Adaptive perception interaction traffic signal electromechanical control method and system

Abstract

The invention discloses a traffic signal electromechanical control method and system based on adaptive perception interaction, and belongs to the technical field of traffic management. Self-adaptive control of traffic signals is realized through four core steps of lane coding and convergence behavior initialization, convergence state matrix construction and path locking, periodic congestion degree calculation, response efficiency evaluation and signal adjustment, and the method comprises the following steps: coding lanes according to the direction of the entrance and exit of an intersection, defining a convergence instruction and building an initial convergence set; identifying and matching a license plate lock vehicle path by using an image device to construct a matrix, counting a confluence congestion degree in a period, evaluating the response efficiency of a confluence instruction, reducing the time length of an efficiency instruction signal, and then recovering to an initial value; the problems that existing fixed period control cannot adapt to traffic flow changes and response lags are solved, through accurate path locking and dynamic signal adjustment, the intersection passing efficiency is improved, and the method can adapt to complex intersections with multiple entrances and multiple exits so as to reduce vehicle idling emission and meet the requirement for green traffic development.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of traffic management, in particular to an electric control method and system of a traffic signal machine with adaptive perception interaction. BACKGROUND

[0002] With the acceleration of urbanization, the traffic volume at urban intersections continues to grow, and the rationality of traffic signal control directly affects the efficiency of traffic operation. The current mainstream traffic signal control mode has the following key problems: first, the fixed cycle timing mode lacks flexibility. This mode sets a fixed signal duration based on historical traffic data, which cannot respond to real-time traffic fluctuations. For example, during the morning rush hour, a fixed duration can easily lead to vehicle queuing congestion. During the flat peak period, the long signal duration also wastes road resources. According to statistics, the loss of intersection traffic efficiency under fixed cycle control can reach 20-25%. Second, the traditional adaptive control has weak path recognition ability. Existing adaptive control relies on devices such as coils and radars to detect the total amount of traffic, which cannot distinguish the differences in traffic from different entry directions converging into the same exit direction. For example, a west exit of a cross intersection may receive traffic from an east entry straight, a south entry left turn, and a north entry left turn. The traditional control only counts the total traffic at the west exit and cannot locate the specific congestion convergence direction, resulting in inaccurate congestion calculation and poor targeted adjustment strategy. In addition, the multi-convergence direction adaptation capability of complex intersections (such as five-way intersections and irregular intersections) is insufficient. Existing solutions need to redesign detection logic for different intersections, which has poor universality and high deployment cost. These problems collectively result in low traffic efficiency and frequent congestion at urban intersections, restricting the smooth operation of urban traffic. SUMMARY

[0003] The purpose of the present application is to provide an electric control method and system of a traffic signal machine with adaptive perception interaction to solve the problems raised in the background.

[0004] To solve the above technical problems, the present application provides the following technical solutions:

[0005] An electric control system of a traffic signal machine with adaptive perception interaction, the system comprises: an adaptive perception module, and a lane and convergence management module, a convergence behavior state matrix construction and path locking module, a convergence congestion calculation module, a response efficiency evaluation and signal adjustment module electrically connected with the adaptive perception module;

[0006] The adaptive perception module is used to coordinate the working timing of each functional module, realize the real-time interaction of data and the synchronization of function execution between modules, and ensure the orderly advancement of the overall control process.

[0007] The lane and convergence management module is used to complete lane coding, convergence instruction generation, and initialization of convergence behavior set construction.

[0008] The confluence behavior state matrix construction and path locking module is used for building the confluence behavior state matrix and locking the vehicle passing path through image acquisition and license plate matching.

[0009] The confluence congestion degree calculation module is used for counting the number of vehicles in a period, assigning values to the matrix and calculating the confluence congestion degree.

[0010] The response efficiency evaluation and signal adjustment module is used for evaluating the response efficiency of the confluence instruction, screening the low-efficiency instruction and dynamically adjusting the signal duration.

[0011] As a preferred scheme of the present application, the lane and confluence management module comprises a lane coding unit, a confluence instruction generation unit and an initialization confluence behavior set construction unit.

[0012] The lane coding unit is used for uniquely coding each lane and adding a direction identifier according to the actual distribution of the entrance direction and the exit direction of the intersection, so as to ensure the accuracy of lane positioning.

[0013] The confluence instruction generation unit is used for identifying the confluence behavior from a single entrance direction to a single exit direction, and assigning a unique confluence instruction to each type of confluence behavior.

[0014] The initialization confluence behavior set construction unit is used for counting the number of all entrance directions corresponding to each exit direction, integrating the confluence instructions and forming an initialization confluence behavior set.

[0015] As a preferred scheme of the present application, the confluence behavior state matrix construction and path locking module comprises a matrix construction unit, an image acquisition and data interaction unit and a license plate recognition and matching unit.

[0016] The matrix construction unit is used for building the confluence behavior state matrix based on the initialization confluence behavior set, taking the lane number of the entrance participating in the confluence as the row index and the exit lane number as the column index.

[0017] The image acquisition and data interaction unit is used for acquiring the license plate images through the image acquisition devices arranged in the entrance and exit directions, and realizing the real-time image data interaction and sharing between the two types of devices through the short-distance Internet of Things.

[0018] The license plate recognition and matching unit is used for recognizing the license plate images acquired at the entrance and matching the license plate images acquired at the exit, so as to lock the passing path of the vehicle from the entrance lane to the exit lane.

[0019] As a preferred scheme of the present application, the confluence congestion degree calculation module comprises a period vehicle counting unit, a matrix assignment unit and a congestion degree calculation unit.

[0020] The periodic vehicle statistical unit is configured to count the number of vehicles entering corresponding exit lanes from each entry lane in units of a traffic signal cycle, thereby ensuring the timeliness of the statistical data.

[0021] The matrix assignment unit is configured to assign the counted number of vehicles to corresponding positions of the merging behavior state matrix, thereby generating a merging behavior state matrix of the current cycle.

[0022] The congestion degree calculation unit is configured to calculate the merging congestion degree corresponding to each merging instruction based on the merging behavior state matrix of the current cycle, thereby representing the degree of traffic concentration.

[0023] As a preferred scheme of the present application, the response efficiency evaluation and signal adjustment module comprises a response efficiency calculation unit, a low-efficiency merging instruction screening unit and a signal duration adjustment unit.

[0024] The response efficiency calculation unit is configured to calculate the response efficiency of the merging instruction based on the merging congestion degrees of the same merging instruction in adjacent two traffic signal cycle indication periods, thereby reflecting the feedback effect of traffic changes on the signal.

[0025] The low-efficiency merging instruction screening unit is configured to compare the response efficiencies of all merging instructions, screen out the merging instruction with the lowest response efficiency, and determine the target to be optimized.

[0026] The signal duration adjustment unit is configured to prolong the initial signal duration of the low-efficiency merging instruction in the first subsequent traffic signal cycle indication period, and the prolonged duration does not exceed the preset maximum signal duration, and the initial signal duration is restored in the second subsequent cycle, thereby avoiding new congestion caused by excessive adjustment.

[0027] An adaptive perception interaction traffic signal machine electric control method, the method comprises the following steps:

[0028] Step S1: According to the entry direction and exit direction of the intersection, each lane is coded and direction mark is attached, the merging behavior from a single entry direction to a single exit direction is determined to correspond to the merging instruction, the number of entry directions corresponding to each exit direction is counted, and an initial merging behavior set is constructed.

[0029] Step S2: Based on the constructed initial merging behavior set, a merging behavior state matrix is built, the row index corresponds to the lane number of each entry direction participating in merging, and the column index corresponds to the lane number of each exit direction; through the image acquisition devices arranged at the entry direction and exit direction of the intersection, license plate images are collected respectively, image data interaction sharing between the two types of image acquisition devices is realized, and through license plate recognition and matching operation, the traffic path of vehicles merging from entry lane to exit lane is locked.

[0030] Step S3: Taking one traffic signal cycle indication period as a statistical period, the number of vehicles entering the corresponding exit lane from each entrance lane is counted, the statistical result is assigned to the corresponding position of the merging behavior state matrix, the merging behavior state matrix is generated to evaluate the merging congestion degree corresponding to each merging instruction;

[0031] Step S4: In the adjacent two traffic signal cycle indication periods, the response efficiency of the merging instruction is evaluated according to the merging congestion degrees in the two periods under the same merging instruction; the merging instruction with the lowest response efficiency is screened out, the initialization signal duration thereof is extended in the subsequent first traffic signal cycle indication period, and the initialization signal duration thereof is restored in the subsequent second traffic signal cycle indication period.

[0032] As a preferred scheme of the present application, the specific implementation process of step S1 comprises:

[0033] Based on the entrance direction and the exit direction of the intersection, the lanes are coded and direction marks are attached, and based on the lane merging indication, the merging behavior from one entrance direction to one exit direction is recorded as one merging instruction;

[0034] Based on the merging instruction, the number of entrance directions when different entrance directions enter the same exit direction at the intersection is counted, and an initialization merging behavior set is formed, denoted as , wherein, represents the i-th exit direction at the intersection, represents the e-th entrance direction at the intersection, represents the exit direction corresponding to the generated initialization merging behavior set, represents the entrance direction participating in the merging of the exit direction , and E represents the number of entrance directions at the intersection.

[0035] As a preferred scheme of the present application, the specific implementation process of step S2 comprises:

[0036] Based on the initialization merging behavior set, a merging behavior state matrix is constructed, the row index of the merging behavior state matrix corresponds to the lane number in the participating lane set of the entrance direction, and the column index of the merging behavior state matrix corresponds to the lane number in the exit direction, so that the merging behavior state matrix between the entrance direction and the exit direction is generated, and the matrix position index of the x-th row and the y-th column in the merging behavior state matrix is denoted as , wherein, represents the x-th lane in the entrance direction , and represents the y-th lane in the exit direction The y-th lane;

[0037] Image acquisition devices are installed at intersections to capture license plate images. Devices deployed at the entrance are used to identify license plate images, while those at the exit are used to match them. The image data is shared between the entrance and exit devices via short-range Internet of Things (SIoT). (The text also mentions lanes at the entrance.) Vehicles in the middle merge into the lane towards the exit. In the middle, image acquisition devices at the entrance and exit directions are used to identify and match license plate images to determine the vehicle's travel path and lane. Merging lane .

[0038] As a preferred embodiment of the present invention, the specific implementation process of step S3 includes:

[0039] Within a traffic signal cycle, statistics are compiled from lanes. Merging lane The number of vehicles locked at the time, and the matrix position index. Assignment, denoted as Based on the assignment results of the matrix position indices, a bus behavior state matrix for a traffic signal cycle is generated, denoted as... Where t is the sequence number of the traffic signal cycle indication period. Indicates the direction of the entrance Pointing towards the exit The confluence command;

[0040] Within one traffic signal cycle, traffic flows from different entrance directions into the same exit direction. At that time, based on the state matrix of the sink behavior Statistical confluence command Convergence congestion In the formula, I represents the total number of export directions.

[0041] As a preferred embodiment of the present invention, the specific implementation process of step S4 includes:

[0042] Based on merging congestion levels, merging instructions are evaluated during two adjacent traffic signal cycles. Response efficiency ,in, This indicates the merging instruction within the (t-1)th traffic signal cycle. The degree of congestion at the confluence;

[0043] Select the minimum response efficiency corresponding to the convergence instruction, and control the initialization signal duration of the extended convergence instruction in the t+1th traffic signal cycle indication period, and the signal duration of the extended convergence instruction does not exceed the maximum value of the preset signal duration, and restores to the initialization signal duration of the convergence instruction in the t+2th traffic signal cycle indication period.

[0044] Compared with the prior art, the beneficial effects achieved by the present application are:

[0045] Through the coding rule of "entry direction-lane type-number", a unique identifier is assigned to each lane, and the corresponding relationship of the convergence instruction (entry→exit) is clear; the number of entries of the exit direction is counted to construct the initialization convergence set, solving the problem of fuzzy correspondence between lanes and convergence directions in traditional control, accurately positioning the specific congestion convergence direction, laying a foundation for subsequent accurate analysis, for example, different convergence instructions such as "east straight-west straight" and "south left-west left" can be clearly distinguished in complex intersections, avoiding confusion;

[0046] Based on the initialization convergence set, a matrix is constructed to convert the abstract convergence behavior into a visual matrix structure, and through the flexible construction of the initialization convergence behavior set, it can adapt to complex intersections such as four-way and five-way intersections without the need to redevelop for different intersections; through the short-distance Internet of Things interaction data of the entry and exit image devices, the vehicle path is locked by means of license plate recognition and matching, ensuring that the convergence trajectory of each vehicle is traceable, so as to overcome the defect of traditional detection "only total quantity, not path", and make the traffic flow state analysis from "macro" to "micro";

[0047] Taking the traffic signal cycle period as the statistical unit, the number of vehicles is assigned to the corresponding position of the matrix to generate a period matrix; through the basic logic of "single convergence instruction vehicle number / total vehicle number of all instructions", the congestion degree is calculated, the abstract traffic flow is converted into a quantitative index of 0-1, and the density of each convergence direction is intuitively reflected, providing data support for adjusting strategy and avoiding the subjectivity of traditional experience judgment;

[0048] The response efficiency is calculated by the difference value of the congestion degree of adjacent periods, the low-efficiency instruction is screened, and the signal duration is dynamically adjusted, such as extending the duration by no more than 1.2 times of the initialization duration, and restoring to the initial value in the next period, and then, the congestion is quickly relieved, and the other direction congestion caused by excessive adjustment is avoided, and the overall traffic balance of the intersection is ensured. BRIEF DESCRIPTION OF DRAWINGS

[0049] The accompanying drawings are used to provide a further understanding of the present application, and constitute a part of the specification, together with the embodiments of the present application, to explain the present application, and do not constitute a limitation on the present application.

[0050] Figure 1is a step schematic diagram of an adaptive perception interaction traffic signal machine electric control method. DETAILED DESCRIPTION

[0051] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of protection of the present application.

[0052] In the first embodiment: an adaptive perception interaction traffic signal machine electric control system is provided, which comprises an adaptive perception module, a lane and merging management module, a merging behavior state matrix construction and path locking module, a merging congestion degree calculation module, and a response efficiency evaluation and signal adjustment module electrically connected with the adaptive perception module;

[0053] The adaptive perception module is used for coordinating the working time sequence of each functional module, realizing real-time interaction of data between modules and synchronization of function execution.

[0054] The lane and merging management module is used for completing lane coding, merging instruction generation and initialization of merging behavior set construction.

[0055] The lane and merging management module comprises a lane coding unit, a merging instruction generation unit and an initialization merging behavior set construction unit.

[0056] The lane coding unit is used for uniquely coding each lane and adding a direction identifier according to the actual distribution of the entrance direction and the exit direction of the intersection.

[0057] The merging instruction generation unit is used for identifying the merging behavior from a single entrance direction to a single exit direction, and assigning a unique merging instruction to each type of merging behavior.

[0058] The initialization merging behavior set construction unit is used for counting the number of all entrance directions corresponding to each exit direction, integrating each merging instruction, and forming an initialization merging behavior set.

[0059] The merging behavior state matrix construction and path locking module is used for building a merging behavior state matrix and locking the vehicle passing path through image acquisition and license plate matching.

[0060] The merging behavior state matrix construction and path locking module comprises a matrix construction unit, an image acquisition and data interaction unit, and a license plate recognition matching unit.

[0061] A matrix construction unit is configured to build a merging behavior state matrix based on the initialized merging behavior set, with the lane number of the entry participating in the merging as the row index and the exit lane number as the column index.

[0062] An image acquisition and data interaction unit is configured to acquire license plate images through image acquisition devices arranged in the entry and exit directions, and to realize real-time image data interaction and sharing between the two types of devices through short-range Internet of Things.

[0063] A license plate recognition and matching unit is configured to recognize the license plate images acquired at the entry and match the license plate images acquired at the exit, so as to lock the passing path of the vehicle from the entry lane to the exit lane.

[0064] A merging congestion degree calculation module is configured to count the number of vehicles in a period, assign values to the matrix, and calculate the merging congestion degree.

[0065] The merging congestion degree calculation module includes a period vehicle counting unit, a matrix assignment unit, and a congestion degree calculation unit.

[0066] The period vehicle counting unit is configured to count the number of vehicles merging into the corresponding exit lane from each entry lane in one traffic signal cycle indication period.

[0067] The matrix assignment unit is configured to assign the counted number of vehicles to the corresponding position of the merging behavior state matrix, to generate the merging behavior state matrix of the current period.

[0068] The congestion degree calculation unit is configured to calculate the merging congestion degree corresponding to each merging instruction based on the merging behavior state matrix of the current period.

[0069] A response efficiency evaluation and signal adjustment module is configured to evaluate the response efficiency of the merging instruction, filter the low-efficiency instruction, and dynamically adjust the signal duration.

[0070] The response efficiency evaluation and signal adjustment module includes a response efficiency calculation unit, a low-efficiency merging instruction filtering unit, and a signal duration adjustment unit.

[0071] The response efficiency calculation unit is configured to calculate the response efficiency of the merging instruction according to the merging congestion degrees of the same merging instruction in two adjacent traffic signal cycle indication periods.

[0072] The low-efficiency merging instruction filtering unit is configured to compare the response efficiencies of all merging instructions and filter out the merging instruction with the lowest response efficiency.

[0073] The signal duration adjustment unit is configured to extend the initialization signal duration of the low-efficiency merging instruction in the first subsequent traffic signal cycle indication period, and the extended duration does not exceed the preset maximum signal duration, and the initialization signal duration is restored in the second subsequent period.

[0074] Please refer to Figure 1 In the second embodiment, an adaptive perception interaction traffic signal machine electrical control method is provided, which is applicable to the above-mentioned first embodiment. In this embodiment, Zhongshan Road and Renmin Road intersection in a core business district of a city is selected as an implementation scene. The intersection is a four-way intersection (east, west, south, and north four entry directions corresponding to four exit directions), which is a key node connecting the business district and the residential area. Each entry direction is provided with 3 lanes (left turn, straight, and right turn), and each exit direction is provided with 3 lanes (matching the entry lane type). The speed limit at the intersection is 50 km / h, the preset traffic signal cycle indicating period is 90 seconds, the initialization signal duration of each merging instruction is 25 seconds for left turn, 30 seconds for straight, and 20 seconds for right turn, and the preset maximum signal duration is 1.2 times the initialization duration (30 seconds for left turn, 36 seconds for straight, and 24 seconds for right turn). The image acquisition device is a 200 million pixel high-definition intelligent camera (frame rate 25 fps, license plate recognition accuracy ≥98%), which is arranged 5 meters in front of the entry lane stop line (to collect entry license plates) and 5 meters at the start of the exit lane (to collect exit license plates). The devices are connected through LoRa short-range Internet of Things (transmission rate 1 Mbps, delay ≤100 ms) to ensure real-time image data interaction.

[0075] The method comprises the following steps:

[0076] Step S1: According to the entry direction and exit direction of the intersection, each lane is coded and labeled with a direction, the merging behavior from a single entry direction to a single exit direction is matched with a merging instruction, and the number of entry directions corresponding to each exit direction is counted to construct an initialization merging behavior set.

[0077] Exemplarily, based on the entry direction and exit direction of the intersection, each lane is coded and labeled with a direction, and the merging behavior from one entry direction to one exit direction is recorded as one merging instruction based on the lane merging indication.

[0078] Based on the merging instruction, the number of entry directions merging into the same exit direction at the intersection is counted, and an initialization merging behavior set is constructed, recorded as wherein, represents the i-th exit direction at the intersection, represents the e-th entry direction at the intersection, represents the exit direction corresponding to the generated initialization merging behavior set, represents the entry direction participating in the merging of the exit direction , E represents the number of entry directions at the intersection.

[0079] For example, in lane coding, it can be coded according to the rule of "entry direction-lane type-number", such as "east-left-1" for the east entry left-turn 1 lane, "east-straight-2" for the east entry straight 2 lane, and "west-straight-2" for the west exit straight 2 lane;

[0080] In the generation of the merging instruction, the merging behavior of a single entry direction to a single exit direction is recorded as a merging instruction, such as the behavior of the east entry straight lane to the west exit straight lane is recorded as an "east-straight-west-straight" instruction;

[0081] In the initialization of the merging behavior set construction, the number of entry directions corresponding to each exit direction is counted, such as the west exit corresponding to the east entry (straight, left, right), the south entry (left), and the north entry (left) a total of 5 entry directions, and the initialization merging behavior set containing "east-straight-west-straight", "east-left-west-left", "east-right-west-right", "south-left-west-left", and "north-left-west-left" is constructed.

[0082] Step S2: Based on the constructed initialization merging behavior set, a merging behavior state matrix is built, and the row index corresponds to the lane number of each lane participating in merging of the entry direction, and the column index corresponds to the lane number of each lane of the exit direction; through the image acquisition devices arranged at the entry direction and the exit direction of the intersection, vehicle license plate images are collected, image data interaction sharing between the two types of image acquisition devices is realized, and through license plate recognition and matching operations, the passing path of the vehicle merging from the entry lane to the exit lane is locked;

[0083] For example, based on the initialization merging behavior set, a merging behavior state matrix is constructed, the row index of the merging behavior state matrix corresponds to the lane number of each lane in the participating lane set of the entry direction, and the column index of the merging behavior state matrix corresponds to the lane number of each lane in the exit direction, then the merging behavior state matrix between the entry direction and the exit direction is generated, then the matrix position index of the xth row and the yth column in the merging behavior state matrix is recorded as , wherein represents the xth lane in the entry direction , and represents the yth lane in the exit direction .

[0084] Through the image acquisition devices arranged at the intersection, vehicle license plate images are collected, wherein the image acquisition devices arranged at the entry direction are used to identify the license plate images, the image acquisition devices arranged at the exit direction are used to match the license plate images, and the image acquisition devices arranged at the entry direction and the exit direction realize image data interaction sharing through short-distance Internet of Things, the vehicle in the lane of the entry direction merges into the lane In the middle, image acquisition devices at the entrance and exit directions are used to identify and match license plate images to determine the vehicle's travel path and lane. Merging lane ;

[0085] For example, when constructing the matrix, the entrance lane number is used as the row index (1-3, corresponding to left, straight, and right), and the exit lane number is used as the column index (1-3). The matrix is ​​constructed for the "East Straight-West Straight" merging instruction, with row index 2 (East entrance straight lane) corresponding to column index 2 (West exit straight lane).

[0086] Step S3: Using one traffic signal cycle as the statistical period, count the number of vehicles merging from each entrance lane into the corresponding exit lane, assign the statistical results to the corresponding position in the merging behavior state matrix, generate the merging behavior state matrix, and evaluate the merging congestion level corresponding to each merging instruction.

[0087] For example, within a traffic signal cycle, statistics are collected from lanes. Merging lane The number of vehicles locked at the time, and the matrix position index. Assignment, denoted as Based on the assignment results of the matrix position indices, a bus behavior state matrix for a traffic signal cycle is generated, denoted as... Where t is the sequence number of the traffic signal cycle indication period. Indicates the direction of the entrance Pointing towards the exit The confluence command;

[0088] Within one traffic signal cycle, traffic flows from different entrance directions into the same exit direction. At that time, based on the state matrix of the sink behavior Statistical confluence command Convergence congestion In the formula, I represents the total number of export directions.

[0089] Step S4: Within two adjacent traffic signal cycle indication periods, evaluate the response efficiency of the merging instruction based on the merging congestion level of the same merging instruction in the two cycles; select the merging instruction with the lowest response efficiency, extend its initial signal duration in the first subsequent traffic signal cycle indication period, and restore it to the initial signal duration in the second subsequent traffic signal cycle indication period.

[0090] For example, based on the merging congestion level of the merging instruction, the merging instruction is evaluated during two adjacent traffic signal cycle indication periods. Response efficiency ,in, represents the merging instruction in the t-1th traffic signal cycle indication period a merging congestion degree of the merging instruction;

[0091] selects the merging instruction corresponding to the minimum response efficiency, and controls the extension of the initial signal duration of the merging instruction in the t+1th traffic signal cycle indication period, and the signal duration of the extended merging instruction does not exceed the maximum value of the preset signal duration, and returns to the initial signal duration of the merging instruction in the t+2th traffic signal cycle indication period.

[0092] It should be noted that in this paper, relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment.

[0093] Finally, it should be noted that: the above only for the preferred embodiments of the present application, and not for limiting the present application, although the present application is described in detail with reference to the foregoing embodiments, for those skilled in the art, it still can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features. Any modification, equivalent replacement, improvement, etc. within the spirit and principles of the present application shall be included in the protection scope of the present application.

Claims

1. A method for adaptive perception-interaction traffic signal machine electrical control, characterized in that, The method comprises the following steps: Step S1: According to the entry direction and the exit direction of the intersection, each lane is coded and direction marks are added, the confluence behavior of a single entry direction pointing to a single exit direction is determined to correspond to a confluence instruction, and the number of entry directions corresponding to each exit direction is counted to construct an initial confluence behavior set; Step S2: Based on the constructed initial confluence behavior set, a confluence behavior state matrix is built, and the row index corresponds to the lane number of each lane participating in confluence in the entry direction, and the column index corresponds to the lane number of each lane in the exit direction; through the image acquisition devices arranged in the entry direction and the exit direction of the intersection, license plate images are collected respectively, image data interaction and sharing between the two types of image acquisition devices are realized, and through license plate recognition and matching operations, the passing path of the vehicle from the entry lane to the exit lane is locked; Step S3: Taking one traffic signal cycle as a statistical cycle, the number of vehicles flowing into the corresponding exit lane from each entry lane is counted, the statistical result is assigned to the corresponding position of the confluence behavior state matrix, and the confluence behavior state matrix is generated to evaluate the confluence congestion degree corresponding to each confluence instruction; Step S4: In the adjacent two traffic signal cycle periods, the response efficiency of the confluence instruction is evaluated according to the confluence congestion degrees of the same confluence instruction in the two periods; the confluence instruction with the lowest response efficiency is selected, and its initial signal duration is extended in the subsequent first traffic signal cycle period, and is restored to the initial signal duration in the subsequent second traffic signal cycle period.

2. The adaptive perception interactive traffic signal machine electrical control method according to claim 1, wherein, The specific implementation process of step S1 comprises: Based on the entry direction and the exit direction of the intersection, the lanes are coded and direction marks are added, and the confluence behavior of one entry direction pointing to one exit direction is recorded as one confluence instruction based on the lane confluence instruction; Based on the convergence instruction, the number of entrance directions converging into the same exit direction at the intersection is counted, and an initial convergence behavior set is formed, denoted as wherein, represents the i-th exit direction at the intersection, represents the e-th entrance direction at the intersection, represents the exit direction corresponding to the generated initial convergence behavior set, represents the entrance direction The participating lane set composed of the lanes participating in the convergence of the exit direction at the intersection, and E represents the number of entrance directions at the intersection.

3. The adaptive perception interactive traffic signal machine electrical control method according to claim 2, wherein, The specific implementation process of step S2 comprises: Based on the initialized set of merging behaviors, a merging behavior state matrix is constructed, the row index of the merging behavior state matrix corresponds to each lane sequence in the participating lane set of the entrance direction, and the column index of the merging behavior state matrix corresponds to each lane sequence in the exit direction. The merging behavior state matrix between the entrance direction and the exit direction is generated, and the matrix position index of the xth row and yth column in the merging behavior state matrix is denoted as , wherein represents the xth lane in the entrance direction , and represents the yth lane in the exit direction . Image acquisition devices are installed at intersections to capture license plate images. Devices deployed at the entrance are used to identify license plate images, while those at the exit are used to match them. The image data is shared between the entrance and exit devices via short-range Internet of Things (SIoT). (The text also mentions lanes at the entrance.) Vehicles in the middle merge into the lane towards the exit. In the middle, image acquisition devices at the entrance and exit directions are used to identify and match license plate images to determine the vehicle's travel path and lane. Merging lane .

4. The adaptive perception interactive traffic signal machine electrical control method according to claim 3, wherein, The specific implementation process of step S3 comprises: The number of vehicles locked in each lane is counted in a traffic signal cycle and the matrix position index is assigned , denoted as . Based on the assigned matrix position index, a merging behavior state matrix in a traffic signal cycle is generated, denoted as , where t is the sequence number of the traffic signal cycle, represents the entry direction and the merging instruction pointing to the exit direction . In a traffic signal cycle, the merging behavior state matrix is based on the merging instruction of different entry directions into the same exit direction , and the merging congestion degree of the merging instruction is counted based on the merging behavior state matrix , wherein I represents the total number of exit directions.

5. The adaptive perception interactive traffic signal machine electrical control method according to claim 4, wherein, The specific implementation process of step S4 comprises: based on the merging instruction, a merging congestion degree of the merging instruction is evaluated during two adjacent traffic signal cycle indication periods response efficiency wherein represents a merging congestion degree of the merging instruction in a t-1th traffic signal cycle indication period ​ The confluence instruction corresponding to the minimum response efficiency is selected, and the initial signal duration of the confluence instruction is controlled to be extended in the t+1th traffic signal cycle period, and the signal duration of the extended confluence instruction does not exceed the maximum value of the preset signal duration, and is restored to the initial signal duration of the confluence instruction in the t+2th traffic signal cycle period.

6. An adaptive perception-interacted traffic signal machine electric control system, which executes an adaptive perception-interacted traffic signal machine electric control method according to any one of claims 1-5, characterized in that, The system comprises: an adaptive perception module, a lane and confluence management module, a confluence behavior state matrix construction and path locking module, a confluence congestion degree calculation module, a response efficiency evaluation and signal adjustment module electrically connected with the adaptive perception module; The adaptive perception module is used for coordinating the working time sequence of each functional module, realizing real-time interaction of data between modules and synchronization of function execution; The lane and confluence management module is used for completing lane coding, confluence instruction generation and initial confluence behavior set construction; The confluence behavior state matrix construction and path locking module is used for building the confluence behavior state matrix, and locking the vehicle passing path through image acquisition and license plate matching; The confluence congestion degree calculation module is used for counting the number of vehicles in a statistical cycle, assigning values to the matrix and calculating the confluence congestion degree. The response efficiency evaluation and signal adjustment module is configured to evaluate the response efficiency of the merging instruction, screen the low-efficiency instruction, and dynamically adjust the signal duration.

7. The adaptive perception interactive traffic signal machine electrical control system according to claim 6, wherein, The lane and merging management module comprises a lane coding unit, a merging instruction generation unit, and an initialization merging behavior set construction unit. The lane coding unit is configured to uniquely code each lane and attach a direction identifier according to the actual distribution of the entrance and exit directions of the intersection. The merging instruction generation unit is configured to identify the merging behavior from a single entrance direction to a single exit direction, and assign a unique merging instruction to each type of merging behavior. The initialization merging behavior set construction unit is configured to count the number of all entrance directions corresponding to each exit direction, integrate the merging instructions, and form an initialization merging behavior set.

8. The adaptive perception interactive traffic signal machine electrical control system according to claim 6, wherein, The merging behavior state matrix construction and path locking module comprises a matrix construction unit, an image acquisition and data interaction unit, and a license plate recognition matching unit. The matrix construction unit is configured to build a merging behavior state matrix based on the initialization merging behavior set, with the lane number of the entrance participating in the merging as the row index and the exit lane number as the column index. The image acquisition and data interaction unit is configured to acquire license plate images through image acquisition devices arranged at the entrance and exit directions, and realize real-time image data interaction and sharing between the two types of devices through short-range Internet of Things. The license plate recognition matching unit is configured to recognize the license plate images acquired at the entrance and match the license plate images acquired at the exit, so as to lock the passing path of the vehicle from the entrance lane to the exit lane.

9. The adaptive perception interactive traffic signal machine electrical control system according to claim 6, wherein, The merging congestion degree calculation module comprises a periodic vehicle statistics unit, a matrix assignment unit, and a congestion degree calculation unit. The periodic vehicle statistics unit is configured to count the number of vehicles merging into the corresponding exit lane from each entrance lane in one traffic signal cycle. The matrix assignment unit is configured to assign the counted number of vehicles to the corresponding position of the merging behavior state matrix, to generate the merging behavior state matrix of the current period. The congestion degree calculation unit is configured to calculate the merging congestion degree corresponding to each merging instruction based on the merging behavior state matrix of the current period.

10. The adaptive perception interactive traffic signal machine electrical control system according to claim 6, wherein, The response efficiency evaluation and signal adjustment module comprises a response efficiency calculation unit, a low-efficiency merging instruction screening unit, and a signal duration adjustment unit. The response efficiency calculation unit is configured to calculate the response efficiency of the merging instruction according to the merging congestion degrees of the same merging instruction in two adjacent traffic signal cycle periods. The low-efficiency merging instruction screening unit is configured to compare the response efficiencies of all merging instructions, and screen the merging instruction with the lowest response efficiency. The signal duration adjustment unit is configured to prolong the initialization signal duration of the low-efficiency merging instruction in the first subsequent traffic signal cycle period, and the prolonged duration does not exceed the preset maximum signal duration, and the initialization signal duration is restored in the second subsequent period.

Images