Traffic signal intelligent regulation and control method and system

Abstract

The invention provides a traffic signal intelligent regulation and control method and system, and relates to the technical field of traffic management, and the method comprises the steps: obtaining radar sensing data, video sensing data and traffic signal lamp sensing data in a monitoring range of a target single intersection, which are correspondingly collected by a plurality of sensors, according to a preset interval time; fusing the radar sensing data, the video sensing data and the traffic light sensing data to obtain multi-source fusion data of the target single intersection; inputting the multi-source fusion data into a pre-trained timing scheme generation model to obtain a timing scheme of the traffic lights at the target single intersection; performing simulation verification on the timing scheme, and issuing the timing scheme to a signal control machine of a target single intersection after the verification is passed; and determining the vehicle patency at the target single intersection based on the issued timing scheme, and optimizing the timing scheme based on the vehicle patency. According to the invention, the timing scheme of the traffic signal is dynamically adjusted, and the flexibility and the control precision of traffic flow control are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic management, and in particular to a method and system for intelligently controlling traffic signals. Background Art

[0002] With the rapid development of cities, traffic is becoming more and more complex. Unreasonable traffic signal control schemes have led to more congested urban traffic and increased the occurrence of traffic accidents. Therefore, it is imperative to seek an efficient and safe urban traffic signal control optimization solution and system.

[0003] Existing signal control schemes are all formulated after analyzing single data, and the generated signal control schemes are fixed and cannot be adjusted according to the actual vehicle operation conditions in different time periods. The control accuracy of vehicle traffic at intersections is not high and the effect is not ideal. Summary of the Invention

[0004] In view of this, the object of the present invention is to provide a method and system for intelligent traffic signal control, so as to dynamically adjust the timing scheme of traffic signals and improve the flexibility and control accuracy of traffic flow control.

[0005] In a first aspect, a method for intelligently controlling traffic signals is provided, the method comprising:

[0006] Acquire radar perception data, video perception data, and traffic light perception data within the monitoring range of the target single intersection collected by multiple sensors at preset intervals;

[0007] Fuse radar perception data, video perception data, and traffic light perception data to obtain multi-source fusion data of the target single intersection;

[0008] Input the multi-source fusion data into the pre-trained timing plan generation model to obtain the timing plan of the traffic lights at the target single intersection;

[0009] The timing plan is simulated and verified, and after passing the verification, it is sent to the signal control machine of the target single intersection;

[0010] The vehicle flow rate at the target single intersection is determined based on the issued timing plan, and the timing plan is optimized based on the vehicle flow rate.

[0011] Optionally, fusing radar perception data, video perception data, and traffic light perception data to obtain multi-source fusion data of a target single intersection includes:

[0012] Time synchronization of radar perception data and video perception data;

[0013] The synchronized radar perception data and video perception data are preliminarily fused to obtain radar-visual fusion data;

[0014] Based on the traffic light perception data and the lane channelization data at the target single intersection, the radar fusion data is corrected to obtain the multi-source fusion data of the target single intersection.

[0015] Optionally, the synchronized radar perception data and video perception data are preliminarily fused to obtain radar-visual fusion data including:

[0016] Perform coordinate conversion from the image coordinate system to the world coordinate system on the video perception data so that it is in the same coordinate system as the radar perception data;

[0017] Perform ROI filtering on the radar perception data and the video perception data after coordinate conversion to determine the overlapping area between the two;

[0018] Associating the target detected by the radar perception data in the overlapping area at the current moment with the target detected by the video perception data in the overlapping area in terms of distance; the target is a vehicle;

[0019] Perform ID association on the targets after distance association; wherein each target is configured with a unique ID;

[0020] The video perception data of the same target after ID association is fused with the radar perception data to obtain radar-visual fusion data.

[0021] Optionally, performing ID association on the distance-associated target includes:

[0022] Based on the video perception data collected at multiple consecutive moments, the same target is tracked by ID, and the video tracking IDs of all targets in the overlapping area detected at the current moment are obtained;

[0023] Based on the radar sensing data collected at multiple consecutive moments, the same target is tracked by ID, and the radar tracking IDs of all targets in the overlapping area detected at the current moment are obtained;

[0024] Targets with the same ID are associated based on the video tracking ID and radar tracking ID.

[0025] Optionally, the process of determining the radar tracking IDs of all targets within the overlapping area detected at the current moment includes:

[0026] Obtain the radar perception data at the last moment, which includes the target's position, velocity, acceleration, and heading angle;

[0027] Input the target's position, velocity, acceleration, and heading angle into the pre-built target motion prediction model to obtain the predicted position of the target at the current moment;

[0028] Calculate the difference between the predicted position of the target at the current moment and the actual position of the target;

[0029] The difference value is compared with the preset difference threshold. If it is less than or equal to the preset difference threshold, it is determined to belong to the same target, and the target ID of the previous moment is used; if it is greater than the preset difference threshold, it is determined to belong to a new target, and a new target ID is assigned.

[0030] Optionally, after acquiring the radar perception data, the method further includes:

[0031] The radar perception data is preprocessed, and the data preprocessing at least includes the coordinated processing of target stillness and target afterimage, and target flickering processing.

[0032] Optionally, the collaborative processing of target stillness and target afterimage is as follows:

[0033] Get the target's stationary time;

[0034] The stationary time is compared with the preset parking holding time; wherein the parking holding time is determined as follows;

[0035] Get the distance between the target and the millimeter-wave radar sensor;

[0036] If the distance is less than the preset distance threshold, the parking holding time is fixed;

[0037] If the distance is greater than the preset distance threshold, the parking hold time is determined by dividing the maximum detection distance of the millimeter-wave radar sensor by the preset adjustable distance parameter;

[0038] If it is less than or equal to the preset parking holding time, the radar perception data and target ID of the target are kept;

[0039] If it is longer than the preset parking holding time, the radar perception data and target ID of the target will be deleted.

[0040] Optionally, determining the vehicle flow rate at the target single intersection based on the issued timing plan includes:

[0041] When executing the issued timing plan, multi-source fusion data of a single intersection is obtained once at a preset interval;

[0042] The multi-source fusion data of multiple single intersections are fused to obtain the regional multi-source fusion data;

[0043] Fusing regional-level multi-source fusion data to obtain city-level multi-source fusion data;

[0044] Determine the vehicle flow rate at the target single intersection based on city-level multi-source fusion data.

[0045] Secondly, a traffic signal intelligent control system is provided, which includes: an intelligent signal perception module, a real-time traffic flow perception module, a video perception module, a high-precision map perception module, a multi-source data fusion module, a data analysis and signal control module, and an evaluation and optimization module;

[0046] The intelligent signal perception module is used to obtain the status of the traffic lights at the target single intersection;

[0047] The real-time traffic flow perception module is used to obtain radar perception data of the target single intersection and pre-process the radar perception data;

[0048] The video perception module is used to obtain video perception data of the target single intersection;

[0049] The high-precision map perception module is used to build a 3D simulated urban model and map the targets detected by radar perception data into the virtual scene of the 3D simulated urban model;

[0050] The multi-source data fusion module is used to fuse radar perception data, video perception data, and traffic light perception data to obtain multi-source fusion data of the target single intersection;

[0051] The data analysis and signal control module is used to input multi-source fusion data into the pre-trained timing plan generation model to obtain the timing plan of the traffic lights at the target single intersection, simulate and verify the timing plan, and send it to the signal control machine at the target single intersection after passing the verification;

[0052] The evaluation and optimization module is used to determine the vehicle flow rate at the target single intersection based on the issued timing plan, and optimize the timing plan based on the vehicle flow rate.

[0053] According to a third aspect, an electronic device is provided, comprising a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus.

[0054] The memory is used to store computer programs;

[0055] The processor is configured to implement any of the method steps described in the first aspect when executing the program stored in the memory.

[0056] The present invention provides a method and system for intelligent traffic signal control, which obtains radar perception data, video perception data, and traffic light perception data within the monitoring range of a target single intersection collected by multiple sensors; fuses the radar perception data, video perception data, and traffic light perception data to obtain multi-source fusion data of the target single intersection; inputs the multi-source fusion data into a pre-trained timing plan generation model to obtain a timing plan for the traffic lights at the target single intersection; simulates and verifies the timing plan, and sends it to the signal control machine of the target single intersection after the verification is passed; determines the vehicle flow rate at the target single intersection based on the sent timing plan, and optimizes the timing plan based on the vehicle flow rate. The present invention performs multi-source fusion of video perception data, radar perception data, and traffic signal light perception data to obtain traffic conditions at intersections from multiple dimensions and in all aspects, thereby ensuring that the generated timing plan is more accurate. The timing plan is regenerated at regular intervals, which satisfies the requirement of dynamically adjusting the timing plan according to changes in traffic conditions and improves the flexibility of traffic flow control. In addition, before the generated timing plan is sent to the signal control machine, simulation verification is also carried out to improve the reliability of the timing plan.

[0057] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0059] Figure 1 A schematic structural diagram of a traffic signal intelligent control system provided by an embodiment of the present invention is shown;

[0060] Figure 2 A flow chart of a traffic signal intelligent control method provided by an embodiment of the present invention is shown;

[0061] Figure 3 A schematic diagram showing a flow chart of a traffic signal intelligent control method provided by another embodiment of the present invention;

[0062] Figure 4 A schematic structural diagram of an electronic device provided by an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0064] Considering that the existing signal control schemes are all formulated after analyzing single data, and the generated signal control schemes are fixed and cannot be adjusted according to the actual vehicle operation conditions in different time periods, the control accuracy of vehicle traffic at intersections is not high and the effect is not ideal.

[0065] Based on this, an embodiment of the present invention provides a method and system for intelligent traffic signal control, which will be described below through embodiments.

[0066] In order to facilitate those skilled in the art to better understand the present application, some terms are explained before the examples:

[0067] Traffic control optimization: that is, under the condition that the basic conditions of the road surface and traffic rules remain unchanged, by adjusting and optimizing the timing, phase, phase sequence and other parameters of traffic lights, and adopting advanced signal control technologies and strategies, the reasonable distribution and guidance of traffic flow can be achieved, thereby improving the traffic capacity and service level of road traffic.

[0068] High-precision map: A high-precision map is a highly accurate and finely defined map with absolute and relative accuracy at the decimeter level. It contains rich map elements such as roads, lanes, roadside traffic signs, and ground markings.

[0069] Traffic signal perception: Traffic signal perception is a crucial component of intelligent transportation systems, autonomous driving, and assisted driving systems. Its purpose is to accurately and in real time determine the status of traffic lights, such as red, green, and yellow. This provides a basis for decision-making by the driving system, ensuring driving safety and improving road efficiency.

[0070] The embodiment of the present invention provides a traffic signal intelligent control method, which is applied to a traffic signal intelligent control system, such as Figure 1As shown, the system includes: intelligent signal perception module 102, real-time traffic flow perception module 101, video perception module 103, high-precision map perception module 104, multi-source data fusion module 105, data analysis and signal control module 106 and evaluation and optimization module 107. The functions of each module are described in detail below:

[0071] The intelligent signal perception module is used to obtain the status of the traffic lights at the target single intersection.

[0072] The intelligent signal sensing module, installed within the traffic control console at the intersection, uses non-contact current mutual sensing to monitor traffic light signals in real time. Specifically, a current transformer is installed externally to the traffic light control circuitry, enabling monitoring of current changes without disconnecting or modifying existing circuitry. This current transformer senses current changes in the traffic light control circuitry through a non-contact method and converts them into processable electrical signals.

[0073] This contactless method for collecting traffic light status improves safety, reduces interference with existing circuit systems, and mitigates safety risks associated with installation and maintenance. It also eliminates the need to disrupt the existing electrical system, simplifying the installation process and making subsequent maintenance easier. Furthermore, it avoids potential failure points caused by direct contact, increasing system stability and reliability.

[0074] In a possible implementation, the intelligent signal sensing unit includes a signal detection unit, a signal filtering and amplifying unit, a clock control unit, and a communication interface unit;

[0075] The signal detection unit may be the above-mentioned non-contact current transformer, which is used to collect the current signal of the signal light in real time;

[0076] The signal filtering and amplifying unit filters and amplifies the acquired weak current signal to obtain real-time traffic light status data.

[0077] The communication interface unit is deployed on a separate network and does not affect the traffic management network. Data is uploaded to the data acquisition and control platform through a specific communication interface and communication protocol. The data acquisition and control platform can perform one-to-many platform-side management and one-to-one parameter configuration for the intelligent signal perception module, and parse and store the data uploaded by the intelligent signal perception module for various application platforms or terminals (for example: vehicle-road cooperative system platform, holographic intersection data display platform, mobile APP application terminal, RSU roadside terminal, OBU roadside terminal, etc., which need to use traffic data for scenario applications) to call.

[0078] The clock control unit ensures its own clock stability in real time through the clock system inside the intelligent signal sensing module, and synchronizes the clock with the data acquisition and control platform or clock server to ensure the accuracy of the clock system.

[0079] In another embodiment, the intelligent signal sensing module further includes an operation status monitoring unit.

[0080] The operating status monitoring unit senses the operating status of traffic light signals in real time, analyzes and calculates the operating status of the signal control machine using a built-in algorithm, and generates data on the machine's operating health. This data is ultimately transmitted to the operation and maintenance platform via the communication interface unit. By acquiring this data on the health of the signal control machine, the equipment operation and maintenance platform enables intelligent and efficient operation and maintenance of the signal control machine.

[0081] The real-time traffic flow perception module is used to obtain radar perception data of the target single intersection and preprocess the radar perception data.

[0082] In this embodiment, a millimeter-wave radar is used to detect targets within the monitoring range of a target single intersection and some motion data of the targets, such as position, speed, acceleration, and heading angle of the targets.

[0083] The video perception module is used to obtain video perception data of the target single intersection.

[0084] The video perception module is a camera installed at the intersection, which collects video data of the intersection.

[0085] The high-precision map perception module is used to establish a three-dimensional simulated urban model and map the targets detected by radar perception data into the virtual scene of the three-dimensional simulated urban model.

[0086] This high-precision map module uses high-precision imagery, vector road network data, and 3D city models as a high-precision foundation to render the movement trajectories of targets reported by radar in real time. Combined with the high-precision foundation, it visualizes radar detection results for easy viewing by personnel. Reports are filtered at the road section, intersection, and equipment levels to avoid wasting unnecessary rendering performance on the front end.

[0087] The coordinate system of this high-precision map adopts the WGS84 coordinate system, which is used to verify the location of the equipment (millimeter wave radar) and the correctness of the monitoring target points and to verify the monitoring coverage of the radar equipment.

[0088] Based on the high-precision base, you can obtain the device's latitude, longitude and elevation by clicking the device's installation location on the map; or use RTK equipment to collect the device's location information on site and enter it into the data acquisition module to provide device location parameters for data analysis, real-time perception and other modules.

[0089] In order to help auxiliary debugging personnel deal with overlapping or missing monitoring ranges, the high-precision map module also distinguishes the observation ranges of nearby radar equipment with different colors, intuitively showing the effective observation range of each radar.

[0090] In addition, a high-precision network time synchronization system is introduced into the high-precision map module to control the time synchronization error of the high-precision base and all targets attached to the high-precision base to the microsecond level, ensuring the time and space synchronization of the targets.

[0091] In a feasible implementation, the high-precision map module also provides auxiliary functions such as point coordinate marking, distance measurement, and height measurement to facilitate the debugging of radar equipment.

[0092] The multi-source data fusion module is used to fuse radar perception data, video perception data and traffic light perception data to obtain multi-source fusion data of the target single intersection.

[0093] The data analysis and signal control module is used to input multi-source fusion data into the pre-trained timing plan generation model to obtain the timing plan of the traffic lights at the target single intersection, simulate and verify the timing plan, and send it to the signal control machine at the target single intersection after verification.

[0094] The evaluation and optimization module is used to determine the vehicle flow rate at the target single intersection based on the issued timing plan, and optimize the timing plan based on the vehicle flow rate.

[0095] Based on the traffic signal intelligent control system constructed in the above embodiment, the embodiment of the present invention provides a traffic signal intelligent control method, such as Figure 2 As shown, the method includes the following steps:

[0096] Step S201: acquiring radar perception data, video perception data, and traffic light perception data within a monitoring range of a target single intersection collected by a plurality of sensors at preset intervals.

[0097] In one example, radar perception data is collected through a millimeter-wave radar sensor, video perception data is collected through a camera, and traffic light perception data is collected through an intelligent signal perception module installed in a signal control machine.

[0098] Millimeter-wave radar can directly detect the position, speed, acceleration, heading angle and other information of the target (vehicle).

[0099] What is detected by the camera is video data, which needs to be further processed to obtain video perception data. This video perception data also includes information such as the target's position, speed, acceleration, and heading angle. The conversion process is as follows:

[0100] Step 1: Segment the video data into continuous frame images and use optical flow method, feature matching and other methods to track and detect the vehicle position on the continuous frame images.

[0101] The second step is to record the position of the vehicle in the continuous frame image. According to the time interval between each frame and the change of the vehicle position, the vehicle speed can be calculated.

[0102] The third step is to calculate the rate of change of velocity over time based on the obtained velocity information to obtain the acceleration information. Specifically, this involves differentiating the velocity value or using numerical difference methods to estimate the acceleration.

[0103] The fourth step is to determine the vehicle's orientation relative to the image coordinate system in each frame. After obtaining the vehicle's orientation in the image coordinate system, it is converted into a heading angle in the world coordinate system.

[0104] Step S202: Fusing radar perception data, video perception data, and traffic light perception data to obtain multi-source fusion data of the target single intersection.

[0105] Existing traffic control systems mostly rely on a single data source and have the following defects: Data limitations: a single sensor is easily affected by environmental interference (such as occlusion and lighting), resulting in inaccurate evaluation; poor dynamic adaptability: the failure to combine traffic light timing predictions with real-time traffic flow makes it difficult to optimize traffic control strategies; insufficient spatial coverage: ground sensors have difficulty covering high-altitude perspectives and cannot fully assess intersection congestion.

[0106] By fusing data from various sensors (such as radar, cameras, and traffic lights), the present invention can more accurately understand real-time traffic conditions, including information such as traffic volume, vehicle types, and speed distribution. This helps provide more comprehensive data for traffic control optimization.

[0107] Step S203: Input the multi-source fusion data into the pre-trained timing plan generation model to obtain the timing plan of the traffic light at the target single intersection.

[0108] In this step, the timing plan generation model can use the long short-term memory network LSTM, and its pre-training process is as follows:

[0109] The first step is to obtain the historical multivariate fusion data sequence {S1, S2, ...S N} as training samples;

[0110] In the second step, the training sample is input into the long short-term memory network LSTM for iterative training until the preset number of iterations is reached to obtain a trained timing plan generation model.

[0111] Step S204: The timing plan is simulated and verified, and after passing the verification, it is sent to the signal control machine of the target single intersection.

[0112] In this step, since the timing plan is to be issued to actual application scenarios to guide vehicle traffic, if its reliability is poor, it will bring serious safety problems. Therefore, before actual application, the feasibility of the generated timing plan is verified through simulation verification to improve the reliability of the timing plan.

[0113] In a feasible implementation, a simulation environment capable of simulating actual traffic conditions must first be established. This typically involves using professional traffic simulation software such as SUMO (Simulation of Urban Mobility) and VISSIM. These software allow users to input parameters such as road networks, traffic flows, and signal control strategies to simulate and analyze traffic behavior.

[0114] After setting all necessary parameters (such as vehicle type distribution, arrival rate, and speed distribution), run simulations to observe the effectiveness of the new scheduling plan. Key metrics of interest may include average delay time, queue length, and traffic efficiency. Simulations can be conducted separately to consider factors such as the impact of different time periods (peak and off-peak) and weather conditions. Adjust the scheduling plan based on the simulation results and repeat the simulation until satisfactory results are achieved.

[0115] Step S205: Determine the vehicle flow rate at the target single intersection based on the issued timing plan, and optimize the timing plan based on the vehicle flow rate.

[0116] In this step, the timing plan is sent to the traffic control computer at the target intersection. Real-time traffic data from the intersection can be collected at regular intervals. The traffic flow rate is then determined based on this data and compared with a preset traffic flow threshold. If the traffic flow rate exceeds the threshold, no adjustments are required. If the traffic flow rate is below the threshold, the timing plan is adjusted, for example, by increasing the red light countdown timer. After the adjustments, the plan is reassessed until the preset traffic flow threshold is reached.

[0117] It can be understood from the above embodiments that the present invention performs multi-source fusion of video perception data, radar perception data and traffic light perception data to obtain the traffic conditions of the intersection from multiple dimensions and in all aspects, thereby ensuring that the generated timing plan is more accurate, and the timing plan is regenerated at regular intervals, which satisfies the dynamic adjustment of the timing plan according to changes in traffic conditions and improves the flexibility of traffic flow control. In addition, before the generated timing plan is sent to the signal control machine, simulation verification is also carried out to improve the reliability of the timing plan.

[0118] Based on the above embodiment, radar perception data, video perception data, and traffic light perception data are fused to obtain multi-source fusion data of the target single intersection, including:

[0119] Step S2021: Time-synchronize the radar perception data and the video perception data.

[0120] Before fusion, the time synchronization of the data collected by each sensor must be ensured first, ensuring that the subsequent processing of the two is carried out under the same time reference, and ensuring the accuracy and consistency of multi-sensor data.

[0121] In a specific example, the time of the two is synchronized by the following formula:

[0122] t sync =t radar +NTP(t camera )+δ trans (1);

[0123] Where, t sync is the synchronization timestamp; t radar is the radar’s timestamp; NTP(t camera ) is the camera timestamp corrected by the Network Time Protocol (NTP); δ trans It is the transmission delay compensation, which is used to compensate for the time difference between the radar and the camera.

[0124] Step S2022: Preliminarily fuse the synchronized radar perception data with the video perception data to obtain radar-visual fusion data.

[0125] In one possible implementation, Figure 3 As shown in the figure, the synchronized radar perception data and video perception data are preliminarily fused to obtain radar-visual fusion data including:

[0126] Step S2022A: performing coordinate conversion from the image coordinate system to the world coordinate system on the video perception data so as to be in the same coordinate system as the radar perception data.

[0127] Video perception data is all based on the image coordinate system, while radar detection data is all based on the physical world coordinate system. By converting the video perception data into the physical world coordinate system, the two can be subsequently processed in the same coordinate system to ensure data consistency.

[0128] In one example, the conversion is performed by the following conversion formula:

[0129]

[0130] Where: is the position vector in the radar coordinate system; is the position vector in the camera coordinate system; R(θ) is the rotation matrix of the calibration plate; is the translation vector used to compensate for the spatial translation between the radar and the camera.

[0131] Step S2022B: performing ROI (Region of Interest) filtering on the radar perception data and the video perception data after coordinate conversion to determine the overlapping area between the two.

[0132] In one feasible implementation, the radar's ROI is first determined, which can be based on the radar's maximum detection range and detection angle. The video's ROI is then determined, which can be implemented using computer vision libraries such as OpenCV. For radar perception data, all targets falling within the radar ROI are filtered out. For video data, all objects within the video ROI are filtered out, and overlapping areas are found by matching their positional information. For example, if a radar point falls within the bounding box of a video object, then this point is considered overlapping. Specifically, point cloud processing libraries such as PCL (Point Cloud Library) can be used to match the spatial positions of radar data with video data.

[0133] Step S2022C: Associating the target detected based on the radar perception data in the overlapping area at the current moment and the target detected based on the video perception data in the overlapping area in terms of distance; the target is a vehicle.

[0134] In the overlapping area, the target detected by the radar is associated with the target detected by the camera at the nearest position. The nearest distance is determined from three dimensions: one is the distance of the spatial position, one is the similarity of the motion state, and the other is the similarity of the heading angle.

[0135] Specifically, the following multimodal feature association algorithm formula is used to determine the correlation between the two in terms of distance and perform the association:

[0136]

[0137] Where: Represents the feature association metric, which is used to evaluate the similarity between the radar target Oir and the visual target Ojv;

[0138] represents the Mahalanobis distance, which measures the spatial similarity of two targets;

[0139] represents the Kullback-Leibler divergence, which measures the difference in motion characteristics between two targets;

[0140] Represents cosine similarity, which measures the similarity of two targets in heading.

[0141] Step S2022D: Perform ID association on the distance-associated targets; wherein each target is configured with a unique ID.

[0142] Determine whether the IDs of the two targets that are similar in distance and associated are the same. If they are the same, associate the two IDs.

[0143] Step S2022E: Fuse the video perception data and radar perception data of the same target after ID association to obtain radar-visual fusion data.

[0144] In one example, data fusion can be performed using methods such as weighted averaging, Kalman filtering, and particle filtering. Weighted averaging is a simple and intuitive data fusion method. It assigns a weight to each sensor's data and then calculates the weighted average of these data to produce the final fusion result. Weights are typically determined based on sensor reliability or accuracy, but this approach fails to consider how data changes over time and may be inaccurate for dynamic systems.

[0145] Therefore, the embodiment of the present application preferably uses the Kalman filter for fusion. The Kalman filter is a recursive algorithm used to estimate the state of a system in the presence of noise. It can handle linear systems and considers the state information of the previous moment in both the prediction and update steps.

[0146] Specifically, before fusion, the radar perception data of target x can be expressed as R x =[p x ,θ x 、v x 、a x ], video perception data is represented as C x =[p x ,θ x 、v x 、a x ]; the fused data is represented by F x =[p x ,θ x 、v x 、a x ]; where p x is the target position; θ x is the heading angle of the target; v x is the speed of the target; a x is the acceleration of the target.

[0147] Step S2023: Based on the traffic light perception data and the lane channelization data at the target single intersection, the radar-visual fusion data is corrected to obtain the multi-source fusion data of the target single intersection.

[0148] In this step, traffic light perception data includes traffic light status and countdown information; lane channelization data includes the direction of each lane (such as straight, left turn, right turn), allowed vehicle types, lane width and other information.

[0149] For example, if the heading angle in the fused data indicates a left turn, but lane channeling data indicates the vehicle is actually in the through lane, the heading angle can be corrected. Another example is when the traffic light is red, a vehicle should not proceed through an intersection. If the fused data indicates the target position has already passed the intersection, this data may be erroneous and requires correction.

[0150] Through the above embodiments, a small amount of data in the overlapping area can be used to quickly associate and fuse radar perception data and video perception data, thereby improving processing efficiency.

[0151] It should be noted that outside the overlapping area, the fused data in the overlapping area can be used for continuous tracking. Specifically, outside the overlapping area, the R x or C x , then the F in the overlapping area x As a result, the target is continuously tracked outside the overlapping area.

[0152] Based on the above embodiment, performing ID association on the distance-associated target includes:

[0153] Step S2022E1: ID tracking is performed on the same target based on the video perception data collected at multiple consecutive moments, and the video tracking IDs of all targets in the overlapping area detected at the current moment are obtained.

[0154] In a specific example, the position of the target in the current frame is predicted based on the position and heading angle of the target in the previous frame image, and the predicted position of the target in the current frame is compared one by one with the actual positions of all targets on the current frame. If the error between the actual position of at least one target and the position of the target predicted in the previous frame is less than a set threshold, it is determined that the target of the previous frame appears in the current frame, and the same target on the current frame continues to use the ID of the target of the previous frame. If there is no target with a value less than the set threshold, it is considered that the target of the previous frame has disappeared and does not appear in the current frame, and the ID of the target of the previous frame is no longer used.

[0155] Step S2022E2: ID tracking is performed on the same target based on the radar sensing data collected at multiple consecutive moments, and the radar tracking IDs of all targets in the overlapping area detected at the current moment are obtained.

[0156] In one feasible implementation, the process of determining the radar tracking IDs of all targets within the overlapping area detected at the current moment includes:

[0157] Step A: Obtain the radar perception data at the previous moment. The radar perception data includes the target's position, velocity, acceleration, and heading angle.

[0158] Step B: Input the target's position, velocity, acceleration, and heading angle into a pre-built target motion prediction model to obtain the predicted position of the target at the current moment.

[0159] In one example, let the radar perception data of target i at time k-1 be Includes: Location speed acceleration Heading angle

[0160] Radar perception data of the jth target detected at the current k moment The real location The sampling time interval is Δt.

[0161] The target motion prediction model is:

[0162]

[0163] Where, is the predicted position of target i at the current moment. If the acceleration is 0, the above target motion prediction model degenerates into a uniform velocity model.

[0164] Step C: Calculate the difference between the predicted position of the target at the current moment and the actual position of the target.

[0165] In this step, the Euclidean distance between the predicted location and the actual location is first calculated:

[0166]

[0167] Then the predicted position of each target detected at the previous moment is matched one by one with the actual position of all targets detected at the current moment to calculate the distance value d ij ;

[0168] All d ij Composition cost matrix Among them, m is the number of predicted targets; n is the number of actual detected targets.

[0169] Step D: Compare the difference value with the preset difference threshold. If it is less than or equal to the preset difference threshold, it is determined to belong to the same target, and the target ID of the previous moment is used; if it is greater than the preset difference threshold, it is determined to belong to a new target, and a new target ID is assigned.

[0170] Continuing from the previous example, based on the cost matrix, the Hungarian algorithm is used for target matching. If d ij ≤D th , then it is believed that and For the same goal. Among them, D th To preset the difference threshold, the Hungarian algorithm is used to find the global optimal match to ensure the minimum total cost.

[0171] like and If the match is successful, the target ID at time k-1 will be used.

[0172] It should be noted that if the predicted target position is not matched for N consecutive moments, the target is removed from the cache.

[0173] Step S2022E3: ID association is performed on targets with the same ID based on the video tracking ID and the radar tracking ID.

[0174] This application is based on the characteristics of millimeter-wave radar, which has a long perception distance and is easy to deploy across the entire area. It uses the perception data of millimeter-wave radar to realize traffic perception, which is more comprehensive and accurate than the perception of conventional visual perception and positioning modules. However, although millimeter-wave radar has accurate perception of speed, it has weak perception of static targets. This is because millimeter-wave radar works based on the principle of Doppler effect and inherently filters stationary targets, so millimeter-wave radar does not have the ability to detect static targets. However, when using millimeter-wave radar for target detection, there are often situations where moving objects stop and continue to move. If the targets in this situation are directly filtered, it will affect the tracking of subsequent IDs. In addition, when identifying targets, there will be false target phenomena such as target afterimages and target flickering (short-term intermittent presence or absence of coordinate point data), which affect the accuracy of subsequent data processing.

[0175] Therefore, based on the above embodiment, after obtaining the radar sensing data, the method further includes:

[0176] Step S206: performing data preprocessing on the radar sensing data, where the data preprocessing at least includes coordinated processing of target stillness and target afterimage, and target flickering processing.

[0177] In one feasible implementation, the collaborative processing of target stillness and target afterimage is as follows:

[0178] Step S2061: Obtain the target's stationary duration.

[0179] Step S2062: Compare the stationary time with the preset parking holding time. The parking holding time is determined as follows:

[0180] Get the distance between the target and the millimeter-wave radar sensor;

[0181] If the distance is less than the preset distance threshold, the parking holding time is fixed;

[0182] If the distance is greater than the preset distance threshold, the parking hold time is determined by dividing the maximum detection distance of the millimeter-wave radar sensor by the preset adjustable distance parameter;

[0183] Step S2063: If the time is less than or equal to the preset parking holding time, the radar sensing data and target ID of the target are kept.

[0184] Step S2064: If the parking holding time is greater than the preset time, the radar sensing data and target ID of the target are deleted.

[0185] After a moving target stops within the preset distance threshold of the millimeter-wave radar, the radar device will retain the target's latitude and longitude coordinates and target ID at the last moment and upload them to the acquisition and control platform. Afterwards, if the target stops for within the preset stop hold time (default 120 seconds, adjustable), the radar will continue to report with the retained ID and track the target's latitude and longitude changes in real time until the target leaves the radar's detection range. If the target stops for more than X seconds, the radar will stop reporting the target's ID and latitude and longitude and discard the latitude, longitude and ID data.

[0186] If the moving target stops within the preset distance threshold from the millimeter-wave radar, the corresponding holding process will be performed, but the holding time will not be too long. This is because the detection rate of far-end targets is lower than that of near-end targets. To prevent the target from leaving but not being detected by the radar and generating a large number of afterimages, in order to avoid the afterimage phenomenon, a parking holding algorithm will be performed when the radar detects the direction of the far end and stops. The specific parking holding time is T d It is inversely proportional to the distance S between the target and the radar and is not a constant.

[0187] The algorithm can be expressed as:

[0188]

[0189] Where L is the maximum detection distance of the radar; C is the preset adjustable distance parameter; and X is the fixed parking holding time.

[0190] The target afterimage phenomenon occurs when a moving object stops and leaves at a certain moment, but the radar does not detect the target movement. In this case, the actual object has left, but the radar will continue to report a target information. In the data acquisition and control platform, it will appear as a stuck target, referred to as afterimage. The target afterimage processing mechanism is designed to compensate for the target sticking problem that may be caused by the target stop processing mechanism. The method to deal with target afterimage is: all stopped targets of the target stop processing module are stopped for more than T d After the time, the radar immediately stops reporting the target information.

[0191] This embodiment coordinates the stop hold function with the afterimage elimination function, and obtains the optimal stop hold time through the above algorithm according to the distance between the target and the radar, thereby effectively improving the radar's accuracy in judging the target stop event and eliminating the target afterimage.

[0192] When the millimeter-wave radar reports data, the target information contained in some data frames within the same second is incomplete, causing the display of the target to appear to "flicker" intermittently.

[0193] In one feasible implementation, the method for processing target flickering is as follows:

[0194] First, determine the number of targets detected at the current moment;

[0195] If the number of targets is less than the set threshold, all target information will be reported;

[0196] When the number of targets exceeds the set threshold, the target data will be stopped and streamlined, and the original speed, acceleration, longitude and latitude, heading angle and other information data will be streamlined into a maintenance mark, but will not be discarded.

[0197] By using the above judgment and streamlining strategies, we can not only reduce the length of data transmission, but also ensure that information is not lost. This not only ensures the integrity of the target data, but also reduces the data pressure in the case of multiple targets.

[0198] In one feasible implementation, the method for determining vehicle traffic flow is as follows:

[0199] Step S2051: When executing the issued timing plan, multi-source fusion data of a single intersection is acquired once at a preset interval.

[0200] The process of multi-source data fusion at a single intersection is described in the above embodiment and will not be repeated here.

[0201] Step S2052: fusing the multi-source fusion data of multiple single intersections to obtain regional-level multi-source fusion data.

[0202] In a specific embodiment, the district-level fusion process is as follows:

[0203] First, determine the correlation between intersections;

[0204] In one example, the formula for the correlation is:

[0205]

[0206] Where A mn Represents the correlation between intersection m and intersection n; Indicates the time when the vehicle arrives at intersection n; represents the time when the vehicle departs from intersection m; τ road represents the theoretical travel time of the road section; σ t represents the time standard deviation; v m , v n are the average vehicle speeds at intersection m and intersection n respectively.

[0207] Then, the data of the intersections with a correlation degree greater than a preset correlation threshold are fused.

[0208] In one example, an average weighted fusion approach may be adopted.

[0209] If there is a correlation between two intersections A and B, it means that the vehicle will pass through intersection B from intersection A. If there is no correlation, it means that the vehicle will not pass through intersection B, then there is no need to fuse them, thereby improving the accuracy of data fusion.

[0210] Step S2053: Fusing the regional-level multi-source fusion data to obtain the city-level multi-source fusion data.

[0211] In this step, the correlation between the areas can also be determined first, that is, whether the vehicle will go from one area to another area can be determined.

[0212] Step S2054: Determine the vehicle flow rate at the target single intersection based on the city-level multi-source fusion data.

[0213] In a feasible implementation, the traffic flow at the target single intersection can be determined based on the fusion data of the target single intersection and the pre-trained traffic prediction model, and the traffic flow at the city level can be determined based on the multi-source fusion data at the city level; the ratio of the traffic flow at the single intersection to the traffic flow at the city level is calculated, and the ratio is determined as a degree of smoothness in the range of 0-1, and it can also be converted into a score representation of 0-10 or 0-100 as needed.

[0214] In one example, the evaluation scores based on city-level multi-source data are:

[0215]

[0216] Among them, Q single is the traffic flow at a single intersection; Q city It is city-level traffic volume.

[0217] This embodiment of the present invention utilizes city-level multi-source fusion data to assess the traffic flow at a specific target intersection. Using city-wide data as a background, combined with data from a specific intersection, the system evaluates traffic flow at that intersection, generating a traffic flow assessment based on city-level traffic conditions, resulting in a more accurate assessment.

[0218] In another embodiment, inspection data of a single intersection can be obtained through drone inspection, the timing plan can be evaluated based on the inspection data, and the timing plan can be comprehensively and collaboratively evaluated based on the evaluation results of the drone and the evaluation results based on multi-source fusion data.

[0219] In a specific embodiment, the drone inspection data includes the vehicle queue length L and the vehicle density D. The smoothness score of the timing plan is calculated by the following formula:

[0220]

[0221] Where β is the data weight of the UAV, which is dynamically adjusted with the flight altitude and lighting conditions; L max is the preset maximum queue length; D max The preset maximum vehicle density.

[0222] Based on the evaluation score S of the drone coreUVN And the evaluation score S based on city-level multi-source data coreGround The final comprehensive score is obtained. The comprehensive scoring formula is as follows:

[0223] S coreTotal =γ·S coreGround +(1-γ)S coreUVN (10);

[0224] Where γ is the ground-hole data balance coefficient, which is adaptively adjusted based on the intersection type and traffic scenario.

[0225] The embodiment of the present invention utilizes a low-altitude drone to observe the traffic conditions at the intersection from the air, and automatically performs secondary fusion with the Leishixin fusion evaluation data, thereby making the assessment of the smoothness of the intersection more accurate, and thus also improving the accuracy of the assessment of the timing plan.

[0226] Based on the same technical concept, an embodiment of the present invention further provides an electronic device, such as Figure 4As shown, it includes a processor 401 , a communication interface 402 , a memory 403 and a communication bus 404 , wherein the processor 401 , the communication interface 402 and the memory 403 communicate with each other via the communication bus 404 .

[0227] Memory 403, used for storing computer programs;

[0228] The processor 401 is configured to implement the steps of the traffic signal intelligent control method when executing the program stored in the memory 403 .

[0229] The communication bus mentioned in the electronic device mentioned above may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used in the figure, but this does not mean that there is only one bus or only one type of bus.

[0230] The communication interface is used for communication between the above electronic device and other devices.

[0231] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage. Alternatively, the memory may be at least one storage device located away from the processor.

[0232] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0233] The computer program product for intelligent traffic signal control provided in the embodiment of the present invention includes a computer-readable storage medium storing program code. The instructions included in the program code can be used to execute the method described in the previous method embodiment. The specific implementation can be found in the method embodiment and will not be repeated here.

[0234] The system for the intelligent traffic signal control method provided in the embodiments of the present invention can be specific hardware on a device or software or firmware installed on the device. The implementation principles and technical effects of the system provided in the embodiments of the present invention are the same as those of the aforementioned method embodiments. For the sake of brief description, any matters not mentioned in the system embodiments can be referred to the corresponding content in the aforementioned method embodiments. Those skilled in the art will clearly understand that for the convenience and brevity of description, the specific working processes of the systems, systems, and units described above can all refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0235] In the embodiments provided by the present invention, it should be understood that the disclosed systems and methods can be implemented in other ways. The system embodiments described above are merely illustrative. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be through some communication interface, the indirect coupling or communication connection of the system or unit may be electrical, mechanical or other forms.

[0236] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of these units may be selected to achieve the purpose of this embodiment according to actual needs.

[0237] In addition, each functional unit in the embodiment provided by the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0238] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0239] It should be noted that similar numbers and letters represent similar items in the following figures. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. In addition, the terms "first", "second", "third", etc. are only used to distinguish the description and are not to be understood as indicating or implying relative importance.

[0240] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or replace some of the technical features therein with equivalents. However, such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. They should all be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A traffic signal intelligent control method, characterized in that: The method comprises: Acquire radar perception data, video perception data, and traffic light perception data within the monitoring range of the target single intersection collected by multiple sensors at preset intervals; Fusing the radar perception data, the video perception data, and the traffic light perception data to obtain multi-source fusion data of the target single intersection; Inputting the multi-source fusion data into a pre-trained timing plan generation model to obtain a timing plan for the traffic light at the target single intersection; Performing simulation verification on the timing plan, and sending it to the signal control machine of the target single intersection after passing the verification; The vehicle flow rate at the target single intersection is determined based on the issued timing plan, and the timing plan is optimized based on the vehicle flow rate.

2. The method according to claim 1, characterized in that The fusing of the radar perception data, the video perception data, and the traffic light perception data to obtain the multi-source fusion data of the target single intersection includes: Time-synchronizing the radar perception data and the video perception data; Preliminarily fusing the synchronized radar perception data with the video perception data to obtain radar-visual fusion data; The radar-visual fusion data is corrected based on the traffic light perception data and the lane channelization data at the target single intersection to obtain multi-source fusion data of the target single intersection.

3. The method according to claim 2, characterized in that The preliminarily fusing the synchronized radar perception data with the video perception data to obtain radar-visual fusion data includes: Performing coordinate conversion on the video perception data from an image coordinate system to a world coordinate system so as to be in the same coordinate system as the radar perception data; Performing ROI filtering on the radar perception data and the video perception data after coordinate conversion to determine an overlapping area between the two; Associating, at a current moment, a target detected based on the radar perception data within the overlapping area and a target detected based on the video perception data within the overlapping area in terms of distance; the target being a vehicle; Perform ID association on the targets after distance association; wherein each target is configured with a unique ID; The video perception data of the same target after ID association is fused with the radar perception data to obtain radar-visual fusion data.

4. The method according to claim 3, characterized in that The ID association of the distance-associated target includes: Based on the video perception data collected at multiple consecutive moments, the same target is tracked by ID, and the video tracking IDs of all targets in the overlapping area detected at the current moment are obtained; Based on the radar sensing data collected at multiple consecutive moments, the same target is tracked by ID, and the radar tracking IDs of all targets in the overlapping area detected at the current moment are obtained; Based on the video tracking ID and the radar tracking ID, targets with the same ID are associated with each other.

5. The method according to claim 4, characterized in that The process of determining the radar tracking IDs of all targets within the overlapping area detected at the current moment includes: Acquire radar sensing data at the last moment, wherein the radar sensing data includes the position, velocity, acceleration, and heading angle of the target; Inputting the position, velocity, acceleration and heading angle of the target into a pre-built target motion prediction model to obtain the predicted position of the target at the current moment; Calculate the difference between the predicted position of the target at the current moment and the actual position of the target; The difference value is compared with the preset difference threshold. If it is less than or equal to the preset difference threshold, it is determined to belong to the same target, and the target ID of the previous moment is used; if it is greater than the preset difference threshold, it is determined to belong to a new target, and a new target ID is assigned.

6. The method according to claim 1, characterized in that After acquiring the radar perception data, the method further includes: The radar sensing data is preprocessed, and the data preprocessing at least includes coordinated processing of target stillness and target afterimage, and target flickering processing.

7. The method according to claim 6, characterized in that The collaborative processing process of target stillness and target afterimage is as follows: Get the target's stationary time; Comparing the stationary time with a preset parking holding time; wherein the process of determining the parking holding time is as follows: Get the distance between the target and the millimeter-wave radar sensor; If the distance is less than the preset distance threshold, the parking holding time is fixed; If the distance is greater than the preset distance threshold, the parking holding time is determined by dividing the maximum detection distance of the millimeter wave radar sensor by a preset adjustable distance parameter; If the target is less than or equal to the preset parking holding time, the radar sensing data and target ID of the target are kept; If it is greater than the preset parking holding time, the radar sensing data and target ID of the target are deleted.

8. The method according to claim 4, characterized in that Determining the vehicle flow rate at the target single intersection based on the issued timing plan includes: When executing the issued timing plan, multi-source fusion data of a single intersection is acquired once at a preset interval; The multi-source fusion data of multiple single intersections are fused to obtain the regional multi-source fusion data; Fusing regional-level multi-source fusion data to obtain city-level multi-source fusion data; The vehicle flow rate at the target single intersection is determined based on the city-level multi-source fusion data.

9. A traffic signal intelligent control system, characterized in that: The system includes: intelligent signal perception module, real-time traffic flow perception module, video perception module, high-precision map perception module, multi-source data fusion module, data analysis and signal control module and evaluation and optimization module; The intelligent signal sensing module is used to obtain the status of the traffic light at the target single intersection; The real-time traffic flow perception module is used to obtain radar perception data of the target single intersection and preprocess the radar perception data; The video perception module is used to obtain video perception data of the target single intersection; The high-precision map perception module is used to establish a three-dimensional simulated urban model and map the target objects detected by radar perception data into the virtual scene of the three-dimensional simulated urban model; The multi-source data fusion module is used to fuse the radar perception data, video perception data and traffic light perception data to obtain multi-source fusion data of the target single intersection; The data analysis and signal control module is used to input the multi-source fusion data into a pre-trained timing plan generation model to obtain a timing plan for the traffic lights at the target single intersection, simulate and verify the timing plan, and send it to the signal control machine at the target single intersection after passing the verification; The evaluation and optimization module is used to determine the vehicle flow rate at the target single intersection based on the issued timing plan, and optimize the timing plan based on the vehicle flow rate.

10. The system according to claim 9, characterized in that The intelligent signal perception module also includes an operation status monitoring unit, The operation status monitoring unit is used to sense the working status of the traffic light signal in real time, analyze and calculate the working status of the signal control machine through the built-in algorithm, generate the working health data of the signal control machine, and finally transmit it to the operation and maintenance platform through the communication interface unit.

Images