Simulation network fast partition processing method and system integrating computing and communication cost

Abstract

The application discloses a simulation road network fast segmentation processing method and system which comprehensively calculates and communication cost, converts a simulation road network with a running simulation vehicle into a gray-scale image, calculates the gravity center of the gray-scale image, quickly locates the position of a segmentation line through the gravity center coordinates of the image, and recursively divides the simulation road network, so that the number of vehicles in each region after segmentation is basically equal, the strong coupling simulation road network segmentation problem facing load balancing is solved, and further, in order to reduce the communication interaction between adjacent sub-regions, when the difference between the number of vehicles in each sub-region meets a pre-defined range, the position of the segmentation line is adjusted, so that the sum of the gray-scale values of the lanes intersected by the segmentation line is minimum, thereby reducing the communication cost. According to the final segmentation result, task scheduling is carried out, load balancing is realized, and the application requirement of a large-scale city road network to a parallel traffic system is better met.

Description

Technical Field

[0001] This invention relates to the field of parallel traffic system technology, and in particular to a method and system for rapid segmentation of simulated road networks that integrates calculation and communication costs. Background Technology

[0002] Parallel traffic systems represent a future direction for intelligent transportation. By modeling people, vehicles, and roads, they virtually construct real traffic scenarios and provide intelligent analysis tools for various traffic planning, management, and control schemes through virtual-real interaction. Because of the need for rapid processing of numerous drivers, vehicles, non-motorized vehicles, pedestrians, and traffic control schemes within the virtual environment, parallel processing is essential for parallel traffic systems. Simulated road network segmentation is a key technology for parallel processing in parallel traffic systems and the foundation for achieving load balancing.

[0003] Due to the strong coupling of simulated road networks, road network segmentation for load balancing is a challenging problem. Research on segmentation methods for simulated road networks is limited; existing methods mainly include area-based or segment-length-based segmentation methods and orthogonal recursive bisection methods. Area-based segmentation methods divide the road network equally according to area; orthogonal recursive bisection methods use intersections as nodes to transform the topology graph, then segment the resulting topology graph, repeatedly performing symmetrical segmentation in both horizontal and vertical directions.

[0004] The limitations of existing methods include: 1) The topology graph is transformed and segmented by taking intersections as the smallest unit. Improved methods mainly use road length or traffic flow as the weight of the topology graph and adjust the segmentation results, which has high computational complexity; 2) Based on the area of ​​the region, the orthogonal recursive bisection method is used to simulate the road network segmentation without considering the number of vehicles running on the road. Although this type of method has a fast processing speed, it is difficult to achieve load balancing.

[0005] To achieve parallel simulation of large-scale urban traffic, adapt to time-varying road network traffic flow conditions, and realize load balancing, it is necessary to study a fast simulation road network segmentation processing method. Summary of the Invention

[0006] The purpose of this invention is to provide a method and system for rapid segmentation of simulated road networks that integrates computation and communication costs. This method can achieve parallel simulation of large-scale urban traffic, adapt to time-varying road network traffic flow conditions, and achieve load balancing.

[0007] The objective of this invention is achieved through the following technical solution:

[0008] A method for fast segmentation of simulated road networks that integrates computational and communication costs includes:

[0009] Read the simulated road network corresponding to the actual traffic scenario, the number of CPUs in each computing node, and the parallel traffic system software, and then transmit the parallel traffic system software to each computing node;

[0010] The system obtains the initial distribution location information of simulated vehicles in the simulated road network and draws the simulated vehicles to the relevant lanes of the simulated road network. Based on the number of simulated vehicles in each lane, the simulated road network is converted into a grayscale image. The centroid of the grayscale image is calculated, and a dividing line is drawn based on the centroid to recursively divide the grayscale image. During each recursive division, under the condition that the difference in the number of simulated vehicles in two adjacent sub-image regions is within a specified range, the sum of the grayscale values ​​of the lanes intersecting the dividing line is minimized. Finally, the number of sub-image regions obtained by recursive division is equal to the total number of CPUs of all computing nodes.

[0011] The simulated road network region corresponding to the segmented sub-image region, the lanes it contains, and the simulated vehicles on the lanes are assigned to the CPUs of each computing node for processing.

[0012] It receives the positions of each simulated vehicle at each time point, calculated by each computing node through the execution of parallel traffic system software.

[0013] A simulated road network fast segmentation processing system that integrates computational and communication costs includes:

[0014] The initialization information acquisition module is used to read the simulated road network corresponding to the actual traffic scenario, the number of CPUs in each computing node, and the parallel traffic system software, and then transmit the parallel traffic system software to each computing node.

[0015] The simulated road network grayscale image conversion and segmentation module is used to obtain the initial distribution location information of simulated vehicles in the simulated road network, draw the simulated vehicles to the relevant lanes of the simulated road network, convert the simulated road network into a grayscale image according to the number of simulated vehicles in each lane, calculate the centroid of the grayscale image, draw segmentation lines according to the centroid, and recursively divide the grayscale image. In each recursive division, under the condition that the difference between the number of simulated vehicles in two adjacent sub-image regions is within a specified range, the sum of the grayscale values ​​of the lanes intersecting the segmentation line is minimized. Finally, the number of sub-image regions obtained by recursive division is equal to the total number of CPUs of all computing nodes.

[0016] The task allocation module is used to allocate the simulated road network area, the lanes contained therein, and the simulated vehicles on the lanes corresponding to the segmented sub-image areas to the CPUs of each computing node for processing.

[0017] The processing result collection module is used to receive the positions of each simulated vehicle at each time point, calculated by each computing node through the execution of the parallel traffic system software.

[0018] A processing device includes: one or more processors; and a memory for storing one or more programs;

[0019] When the one or more programs are executed by the one or more processors, the one or more processors implement the aforementioned method.

[0020] A readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned method.

[0021] As can be seen from the technical solution provided by the present invention, converting the simulated road network with running simulated vehicles into a grayscale image, calculating the centroid of the grayscale image, and quickly locating the segmentation line position using the image centroid coordinates, recursively dividing the simulated road network, achieves that the number of vehicles in each segmented area is basically equal, solving the problem of segmenting strongly coupled simulated road networks for load balancing. Furthermore, in order to reduce communication interaction between adjacent sub-areas, within a predefined range where the difference in the number of vehicles in each sub-area meets the requirements, the position of the segmentation line is adjusted to minimize the sum of the grayscale values ​​of the lanes where the segmentation lines intersect, thereby reducing communication costs. Task scheduling is performed based on the final segmentation result, achieving load balancing and better adapting to the application requirements of parallel traffic systems in large-scale urban road networks. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A basic flowchart of a simulated road network fast segmentation processing method that integrates calculation and communication costs is provided in an embodiment of the present invention;

[0024] Figure 2 A flowchart illustrating a method for rapid segmentation of simulated road networks that integrates computation and communication costs, provided in an embodiment of the present invention.

[0025] Figure 3 A schematic diagram illustrating the segmentation result of the area-based segmentation method provided in an embodiment of the present invention;

[0026] Figure 4 A schematic diagram of the segmentation result of the orthogonal recursive segmentation method provided in an embodiment of the present invention;

[0027] Figure 5 A schematic diagram of the segmentation result provided in this embodiment of the invention;

[0028] Figure 6A schematic diagram of a simulated road network fast segmentation processing system that integrates computation and communication costs, provided as an embodiment of the present invention;

[0029] Figure 7 This is a schematic diagram of a processing device provided in an embodiment of the present invention. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of the present invention.

[0031] First, the following explanations are provided for the terms that may be used in this article:

[0032] The terms “including,” “comprising,” “containing,” “having,” or other similar semantic descriptions should be interpreted as non-exclusive inclusion. For example, “including a technical feature element (such as raw material, component, ingredient, carrier, dosage form, material, size, part, component, mechanism, device, step, process, method, reaction conditions, processing conditions, parameter, algorithm, signal, data, product or article of manufacture, etc.)” should be interpreted as including not only the expressly listed technical feature element, but also other technical feature elements that are not expressly listed and are well-known in the art.

[0033] The following is a detailed description of a simulated road network fast segmentation processing scheme that integrates computation and communication costs, provided by this invention. Contents not described in detail in the embodiments of this invention are prior art known to those skilled in the art. Where specific conditions are not specified in the embodiments of this invention, they should be performed according to conventional conditions in the art or conditions recommended by the manufacturer.

[0034] Example 1

[0035] This invention provides a method for fast segmentation of simulated road networks that integrates computation and communication costs, such as... Figure 1 As shown, the main steps include the following:

[0036] Step 11: Read the simulated road network corresponding to the actual traffic scenario, the number of CPUs in each computing node, and the parallel traffic system software, and transfer the parallel traffic system software to each computing node.

[0037] Step 12: Obtain the initial distribution location information of the simulated vehicles in the simulated road network, and draw the simulated vehicles to the relevant lanes of the simulated road network. Convert the simulated road network into a grayscale image according to the number of simulated vehicles in each lane, calculate the centroid of the grayscale image, draw a dividing line according to the centroid, and recursively divide the grayscale image. In each recursive division, under the condition that the difference in the number of simulated vehicles in two adjacent sub-image regions is within a specified range, minimize the sum of the grayscale values ​​of the lanes intersecting the dividing line. Finally, through recursive division, make the number of sub-image regions obtained by segmentation equal to the total number of CPUs of all computing nodes.

[0038] In this step, at the beginning, horizontal or vertical dividing lines are drawn using the centroid of the grayscale image to bisect the image vertically or horizontally, forming two sub-image regions. Then, the position of the dividing lines is adjusted to minimize the sum of grayscale values ​​of lanes intersecting the dividing lines, provided that the difference in the number of simulated vehicles between the two adjacent sub-image regions is within a specified range. Afterward, the two sub-image regions are bisected again in the same way as in the beginning step: the centroid is calculated for each sub-image region, and horizontal or vertical dividing lines are drawn using the centroid. The position of the dividing lines is then adjusted to minimize the sum of grayscale values ​​of lanes intersecting the dividing lines, provided that the difference in the number of simulated vehicles between the two adjacent sub-image regions is within a specified range. This recursive bisection is performed until the number of sub-image regions obtained by the segmentation equals the total number of CPUs in all computing nodes.

[0039] Step 13: The simulated road network region corresponding to the segmented sub-image region, the lanes it contains, and the simulated vehicles on the lanes are assigned to the CPUs of each computing node for processing.

[0040] Step 14: Receive the positions of each simulated vehicle at each time point calculated by each computing node through the execution of the parallel traffic system software.

[0041] Compared with existing solutions, the above method is a simulated road network segmentation method that takes into account the moving vehicle information of the road network and is suitable for parallel processing of dynamic traffic flow. Furthermore, the simulated road network is converted into a grayscale image and image processing methods are introduced for simulated road network segmentation. This not only enables rapid segmentation of strongly coupled road networks, but also reduces the number of vehicles interacting between adjacent sub-image regions, thus achieving load balancing.

[0042] To more clearly demonstrate the technical solution and its effects provided by the present invention, the methods provided by the embodiments of the present invention will be described in detail below with reference to specific examples.

[0043] Parallel traffic systems (PCS) are effective analytical and decision-making tools for traffic planning, design, management, and control. They can avoid the risks and huge economic losses caused by the direct implementation of various traffic planning, management, organization, and control schemes. PCS can realize the virtual construction and online intelligent analysis of large-scale urban road networks. Its parallel processing requires the segmentation of the simulated road network. However, the segmentation of the strongly coupled simulated road network is a challenging problem. In order to meet the load balancing application requirements of parallel traffic systems for parallel micro-simulation of large-scale urban road networks, this invention provides a fast segmentation method for simulated road networks that integrates computation and communication costs, addressing the strong coupling characteristics of simulated road networks. First, the simulated road network is transformed into a grayscale image based on the simulated vehicles in each lane. Then, the centroid of the grayscale image is calculated, and the grayscale image is recursively bisected based on the centroid coordinates. After the grayscale image is segmented, in order to reduce communication interaction between adjacent sub-image regions, the position of the dividing line is adjusted within a predefined range, such that the sum of the grayscale values ​​of the lanes intersecting the dividing line is minimized, thereby reducing communication costs. Based on the obtained simulated road network segmentation results, the management node allocates simulation tasks to the CPUs of each computing node for processing. This method not only enables rapid segmentation of the simulated road network but also reduces communication costs and achieves load balancing.

[0044] The method provided in this invention can use a small server cluster as the computing platform. For example, the computing platform can be a Loongson big data appliance, which has one management node and four computing nodes, with a total of 2 CPUs per computing node. P P is a positive integer. Figure 2 The overall process of the above method is shown, mainly including:

[0045] Step 21: The management node first initializes itself by reading the simulated road network corresponding to the actual traffic scenario, the number of CPUs in each computing node, and the parallel traffic system software, and then transmits the parallel traffic system software to each computing node for execution.

[0046] Step 22: The management node obtains the initial distribution location information of the simulated vehicles in the road network, that is, the coordinates of the simulated vehicles, and draws the simulated vehicles to each lane of the road network.

[0047] Step 23: The management node converts the simulated road network into a grayscale image based on the number of simulated vehicles in each lane, calculates the centroid of the grayscale image, and performs recursive binary division on the grayscale image based on the centroid coordinates.

[0048] In this embodiment of the invention, the road occupancy rate of a single lane is set to [0, A], where the road occupancy rate is the number of vehicles per kilometer of road, in units of vehicles / kilometer. Here, A represents the maximum road occupancy rate (e.g., A = 170); the grayscale image pixel value range is [0, 255], and an image pixel value of 0 represents the background; the grayscale value of a single lane is set according to the following formula:

[0049] f:[0,A]→[0,255].

[0050] Where f represents the mapping function that converts road occupancy rate into grayscale value. The larger the number of vehicles in a lane, the larger the corresponding grayscale value. The meaning of the above formula can be intuitively described as: grayscale value of a single lane = 255 * road occupancy rate / A, that is, a single lane uses a uniform grayscale value.

[0051] In this embodiment of the invention, the method for calculating the centroid of a grayscale image includes: setting the size of the grayscale image to Ml2N, where M and N represent the number of image pixels in the horizontal and vertical directions, respectively; setting f(i,j) as the grayscale value of the image at pixel (i,j), and calculating the 0th and 1st moments using the following formula:

[0052]

[0053]

[0054]

[0055] Among them, H 00 H is the zeroth moment. 10 With H 01 Let i be the first-order moment, and j represent the position numbers of the image pixels in the horizontal and vertical directions, respectively.

[0056] Calculate the image centroid by combining the 0th and 1st moments:

[0057]

[0058]

[0059] Where (x0, y0) are the coordinates of the image centroid.

[0060] In this embodiment of the invention, the image is divided vertically or horizontally using the image centroid coordinates (x0, y0). The vertical or horizontal division is based on whether the areas of the two sub-image regions after the division are similar. Specifically:

[0061] Let the two sub-image regions that are vertically bisected be: (a 1v ,a 2vThe two sub-image regions, which are divided horizontally, are: (a 1h ,a )h ), calculate the area of ​​each sub-image region, and express the area of ​​the above four sub-image regions as: S(a 1v ),S(a 2v ),S(a 1h ),S(a 2h ); Determine whether the following condition is met: S(a) 1v ) / S(a 2v )≥S(a 1h ) / S(a 2h If so, use horizontal bisection; otherwise, use vertical bisection.

[0062] Step 24: By adjusting the position of the dividing line, under the condition that the difference in the number of simulated vehicles in two adjacent sub-image regions is within a specified range, the sum of the gray values ​​of the roads intersecting with the dividing line is minimized until the number of sub-image regions obtained by segmentation is equal to the total number of CPUs of all computing nodes.

[0063] In this embodiment of the invention, each stage is executed according to steps 23 to 24. The object of operation in the first stage is the complete grayscale image, and the object of operation in each subsequent stage is the sub-image region obtained by bisection in the previous stage. The process of each bisection mainly includes: first calculating the centroid, then determining whether to use a vertical or horizontal method for bisection, and then adjusting the position of the dividing line. Under the condition that the difference in the number of simulated vehicles in two adjacent sub-image regions is within a specified range, the sum of the grayscale values ​​of the lanes intersecting the dividing line is minimized, thereby reducing the communication interaction cost between adjacent sub-image regions. The condition that the sum of the grayscale values ​​of the lanes intersecting the dividing line is minimized under the condition that the difference in the number of simulated vehicles in two adjacent sub-image regions is within a specified range is expressed as:

[0064]

[0065]

[0066] in, and This represents two adjacent sub-image regions obtained by bisecting a single sub-image region. and express and The number of simulated vehicles, ε represents a specified range (which can be set by the user based on actual conditions or experience, for example: ε = 100); L represents the total number of lanes intersecting the dividing line, f(e k ) represents the grayscale value of the k-th lane that intersects the dividing line, obtained using the number of simulated vehicles in the k-th lane.

[0067] Repeat the above steps until the number of sub-image regions equals 2 times the number of CPUs. P The event will end at that time.

[0068] Step 25: The management node, through its allocation module, allocates the simulated road network area, the lanes it contains, and the simulated vehicles on the lanes corresponding to each sub-image area to the CPU of each computing node for processing.

[0069] Step 26: The computing node starts the parallel traffic system software and calculates the position of each simulated vehicle at each time by executing the vehicle micro-car-following model, lane-changing model and signal timing module in the software until the simulation cycle ends. The simulation results are then transmitted to the management node through the feedback module to wait for the next task to be assigned.

[0070] Step 27: The management node receives the transmission results from the compute node's feedback module and stores them in the local database for user query.

[0071] Through the above description of the embodiments, those skilled in the art can clearly understand that the above embodiments can be implemented by software, or by using software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solutions of the above embodiments can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.), including several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0072] To verify the effectiveness of this invention, a comparative experiment was conducted to compare the area-based segmentation method, the orthogonal recursive segmentation method, and the method of this invention. The simulated road network used in the comparative experiment consisted of 150 intersections and simulated over 36,000 vehicles. The evaluation metrics were the number of vehicles in each sub-region and the sum of the number of vehicles in all lanes intersecting the segmentation line. The segmentation results in the experiment are as follows: Figure 3 , 4 As shown in Figure 5. Figure 3 The segmentation result is from an area-based segmentation method. Figure 4 The segmentation result is from the orthogonal recursive segmentation method. Figure 5 This is the segmentation result of the present invention. Figures 3-5 In the diagram, 1, 2, and 3 represent the dividing lines in three separate partitions, with the number of partitions being merely an example. The experimental results for the three partitioning schemes are shown in Table 1.

[0073] Table 1: Experimental Results

[0074]

[0075] Experimental results show that, compared with existing simulated road network segmentation methods, the method of the present invention can not only achieve rapid segmentation of simulated road networks, but also reduce communication costs.

[0076] Example 2

[0077] This invention also provides a simulated road network fast segmentation processing system that integrates computation and communication costs, which is mainly based on the method provided in the foregoing embodiments, such as... Figure 6 As shown, the system mainly includes:

[0078] The initialization information acquisition module is used to read the simulated road network corresponding to the actual traffic scenario, the number of CPUs in each computing node, and the parallel traffic system software, and then transmit the parallel traffic system software to each computing node.

[0079] The simulated road network grayscale image conversion and segmentation module is used to obtain the initial distribution location information of simulated vehicles in the simulated road network, draw the simulated vehicles to the relevant lanes of the simulated road network, convert the simulated road network into a grayscale image according to the number of simulated vehicles in each lane, calculate the centroid of the grayscale image, draw segmentation lines according to the centroid, and recursively divide the grayscale image. In each recursive division, under the condition that the difference between the number of simulated vehicles in two adjacent sub-image regions is within a specified range, the sum of the grayscale values ​​of the lanes intersecting the segmentation line is minimized. Finally, the number of sub-image regions obtained by recursive division is equal to the total number of CPUs of all computing nodes.

[0080] The task allocation module is used to allocate the simulated road network area, the lanes contained therein, and the simulated vehicles on the lanes corresponding to the segmented sub-image areas to the CPUs of each computing node for processing.

[0081] The processing result collection module is used to receive the positions of each simulated vehicle at each time point, calculated by each computing node through the execution of the parallel traffic system software.

[0082] Those skilled in the art will understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the system can be divided into different functional modules to complete all or part of the functions described above.

[0083] Example 3

[0084] The present invention also provides a processing device, such as Figure 7 As shown, it mainly includes: one or more processors; a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the method provided in the foregoing embodiments.

[0085] Furthermore, the processing device also includes at least one input device and at least one output device; in the processing device, the processor, memory, input device, and output device are connected via a bus.

[0086] In this embodiment of the invention, the specific types of the memory, input device, and output device are not limited; for example:

[0087] Input devices can be touchscreens, image acquisition devices, physical buttons, or mice, etc.

[0088] The output device can be a display terminal;

[0089] The memory can be random access memory (RAM) or non-volatile memory, such as disk storage.

[0090] Example 4

[0091] The present invention also provides a readable storage medium storing a computer program that, when executed by a processor, implements the method provided in the foregoing embodiments.

[0092] In this embodiment of the invention, the readable storage medium is a computer-readable storage medium and can be disposed in the aforementioned processing device, for example, as a memory in the processing device. Furthermore, the readable storage medium can also be any medium capable of storing program code, such as a USB flash drive, portable hard drive, read-only memory (ROM), magnetic disk, or optical disk.

[0093] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A simulation road network fast partitioning method that integrates computing and communication cost, characterized in that, The method comprises the following steps: reading a simulation road network corresponding to an actual traffic scene, the number of CPUs of each computing node and parallel traffic system software, and transmitting the parallel traffic system software to each computing node; obtaining initial distribution position information of simulation vehicles on the simulation road network, and drawing the simulation vehicles to relevant lanes of the simulation road network, converting the simulation road network into a gray-scale image according to the number of simulation vehicles on each lane, calculating the gravity center of the gray-scale image, and drawing a segmentation line according to the gravity center to perform recursive bisection on the gray-scale image, wherein each time the recursive bisection is performed, the sum of the gray-scale values of the lanes intersecting the segmentation line is minimized under the condition that the difference between the number of simulation vehicles in two adjacent sub-image regions is within a specified range, and finally the number of sub-image regions obtained by segmentation is equal to the total number of CPUs of all computing nodes through recursive bisection; processing the simulation road network region corresponding to the sub-image region obtained by segmentation, the lanes contained therein and the simulation vehicles on the lanes according to the CPUs of each computing node; receiving the positions of each simulation vehicle at each time calculated by each computing node by executing the parallel traffic system software.

2. The method of claim 1, wherein the method further comprises: The method of converting the simulation road network into a gray-scale image according to the number of simulation vehicles on each lane comprises: setting the road occupancy rate range of a single lane as [0, A], wherein A represents the maximum value of the road occupancy rate, and the road occupancy rate is the number of vehicles per kilometer of road; and setting the gray-scale value range of the gray-scale image as [0, 255]; the gray-scale value of a single lane = 255*road occupancy rate / A.

3. The method of claim 1, wherein the method further comprises: The method of calculating the gravity center of the gray-scale image comprises: setting the size of the gray-scale image as MxN, wherein M and N respectively represent the number of image pixels in the horizontal and vertical directions; setting f(i, j) as the gray-scale value of the image pixel at position (i, j), and calculating the 0th moment and the 1st moment by the following formula: Among them, H 00 H is the zeroth moment. 10 With H 01 Let i be a first-order moment, and j represent the position numbers of image pixels in the horizontal and vertical directions, respectively. calculating the gravity center of the image in combination with the 0th moment and the 1st moment: wherein (x0, y0) is the gravity center coordinate of the image.

4. The simulation network fast partitioning method of claim 1, wherein, The method of drawing a segmentation line according to the gravity center to perform recursive bisection on the gray-scale image, wherein each time the recursive bisection is performed, the sum of the gray-scale values of the lanes intersecting the segmentation line is minimized under the condition that the difference between the number of simulation vehicles in two adjacent sub-image regions is within a specified range, comprises: in the initial stage, a horizontal or vertical segmentation line is drawn according to the gravity center of the gray-scale image to perform vertical or horizontal bisection on the gray-scale image, then the position of the segmentation line is adjusted to minimize the sum of the gray-scale values of the lanes intersecting the segmentation line under the condition that the difference between the number of simulation vehicles in two adjacent sub-image regions is within a specified range, and then the two sub-image regions are respectively bisectioned in the same way as in the initial stage, i.e., the gravity center of each sub-image region is calculated, and a horizontal or vertical segmentation line is drawn according to the gravity center, then the position of the segmentation line is adjusted to minimize the sum of the gray-scale values of the lanes intersecting the segmentation line under the condition that the difference between the number of simulation vehicles in two adjacent sub-image regions is within a specified range; the recursive bisection is continuously performed until the number of sub-image regions obtained by segmentation is equal to the total number of CPUs of all computing nodes.

5. The simulation network fast partitioning method of claim 4, wherein, The sum of lane gray values intersecting with the segmentation line is minimized under the condition that the difference between the number of simulation vehicles in two adjacent sub-image regions is in a specified range, which is expressed as: wherein, with denotes two adjacent sub-image regions obtained by binary division of a single sub-image region; with denotes with the number of simulated vehicles, ε denotes a specified range; L denotes the total number of lanes intersecting the division line, f(e k ) denotes the gray value of the kth lane intersecting the division line.

6. The simulation network fast partitioning method of claim 4, wherein, The basis for the vertical or horizontal bisection of the gray image includes: Let two sub-image regions using vertical bisection be: (a 1v , a 2v ), two sub-image regions using horizontal bisection be: (a 1h , a 2h ), the area of each sub-image region is calculated, and the area of the above four sub-image regions be expressed as: S(a 1v ), S(a 2v ), S(a 1h ), S(a 2h ); determining whether or not S(a 1v ) / S(a 2v ) ≥ S(a 1h ) / S(a 2h ) holds, if yes, horizontal bisection is used, otherwise vertical bisection is used.

7. The method of claim 1, wherein the method further comprises: The manner in which each computing node calculates the position of each simulation vehicle at each time point by executing the parallel traffic system software includes: Each computing node calculates the position of each simulation vehicle at each time point by executing a vehicle micro-following model, a lane-changing model and a signal timing model in the parallel traffic system software.

8. A system for fast partitioning of a simulated road network that integrates computation and communication cost, characterized in that, The system is realized based on the method of any one of claims 1-7, and the system comprises: An initialization information acquisition module configured to read a simulation road network corresponding to an actual traffic scene, a CPU number of each computing node and a parallel traffic system software, and transmit the parallel traffic system software to each computing node; A simulation road network gray image conversion and segmentation module configured to acquire initial distribution position information of simulation vehicles on the simulation road network, draw the simulation vehicles to relevant lanes of the simulation road network, convert the simulation road network into a gray image according to the number of simulation vehicles on each lane, calculate a gravity center of the gray image, draw a segmentation line according to the gravity center to perform recursive bisection on the gray image, minimize the sum of lane gray values intersecting with the segmentation line under the condition that the difference between the number of simulation vehicles in two adjacent sub-image regions is in a specified range each time the recursive bisection is performed, and finally make the number of sub-image regions obtained by segmentation equal to the total number of CPUs of all computing nodes through recursive bisection; A task allocation module configured to allocate a simulation road network region corresponding to each sub-image region obtained by segmentation, a contained lane and simulation vehicles on the lane to a CPU of each computing node for processing; A processing result comprehensive collection module configured to receive the position of each simulation vehicle at each time point calculated by each computing node by executing the parallel traffic system software.

9. A processing device, characterized by Comprise: One or more processors; A memory configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the method of any one of claims 1-7.

10. A readable storage medium, storing a computer program, characterized in that, When the computer program is executed by the processor, the method of any one of claims 1-7 is implemented.

Images