoring data of the traffic light is outside the safe range, it can be preliminarily determined that the traffic light is in a faulty state or an abnormal state, and the corresponding degree of abnormality can be determined based on the specific data values ​​and safe ranges.

[0043] In some embodiments, when the first monitoring data of the traffic light triggers an abnormal triggering condition, it can be preliminarily determined that the traffic light is faulty or that the traffic light is abnormal.

[0044] For example, the emergency monitoring and management platform can use historical data to establish a historical brightness baseline, which is the average historical brightness of traffic lights under different ambient light intensities when they are in normal operation. The platform can control the ECU to compare the real-time brightness data of each light panel of the traffic light with the current real-time ambient light intensity and the historical brightness baseline of that traffic light under the corresponding ambient light intensity. If the real-time brightness data is consistently lower than the historical brightness baseline, the traffic light is marked as having abnormal brightness, and the degree of abnormality is determined based on the deviation between the real-time brightness data and the historical brightness baseline. The historical brightness baseline can be preset by technicians based on experience.

[0045] For example, the emergency monitoring and management platform can control the ECU to monitor the current and voltage data of each light group of the traffic lights in real time. If the current of a certain light group of the traffic lights suddenly drops to zero, it is initially determined to be an open circuit fault; if the current of a certain light group of the traffic lights continues to be higher than the safe range, it is determined that the traffic lights are in a faulty state (e.g., overload or short circuit). The historical brightness baseline can be preset by technicians based on experience.

[0046] For example, the emergency monitoring and management platform can control the ECU to monitor the temperature inside the signal light box and the LED module. If the temperature inside the light box and the LED module continues to be higher than the safety threshold, the signal light is marked as overheating abnormal, indicating a failure of the heat dissipation system or an abnormality of internal components.

[0047] For example, an emergency monitoring and management platform can control the ECU to monitor traffic light attitude data. If a traffic light is detected to be tilting or vibrating continuously beyond a preset attitude threshold, it is determined to be a structural abnormality of the traffic light, possibly caused by a collision or a loose pole. The preset attitude threshold can be set by technicians based on experience.

[0048] In some embodiments, the emergency monitoring and management platform can control the ECU to compare the first monitoring data of the same type of traffic lights (e.g., all red traffic lights) at the same intersection, in the same direction, or in adjacent directions. If the first monitoring data of any traffic light deviates from the average value of the same type of traffic lights by more than a preset proportion (e.g., 30%), it can be preliminarily determined that the traffic light is in a faulty state or in an abnormal state, and the corresponding degree of abnormality can be determined. The preset proportion can be preset by technicians based on experience.

[0049] In some embodiments, the emergency monitoring and management platform can compare and verify the initial assessment results generated by the ECU with the second monitoring data. For example, the emergency monitoring and management platform can perform visual verification and traffic flow verification of the initial assessment results and the second monitoring data.

[0050] Visual verification refers to the process of quantitatively and objectively verifying the initial evaluation results generated by the ECU using video or image information from the second monitoring data.

[0051] In some embodiments, the emergency monitoring and management platform can use computer vision algorithms based on the smart cameras at intersections to identify the actual light emission status (color, brightness) of traffic lights in real time, and then visually verify the initial assessment results and the second monitoring data.

[0052] For example, if the initial assessment results show that the brightness of the green light in a certain traffic light group is too low, and the visual algorithm analysis results show that the brightness of the green light in that group is too low, then the initial assessment results are verified. If the visual algorithm analysis results show that the brightness of the green light in that group is normal, then the initial assessment results are not verified, and the initial assessment results are marked as questionable.

[0053] For example, if the initial evaluation result shows that the current of a certain light group of traffic lights is 0 (i.e., the traffic light is not lit), and the visual algorithm analysis result shows that the light group is not lit, then the initial evaluation result is verified; if the visual algorithm analysis result shows that the light group has brightness, then the initial evaluation result is not verified, and the initial evaluation result is marked as questionable.

[0054] Traffic flow verification refers to the quantitative and objective verification process of the initial evaluation results generated by the ECU by using traffic flow data from the second monitoring data and through data comparison and logical analysis.

[0055] In some embodiments, the emergency monitoring and management platform can monitor traffic flow data using roadside radar, and then verify the traffic flow between the initial assessment results and the second monitoring data.

[0056] For example, if the initial assessment shows that the brightness of the green light in a certain traffic light group is abnormal, and the traffic flow data shows that the increase in queue length in the direction of the green light exceeds the length threshold and the decrease in traffic efficiency exceeds the efficiency threshold, then the initial assessment is verified. If the traffic flow data shows that there is no significant change in queue length and traffic efficiency in the direction of the green light (i.e., the increase in queue length does not exceed the length threshold and the decrease in traffic efficiency does not exceed the efficiency threshold), then the initial assessment is not verified and the initial assessment is marked as questionable.

[0057] In some embodiments, the emergency monitoring and management platform may determine the status assessment result based on the initial assessment result and the verification result of the second monitoring data.

[0058] In some embodiments, if the initial evaluation result is verified and the initial evaluation result shows that the signal light is in a faulty state (e.g., the current of a certain lamp panel of the signal light is 0), then the signal light is marked as a faulty signal light; if the initial evaluation result is verified and the initial evaluation result shows that the signal light is in an abnormal state (e.g., the brightness of a certain lamp panel of the signal light is too low), then the signal light is marked as an abnormal signal light, and the specific degree of abnormality is uploaded.

[0059] In some embodiments, if the initial assessment result fails verification, the initial assessment result is marked as questionable, the data collection frequency of the signal light is increased, and an alarm is sent to the emergency monitoring and management platform to request further in-depth diagnosis (e.g., human intervention or intervention of more complex machine learning models).

[0060] In some embodiments of this specification, on the one hand, the hierarchical verification evaluation process can output accurate and reliable traffic light status; on the other hand, by using external, independent sensor data to cross-verify the traffic light's internal data, the accuracy of fault diagnosis is greatly improved, and false alarms and missed alarms caused by damage to a single sensor or data drift can be effectively avoided, ensuring the reliability of the final status evaluation result.

[0061] Step 220: Determine the target traffic light based on the condition assessment results.

[0062] Target traffic lights refer to traffic lights that are identified as requiring attention during condition assessment. In some embodiments, target traffic lights include fault traffic lights and abnormal traffic lights.

[0063] A fault indicator light is a light that has malfunctioned. In some embodiments, a fault indicator light is a light whose condition assessment result indicates a fault state.

[0064] An abnormal indicator light is a light that displays abnormal data but for which a fault has not yet been confirmed. For example, parameters such as brightness, chromaticity, current, or temperature may exceed normal ranges, but the light itself has not lost its illumination function. Abnormal data refers to monitoring data that exceeds normal ranges. In some embodiments, an abnormal indicator light is one whose status assessment result indicates an abnormal state.

[0065] In some embodiments, the emergency monitoring and management platform can determine the target traffic lights based on the status assessment results, identifying those with fault or abnormal statuses.

[0066] In some embodiments, the emergency monitoring and management platform may identify abnormal signal lights through the following steps.

[0067] Step 1: Data preprocessing and data fusion.

[0068] Preprocessing refers to operations such as data cleaning, data formatting, and time-series alignment of the raw data collected by various sensors.

[0069] Data fusion refers to the process of collaboratively processing and integrating data from multiple sensors or information sources.

[0070] Step 2, Real-time Status Monitoring. Real-time status monitoring involves collecting and analyzing real-time data on traffic lights and their related parameters to promptly determine if any abnormalities have occurred.

[0071] In some embodiments, the emergency monitoring and management platform can compare the first monitoring data and the second monitoring data with preset threshold ranges to identify abnormal data. Abnormal data refers to data that does not meet the preset threshold ranges. The preset threshold ranges include the normal range values ​​corresponding to brightness data, chromaticity data, electrical data, temperature data, posture data, ambient light data, visual data, and traffic flow data, respectively.

[0072] In some embodiments, the emergency monitoring and management platform can determine a preset threshold range based on historical data and statistical analysis of normal traffic light operation using an edge computing unit (ECU). An edge computing unit is an embedded device deployed on the traffic light body that possesses data processing and communication capabilities. In some embodiments, traffic lights at multiple adjacent intersections can share a single ECU for control, or one ECU can be assigned to one traffic light.

[0073] Step 3: Local Fault Prediction. In some embodiments, when real-time status monitoring detects abnormal data, the emergency monitoring and management platform can determine that the signal light corresponding to the abnormal data is an abnormal signal light.

[0074] In some embodiments, for abnormal traffic lights, the emergency monitoring and management platform can determine the degree of abnormality of the abnormal traffic lights based on the difference between the abnormal data and the preset threshold range.

[0075] For example, an emergency monitoring and management platform can compare the actual brightness data of an abnormal traffic light with its corresponding preset threshold range of brightness data, and determine the degree of abnormality of the traffic light by the absolute value of the minimum difference between the actual brightness data and the minimum value at the end of the range. The method for determining the degree of abnormality of other data such as chromaticity data and electrical data is similar.

[0076] In some embodiments, the emergency monitoring and management platform can fuse the anomaly levels corresponding to various data points of the abnormal traffic lights to determine the final anomaly level of the abnormal traffic lights. For example, the emergency monitoring and management platform can first normalize the anomaly levels corresponding to various data points of the abnormal traffic lights, then perform weighted summation, and use the summed value as the final anomaly level of the abnormal traffic lights.

[0077] In some embodiments, the determination of abnormal traffic lights is related to the traffic flow and the number of historical accidents at the intersection where the abnormal traffic light is located.

[0078] Traffic flow at an intersection refers to the number of motor vehicles passing through per unit of time. In some embodiments, traffic flow at an intersection can be obtained using devices such as loop detectors, high-definition cameras, millimeter-wave radar, and lidar.

[0079] Historical accident count refers to the number of traffic accidents that occur at an intersection within a preset statistical period. In some embodiments, the historical accident count can be obtained from the traffic accident database or command and dispatch platform of the traffic management department.

[0080] In some embodiments, an abnormal traffic light is one that has not experienced a fault but whose abnormality level in the status assessment exceeds a second abnormality threshold but is lower than a first abnormality threshold. Traffic lights located at the same intersection have the same first and second abnormality thresholds. The first abnormality threshold is greater than the second abnormality threshold.

[0081] In some embodiments, the first anomaly threshold may be automatically generated after the second anomaly threshold is determined, based on the difference between a first anomaly threshold and a second anomaly threshold preset by a person skilled in the art based on experience.

[0082] In some embodiments, the emergency monitoring and management platform can derive the second abnormal threshold of the abnormal signal light according to the following formula (1): (1) In equation (1), Indicates the second abnormal threshold; This represents a baseline threshold, which in some embodiments is preset manually. and These represent the historical traffic flow and historical average vehicle speed of the intersection where the abnormal traffic light is located during a preset historical time period, respectively. and These represent the traffic flow and average speed at similar intersections where the abnormal traffic light is located; This indicates the number of historical accidents at the intersection where the abnormal traffic light is located. Similar intersections refer to intersections with similar road classifications, lane structures, etc., to the intersection where the abnormal traffic light is located.

[0083] In some embodiments of this specification, the first and second abnormal thresholds are dynamically adjusted according to the traffic conditions at the intersection, making the traffic signal fault emergency management system based on the Internet of Things big data model more intelligent and able to more accurately identify potential faults at high-risk intersections.

[0084] Step 230: Disconnect the power to the fault indicator light and issue a detour suggestion to the target personnel.

[0085] In some embodiments, after the emergency monitoring and management platform identifies a faulty traffic light, it sends a cut-off command to the ECU corresponding to that traffic light, cutting off the power supply to the faulty traffic light to eliminate traffic congestion and safety hazards caused by the extreme fault of the traffic light. An extreme fault refers to a situation such as the traffic light continuously displaying only one color.

[0086] In some embodiments, after performing a power-off operation, the ECU will send a power-off status message to the emergency monitoring and management platform via a communication network such as 5G, Narrowband Internet of Things (NB-IoT), or fiber optics. The emergency monitoring and management platform will then generate a maintenance work order and notify maintenance personnel via system messages, SMS, or email.

[0087] The target personnel refer to drivers of vehicles that may be heading to or approaching the intersection where the faulty traffic light is located. For example, drivers of vehicles located upstream of the intersection and drivers currently traveling through the intersection.

[0088] Detour suggestions are based on real-time traffic data and provide affected road users with alternative route guidance to avoid intersections with faulty traffic lights and optimize traffic flow in the area.

[0089] In some embodiments, the emergency monitoring and management platform can issue detour suggestions through multiple channels in parallel. These include, for example, navigation service platforms, in-vehicle terminals, traffic guidance screens, and mobile navigation apps.

[0090] Step 240: Based on the first monitoring data, the second monitoring data, and the status assessment results of the abnormal signal light, determine the fault prediction status of the abnormal signal light.

[0091] For more information on the first monitoring data, the second monitoring data, and the assessment results, please refer to step 210 and its related description.

[0092] Fault prediction refers to predicting the possible fault type and probability of an abnormal indicator light. In some embodiments, the fault prediction includes fault type and fault probability. Fault types include component aging or communication malfunctions, etc.

[0093] Component aging refers to the performance degradation of signal light components (such as bulbs and power modules) during long-term operation, but not to the point of affecting normal lighting or control functions. Component aging can include bulb aging, driver power supply performance deterioration, clock crystal drift, or seal aging.

[0094] In some embodiments, communication anomalies include signal loss, increased packet error rate, excessive delay, or periodic interruption.

[0095] In some embodiments, the emergency monitoring and management platform can determine the fault prediction status of abnormal traffic lights based on the first monitoring data, the second monitoring data, and the status assessment results of the abnormal traffic lights through a fault table.

[0096] In some embodiments, the emergency monitoring and management platform can construct a fault table based on historical first monitoring data, historical second monitoring data, and historical status assessment results manually confirmed during maintenance when multiple traffic lights exhibit abnormal data in historical data. The fault table includes the correspondence between different historical first monitoring data, historical second monitoring data, and historical status assessment results.

[0097] The emergency monitoring and management platform can determine the current fault prediction status of traffic lights by querying the fault table based on the first monitoring data, the second monitoring data, and the status assessment results. For example, if historical data shows that 95% of traffic lights experience a decrease in brightness to range C when the bulbs age, then range C is written into the fault table as the basis for judging "bulb aging," and the corresponding fault probability can be 95%.

[0098] In some embodiments, the emergency monitoring and management platform can integrate the acquired multiple faults and their corresponding fault probabilities, and select the fault type with the highest fault probability as the fault type in the fault prediction scenario, and its corresponding fault probability is the fault probability in the fault prediction scenario.

[0099] For more information on determining fault types and fault probabilities, see [link to relevant documentation]. Figure 3 And its related descriptions.

[0100] Step 250: Based on the fault prediction, determine the maintenance parameters and send them to the maintenance personnel.

[0101] Maintenance parameters refer to the various data used when performing signal light maintenance and repair work. In some embodiments, maintenance parameters include maintenance sequence, maintenance time, remote fine-tuning parameters, and frequency adjustment parameters.

[0102] The maintenance sequence refers to the planned order of execution for multiple traffic light maintenance tasks to optimize maintenance efficiency and ensure traffic operation. Maintenance time refers to the specific time window planned for performing maintenance on abnormal traffic lights. A time window is a period of time with a clearly defined start and end point. Remote fine-tuning parameters refer to the set of instructions and values ​​used to configure, calibrate, or control traffic lights or their control equipment via remote connection. For example, increasing the drive current of traffic light A to 250mA or activating the third phase controller. Frequency adjustment parameters refer to the parameters used to adjust the monitoring strategy to closely monitor the status of abnormal traffic lights. For example, increasing the data acquisition frequency of abnormal traffic lights from once every 10 minutes to once every 30 seconds.

[0103] In some embodiments, the emergency monitoring and management platform can determine maintenance parameters and send them to maintenance personnel based on the fault prediction of abnormal traffic lights through various methods. For example, the emergency monitoring platform can sort multiple abnormal traffic lights in descending order of their fault probability to form a maintenance priority queue, and calculate the maintenance order, maintenance time, and other maintenance parameters based on this priority queue using existing path planning algorithms.

[0104] Maintenance personnel refers to the technical personnel or teams responsible for the operation and maintenance of traffic lights.

[0105] In some embodiments, maintenance parameters can be sent to relevant maintenance personnel via work order systems, mobile application push notifications, or SMS messages, providing detailed maintenance instructions to guide them in completing on-site operations.

[0106] In some embodiments, the emergency monitoring and management platform is further configured to: in response to fault prediction conditions, including component aging or communication anomalies, determine remote fine-tuning parameters based on the fault prediction conditions; send remote commands to the abnormal indicator lights to adjust the drive current of the abnormal indicator lights based on the remote fine-tuning parameters, and / or reset the communication module of the abnormal indicator lights. For more information on component aging and communication anomalies, see step 240 and its related description.

[0107] In some embodiments, the emergency monitoring and management platform can determine remote fine-tuning parameters through various methods based on fault prediction. In some embodiments, the emergency monitoring and management platform can determine remote fine-tuning parameters based on a fault strategy mapping table, according to fault prediction.

[0108] The fault policy mapping table includes the correspondence between different fault types and different remote fine-tuning parameters. The emergency monitoring and management platform can determine the fault policy mapping table based on historical remote fine-tuning parameters issued for different fault types that extended the lifespan of traffic lights (e.g., traffic lights functioned normally within a preset continuous period and were not identified as faulty). The emergency monitoring and management platform can determine the remote fine-tuning parameters corresponding to the currently abnormal traffic light by querying the fault policy mapping table based on the fault type corresponding to the abnormal traffic light.

[0109] Remote commands refer to control commands sent by the emergency monitoring and management platform to the fault indicator light to achieve status adjustment or function execution. In some embodiments, the emergency monitoring and management platform sends a command packet containing remote fine-tuning parameters to the ECU of the fault indicator light via wired or wireless communication networks. The command packet adopts a predefined protocol format to ensure accurate parsing and safe execution of the command.

[0110] Drive current refers to the operating current output by the drive power supply and applied to the lamp assembly to ensure that the signal lights can be lit normally and operate stably.

[0111] In some embodiments, after receiving a remote command, the ECU of the fault indicator light adjusts the output power of the drive circuit connected to the LED beads through its digital or analog output interface, so that the drive current is stabilized at the target value specified by the remote fine-tuning parameters. For example, the current of the LED beads whose brightness has decreased due to aging is fine-tuned from the rated 200mA to 220mA to compensate for light decay and restore the rated brightness.

[0112] In some embodiments, the emergency monitoring and management platform can determine the adjusted drive current of the abnormal signal light according to the following formula (2): (2) In equation (2), This indicates the drive current after the abnormal indicator light has been adjusted; This indicates the drive current before the abnormal indicator light was adjusted; This indicates the degree of abnormality of the faulty indicator light. In some embodiments, after receiving a remote command, the ECU of the faulty indicator light performs a power-off restart, software reset, or factory network configuration restoration operation on the integrated or external communication module to clear its temporary software faults or state lock-up and re-establish a stable communication connection with the emergency monitoring and management platform.

[0113] In some embodiments of this specification, the emergency monitoring and management platform can fine-tune abnormal signal lights by remotely adjusting parameters, such as increasing the drive current or resetting the communication module, which can temporarily reset the working status or extend the effective working life to buy time for maintenance personnel.

[0114] In some embodiments, the maintenance parameters include frequency adjustment parameters. The emergency monitoring and management platform is further configured to: determine the frequency adjustment parameters based on the maintenance time; and send acquisition instructions to the edge computing unit based on the frequency adjustment parameters to adjust the data acquisition frequency of the abnormal signal lights.

[0115] In some embodiments, the emergency monitoring and management platform can determine the adjusted collection frequency parameters according to the following formula (3): (3) In equation (3), This indicates the sampling frequency parameters after the abnormal signal light has been adjusted; This indicates the sampling frequency parameters before the abnormal signal light was adjusted; This indicates the time interval between the start time of the maintenance period and the current time.

[0116] A data acquisition command is a structured command generated by an emergency monitoring and management platform that encapsulates frequency adjustment parameters. In some embodiments, the data acquisition command specifies the abnormal indicator light, the data type to be adjusted (such as current or voltage), and the new data acquisition and / or reporting frequency value.

[0117] For more information on edge computing units, see step 240 and its related description.

[0118] In some embodiments, after receiving a data acquisition command, the edge computing unit's embedded agent program parses the command content and dynamically adjusts the scheduling cycle of its data acquisition task accordingly. For example, the edge computing unit may reconfigure the timer according to a new frequency parameter (such as a 30-second interval) and read data from the specified sensor or device register at the frequency of the frequency adjustment parameter, thereby achieving refined and real-time monitoring of the abnormal signal light's operating status.

[0119] In some embodiments of this specification, the emergency monitoring and management platform can effectively capture more transient abnormal data and performance fluctuation characteristics by increasing the acquisition frequency within the equipment maintenance time window, providing more complete and timely data support for fault diagnosis and analysis. At the same time, this adaptive adjustment mechanism avoids the waste of resources caused by continuous high-frequency acquisition, enabling the edge computing unit to maintain efficient operation while ensuring monitoring effectiveness, thus achieving a balance between accurate monitoring and resource utilization.

[0120] In some embodiments of this specification, the emergency monitoring and management platform integrates first and second monitoring data collected by different sensing devices to achieve accurate diagnosis and classification of traffic light operating status: for traffic lights that have malfunctioned, it can immediately cut off power and issue detour suggestions, effectively ensuring road safety; for abnormal traffic lights with potential risks, it performs fault prediction based on multi-dimensional data and generates accurate maintenance parameters to guide maintenance work. This tiered handling mechanism not only ensures rapid response to serious faults but also achieves early warning and proactive intervention for potential faults, thereby significantly improving the intelligence level and emergency response efficiency of urban traffic management, and ultimately building a safe, efficient, and reliable urban traffic light operation and maintenance system.

[0121] Figure 3 This is an exemplary schematic diagram of a first machine learning model shown according to some embodiments of this specification.

[0122] In some embodiments, the emergency monitoring and management platform can also predict the failure probability 360 and failure type 350 of the abnormal signal light based on the first monitoring data 310, the second monitoring data 320, and the status assessment result 330 of the abnormal signal light through the first machine learning model 340.

[0123] For more information on abnormal indicator lights, first monitoring data, second monitoring data, condition assessment results, and fault types, please refer to [link / reference needed]. Figure 2 And its related descriptions.

[0124] In some embodiments, the first machine learning model selects only the fault type with the highest probability among the predicted fault types that may occur within a preset future time period as its output. Examples include LED module aging and controller communication malfunctions. The preset time period can be predetermined by technicians based on experience.

[0125] The failure probability refers to the probability predicted by the first machine learning model that a faulty traffic light may occur within a preset future time period. The failure probability can be expressed as a percentage. For example, the first machine learning model predicts that the probability of a controller communication failure occurring in a faulty traffic light within a preset future time period is 50%.

[0126] The first machine learning model refers to a model used to predict the failure probability and failure type of abnormal traffic lights. In some embodiments, the duration determination model is a machine learning model. For example, the first machine learning model may include one or more combinations of a Recurrent Neural Network (RNN) model, a Long Short-Term Memory Network (LSTM) model, or other custom models.

[0127] In some embodiments, the input to the first machine learning model may include first monitoring data, second monitoring data, and status assessment results of the abnormal signal light, and the output of the first machine learning model may include the failure probability and failure type of the abnormal signal light.

[0128] In some embodiments, the first machine learning model can be trained on an initial first machine learning model using first training samples with a first label. The first training samples may include first monitoring data corresponding to a sample traffic light at a first historical time point, second monitoring data, and a status evaluation result. The first label may include, under the first training samples, the sample fault type that occurs within a preset period after the first historical time point, and the sample fault probability corresponding to the sample fault type.

[0129] In some embodiments, the first training sample and the first label can be obtained based on historical data. For example, the emergency monitoring and management platform can select multiple sample traffic lights that have the same or similar first monitoring data, second monitoring data, and status assessment results at a first historical time point from historical data. The fault types that occur in these sample traffic lights within a preset period after the first historical time point are taken as sample fault types. The ratio of the number of the same fault type occurring in these sample traffic lights within the preset period after the first historical time point to the total number of sample traffic lights is taken as the sample fault probability of that fault type.

[0130] In some embodiments, the emergency monitoring and management platform can input multiple first training samples with first labels into an initial first machine learning model to obtain the sample fault types and corresponding sample fault probabilities output by the initial first machine learning model. A loss function is constructed using the labels and the output of the initial first machine learning model. Based on the loss function, the parameters of the initial first machine learning model are iteratively updated using gradient descent, simulated annealing, or other methods. When preset conditions are met, model training is complete, resulting in a trained first machine learning model. These preset conditions may include loss function convergence, the number of iterations reaching a threshold, etc.

[0131] In some embodiments of this specification, by introducing a first machine learning model, the traffic signal light fault emergency management system based on the Internet of Things big model can predict the specific problems that may occur with the traffic lights. This fundamentally changes the traditional passive maintenance mode after a fault is discovered, and transforms it into a preventive maintenance mode that actively intervenes after predicting potential faults, thus significantly improving operation and maintenance efficiency.

[0132] In some embodiments, the input to the first machine learning model may also include first monitoring data of other traffic lights (if any) at the intersection where the abnormal traffic light is located, and the training samples of the first machine learning model may also include sample first monitoring data of other traffic lights (if any) at the intersection where the sample traffic light is located at a first historical time point.

[0133] In some embodiments of this specification, cross-traffic light collaborative analysis at the same intersection can identify systemic or related problems that cannot be reflected by single-light data, and can more accurately determine whether the anomaly comes from the traffic light itself or from shared equipment at the intersection (such as power cabinets or controllers), thereby improving the accuracy of fault location and reducing the possibility of misdiagnosis.

[0134] The traffic lights adjacent to the abnormal traffic light at the intersection refer to traffic lights at intersections that are adjacent to each other.

[0135] In some embodiments, the input to the first machine learning model may further include first monitoring data of adjacent traffic lights at the intersection (if any) of the abnormal traffic light, and the training samples of the first machine learning model may also include sample first monitoring data of adjacent traffic lights at the intersection (if any) of the sample traffic light at a first historical time point. In some embodiments of this specification, cross-traffic light collaborative analysis at different intersections enables the traffic light fault emergency management system based on the Internet of Things large model to have a global perspective, enabling the analysis of anomalies of individual traffic lights within a more macroscopic network. This effectively eliminates environmental interference, improves the robustness of fault prediction, and ensures the reliability and practicality of the prediction results.

[0136] In some embodiments, the output of the first machine learning model also includes the fault time; the maintenance parameters include the maintenance sequence, maintenance time, and data acquisition frequency of the abnormal signal lights; the emergency monitoring and management platform can also determine the maintenance sequence and maintenance time of the abnormal signal lights based on the fault time; and adjust the data acquisition frequency of the abnormal signal lights based on the maintenance time.

[0137] In some embodiments, the label of the first machine learning model further includes sample failure times. Sample failure times can be obtained from historical data. For example, the emergency monitoring and management platform can select multiple sample traffic lights that have the same or similar first monitoring data, second monitoring data, and status assessment results at a first historical time point, obtain the failure times of the same sample failure type that occurred in these sample traffic lights within a preset period after the first historical time point, and use the average of these failure times as the sample failure time corresponding to the sample failure type.

[0138] For more information on maintenance sequence, maintenance time, and data acquisition frequency, please refer to [link / reference]. Figure 2 And its related descriptions.

[0139] Failure time refers to the time interval between the predicted failure time by the first machine learning model and the current moment.

[0140] In some embodiments, the shorter the failure time of the fault indicator light, the higher the priority of its maintenance sequence.

[0141] In some embodiments, the emergency monitoring and management platform can also perform weighted sorting based on the importance of intersections to reconfirm the maintenance sequence.

[0142] The importance of an intersection refers to a quantitative indicator used to determine its maintenance priority. It can be defined by comprehensively assessing an intersection's criticality within the traffic network, and may include factors such as traffic volume and historical accident rate. For example, the higher the traffic volume at an intersection, the higher its maintenance priority. Similarly, a higher historical accident rate at an intersection also indicates a higher maintenance priority.

[0143] In some embodiments, when the maintenance time is approaching, the emergency monitoring and management platform can automatically send instructions to the ECU to increase the data collection frequency of the abnormal indicator light from the minute level to the second level, so that maintenance personnel can obtain more detailed and richer real-time data before arriving at the site, which can help with rapid diagnosis.

[0144] In some embodiments, based on the predicted fault type, the emergency monitoring and management platform can also send remote diagnostic commands to the ECU. These commands can trigger the traffic light controller to perform deeper self-test procedures, read lower-level operating logs and parameters, and even perform remote fine-tuning of certain parameters. For example, the traffic light controller can perform parameter optimization and communication protocol optimization.

[0145] In some embodiments, if the predicted fault type is "light aging of LED module" causing abnormal brightness, the emergency monitoring and management platform can send remote instructions to the LED driver module without affecting visibility, and increase the output current according to the degree of brightness abnormality to compensate for brightness decay, extend its effective working life, and gain maintenance time window.

[0146] In some embodiments, if the predicted fault type is "controller communication anomaly", the emergency monitoring and management platform can remotely reset the communication module or adjust the communication parameters of the communication module (such as baud rate, retransmission mechanism, etc.) in an attempt to restore a stable connection of the communication module.

[0147] In some embodiments of this specification, the first machine learning model can not only output where a fault might occur, but also when and how to perform maintenance, thus achieving fine-grained scheduling of predictive maintenance and a closed loop from prediction to action. Simultaneously, remote fine-tuning can alleviate fault problems to some extent without sending personnel to the site, gaining valuable maintenance time windows, reducing the frequency and difficulty of on-site maintenance, and greatly improving operational efficiency and resource utilization.

[0148] In some embodiments, the emergency monitoring and management platform is further configured to: determine the coordination parameters of the associated traffic lights at the surrounding intersections of the faulty traffic light, and send them to the associated traffic lights, wherein the coordination parameters include the timing parameters of the associated traffic lights; the associated traffic lights are configured to: trigger a sub-clock at the start time of each color light according to the timing parameters, and drive the bulbs to light up alternately by cyclically controlling the relay group.

[0149] The surrounding related intersections refer to adjacent intersections that have a direct traffic flow connection with the intersection where the faulty traffic light is located. For example, intersections located on the same main road as the faulty traffic light, or intersections connected by a side road.

[0150] In some embodiments, the emergency monitoring and management platform can identify surrounding intersections that have a direct or indirect traffic connection with a faulty traffic light based on geographic information, road network topology, and historical traffic data.

[0151] Associated traffic lights refer to traffic lights installed at adjacent intersections. In some embodiments, the operation of associated traffic lights needs to be coordinated with the traffic lights at intersections that are malfunctioning.

[0152] Coordination parameters are instruction parameters used to achieve coordinated operation of traffic lights in a region. Examples include phase difference, cycle duration, and green light ratio.

[0153] In some embodiments, the emergency monitoring and management platform can generate coordination parameters based on prior experience, such as vehicle flow, average speed, queue length, and vehicle type distribution at surrounding intersections.

[0154] In some embodiments, the emergency monitoring and management platform can send encrypted data packets to the associated traffic light ECU via a city private network or wireless communication network (such as 5G) based on coordination parameters. The sending process employs a "request-confirmation" mechanism to ensure reliable delivery of instructions, and updates the traffic light status log in the emergency monitoring and management platform upon successful delivery.

[0155] In some embodiments, the coordination parameters include timing parameters, phase parameters, and special modes.

[0156] Timing parameters refer to the start time and duration of traffic lights in different color states (such as red, green, and yellow). In some embodiments, the emergency monitoring and management platform can dynamically adjust timing parameters according to different traffic conditions, including but not limited to the following methods: Diversion: By appropriately shortening the green light duration of the associated traffic lights upstream of the faulty intersection, traffic flow entering the faulty intersection is reduced, thereby guiding vehicles to choose alternative routes in advance; Traffic management: By extending the green light duration of the associated traffic lights downstream of the faulty intersection, the passage speed of vehicles already entering the intersection or adjacent area is accelerated, improving evacuation efficiency; Balance: By comprehensively considering the timing schemes of associated traffic lights at multiple adjacent intersections within the area, the overall balance and stability of traffic flow is achieved by coordinating the ratio of green lights to red lights at each intersection.

[0157] Phase parameters refer to parameters that control the release order and switching mode of traffic in each direction. In some embodiments, by adjusting the phase parameters, the passage order of vehicles and pedestrians can be changed, optimizing traffic flow organization. For example, based on the control of associated traffic lights at surrounding intersections, priority can be given to the flow of traffic in the direction of the main road, or the order of left turns, straight ahead, and right turns at intersections can be adjusted to alleviate local congestion.

[0158] Special modes refer to unconventional signal control methods activated under specific circumstances to meet special traffic needs or emergency management objectives. For example, activating green waves to improve the efficiency of main roads, initiating two-way traffic to temporarily alleviate traffic pressure on one-way roads, or opening green lanes for emergency vehicles (such as ambulances and fire trucks) to ensure rapid and safe passage.

[0159] In some embodiments, the emergency monitoring and management platform can construct a first feature vector based on the traffic flow of the surrounding related intersections of the currently faulty traffic light, the fault type of the currently faulty traffic light, and the fault time; and perform vector matching in the first vector database based on the first feature vector to determine the first associated vector.

[0160] In some embodiments, the first feature vector can be constructed in various ways. For example, it can be constructed using methods such as TF-IDF (Term Frequency-Inverse Document Frequency), One-Hot, and Word2Vec. The first vector database may include multiple first reference vectors and corresponding reference timing parameters. Each first reference vector can be constructed based on historical traffic flow, historical fault type, and historical fault time. The construction method of the first reference vector is similar to that of the first feature vector. The reference timing parameters can be the actual timing parameters corresponding to the first reference vector. The actual timing parameters refer to the historical timing parameters corresponding to the first reference vector in cases of successful traffic diversion and no congestion.

[0161] In some embodiments, the emergency monitoring and management platform can determine a first reference vector that meets a first preset condition as a first associated vector through vector matching. The first preset condition may refer to preset conditions used to determine the first associated vector. In some embodiments, the first preset condition may include vector distance being less than a distance threshold, minimum vector distance, etc. The emergency monitoring and management platform can use the reference timing parameters corresponding to the first associated vector as timing parameters in the coordination parameters sent to the associated traffic lights.

[0162] Figure 4 This is an exemplary schematic diagram illustrating the control of associated signals according to some embodiments of this specification.

[0163] In some embodiments, the emergency monitoring and management platform is further configured to: generate candidate timing parameters based on a historical database; predict the traffic congestion duration corresponding to the candidate timing parameters in a future time period using a second machine learning model; and use the candidate timing parameters whose traffic congestion duration meets preset conditions as the timing parameters of the associated traffic lights.

[0164] A historical database is a structured collection of data storing multi-dimensional historical data related to traffic operations. In some embodiments, the historical database includes historical traffic flow, historical fault types, and historical fault times at each intersection where a traffic light is located. In some embodiments, the historical database includes the aforementioned first vector database. Candidate timing parameters refer to reference timing parameters corresponding to multiple first correlation vectors that satisfy a first preset condition.

[0165] The duration of traffic congestion in the future time period refers to the cumulative duration predicted by the traffic lights at the surrounding intersections of the faulty traffic light to be below a set threshold within a specific future time period (such as the next 30 minutes) when a certain candidate timing parameter is applied. The set threshold can be set based on the experience of those skilled in the art. For more information on traffic flow data, see step 210 and its related description.

[0166] The second machine learning model refers to a machine learning model used to predict candidate timing parameters, corresponding to the duration of traffic congestion in future time periods. For example, the second machine learning model may include one or more combinations of Recurrent Neural Network (RNN) models, Long Short-Term Memory (LSTM) models, or other custom models.

[0167] In some embodiments, the input to the second machine learning model may be candidate timing parameters and traffic flow data at the associated intersection of the faulty signal light. The output of the second machine learning model may be the traffic congestion duration corresponding to the candidate timing parameters. For more information on traffic flow data, see step 210 and its related description.

[0168] In some embodiments, the second machine learning model can be obtained by training an initial second machine learning model using multiple sets of second training samples with second labels. The second training samples may include sample timing parameters executed by the associated signal lights of the sample faulty signal light, and sample traffic flow data of the associated signal lights. The second label may include the traffic congestion duration corresponding to the execution of the sample timing parameters by the associated signal light.

[0169] In some embodiments, the second training sample and the second label may be obtained based on historical data.

[0170] The training process of the second machine learning model is similar to that of the first machine learning model. For details, please refer to the training process of the first machine learning model.

[0171] Preset criteria refer to the evaluation standards set for selecting the optimal candidate timing parameters. For example, a preset criterion could be to minimize traffic congestion duration.

[0172] In some embodiments, the emergency monitoring and management platform can compare multiple candidate timing parameters output by the second machine learning model with their corresponding traffic congestion durations, and select the candidate timing parameter with the shortest traffic congestion duration as the timing parameter of the associated traffic light.

[0173] In some embodiments of this specification, the emergency monitoring and management platform, through the simulation and prediction capabilities of a second machine learning model, can proactively select the optimal timing scheme, rather than simply adjusting according to preset rules. This ensures that, under complex traffic conditions, the most effective solution can be found to minimize congestion, achieving true intelligent traffic collaborative scheduling.

[0174] The start time refers to the initial moment when each color light is illuminated within a complete cycle of the traffic lights.

[0175] A sub-clock is a logic function unit that generates independent timing for a single signal light group and can control the phase difference by setting a start delay.

[0176] A relay group is a collection of electrical switching components used to safely drive high-power signal lights via low-power control circuitry. In some embodiments, the relay group connects an external high-voltage circuit by energizing a coil to activate a physical switch.

[0177] In some embodiments, the emergency monitoring and management platform triggers a sub-clock at the start time of each color light based on timing parameters. This sub-clock then cyclically controls the relay group to activate the bulbs, driving different bulbs to light up alternately. For example, after receiving the timing parameters, the associated signal light parses them to obtain the start time and duration of each color light and configures the corresponding sub-clock accordingly. When the master clock reaches the green light start time, it triggers the green light sub-clock and activates the green light relay, illuminating the green light. After the green light sub-clock finishes its countdown, it triggers the yellow light sub-clock and switches to the yellow light relay, illuminating the yellow light. After the yellow light sub-clock finishes its countdown, it triggers the red light sub-clock and switches to the red light relay, illuminating the red light. This cycle repeats continuously, forming a complete signal cycle and ensuring that each color bulb lights up precisely and orderly according to the timing parameters.

[0178] In some embodiments of this specification, when a fault occurs, the emergency monitoring and management platform sends coordination parameters to the associated traffic lights. The associated traffic lights then precisely execute the new timing scheme through sub-clock triggering and relay group control. This mechanism can effectively disperse traffic pressure at the fault point, guide vehicle diversion, and avoid local faults causing regional traffic congestion, thereby significantly improving the overall traffic efficiency of the road network.

[0179] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification and therefore remain within the spirit and scope of the exemplary embodiments described herein.

[0180] Finally, it should be understood that the embodiments described in this specification are merely illustrative of the principles of the embodiments described herein. Other variations may also fall within the scope of this specification. Therefore, alternative configurations of the embodiments described herein are intended to be illustrative rather than limiting, and should be considered consistent with the teachings of this specification. Accordingly, the embodiments described herein are not limited to those explicitly introduced and described herein.

Claims

1. A traffic signal light malfunction emergency management system based on an Internet of Things (IoT) big data model, characterized in that, The system includes an emergency monitoring and management platform, which is configured as follows: Based on the first and second monitoring data of the traffic light, the status assessment result of the traffic light is determined. The first and second monitoring data are collected through different sensing devices. Target traffic lights are determined based on the status assessment results, and the target traffic lights include fault traffic lights and abnormal traffic lights; Cut off the power to the fault indicator light and issue detour suggestions to the personnel in the target area; Based on the first monitoring data, the second monitoring data, and the status assessment result of the abnormal signal light, the fault prediction status of the abnormal signal light is determined; Based on the fault prediction, maintenance parameters are determined and sent to maintenance personnel.

2. The system according to claim 1, characterized in that, The determination of the abnormal traffic light is related to the traffic flow and the number of historical accidents at the intersection where the abnormal traffic light is located.

3. The system according to claim 1, characterized in that, The status assessment results include whether there is a fault and the degree of abnormality. The emergency monitoring and management platform is further configured as follows: Based on the first monitoring data, the initial evaluation result of the traffic light is determined; The initial assessment results are verified based on the second monitoring data; The initial evaluation result that passes the verification is used as the state evaluation result.

4. The system according to claim 1, characterized in that, The emergency monitoring and management platform is further configured as follows: Based on the first monitoring data, the second monitoring data, and the status assessment results of the abnormal signal light, the failure probability and failure type of the abnormal signal light are predicted using a first machine learning model.

5. The system according to claim 1, characterized in that, The emergency monitoring and management platform is further configured as follows: Determine the coordination parameters of the associated traffic lights at the surrounding intersections of the faulty traffic light, and send them to the associated traffic light. The coordination parameters include the timing parameters of the associated traffic lights. The associated signal light is configured to trigger a sub-clock at the start time of each color light according to the timing parameters, and to drive different light bulbs to light up alternately by cyclically controlling the relay group to turn on the bulbs.

6. A traffic signal light malfunction emergency management method based on an Internet of Things (IoT) big data model, characterized in that, The method is executed by the emergency monitoring and management platform, and the method includes: Based on the first and second monitoring data of the traffic light, the status assessment result of the traffic light is determined. The first and second monitoring data are collected through different sensing devices. Target traffic lights are determined based on the status assessment results, and the target traffic lights include fault traffic lights and abnormal traffic lights; Cut off the power to the fault indicator light and issue detour suggestions to the personnel in the target area; Based on the first monitoring data, the second monitoring data, and the status assessment result of the abnormal signal light, the fault prediction status of the abnormal signal light is determined; Based on the fault prediction, maintenance parameters are determined and sent to maintenance personnel.

7. The method according to claim 6, characterized in that, The determination of the abnormal traffic light is related to the traffic flow and the number of historical accidents at the intersection where the abnormal traffic light is located.

8. The method according to claim 6, characterized in that, The status assessment result includes whether there is a fault and the degree of abnormality. The determination of the status assessment result of the traffic light based on the first and second monitoring data of the traffic light also includes: Based on the first monitoring data, the initial evaluation result of the traffic light is determined; The initial assessment results are verified based on the second monitoring data; The initial evaluation result that passes the verification is used as the state evaluation result.

9. The method according to claim 6, characterized in that, The step of determining the fault prediction status of the abnormal signal light based on the first monitoring data, the second monitoring data, and the status assessment result of the abnormal signal light further includes: Based on the first monitoring data, the second monitoring data, and the status assessment results of the abnormal signal light, the failure probability and failure type of the abnormal signal light are predicted using a first machine learning model.

10. The method according to claim 6, characterized in that, The method further includes: Determine the coordination parameters of the associated traffic lights at the surrounding intersections of the faulty traffic light, and send them to the associated traffic light. The coordination parameters include the timing parameters of the associated traffic lights. The associated signal light is configured to trigger a sub-clock at the start time of each color light according to the timing parameters, and to drive different light bulbs to light up alternately by cyclically controlling the relay group to turn on the bulbs.

Images