A multi-mode reconfigurable intelligent traffic signal control method and its holographic sensing system

Abstract

This invention discloses a multi-mode reconfigurable intelligent traffic signal control method and its holographic perception system, belonging to the field of intelligent traffic control technology. The invention collects traffic flow data through a holographic perception module and lamp operating status data through a single-lamp controller, achieving dual-domain perception of traffic flow and equipment health. A multi-mode decision module selects or generates the current control mode from preset schemes based on the perceived data. When a streetlight malfunctions, it automatically adjusts the signal timing and simultaneously releases guidance information. The system integrates a closed-loop operation and maintenance process for work orders, enabling automatic fault detection, work order dispatch, and asset status updates. The strategy configuration platform supports independent configuration of control schemes by region, group, and time period. This invention solves the technical problems of single perception dimension, lack of decision-making collaboration, and broken operation and maintenance loops in existing technologies, improving the safety, operation and maintenance efficiency, and refined management capabilities of the traffic system.

Description

Technical Field

[0001] This invention relates to the field of intelligent traffic control technology, specifically to a multi-mode reconfigurable intelligent traffic signal control method and its holographic perception system. Background Technology

[0002] With the acceleration of urbanization and the continuous growth of motor vehicle ownership, urban traffic congestion and traffic accidents are becoming increasingly prominent, which puts forward higher requirements for the intelligence level of traffic signal control systems. At present, urban traffic signal control systems have gradually developed from traditional fixed timing control to inductive control and adaptive control. Some systems have begun to introduce artificial intelligence algorithms for signal timing optimization or deploy multimodal sensors for traffic flow perception.

[0003] Existing technologies disclose methods for traffic signal control through a cloud-edge-device collaborative architecture. These methods configure corresponding signal control strategies for different intersections using intersection clustering and conditional judgment logic, and optimize the control strategies using deep reinforcement learning models. This type of solution intelligently distributes signal control tasks to edge stations for processing, reducing network latency and improving the responsiveness of signal timing to real-time traffic flow. In addition, other technical solutions disclose intelligent sensing control methods based on multimodal sensor fusion. These methods fuse traffic data collected by visual sensors, millimeter-wave radar, and geomagnetic sensors, use an adaptive weighted fusion algorithm to generate fused feature information, and generate intelligent control decisions based on fuzzy logic, reinforcement learning, or event rules to achieve dynamic adjustment of traffic signal timing. However, the aforementioned existing technologies still have the following shortcomings: The current technology focuses on traffic flow data, such as vehicle queue length, vehicle speed, vehicle type, and number of pedestrians, but fails to perceive the health status of the traffic infrastructure equipment. For example, streetlights are an important facility for ensuring nighttime traffic safety, and their working status directly affects the vision and driving safety of traffic participants. When streetlights malfunction and cause insufficient lighting, the current system cannot detect the malfunction, and the traffic lights continue to operate according to the original strategy. It cannot adjust the timing strategy or issue safety warnings in a timely manner according to the deterioration of lighting conditions, which poses a traffic safety hazard. Traffic control and equipment operation and maintenance are disconnected and lack closed-loop management. Existing technologies focus on optimizing signal timing and generating control strategies, but do not address the operation and maintenance handling mechanism after equipment failure. When equipment such as sensors, traffic lights or streetlights fail, the system cannot automatically generate maintenance work orders and track maintenance progress. The discovery and handling of equipment failures still rely on manual inspections or public reports, resulting in delayed operation and maintenance response and affecting the operational reliability of the traffic system. The configuration flexibility of the control strategy is insufficient. Although the existing technology supports the dynamic adjustment of the control strategy according to the traffic conditions, its strategy configuration is usually for a single intersection or adopts a uniform rule. It cannot be configured in a refined and independent manner according to the region, group or time period, and it is difficult to adapt to the differentiated needs of traffic control in different functional areas and at different times. In summary, existing intelligent traffic signal control systems are limited to traffic flow data in terms of perception, failing to acquire the health status of traffic infrastructure such as streetlights. In terms of decision-making coordination, signal control and equipment maintenance are disconnected, making it impossible to dynamically adjust timing strategies to ensure traffic safety when equipment malfunctions. Regarding maintenance management, there is a lack of automated fault detection and work order closed-loop mechanisms, relying on manual inspections with delayed responses. In terms of strategy configuration, it is difficult to achieve refined independent control by region, group, or time period. Therefore, there is an urgent need for an intelligent traffic signal control method and system that can integrate dual-domain perception of traffic flow and equipment status, achieve coordinated execution of signal control and information dissemination, possess automated maintenance closed-loop management capabilities, and support software-defined reconfigurable configuration. Summary of the Invention

[0004] To overcome the problems of the prior art, this invention discloses a multi-mode reconfigurable intelligent traffic signal control method and its holographic perception system.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: A multi-mode reconfigurable intelligent traffic signal control method includes the following steps: S1, Multiple control schemes are preset for at least one target area through the strategy configuration platform. Each control scheme includes a signal timing strategy and an information release strategy associated with the signal timing strategy. S2, real-time acquisition of holographic perception data of the target area, including traffic flow data and lamp working status data collected by the single lamp controller deployed on the street lamp pole; S3, based on holographic perception data, selects or dynamically generates the current control mode from multiple control schemes; S4, execute the current control mode, including: Send timing commands to the traffic light controller to adjust the phase and duration of the traffic lights; Send a release command to the information release device and output guidance information that matches the current control mode; When the lighting status data indicates a street light malfunction, the signal timing strategy at the associated intersection is automatically adjusted.

[0006] Preferably, the information dissemination equipment includes an LED display screen and / or broadcasting equipment.

[0007] Preferably, it also includes work order generation and closed-loop management steps: When a device malfunction is detected, a repair work order is automatically generated and dispatched to the designated repair personnel. Maintenance personnel receive work orders, complete repairs, and upload repair records via mobile devices. The system automatically updates the status of equipment assets based on maintenance records.

[0008] Preferably, the control schemes are configured independently by region or group, and an activation time period is configured for each control scheme.

[0009] Preferably, the single-lamp controller monitors the operating current, voltage, power and fault status of each street lamp in real time through built-in current detection circuit and voltage detection circuit, and reports the monitoring data to the strategy configuration platform through a wireless communication module.

[0010] Preferably, in step S3, the holographic perception data is processed based on preset rules or artificial intelligence models to select or generate the current control mode.

[0011] Preferably, the traffic flow data includes at least one of vehicle queue length, vehicle speed, vehicle type, and number of pedestrians.

[0012] Preferably, the holographic perception data also includes event data, which includes pedestrian crossing requests, emergency vehicle passage information, or traffic accident information.

[0013] Preferably, the guidance information in step S4 includes text prompts or voice prompts that match the current control mode.

[0014] On the other hand, a multi-mode reconfigurable intelligent traffic signal holographic perception system is provided, including: Holographic sensing modules are deployed at intersections or on the roadside to collect traffic flow data in real time. A single-lamp controller, deployed on a streetlight pole, has built-in current and voltage detection circuits for real-time acquisition of lamp operating status data; The strategy configuration platform provides a human-machine interface and supports multiple preset control schemes for target areas. Each control scheme includes a signal timing strategy and an information release strategy associated with the signal timing strategy. The multi-mode decision-making module is connected to the holographic perception module, the single lamp controller and the strategy configuration platform respectively, and is used to dynamically select or generate the current control mode based on traffic flow data and lamp working status data. The execution module, connected to the multi-mode decision module, includes a traffic light controller and an information dissemination device controller, and is used to synchronously perform signal timing adjustment and information dissemination according to the current control mode; When the lamp working status data indicates a street light malfunction, the multi-mode decision module automatically adjusts the signal timing strategy of the associated intersection and simultaneously controls the information dissemination device to output guidance information.

[0015] The beneficial effects of this invention are as follows: Compared with the prior art, the technical solution provided by the present invention has the following significant advantages: To improve traffic safety, this invention monitors the working status of streetlights in real time through a single-lamp controller. When a streetlight malfunction is detected, the multi-mode decision module automatically adjusts the signal timing strategy of the associated intersection, such as extending the green light time for pedestrians crossing the street, and simultaneously publishes safety reminder information through LED displays or broadcasting equipment. This mechanism enables traffic signal control to sense the health status of the equipment and respond dynamically, effectively reducing the risk of traffic accidents caused by insufficient lighting. To achieve closed-loop operation and maintenance management, this invention automatically links equipment fault detection with work order generation. When a fault occurs, a repair work order is automatically dispatched to designated maintenance personnel. Maintenance personnel use mobile terminals to accept the order, sign in, upload repair records, and register material issuance. After repairs are completed, the system automatically updates the equipment asset status. This closed-loop mechanism upgrades traditional manual repair reporting to automated operation and maintenance, significantly shortening fault response time and improving the operational reliability of transportation equipment. To enhance the flexibility and precision of control, this invention supports the division of intersections by administrative division, road segment, or logical grouping through a strategy configuration platform. Multiple control schemes are preset independently for each area or group, and an activation time period is configured for each scheme. The system automatically calls the corresponding scheme according to the time, realizing reconfigurable control that is time- and location-specific and software-defined, meeting the differentiated traffic management needs of different functional areas such as commercial areas, school areas, and residential areas. To achieve deep coordination between signal control and information dissemination, this invention integrates a traffic light controller and an information dissemination device controller in the execution module. When the control mode is switched, the signal timing command and information dissemination command are output synchronously, enabling traffic participants to obtain guidance information that matches the current control mode in a timely manner. This avoids misjudgment caused by information lag and improves the overall traffic efficiency and safety of the intersection. Attached Figure Description

[0016] Figure 1 This is a flowchart of a multi-mode reconfigurable intelligent traffic signal control method provided in Embodiment 1 of the present invention; Figure 2 The flowchart shows the framework of a multi-mode reconfigurable holographic sensing system provided in Embodiment 1 of the present invention. Detailed Implementation

[0017] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, exemplary embodiments will be described in detail below, examples of which are illustrated in the accompanying drawings. In the following description relating to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of methods and systems consistent with some aspects of this application as detailed in the appended claims.

[0018] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0019] The following detailed description of the specific implementation methods, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided in detail.

[0020] Please refer to Figure 1 This invention relates to a multi-mode reconfigurable intelligent traffic signal control method and its holographic perception system. As the complexity of urban traffic management increases, existing technologies often only focus on optimizing traffic flow itself, while neglecting the impact of the health status of traffic infrastructure on traffic safety and operational efficiency. When street light failures lead to insufficient lighting, traffic lights still operate according to the original strategy, which can easily cause traffic accidents. At the same time, equipment failures rely on manual repair reporting, resulting in delayed operation and maintenance response and the inability to form closed-loop management. This invention overcomes the limitations of the unidirectional link between perception, decision-making, and execution in existing technologies by constructing an integrated technical architecture that enables cross-domain perception, collaborative decision-making, and closed-loop operation and maintenance. The system includes a holographic sensing module deployed at intersections to collect traffic flow data, a single-lamp controller deployed on streetlight poles to collect lamp working status data, a strategy configuration platform that supports multiple preset schemes by region / group, a multi-mode decision module that dynamically selects or generates control modes based on sensing data, and an execution module, including a traffic light controller and an information publishing device controller. When the lighting status data indicates a street light malfunction, the multi-mode decision module automatically adjusts the signal timing strategy of the associated intersection and simultaneously issues safety alerts via broadcast or LED screen, forming a dual-domain perception and collaborative decision-making for traffic flow and equipment health. The system integrates a closed-loop operation and maintenance process for work orders, enabling automatic detection of equipment faults, work order dispatch, maintenance execution, and asset status updates. This upgrades traditional manual repair reporting to intelligent closed-loop management based on awareness-based operation and maintenance. Through a software-defined, reconfigurable architecture, this invention supports flexible configuration of control schemes by region, group, and time period, significantly improving the safety, operational efficiency, and refined management capabilities of the transportation system.

[0021] Example 1 This embodiment takes an urban main road intersection as an example to explain in detail the specific implementation process of a multi-mode reconfigurable intelligent traffic signal control method. The method includes four core steps: preset multiple schemes, real-time acquisition of holographic perception data, dynamic selection or generation of control modes, and collaborative execution.

[0022] S1. During the system initialization phase, the administrator presets multiple control schemes for the intersection through the strategy configuration platform. Each control scheme includes a signal timing strategy and an information release strategy associated with the signal timing strategy. The signal timing strategy defines parameters such as the green light duration, cycle duration, and phase difference for each phase. The information release strategy defines the guidance information content output through LED displays or broadcasting equipment at the same time as the signal timing is adjusted. For example, for the morning rush hour, the administrator presets a morning rush hour plan, with the signal timing strategy being to extend the green light for north-south traffic by 10 seconds and shorten the green light for east-west traffic by 5 seconds to cope with tidal traffic flow. The associated information dissemination strategy is to display the morning rush hour on the LED display screen at the intersection, asking people to pass in an orderly manner. For intersections around schools, a pedestrian priority scheme is preset, with a signal timing strategy that extends the green light for pedestrians by 5 seconds. The associated information dissemination strategy is to announce pedestrians crossing the street and ask them to give way to pedestrians via broadcast. Each scheme can also be configured independently by area or group and the activation time period can be set. When the system clock reaches the preset time, the corresponding scheme will be automatically called without manual intervention. S2, the system acquires multi-source sensing data of the intersection in real time, including traffic flow data and lighting status data; Traffic flow data is collected by a holographic perception module deployed in all directions of the intersection, which includes AI cameras and millimeter-wave radar. The AI ​​cameras are used to identify vehicle queue length, vehicle type, number of pedestrians, and non-motorized vehicle flow; the millimeter-wave radar is used to accurately measure vehicle speed, vehicle spacing, and acceleration in each lane. These sensors continuously collect data at millisecond intervals, forming a dynamic sequence of traffic flow characteristics; The operating status data of the lamps is collected by the single lamp controller deployed in each street light pole. The single lamp controller integrates current detection circuit and voltage detection circuit. Specifically, a high-precision sampling resistor is connected in series in the lamp power supply circuit. The voltage drop across the sampling resistor is read through the analog-to-digital converter, and the operating current of the lamp is calculated in real time according to Ohm's law. At the same time, the operating voltage of the lamp is collected through the voltage divider circuit. The single-lamp controller also has built-in fault diagnosis logic: When the detected current is lower than the preset threshold and continues for more than the set time, it is determined to be an open circuit fault, such as a damaged lamp. When the detected current exceeds a preset threshold, it is determined to be a short circuit fault; When abnormal voltage fluctuations are detected, it is determined to be a power supply failure; The single lamp controller reports the collected current, voltage, power and fault status data to the multi-mode decision module through the wireless communication module at a cycle of 30 seconds or in an event-triggered manner. In a specific scenario of this embodiment, during a certain nighttime period, the single-lamp controller deployed on the street lamp pole on the south side of the intersection detects a sudden drop in current value to zero, determines that the lamp has an open circuit fault, and immediately reports the fault type, fault location, fault occurrence time and other data to the multi-mode decision module. S3, the multi-mode decision-making module receives real-time data from the holographic perception module and the single-lamp controller, and performs comprehensive analysis. On the one hand, the module analyzes traffic flow data: The AI ​​camera detected that the north-south traffic queue length reached 35 meters, exceeding the preset threshold, such as 20 meters. At the same time, the millimeter-wave radar detected that the average vehicle speed in this direction dropped to 15 km / h, indicating that this direction was congested. On the other hand, the module analyzes the working status data of the lighting fixtures: the streetlights on the south side are malfunctioning, resulting in insufficient lighting in that direction and posing a safety hazard. The multi-mode decision-making module has a built-in preset rule engine and artificial intelligence model. In this embodiment, the module makes judgments based on preset rules: If a street light malfunctions in a certain direction and there is heavy traffic in that direction, a fail-safe mode is triggered. This mode dynamically selects or generates the current control mode from a preset scheme library in real time, specifically including: The signal timing strategy appropriately shortens the green light duration for southbound traffic lights from 30 seconds to 25 seconds, while extending the green light duration for pedestrian crossings by 3 seconds to reduce the risk of pedestrians crossing the street due to insufficient lighting. Information dissemination strategy: Display a message on the LED screen indicating a street light malfunction on the south side, please observe carefully and slow down; also broadcast a message about a street light malfunction, please drive with caution. If the system is equipped with an artificial intelligence model, such as a deep Q-network based on reinforcement learning, the module can also input the current perception data into the model, and the model outputs the optimal control mode. The input state space of the model includes traffic flow characteristics and equipment health characteristics, and the output action space includes multiple control mode options. S4, the multi-mode decision module sends the generated current control mode to the execution module. The execution module includes a traffic light controller and an information publishing device controller. Both receive instructions in parallel and execute them synchronously to ensure that the signal timing adjustment and information publishing are highly consistent in time, with a delay of less than 100 milliseconds. The specific actions to be performed include: Send timing instructions to the traffic light controller: adjust the green light duration for southbound to northbound traffic to 25 seconds, extend the green light duration for pedestrian crossings by 3 seconds, and make corresponding minor adjustments to the timing for other directions; Send a release command to the LED display controller: The screen displays a message indicating that the streetlights on the south side are faulty. Please pay attention and slow down. The message uses bright red text to enhance the warning effect. Send a broadcast command to the broadcast equipment controller: The loudspeaker will announce a street light malfunction. Please drive carefully. The voice should be clear and the volume should be moderate to ensure that pedestrians and drivers within the intersection area can hear it clearly. The above three actions are executed simultaneously to form an integrated collaborative mechanism for control and guidance. When the single lamp controller reports that the fault has been repaired, that is, the current returns to normal, the multi-mode decision module automatically exits the fault-safe mode and restores the control scheme corresponding to the original time period. Specifically, this embodiment incorporates the operating status data of the lighting fixtures into the sensing range and automatically adjusts the signal timing strategy and simultaneously releases guidance information when the equipment fails. This solves the safety hazard of the prior art where the signal remains unchanged even when the equipment fails. It realizes the dynamic reconstruction of traffic signal control from efficiency priority to safety priority. At the same time, the synchronous execution of signal timing adjustment and information release enables traffic participants to obtain guidance information that matches the current control mode in a timely manner, thereby improving the overall safety and traffic efficiency of the intersection.

[0023] Another embodiment provides a closed-loop operation and maintenance process for work orders, further illustrating how to automatically generate maintenance work orders and complete the complete closed-loop management process when the system detects equipment failure. This process links the detection of equipment failure, the generation and dispatch of work orders, the on-site operation of maintenance personnel, the review of maintenance results, and the updating of asset status into an automated closed loop, realizing full digital control from the occurrence of failure to the completion of maintenance. When any device deployed at the intersection malfunctions, the system automatically triggers the work order generation mechanism. The sources of malfunction include lamp malfunctions reported by the single lamp controller, offline or communication abnormalities detected by the holographic perception module, and inspection abnormalities manually reported by the inspection personnel on the mobile terminal. Taking the open circuit malfunction of the south street light reported by the single lamp controller as an example, after receiving the fault data and performing the corresponding signal timing adjustment, the multi-mode decision module pushes the fault information to the work order maintenance module. After receiving the fault information, the work order maintenance module first identifies the fault type and determines its priority. For street light open circuit faults, the system automatically identifies the fault type as a lamp fault and, based on the fault occurrence time, location, and historical fault data of the road section, determines the priority as high risk, because insufficient nighttime lighting directly affects traffic safety. The system automatically assigns maintenance work orders to the street light maintenance personnel responsible for that section of road based on preset work assignment rules, regional location, maintenance personnel skill level, and current task load. After a maintenance work order is generated, the system pushes the work order notification to the designated maintenance personnel through the mobile application. After logging into the mobile terminal application, the maintenance personnel can view the work order details in the list of pending work orders, including the specific location of the faulty equipment, the fault type, the time of the fault, the equipment number, and historical maintenance records. After the maintenance personnel click to accept the order, the system records the time of acceptance and updates the work order status to "accepted". Maintenance personnel arrive at the fault site using the navigation function built into the mobile terminal, record the arrival time using the automatic check-in function of the mobile terminal, and take photos of the condition before maintenance using the photo function and upload them to the system. During the maintenance process, maintenance personnel scan the QR code or barcode on the faulty equipment to confirm that the equipment information matches the work order record. If materials need to be replaced, maintenance personnel can scan the outbound QR code of the new materials using a mobile terminal. The system will automatically complete the outbound registration of the materials and record the model, batch and quantity of the replaced materials. After the repair is completed, the repair personnel will use the mobile terminal's photo function to take photos of the repair status and fill in the repair record, including the fault cause analysis, information on replaced parts, and repair time. After the repair personnel submit the completion application, the work order status will change to pending review. The system automatically compares the photos before and after the repair to help the reviewers confirm the repair quality. The reviewers review the repair records through the strategy configuration platform. Once the repair results are confirmed to meet the requirements, the work order status is changed to completed. If the review fails, the system will return the work order with a note explaining the reason for the return, requiring the maintenance personnel to reprocess it; After the work order is approved, the system automatically updates the equipment asset status. In the case of replacing streetlights in this embodiment, the system updates the asset status of the original faulty light fixture to scrapped, enters the asset information of the newly installed light fixture into the asset database, and associates it with the corresponding light pole number. At the same time, the system updates the equipment's operation file based on the maintenance records, including information such as maintenance time, maintenance personnel, and replaced parts, forming a complete equipment lifecycle history; Specifically, this embodiment automatically links fault detection with work order generation, synchronizes on-site maintenance operations with system data in real time, and seamlessly connects maintenance results with asset status updates, thus constructing an automated closed-loop management process from fault perception to maintenance completion. This process changes the inefficient traditional model that relies on manual repair reporting and paper work order circulation, and achieves minute-level fault response, full-process traceability of maintenance, and dynamic updates of asset status, significantly improving the efficiency and accuracy of transportation equipment operation and maintenance.

[0024] Another embodiment provides a scheme management method based on independent configuration of regional groups and time period scheduling, further illustrating how to achieve fine-grained configuration and management of control schemes through a strategy configuration platform. Specifically, it includes configuring control schemes independently by region or group, and configuring an activation time period for each control scheme, thereby meeting the differentiated traffic control needs of different regions and different time periods. In urban traffic management, there are significant differences in the needs for traffic signal control among different functional areas, road grades, and time periods. To adapt to these differentiated needs, this invention provides a flexible scheme management mechanism through a strategy configuration platform. Administrators can divide intersections according to administrative divisions, road segments, or custom logical groups, and configure control schemes independently for each area or group. At the same time, specific activation time periods are set for each control scheme to achieve precise control based on time and location. Taking a typical application scenario in a certain city as an example, the city is divided into four different types of control areas: commercial core area, school-dense area, residential area and urban main road. Each area is configured with a corresponding control scheme independently according to its traffic characteristics and control objectives. For the core business area, the area has a large flow of people and vehicles during the day, especially at noon and in the evening, but it is relatively quiet at night. The administrator created three independent plans for the core business area through the strategy configuration platform. The weekday midday plan is configured to appropriately shorten the traffic light cycle to improve traffic efficiency. The associated information dissemination strategy is to display the midday peak in the commercial area on the LED screen, urging people to travel in a civilized manner. The plan is set to be activated from 11:30 to 13:30 on weekdays. The evening plan is configured to extend the green light time for pedestrians to accommodate dense pedestrian flow. The associated information dissemination strategy is to broadcast announcements that pedestrians should use crosswalks and vehicles should yield to pedestrians. The activation time is set from 5:00 PM to 7:00 PM daily. The nighttime plan is configured to shorten the traffic light cycle and reduce the brightness of information display equipment, with the activation period set from 10 PM to 6 AM the next day. For densely populated school areas, there is a large demand for students and parents to cross the street during school hours, while traffic flow is relatively low at other times. The administrator has created a priority plan for school hours in densely populated school areas. The signal timing strategy of this plan extends the green light time for pedestrians in all directions by eight seconds and sets a phase difference to ensure that students can safely cross multiple intersections in succession. The associated information dissemination strategy is to broadcast a message on the radio during school hours, asking vehicles to slow down and give way to students crossing the street, and to display the peak dismissal time on LED screens during dismissal time, asking vehicles to give way to students. The activation time for this scheme is set to 7:00 to 8:00 and 16:00 to 17:30 on weekdays. Outside of these time periods, the system will automatically switch to the regular time scheme and restore the standard signal timing. For residential areas, the traffic flow during morning and evening rush hours is mainly commuter traffic, while there is less traffic during off-peak hours. At night, there is a high requirement for noise reduction. The administrator created a morning rush hour commuter plan, an evening rush hour commuter plan, and an off-peak energy-saving plan for the residential areas. The morning rush hour commuter plan extends the green light for outbound traffic by 12 seconds. The associated information dissemination strategy is to display on the LED screen that there is a lot of traffic in the outbound direction during the morning rush hour and please keep a safe distance. The activation time is from 7:00 to 9:00 on weekdays. The evening rush hour commuting plan will extend the green light for inbound traffic by 12 seconds, and will be in effect from 5:00 to 7:00 on weekdays. The off-peak energy-saving plan will shorten the traffic light cycle when traffic flow is low, and at the same time reduce the operating power of information display equipment, and will be in effect from 9:00 to 5:00 daily. For urban arterial roads, multiple intersections need to coordinate and control green wave. The administrator created a green wave group for the arterial road, which includes eight consecutive intersections along the arterial road, and configured a green wave coordination scheme for the group. The signal timing strategy of this scheme calculates the phase difference between each intersection based on historical traffic flow data, so that vehicles can pass through multiple intersections continuously without encountering red lights when traveling at a set speed. The associated information dissemination strategy is to display the green wave speed recommendation of 60 kilometers per hour on the LED screen at the intersections entering the green wave. The plan is set to be implemented during the morning rush hour from 7:30 to 9:00 and the evening rush hour from 5:00 to 6:30. During other times, the plan will be implemented with independent control for each intersection. In actual operation, the system clock monitors the current time in real time. When the time enters the preset activation period of a certain scheme, the system automatically calls the scheme and sends it to all intersections in the corresponding area or group for execution. When the time leaves the activation period, the system automatically switches back to the default scheme or enters the preset scheme for the next period. The entire switching process does not require manual intervention, realizing the automatic scheduling of control schemes. Specifically, this embodiment achieves refined management and automated scheduling of traffic signal control by configuring control schemes independently by region or group and assigning each scheme an independent activation time period. This software-defined reconfigurable architecture enables traffic management departments to flexibly preset multiple control strategies according to the actual needs of different functional areas, different road grades and different time periods. The system automatically calls and executes these strategies based on time, significantly improving the flexibility and adaptability of traffic control and avoiding the inefficient method of relying on real-time manual intervention in the traditional model.

[0025] Another embodiment provides a decision-making method based on preset rules and artificial intelligence models, further illustrating how the multi-mode decision-making module processes holographic perception data based on preset rules or artificial intelligence models to select or generate the current control mode. This method takes traffic flow data, lighting status data, and event data as inputs, performs comprehensive analysis through a preset rule engine or deep reinforcement learning model, and outputs the optimal control mode adapted to the current scenario. During actual intersection operation, the multi-mode decision module continuously receives real-time data from the holographic perception module and the single-lamp controller. The traffic flow data collected by the holographic perception module includes various parameters such as vehicle queue length, vehicle speed, vehicle type, and number of pedestrians. Among them, the length of the vehicle queue is obtained by visually recognizing vehicles waiting in front of the stop line using an AI camera, with an accuracy down to the meter level. Vehicle speed is obtained by measuring the average speed of vehicles in each lane using millimeter-wave radar, with the unit being kilometers per hour. Vehicle types are categorized into small cars, large cars, and buses using image recognition algorithms; The number of pedestrians is obtained by counting the flow of people in the waiting area of ​​the intersection using AI cameras; The operating status data of the lamps collected by the single lamp controller includes the operating current, operating voltage, power and fault status of each street lamp. The fault status is specifically divided into three types: open circuit fault, short circuit fault and power supply fault. The multi-mode decision-making module has a built-in preset rule engine. This engine matches the aforementioned perception data with predefined rule conditions and triggers the corresponding control mode. Taking a main urban road intersection as an example, the preset rule engine is configured with multiple rules based on traffic flow data. When the queue length of vehicles in the north-south direction exceeds 40 meters and the duration exceeds 30 seconds, the rule engine determines that the direction is congested and triggers the congestion relief mode. This mode extends the green light duration in the north-south direction by 10 seconds based on the original cycle. When the number of pedestrians waiting exceeds 15 and the pedestrian red light waiting time exceeds 60 seconds, the rule engine determines that the demand for pedestrians to cross the street is backlogged and triggers the pedestrian priority mode. In this mode, the pedestrian green light duration is extended by 5 seconds and a pedestrian announcement is broadcast simultaneously, asking pedestrians to cross the street quickly. The preset rule engine is also configured with rules that integrate device health data. When a single lamp controller reports an open circuit fault in a street lamp and the fault occurs at night, the rule engine determines that there is a safety hazard of insufficient lighting in the road section and triggers the fault safety mode. The specific execution logic of this mode is as follows: if the faulty street light is located near a pedestrian crossing, the green light duration for pedestrian crossing will be extended by three seconds, and the faulty street light will be displayed on the LED screen. Please pay attention to the road surface. If a faulty street light is located directly above a motor vehicle lane, the recommended speed for the corresponding direction will be reduced by 10 kilometers per hour, and a speed limit warning will be issued via a variable message sign. Once the fault is repaired, the single-light controller will report that the vehicle has returned to normal, and the rules engine will automatically exit this mode. In addition to the preset rule engine, the multi-mode decision module also integrates an artificial intelligence model based on deep reinforcement learning. This model adopts a deep Q-network architecture and is trained with the long-term traffic efficiency and safety of the intersection as the optimization goal. The model's state space is designed as a multi-dimensional vector, which includes more than 30 dimensions such as vehicle queue length in each direction, average vehicle speed in each lane, number of pedestrians waiting, operating current value of each street light, fault status code of each street light, and current time period type. The model's action space is defined as a variety of selectable signal timing adjustment actions, including the increase or decrease in green light duration for each phase, the adjustment value of pedestrian green light duration, and the index number of the information dissemination strategy. The model's reward function comprehensively considers the reduction in average intersection delay time, the reduction in the number of vehicle stops, the reduction in pedestrian waiting time, and the safety risk coefficient during equipment failure, and forms a comprehensive reward value through weighted summation. In actual operation, the deep Q-network model performs a decision calculation every five seconds. The model takes the current perception data as the state input, calculates the expected cumulative reward value of each action through the forward propagation of the neural network, and selects the action with the highest reward value as the output. For example, at a certain decision-making moment, the model receives status inputs such as a north-south queue length of 32 meters, an east-west vehicle speed of 28 kilometers per hour, and a south-side street light operating current of zero amperes with an open-circuit fault code. The optimal action output by the model is to shorten the north-south green light duration by five seconds, extend the pedestrian green light duration by three seconds, display a south-side street light fault on the LED screen, and slow down. This output is converted into specific timing instructions and release instructions by the instruction generation unit of the multi-mode decision module and sent to the execution module. The deep Q-network model has continuous learning capabilities. The system will form an experience sample by taking the state at each decision, the selected action, the reward value observed after execution, and the state at the next moment, and store it in the experience replay pool. Every hundred decision cycles, the system will randomly sample a batch of samples from the experience replay pool and update the neural network parameters with gradients, so that the model's decision strategy will be continuously optimized as traffic flow characteristics and equipment operating status change. In event data-triggered scenarios, the multi-mode decision-making module prioritizes responding to preset event rules. When the holographic perception module detects emergency vehicle passage information, the system immediately interrupts the current routine decision-making process based on rules or artificial intelligence models and switches to emergency priority mode. This mode forcibly extends the green light duration for emergency vehicles until they have completely passed through. At the same time, it broadcasts an announcement that emergency vehicles are passing through and asks people to give way. It also displays the direction of emergency vehicle passage on LED screens. When traffic accident information is detected, the system switches to accident handling mode. This mode adjusts the timing of traffic lights at intersections around the accident site to facilitate the evacuation of vehicles from the accident area and simultaneously publishes detour suggestions through information dissemination devices. Specifically, this embodiment combines a preset rule engine with an artificial intelligence model to achieve comprehensive analysis and intelligent decision-making of multi-source perception data. The preset rule engine ensures the certainty and reliability of decisions in key scenarios, while the deep reinforcement learning model gives the system adaptive optimization capabilities in complex traffic environments. Both the rule engine and the artificial intelligence model use the lighting status data as input factors in their decision-making process, enabling the system to perceive the health status of the equipment and dynamically adjust the control strategy accordingly. This achieves an intelligent leap from simply optimizing traffic efficiency to taking into account both safety and efficiency.

[0026] Another embodiment provides a traffic flow and event collaborative control method based on multi-source data fusion, further explaining the specific types of traffic flow data collected by the holographic perception module, and how event data participates in the dynamic reconstruction of the control mode as a trigger condition. By combining refined traffic flow parameters such as vehicle queue length, vehicle speed, vehicle type, and number of pedestrians with event data such as pedestrian crossing requests, emergency vehicle passage information, and traffic accident information, the system can more accurately perceive the traffic status and achieve rapid mode switching and collaborative response when an event occurs. At the intersection of main roads and secondary roads in the city, holographic perception modules are deployed in all directions of the intersection, including four AI cameras and four millimeter-wave radars. AI cameras use deep learning algorithms to analyze traffic scenes in real time and can accurately output various traffic flow data. The vehicle queue length is obtained by detecting the queue length of vehicles waiting in front of the stop line. The system outputs this value in meters. When the queue length exceeds 30 meters, a congestion warning is triggered. Vehicle speed is calculated by the changes in vehicle position between consecutive frames. The system outputs the average speed of vehicles in each lane, in kilometers per hour. Vehicle type recognition classifies vehicles into small passenger cars, large passenger cars, trucks, and buses based on their external features. This data is used to differentiate the traffic needs of different vehicles and their sensitivity to signal timing. The number of pedestrians is obtained by detecting the pedestrian density in the waiting area of ​​the intersection. The system outputs the total number of pedestrians waiting to cross the street in four directions. When the number of people waiting in a certain direction exceeds ten, a pedestrian backlog warning is triggered. Millimeter-wave radar works in conjunction with AI cameras. The radar provides more accurate vehicle distance and speed data. The radar measures the real-time distance, speed and acceleration of each vehicle by emitting continuous frequency modulated waves and receiving the echoes. The system integrates radar data with visual data to form a complete description of the vehicle's motion status in each lane, including the distance between the vehicle and the stop line, instantaneous speed, deceleration, and estimated time to reach the stop line. Based on the traffic flow data mentioned above, the system simultaneously collects and processes multiple event data, which come from the event detection unit of the holographic perception module and messages pushed by external systems. The pedestrian crossing request event is triggered by the pedestrian crossing button installed on the street light pole at the intersection. When a pedestrian presses the button, the system receives a crossing request signal with a location identifier. Emergency vehicle traffic information is received through vehicle-road cooperative communication. When an ambulance, fire truck, or police car turns on its emergency lights and approaches an intersection, the on-board terminal sends information such as vehicle location, direction of travel, and estimated arrival time to the roadside equipment through dedicated short-range communication or cellular vehicle-to-everything (V2X) communication. Traffic accident information is obtained through visual recognition of abnormal events on the road by AI cameras. When a vehicle collision, abnormal vehicle stillness, or a person falling to the ground is detected, the system automatically determines that a traffic accident has occurred and generates information with the event type and location. When the system receives a pedestrian crossing request event, the multi-mode decision module immediately triggers the event response process. The module first obtains traffic flow data in each direction of the current intersection, including the queue length of vehicles in the north-south direction, the vehicle speed in the east-west direction, and the number of pedestrians waiting in each direction. Based on preset event handling rules, the module determines whether the pedestrian priority conditions are met. If the number of pedestrians waiting in the direction of the pedestrian crossing request exceeds five and the queue length of vehicles in the motor vehicle lane in that direction does not exceed fifty meters, the system automatically switches to pedestrian priority mode. In this mode, the signal timing strategy extends the green light duration for pedestrians crossing the street by five seconds from the original cycle and sets a clear phase to ensure that pedestrians can safely cross the entire crosswalk. At the same time, the execution module sends guidance information that matches the current control mode to the information display device. The LED screen displays that the pedestrian crossing request has been received, the green light has been extended, and the broadcast announces that pedestrians should cross quickly and vehicles should wait patiently. When the system receives emergency vehicle passage information, the multi-mode decision module activates a high-priority event response mechanism; Emergency vehicle traffic information includes vehicle type, direction of travel, current location, and estimated time of arrival at the intersection; The module calculates the approach lanes to the intersection the emergency vehicle will pass through based on its direction of travel and assesses the current signal status of each phase. The system immediately switches to emergency priority mode, the specific execution logic of which is as follows: If an emergency vehicle is expected to arrive at the intersection within ten seconds and the current direction of travel has a red light, the system will immediately switch the traffic light in that direction to a green light and maintain the green light until the emergency vehicle has completely passed through the intersection; if the current direction of travel has a green light, the system will extend the green light duration in that direction to ensure that the emergency vehicle does not need to slow down to pass through. While performing signal timing adjustments, the system simultaneously triggers information dissemination equipment. The LED screen displays "Emergency vehicle is passing, please give way," and the broadcast announces "Ambulance is passing, please move to the right to give way." After the emergency vehicle passes, the system automatically returns to the control mode before the incident. When the system detects traffic accident information, the multi-mode decision-making module triggers a more complex collaborative response mechanism; The AI ​​camera detected a rear-end collision between two vehicles in the middle of the intersection, and the system generated traffic accident data, including the type of accident, the location of the accident, and the number of vehicles involved. The module first assesses the impact on traffic flow based on the location of the accident. If the accident occupies two lanes in the north-south direction, it is determined to be a serious congestion event. The system switches to the accident handling mode. In this mode, the signal timing strategy aims to quickly evacuate vehicles from the accident area. The green light duration at the upstream intersection in the direction of the accident is extended by 15 seconds, while the green light duration at the downstream intersection in the direction of the accident is shortened to prevent further vehicle backlog. The associated information dissemination strategies include displaying the accident at the intersection ahead on LED screens at three intersections around the accident site and suggesting detours, and broadcasting an announcement at the accident site that the accident is being handled and that drivers should slow down. At the same time, the system automatically pushes accident information to the traffic management department, triggering the work order generation process, and achieving seamless connection from incident detection to response. In complex scenarios where multiple events occur simultaneously, the system arbitrates according to preset priorities, with emergency vehicle passage events having the highest priority. When an emergency vehicle passage event and a pedestrian crossing request event occur simultaneously, the system prioritizes responding to emergency vehicle passage and immediately responds to pedestrian crossing requests after the emergency vehicle has passed. When a traffic accident and a pedestrian crossing request occur simultaneously, the system will make a comprehensive judgment based on the correlation between the accident location and the pedestrian crossing intersection. If the accident point is located on the pedestrian's necessary crossing route, the system will prioritize handling the traffic accident and guide the pedestrian to cross the street through other passages. Specifically, by combining refined traffic flow data such as vehicle queue length, vehicle speed, vehicle type, and number of pedestrians with event data such as pedestrian crossing requests, emergency vehicle passage information, and traffic accident information, a multi-source data-driven collaborative control mechanism is constructed. This mechanism enables the system not only to perceive the regular traffic flow status but also to respond to various sudden traffic events in real time. It also dynamically reconstructs signal timing strategies and information dissemination content according to the event type, achieving full-scenario coverage from regular traffic optimization to emergency event response, and significantly improving the traffic system's ability to cope with complex road conditions and emergencies.

[0027] Another embodiment provides a specific implementation method for information release linkage, further illustrating how the execution module sends a release instruction to the information release device according to the current control mode, and outputs guidance information that matches the current control mode. The guidance information includes two forms: text prompts and voice prompts, which are output through an LED display screen and a broadcasting device, respectively, to achieve synchronous coordination between signal control and information guidance. At intersections on main urban roads, the execution module includes traffic light controllers, LED display controllers, and broadcasting equipment controllers. The LED display screens are installed on the lampposts in all directions of the intersection. They use high-brightness full-color screens and can clearly display text and graphic information under various lighting conditions. The broadcasting equipment is installed on the top of the lamppost and uses a directional speaker array, which can broadcast clear voice information in all directions of the intersection without generating excessive noise to disturb nearby residents; After the multi-mode decision module selects or generates the current control mode, the execution module receives the signal timing instruction and information release instruction corresponding to the mode in parallel. Taking the pedestrian priority mode as an example, when the system detects that the number of pedestrians waiting in the north-south direction exceeds 15 and the pedestrian red light waiting time exceeds 60 seconds, the multi-mode decision module generates the mode and outputs the instruction. The signal timing instruction extends the green light duration for pedestrians in the north-south direction from the original 25 seconds to 35 seconds, and correspondingly shortens the green light duration for vehicles in the north-south direction by 10 seconds to maintain cycle balance. The information release command consists of two independent commands, one sent to the LED display controller and the other sent to the broadcasting equipment controller; After receiving the instruction, the LED display controller parses the text content field in the instruction, which contains text prompts that match the current control mode. In pedestrian priority mode, the text prompts are: pedestrians please cross the street quickly, and vehicles please give way to pedestrians; The LED display controller converts text information into screen display data, which is then displayed on the screen in a scrolling manner through the internal drive circuit. The font is highlighted in red to enhance the warning effect. The display duration is set to ten seconds, after which the default display content is restored. After receiving the instruction, the broadcast equipment controller parses the voice content field and volume field in the instruction. In pedestrian priority mode, the voice prompt message is: Pedestrians are crossing the street, please cross quickly; vehicles should slow down and give way to pedestrians. The volume field is set to 80 decibels to ensure that pedestrians and drivers within the intersection area can hear it clearly. The broadcast equipment controller converts the voice content into an audio signal, which is then used by a power amplifier to drive the speaker for broadcasting. The broadcast is set to be twice, with a three-second interval between the two broadcasts. In emergency priority mode, when the system detects an emergency vehicle approaching an intersection, the execution module simultaneously outputs signal timing instructions and information release instructions. The signal timing instructions switch the traffic light in the direction of the emergency vehicle to green and lock it to ensure that the emergency vehicle can pass through without obstruction. The text message in the information release instruction is "Emergency vehicle is passing, please give way," and is displayed in full screen in flashing red font until the emergency vehicle has completely passed the intersection. The voice prompt message is: "The ambulance is on its way. Vehicles ahead, please move to the right and give way." The volume of the announcement is increased to 85 decibels, and the frequency is set to once every five seconds until the emergency vehicle passes. In fail-safe mode, when a single-lamp controller reports a streetlight malfunction, the multi-mode decision module generates a fail-safe mode. After the execution module receives the instruction, the signal light controller extends the pedestrian green light in the direction of the malfunction by three seconds. The LED display controller displays the text message "Streetlight malfunction on the south side, please observe the road surface and slow down." The broadcast equipment controller broadcasts the voice message "Streetlight malfunction, please drive carefully and observe the road surface." Both text and voice prompts highlight the core safety warning of street light malfunction, enabling traffic participants to be promptly aware of the abnormal situation and take appropriate preventive measures. The generation mechanism of text prompts and voice prompts is deeply bound to the current control mode. When the strategy configuration platform presets each control scheme, it also configures the corresponding text prompt template and voice prompt template. The template reserves variable fields, such as direction, fault type, and emergency vehicle type, which are filled in by the execution module according to the actual situation when the instruction is issued; Information dissemination instructions and signal timing instructions are issued through the same communication channel. The execution module uses a hardware timer to ensure that the effective time of the two instructions is synchronized, with the time deviation controlled within 50 milliseconds. This synchronization mechanism ensures that when the traffic lights change, the information dissemination device can output matching guidance information almost simultaneously, avoiding confusion or misjudgment by traffic participants due to information lag.

[0028] Another embodiment provides a collaborative deployment scheme for a holographic perception module and a single-lamp controller, further illustrating the specific composition of the holographic perception module and the single-lamp controller and their collaborative working method. The holographic perception module includes an AI camera and a millimeter-wave radar, used to collect various traffic flow data such as vehicle queue length, vehicle speed, vehicle type, and number of pedestrians; the single-lamp controller is deployed inside the streetlight pole, used to collect real-time lamp operating status data. Together, they constitute a complete perception layer, providing comprehensive data input for the multi-mode decision-making module. At the intersection of main urban roads, holographic perception modules are deployed on the cross arms of light poles in four directions of the intersection. Each direction is equipped with an AI camera and a millimeter-wave radar. The AI ​​camera uses an industrial-grade high-definition camera with a built-in deep learning image processing chip, which can complete image recognition locally without relying on cloud computing power. The camera's field of view covers all lanes from the stop line to within 80 meters upstream, and can accurately identify the end position of the vehicle queue in each lane. The millimeter-wave radar uses frequency-modulated continuous wave technology and transmits at a frequency of 77 GHz, which can penetrate adverse weather conditions such as rain and fog, and stably measure the distance, speed and angle of targets within a range of up to 200 meters. The AI ​​camera continuously collects images of the intersection during operation and performs real-time analysis of the images through a built-in convolutional neural network model. The model outputs four core traffic flow data. The vehicle queue length is obtained by detecting the length of the stationary vehicle queue in front of the stop line. The system outputs this value in meters and calculates the independent queue length of the straight, left-turn, and right-turn lanes in each direction. The vehicle speed is calculated by the changes in vehicle position between consecutive frames. The system outputs the average speed and minimum speed of all vehicles in each lane in kilometers per hour. Vehicle types are classified into four categories based on their exterior features: small passenger cars, large passenger cars, trucks, and buses. The system outputs statistics on the number of each type of vehicle in each lane. The number of pedestrians is obtained by detecting the pedestrian flow density in the waiting area of ​​the intersection. The system outputs the real-time number of pedestrians in the waiting area in each direction and marks whether there are elderly people, children, or other groups that require extra attention. The millimeter-wave radar works in sync with the AI ​​camera. The radar provides supplementary dynamic measurement data. The radar measures the distance, radial velocity, and angle of each target by transmitting continuous frequency modulated waves and receiving the echoes. The system spatiotemporally aligns the point cloud data output by the radar with the image data output by the AI ​​camera to form a complete description of the vehicle motion state in each lane. The fused data includes the precise distance between each vehicle and the stop line, instantaneous speed, deceleration, and estimated time to reach the stop line. This data is used to accurately calculate the timing and magnitude of signal timing adjustments. The single-lamp controller is deployed inside each street lamp pole at the intersection and installed between the lamp power supply circuit and the main power supply. Each single-lamp controller includes a power management unit, a microcontroller unit, a current detection circuit, a voltage detection circuit, and a wireless communication unit. The power management unit converts AC mains power into low DC voltage to power the internal circuitry of the controller. The microcontroller unit uses a low-power chip to run an embedded control program and is responsible for data acquisition, fault diagnosis and communication management. The current detection circuit uses a high-precision sampling resistor connected in series in the lamp power supply circuit. The resistance value of the sampling resistor is 0.01 ohms and the accuracy is 1%. The microcontroller reads the voltage drop across the sampling resistor at a frequency of 100 times per second through the analog-to-digital converter and calculates the lamp's operating current in real time according to the formula of voltage drop divided by resistance value. The voltage detection circuit is connected in parallel to the lamp power supply circuit through a voltage divider resistor with a voltage division ratio of one percent. The microcontroller reads the voltage value after voltage division through an analog-to-digital converter and calculates the lamp's operating voltage in real time according to the formula of multiplying the voltage value by the reciprocal of the voltage division ratio. The microcontroller further calculates the lamp's instantaneous power according to the formula of multiplying voltage by current. The fault diagnosis logic is implemented in software within the microcontroller. The microcontroller monitors the current, voltage, and power values ​​in real time and compares them with the preset normal operating range. The normal operating current range is 0.5 amperes to 3.5 amperes, and the normal operating voltage range is 220 volts to 240 volts. When the current value is below 0.1 amperes and lasts for more than three seconds, it is determined to be an open circuit fault, corresponding to a damaged lamp or a broken circuit. When the current value is above 4.0 amperes and lasts for more than one second, it is determined to be a short circuit fault, corresponding to an internal short circuit in the lamp or a short circuit in the circuit. When the voltage value is below 190 volts or above 260 volts and lasts for more than 10 seconds, it is determined to be a power supply fault, corresponding to an abnormal power supply line. When any fault is detected, the microcontroller generates a fault status code. Open circuit fault corresponds to status code 01, short circuit fault corresponds to status code 02, and power supply fault corresponds to status code 03. The wireless communication unit adopts a narrowband IoT communication module, supports cellular network access, has low power consumption and wide coverage. The single lamp controller actively reports the collected data every thirty seconds. The data packet includes the lamp post number, timestamp, current value, voltage value, power value and current fault status code. When a fault is detected, the wireless communication unit immediately triggers event reporting and sends the fault data to the multi-mode decision module with the highest priority. The reporting delay is less than one second. Traffic flow data collected by the holographic sensing module and lamp operating status data collected by the single lamp controller are aggregated through a unified IoT access platform, and then input into the multi-mode decision module after being aligned according to a unified data format and time sequence. The multi-mode decision-making module uses both types of data when making decisions. For example, during nighttime, if a single-lamp controller reports a street light malfunction in a certain direction, and the AI ​​camera detects a large number of pedestrians waiting in that direction, the module will comprehensively judge it as a high-risk scenario, trigger the fault safety mode, and extend the pedestrian green light duration.

[0029] Please refer to Figure 2 This embodiment provides a specific implementation of a multi-mode reconfigurable intelligent traffic signal holographic perception system, which includes a holographic perception module, a single-lamp controller, a strategy configuration platform, a multi-mode decision-making module, and an execution module. The holographic perception module is deployed in all directions of the intersection and consists of AI cameras and millimeter-wave radar. It is used to collect traffic flow data such as vehicle queue length, vehicle speed, vehicle type, and number of pedestrians in real time. The single lamp controller is deployed inside each street lamp pole and has built-in current detection circuit and voltage detection circuit. It is used to collect the working current, working voltage, power and fault status data of each street lamp in real time and report them to the multi-mode decision module through the wireless communication module. The strategy configuration platform is deployed on a cloud server, providing a human-computer interaction interface. It supports the division of intersections by administrative division, road segment or logical group, and independently presets multiple control schemes for each area or group. Each control scheme includes a signal timing strategy and an information release strategy associated with the signal timing strategy. It also configures the activation time period for each control scheme. The multi-mode decision module is deployed on the edge computing node and communicates with the holographic perception module, the single lamp controller and the strategy configuration platform respectively. Based on the real-time traffic flow data and lamp working status data, the multi-mode decision module selects or dynamically generates the current control mode from multiple preset control schemes. When the lamp working status data indicates a street lamp malfunction, the multi-mode decision module automatically adjusts the signal timing strategy of the associated intersection. The execution module is connected to the multi-mode decision module, including a traffic light controller and an information publishing device controller. The traffic light controller is used to perform signal timing adjustment according to the current control mode. The information publishing device controller includes an LED display controller and a broadcasting device controller, which are used to synchronously control the LED display to output text prompts or control the broadcasting device to output voice prompts according to the current control mode. Taking the typical operation of the system as an example, the holographic perception module collects the queue length of vehicles in the north-south direction, the speed of vehicles in the east-west direction, and the number of pedestrians in each direction in real time. The single-lamp controller reports the current, voltage and fault status of each street light in a 30-second cycle. When a single lamp controller reports an open circuit fault in the south street light at a certain moment, the multi-mode decision module recognizes the fault and automatically calls the fail-safe mode from the preset solution library of the strategy configuration platform. This mode extends the green light for pedestrians on the south side by three seconds, while simultaneously displaying a message on the LED screen indicating a street light malfunction on the south side, urging drivers to observe the road surface and broadcasting a message about the street light malfunction, urging drivers to drive with caution. The traffic light controller, LED display controller, and broadcasting equipment controller execute their respective instructions synchronously, completing a closed loop from fault detection to coordinated response. Specifically, in this embodiment, a dual-domain perception layer is formed by a holographic perception module and a single-lamp controller. Software-defined reconfigurable control is achieved through a strategy configuration platform. Collaborative decision-making on traffic flow and equipment health is achieved through a multi-mode decision module. Synchronous execution of signal control and information dissemination is achieved through an execution module. All modules work together to form a complete intelligent traffic signal holographic perception system.

[0030] The above-described embodiments are merely preferred embodiments of the present invention, and the scope of protection of the present invention is not limited thereto. Based on the basic concept of the present invention, those skilled in the art can make various extensions and modifications to the above embodiments without departing from the principle of the present invention.

[0031] For example, in addition to current and voltage detection and fault diagnosis functions, the single-lamp controller can be further expanded to support color temperature adjustment and energy consumption statistics. Through the built-in color temperature adjustment circuit, the single-lamp controller can automatically adjust the color temperature of the street lamp according to the ambient light or time period to improve lighting comfort. Through the built-in power metering chip, the power consumption of each street lamp can be accurately counted, providing data support for energy consumption management and energy-saving optimization. For example, in addition to LED displays and broadcasting equipment, information dissemination equipment may also include vehicle-road cooperative roadside units. Through vehicle-road cooperative roadside units, the system can directly push traffic light timing information, guidance prompts, and equipment fault warnings to the on-board terminals of connected vehicles, realizing real-time interaction of vehicle-road information and providing roadside cooperative support for high-level autonomous driving. For example, in addition to being deployed on the central server, the multi-mode decision-making module can also be deployed on edge computing nodes. By bringing the decision-making module down to the intersection side, the transmission delay between the perceived data and the decision command can be further reduced, achieving millisecond-level control response, which is particularly suitable for traffic event response scenarios with high real-time requirements. The above-mentioned variations are all equivalent substitutions or simple extensions under the technical concept of this invention. Their implementation principle is the same as that of the embodiments of this invention, and they should fall within the protection scope of this invention.

Claims

1. A multi-mode reconfigurable intelligent traffic signal control method, characterized in that, Includes the following steps: S1, Multiple control schemes are preset for at least one target area through a strategy configuration platform. Each control scheme includes a signal timing strategy and an information publishing strategy associated with the signal timing strategy. S2, acquire holographic perception data of the target area in real time, the holographic perception data includes traffic flow data and lamp working status data collected by the single lamp controller deployed on the street lamp pole; S3, based on the holographic perception data, select or dynamically generate the current control mode from the multiple control schemes; S4, Execute the current control mode, including: Send timing commands to the traffic light controller to adjust the phase and duration of the traffic lights; Send a release command to the information release device and output guidance information that matches the current control mode; When the lamp operating status data indicates a street light malfunction, the signal timing strategy of the associated intersection is automatically adjusted.

2. The method according to claim 1, characterized in that, The information dissemination equipment includes LED displays and / or broadcasting equipment.

3. The method according to claim 1, characterized in that, It also includes the steps of work order generation and closed-loop management: When a device malfunction is detected, a repair work order is automatically generated and dispatched to the designated repair personnel. Maintenance personnel receive work orders, complete repairs, and upload repair records via mobile devices. The system automatically updates the status of equipment assets based on maintenance records.

4. The method according to claim 1, characterized in that, The control schemes are configured independently by region or group, and an activation time period is configured for each control scheme.

5. The method according to claim 1, characterized in that, The single-lamp controller monitors the operating current, voltage, power, and fault status of each street lamp in real time through built-in current and voltage detection circuits, and reports the monitoring data to the strategy configuration platform through a wireless communication module.

6. The method according to claim 1, characterized in that, In step S3, the holographic perception data is processed based on preset rules or artificial intelligence models to select or generate the current control mode.

7. The method according to claim 1, characterized in that, The traffic flow data includes at least one of the following: vehicle queue length, vehicle speed, vehicle type, and number of pedestrians.

8. The method according to claim 1, characterized in that, The holographic perception data also includes event data, which includes pedestrian crossing requests, emergency vehicle passage information, or traffic accident information.

9. The method according to claim 1, characterized in that, The guidance information mentioned in step S4 includes text prompts or voice prompts that match the current control mode.

10. A multi-mode reconfigurable intelligent traffic signal holographic perception system, characterized in that, include: Holographic sensing modules are deployed at intersections or on the roadside to collect traffic flow data in real time. A single-lamp controller, deployed on a streetlight pole, has built-in current and voltage detection circuits for real-time acquisition of lamp operating status data; A strategy configuration platform is used to provide a human-computer interaction interface and support multiple preset control schemes according to a target area. Each control scheme includes a signal timing strategy and an information release strategy associated with the signal timing strategy. A multi-mode decision-making module is connected to the holographic perception module, the single-lamp controller, and the strategy configuration platform, respectively, and is used to dynamically select or generate the current control mode based on the traffic flow data and the lamp working status data. An execution module, connected to the multi-mode decision module, includes a traffic light controller and an information dissemination device controller, used to synchronously perform signal timing adjustment and information dissemination according to the current control mode; When the lamp working status data indicates a street light malfunction, the multi-mode decision module automatically adjusts the signal timing strategy of the associated intersection and simultaneously controls the information publishing device to output guidance information.

Images