A method, system, equipment, and medium for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism.

By using a spatial index-based grid partitioning and multi-threaded parallel computing method, the problem of low efficiency in traditional traffic simulation calculations is solved, and efficient index calculation and stable simulation operation are achieved under large-scale road networks.

CN123116706APending Publication Date: 2026-11-17四川易方智慧科技有限公司 +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611395913.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-09
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Traditional microscopic traffic simulation calculation methods are inefficient and resource-intensive in large-scale road networks, traffic flows of tens of thousands of vehicles, and long-term evolution scenarios. They cannot achieve real-time evaluation and automated analysis, nor can they quickly locate the road where the vehicle is located and perform parallel computing.

Method used

A spatial index-based grid partitioning method is adopted to divide the map data into coarse-grained large grids and fine-grained small grids, establish a dictionary structure of vehicles and roads, utilize multi-threaded parallel computing of evaluation indicators, and quickly locate the road where the vehicle is located and perform local retrieval through spatial indexing.

Benefits of technology

It significantly improves vehicle matching efficiency, reduces computational time complexity and resource consumption, achieves efficient index calculation and stable simulation operation, and supports real-time evaluation and automated analysis.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123116706A_ABST
    Figure CN123116706A_ABST
Patent Text Reader

Abstract

This invention relates to the field of traffic simulation technology, specifically to a method, system, device, and medium for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism. The main components include dividing the map into coarse-grained large grids, each of which is further divided into fine-grained small grids; all roads are pre-assigned to corresponding sub-grids, establishing a mapping; during simulation, the vehicle's location within its sub-grid is quickly determined based on its coordinates, requiring only a small number of roads to be traversed within that grid to identify the vehicle's location. Using the method provided by this invention, the traditional global traversal retrieval is optimized to a local, minimal-range matching within a grid, requiring only the traversal of a few candidate roads, significantly improving matching efficiency. Furthermore, the hierarchical retrieval mechanism of first large-scale contraction and then precise positioning results in more accurate spatial positioning, shorter calculation paths, and significantly better overall retrieval efficiency than traditional single-layer indexes or indexless structures.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of traffic simulation technology, and more specifically, to a method, system, device, and medium for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism. Background Technology

[0002] Microscopic traffic simulation (such as SUMO) is a core tool for urban traffic analysis, signal control optimization, vehicle-road cooperative testing, and road network capacity assessment. During the simulation process, key evaluation indicators such as road occupancy, traffic density, average vehicle speed, delay, queue length, and flow rate need to be calculated in real time. These indicators are the core basis for traffic state judgment, scheme comparison, and algorithm verification.

[0003] As simulation scenarios evolve towards large-scale road networks, traffic flows of tens of thousands of vehicles, and long-term evolution, traditional indicator calculation methods are showing significant bottlenecks: Vehicle and road matching relies on global traversal, which is inefficient and costly; it lacks a spatial index structure, making it impossible to quickly locate the road where the vehicle is located; the statistical process is executed serially and cannot be accelerated in parallel; and it consumes a lot of resources, resulting in unstable simulation operation and difficulty in supporting real-time evaluation and automated analysis.

[0004] Therefore, there is an urgent need for an efficient method for calculating evaluation indicators based on spatial indexing, local retrieval, and data parallelism. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, device, and medium for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism, in order to solve the above-mentioned problems in the prior art.

[0006] This invention is achieved through the following technical solution: In a first aspect, the present invention provides a method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism, including: The current map data is acquired and divided into subdivided grids. The acquired target road data is then mapped to the subdivided grids. Obtain vehicle coordinates, locate the vehicle in the target subdivision grid based on the vehicle coordinates, determine the road where the vehicle is currently located in the target subdivision grid, and establish a vehicle-road dictionary; Read the vehicle sets of each road, obtain the various indicators of the current road based on the vehicle sets, and output the indicators to the global result set.

[0007] Preferably, the step of acquiring the current map data and dividing it into subdivided grids includes: To obtain the global map range, read the simulation road network file, extract the minimum or maximum coordinates of the entire road network, and determine the rectangular spatial range covering the entire scene; Set the parameters for the primary grid and the subdivision grid according to the road network density and the simulation scale. Set the side length of the primary grid G ​​and the side length of the subdivision grid S, where G is greater than S.

[0008] The global first-level grid is divided into several rows and several columns of first-level grids according to the step size G, and each first-level grid is assigned a unique number. The subdivision of the mesh is further subdivided into smaller subdivisions within each first-level mesh, with a step size S, and subdivision numbers are assigned. An index table is established to construct a fast mapping relationship between two-level grids, generating a static spatial partitioning structure grid.

[0009] Preferably, assigning a unique number to each first-level grid includes: For any vehicle or road coordinate point The first-level grid number to which it belongs is calculated by the following formula:

[0010] Its subdivision grid number is calculated by the following formula:

[0011] In the formula, and The row and column numbers are for the first-level grid. and These are coordinate values. and These are the minimum x and y coordinates in the global road network. and The row and column numbers for the subdivided grid.

[0012] Preferably, mapping the acquired target road data to a subdivided grid includes: Obtain all geometric and topological information, including road, lane, centerline coordinates, and road extent. Using all subdivision grid numbers as keys, initialize the road list corresponding to each subdivision grid to be empty, and construct an empty binding table; Process each road in turn, extract the coordinate sequence of all centerline points of the road, and calculate the subdivision grid number of each point; Add the current road to the road list of all subdivision grids containing its centerline points, keeping the same road only once in the same grid.

[0013] Preferably, the process of establishing the vehicle-road dictionary includes: Traverse all roads, create an empty vehicle list for each road, and construct an initial dictionary structure; Read the real-time coordinates of the vehicle, calculate the subdivided grid to which it belongs, and obtain the list of candidate roads within that grid. Traverse the candidate roads, determine whether the coordinates are within the road's passability range, and identify the road where the vehicle is located; Add the vehicle ID to the vehicle list for the corresponding road to complete real-time dictionary maintenance.

[0014] Preferably, the indicators for obtaining the current road based on the vehicle set include: Create a thread pool to perform the calculation of various grid metrics in parallel; Traverse all first-level grids, encapsulate the road set within each first-level grid into an independent computation task, and submit it to the thread pool.

[0015] Each thread reads all roads within its respective first-level grid, retrieves the vehicle list from the vehicle and road dictionary, and calculates the indicators for each item. After each thread completes its calculation, it writes the result to the global result set.

[0016] Preferably, the indicators include: road occupancy rate, traffic density, average speed, traffic flow, queue length, and delay time.

[0017] Secondly, the present invention also provides a vehicle simulation evaluation index calculation system based on spatial partitioning and data parallelism, used to execute the aforementioned vehicle simulation evaluation index calculation method based on spatial partitioning and data parallelism, comprising: The grid division module is configured to acquire the current map data and divide it into subdivided grids, and then map the acquired target road data into the subdivided grids. The dictionary building module is configured to obtain vehicle coordinates, locate the target subdivision grid based on the vehicle coordinates, determine the road where the current vehicle is located in the target subdivision grid, and build a vehicle-road dictionary; The indicator calculation module is configured to read the vehicle set of each road, obtain various indicators of the current road based on the vehicle set, and output the indicators to the global result set.

[0018] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism.

[0019] Fourthly, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism.

[0020] The technical solution of the present invention has at least the following advantages and beneficial effects: The method provided by this invention mainly includes dividing the map into coarse-grained large grids, and then further dividing each large grid into fine-grained small grids. All roads are pre-assigned to corresponding sub-grids, establishing a mapping. During simulation, the vehicle's location within a sub-grid is quickly determined based on its coordinates, requiring only a small number of roads to be traversed within that grid to identify the vehicle's location. This method optimizes traditional global traversal retrieval to a localized, minimal-range matching within a single grid, significantly improving matching efficiency by requiring only a small number of candidate roads to be traversed. Furthermore, the hierarchical retrieval mechanism, which first shrinks the large area and then precisely locates the target, results in more accurate spatial positioning, shorter computational paths, and significantly better overall retrieval efficiency than traditional single-layer indexes or unindexed structures.

[0021] By binding roads to subdivided meshes during the initialization phase, there is no need to repeatedly calculate road coverage relationships during simulation, resulting in no redundant overhead and higher utilization of computational resources.

[0022] The mapping relationship between roads and their associated vehicles is maintained using a dictionary structure. When calculating indicators, the vehicle list is read directly, eliminating the need for repeated retrieval, matching, and traversal, resulting in faster statistical response.

[0023] It executes in parallel with the first-level grid as an independent computing unit. There are no data conflicts or lock waiting between units. The computing power can be linearly expanded with the number of hardware cores, and the acceleration effect is obvious.

[0024] The three steps of this invention are sequentially linked and progressively advanced in terms of temporal logic: the road network topology connectivity graph provides backend data support, the end-point completion trigger judgment is the logical start gate, and the multi-dimensional road traffic cost optimization technology is the core of path decision-making. The entire calculation process requires no manual intervention and automatically completes the end-point extension while retaining the preset path, completely solving the technical problems of vehicles disappearing midway, cumbersome manual configuration, and missing end-point trajectories. Attached Figure Description

[0025] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0026] Figure 1 This is a schematic diagram of the process of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0028] The module division in this invention is a logical division. In practical applications, other division methods may be used. For example, multiple modules may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the connections, couplings, or communications in this invention can be direct connections, couplings, or communications between related objects, or indirect connections, couplings, or communications through other devices. Moreover, the connections, couplings, or communications between objects can be electrical or other similar forms, which are not limited in this invention. Independently described modules or sub-modules may or may not be physically separated: they may be implemented in software or hardware, and some modules or sub-modules may be implemented in software, with the processor calling the software to implement the functions of these modules or sub-modules, while other modules or sub-modules are implemented in hardware, such as through hardware circuits. Furthermore, some or all of the modules can be selected to achieve the purpose of this invention according to actual needs.

[0029] Please refer to Figure 1 The present invention also provides a method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism, including: S1: Obtain the current map data and divide it into subdivided grids, then map the obtained target road data into the subdivided grids; The two-level grid spatial partitioning of the map is the core foundational module supporting this invention for achieving rapid vehicle location, local road retrieval, and parallel index computation. Its purpose is to: transform the continuous, unstructured geographic space into a structured spatial indexing system through two layers of regularized grid partitioning, enabling any vehicle coordinate to be located in its assigned spatial region within a constant time. This transforms the traditional full-network traversal matching into a local retrieval within a very small area, fundamentally reducing the time complexity of vehicle and road attribution; simultaneously, using the first-level grid as a natural task partition provides stable, conflict-free, and scalable task units for subsequent multi-threaded parallel computation, achieving high-speed, parallel, and low-cost computation of evaluation indicators in large-scale traffic scenarios.

[0030] Road pre-binding to subdivided grids is the core supporting element for achieving rapid vehicle-road matching in this invention. Its function is to pre-allocate and bind the geometric spatial extent of all roads in the road network to corresponding subdivided grids during the simulation initialization phase, establishing a static mapping relationship between subdivided grids and the road list. Through this preprocessing operation, when a vehicle locates its assigned road, it does not need to traverse all roads; it only needs to locate its current subdivided grid based on its coordinates and then traverse the very few bound roads within that grid to complete the assignment determination, fundamentally reducing the time complexity and computational overhead of vehicle positioning.

[0031] S2: Obtain vehicle coordinates, locate the vehicle in the target subdivision grid based on the vehicle coordinates, determine the road where the current vehicle is located in the target subdivision grid, and establish a vehicle-road dictionary; After completing the two-level grid space division and road pre-binding, this module uses real-time vehicle coordinates and a two-level grid index to quickly narrow the search range. It only needs to traverse a very small number of pre-bound roads within the subdivided grid to determine the road where the vehicle is located in constant time. This completely abandons the traditional full road network traversal matching method, significantly improves the vehicle attribution efficiency in large-scale traffic scenarios, and provides an accurate data foundation for subsequent parallel calculation of indicators.

[0032] S3: Read the vehicle set of each road, obtain the various indicators of the current road based on the vehicle set, and output the indicators to the global result set.

[0033] Based on the previously constructed two-level grid spatial partitioning and road-vehicle dictionary structure, and using large grids as independent parallel computing units, a multi-threaded parallel approach is adopted to synchronously solve traffic indicators for roads within each grid. Its core function is to transform traditional global serial statistics into regional parallel computing, fully utilizing multi-core hardware resources. Without intruding on the simulation kernel or adding redundant calculations, it achieves low-latency, high-throughput, and stable output of key indicators such as road occupancy, traffic density, average speed, flow rate, and queue length, completely solving the technical bottlenecks of low efficiency, stuttering, and latency in indicator calculations under large-scale traffic flow.

[0034] The method provided by this invention mainly includes dividing the map into coarse-grained large grids, and then further dividing each large grid into fine-grained small grids. All roads are pre-assigned to corresponding sub-grids, establishing a mapping. During simulation, the vehicle's location within a sub-grid is quickly determined based on its coordinates, requiring only a small number of roads to be traversed within that grid to identify the vehicle's location. This method optimizes traditional global traversal retrieval to a localized, minimal-range matching within a single grid, significantly improving matching efficiency by requiring only a small number of candidate roads to be traversed. Furthermore, the hierarchical retrieval mechanism, which first shrinks the large area and then precisely locates the target, results in more accurate spatial positioning, shorter computational paths, and significantly better overall retrieval efficiency than traditional single-layer indexes or unindexed structures.

[0035] In one exemplary embodiment of the present invention, the step of acquiring current map data and performing grid division to obtain a subdivided grid includes: This invention employs a two-dimensional planar two-level equally spaced regular grid partitioning model to perform a full-coverage, non-overlapping, and non-omission discretization of the global space where the simulated road network is located.

[0036] The two levels of grids each perform different functions: Level 1 grid: As a parallel computing unit, it is used for multi-threaded task allocation during the indicator statistics phase.

[0037] Subdivided grid: As a fast spatial retrieval unit, it is used for rapid vehicle coordinate localization and road retrieval range narrowing.

[0038] Let the global spatial extent of the simulated road network be:

[0039] In the formula: Represents the minimum x-coordinate in the global road network. Represents the maximum x-coordinate in the global road network; Represents the minimum ordinate in the global road network. This represents the maximum ordinate in the global road network.

[0040] Simultaneously define: Large grid side length: G (unit: meters); Subdivision grid side length: S (unit: meters); Subdivision factor: k (integer k>=2); Constraints must be satisfied: .

[0041] Assigning a unique number to each first-level grid includes: For any vehicle or road coordinate point The first-level grid number to which it belongs is calculated by the following formula:

[0042] Its subdivision grid number is calculated by the following formula:

[0043] In the formula, and The row and column numbers are for the first-level grid. and These are coordinate values. and These are the minimum x and y coordinates in the global road network. and The row and column numbers for the subdivided grid.

[0044] Specifically, the global map range is obtained by reading the simulation road network file, extracting the minimum or maximum coordinates of the entire road network, and determining the rectangular spatial range covering the entire scene; Set the parameters for the primary grid and the subdivision grid according to the road network density and the simulation scale. Set the side length of the primary grid G ​​and the side length of the subdivision grid S, where G is greater than S.

[0045] The global first-level grid is divided into several rows and several columns of first-level grids according to the step size G, and each first-level grid is assigned a unique number. The subdivision of the mesh is further subdivided into smaller subdivisions within each first-level mesh, with a step size S, and subdivision numbers are assigned. An index table is established to construct a fast mapping relationship between two-level grids, generating a static spatial partitioning structure grid.

[0046] In one exemplary embodiment of the present invention, mapping the acquired target road data to a subdivided grid includes: Suppose that a two-level mesh has been generated, and the global space is divided into several subdivided meshes with side length S. For any point in the space... The subdivision grid number is:

[0047] A road is composed of a continuous sequence of shape points, and its spatial coverage can be marked by traversing all shape points or the subdivision grid containing the centerline. For a road r, traverse all its center subdivision points. Calculate the subdivision grid to which each point belongs ( , The road r is then added to the road list of the subdivided grid, ultimately forming a global road-grid binding relationship.

[0048] Specifically, it acquires all geometric and topological information, including the coordinates of all roads, lanes, centerlines, and road extents. Using all subdivision grid numbers as keys, initialize the road list corresponding to each subdivision grid to be empty, and construct an empty binding table; Process each road in turn, extract the coordinate sequence of all centerline points of the road, and calculate the subdivision grid number of each point; Add the current road to the road list of all subdivision grids containing its centerline points. The same road is retained only once in the same grid to avoid duplicate binding.

[0049] In one exemplary embodiment of the present invention, establishing a vehicle-road dictionary includes: Based on the established two-level grid index and subdivided grid-road binding table, vehicle localization follows the principle of spatial localization first, followed by local judgment: From vehicle coordinates Quickly calculate the subdivision mesh ( , The process reads the candidate road list within the grid directly from the binding table; sequentially checks whether the vehicle coordinates are within the spatial range of the candidate roads to complete the matching of vehicles and roads; and dynamically adds the successfully matched vehicles to the vehicle list of the corresponding road to form a dictionary structure of road-vehicle set.

[0050] Specifically, iterate through all roads, create an empty vehicle list for each road, and construct an initial dictionary structure; Read the real-time coordinates of the vehicle, calculate the subdivided grid to which it belongs, and obtain the list of candidate roads within that grid. Traverse the candidate roads, determine whether the coordinates are within the road's passability range, and identify the road where the vehicle is located; Add the vehicle ID to the vehicle list for the corresponding road to complete real-time dictionary maintenance.

[0051] An exemplary embodiment of the present invention, which obtains various indicators of the current road based on a set of vehicles, includes: Using a first-level grid as the smallest parallel task granularity, each large grid contains several sub-grids and corresponding roads. By utilizing the already constructed road-vehicle dictionary D, the vehicle set of each road can be directly read without repeated localization and matching.

[0052] Specifically, a thread pool is created to perform the calculation of various grid metrics in parallel; Traverse all first-level grids, encapsulate the road set within each first-level grid into an independent computation task, and submit it to the thread pool.

[0053] Each thread reads all roads within its respective first-level grid, retrieves the vehicle list from the vehicle and road dictionary, and calculates the indicators for each item. After each thread completes its calculation, it writes the result to the global result set.

[0054] The metrics include: road occupancy, traffic density, average speed, traffic flow, queue length, and delay time.

[0055] Based on the above method, the three steps are linked sequentially and progressively in a temporal logic: the road network topology connectivity graph provides backend data support, the end-point completion trigger judgment is the logical start gate, and the multi-dimensional road traffic cost optimization technology is the core of path decision-making. The entire calculation process requires no manual intervention and automatically completes the end-point extension while retaining the preset path, completely solving the technical problems of vehicles disappearing midway, cumbersome manual configuration, and missing end-point trajectories.

[0056] Secondly, the present invention also provides a vehicle simulation evaluation index calculation system based on spatial partitioning and data parallelism, used to execute the aforementioned vehicle simulation evaluation index calculation method based on spatial partitioning and data parallelism, comprising: The grid division module is configured to acquire the current map data and divide it into subdivided grids, and then map the acquired target road data into the subdivided grids. The dictionary building module is configured to obtain vehicle coordinates, locate the target subdivision grid based on the vehicle coordinates, determine the road where the current vehicle is located in the target subdivision grid, and build a vehicle-road dictionary; The indicator calculation module is configured to read the vehicle set of each road, obtain various indicators of the current road based on the vehicle set, and output the indicators to the global result set.

[0057] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0058] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. This computer software product, stored in a storage medium, includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0059] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism, characterized in that, include: The current map data is acquired and divided into subdivided grids. The acquired target road data is then mapped to the subdivided grids. Obtain vehicle coordinates, locate the vehicle in the target subdivision grid based on the vehicle coordinates, determine the road where the vehicle is currently located in the target subdivision grid, and establish a vehicle-road dictionary; Read the vehicle sets of each road, obtain the various indicators of the current road based on the vehicle sets, and output the indicators to the global result set.

2. The method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism according to claim 1, characterized in that, The process of acquiring the current map data and dividing it into subdivided grids includes: Obtain the global map extent, read the simulated road network file, extract the minimum or maximum coordinates of the entire road network, and determine the rectangular spatial extent covering the entire scene; Set the parameters for the primary grid and the subdivision grid. Based on the road network density and the simulation scale, set the side length G of the primary grid and the side length S of the subdivision grid, where G is greater than S. The global first-level grid is divided into several rows and several columns of first-level grids according to the step size G, and each first-level grid is assigned a unique number. The subdivision of the mesh is further subdivided into smaller subdivisions within each first-level mesh, with a step size S, and subdivision numbers are assigned. An index table is established to construct a fast mapping relationship between two-level grids, generating a static spatial partitioning structure grid.

3. The method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism according to claim 2, characterized in that, The process of assigning a unique number to each first-level grid includes: For any vehicle or road coordinate point The first-level grid number to which it belongs is calculated by the following formula: Its subdivision grid number is calculated by the following formula: In the formula, and The row and column numbers are for the first-level grid. and These are coordinate values. and These are the minimum x and y coordinates in the global road network. and The row and column numbers for the subdivided grid.

4. The method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism according to claim 1, characterized in that, The process of mapping the acquired target road data to a subdivided grid includes: Obtain all road, lane, centerline coordinates, road range geometry and topology information; Using all subdivision grid numbers as keys, initialize the road list corresponding to each subdivision grid to be empty, and construct an empty binding table; Process each road in turn, extract the coordinate sequence of all centerline points of the road, and calculate the subdivision grid number of each point; Add the current road to the road list of all subdivision grids containing its centerline points, keeping the same road only once in the same grid.

5. The method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism according to claim 1, characterized in that, The establishment of the vehicle and road dictionary includes: Traverse all roads, create an empty vehicle list for each road, and construct an initial dictionary structure; Read the real-time coordinates of the vehicle, calculate the subdivided grid to which it belongs, and obtain the list of candidate roads within that grid. Traverse the candidate roads, determine whether the coordinates are within the road's passability range, and identify the road where the vehicle is located; Add the vehicle ID to the vehicle list for the corresponding road to complete real-time dictionary maintenance.

6. The method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism according to claim 2, characterized in that, The indicators for the current road obtained based on the vehicle set include: Create a thread pool to perform the calculation of various grid metrics in parallel; Traverse all first-level grids, encapsulate the road set within each first-level grid into an independent computation task, and submit it to the thread pool; Each thread reads all roads within its respective first-level grid, retrieves the vehicle list from the vehicle and road dictionary, and calculates the indicators for each item. After each thread completes its calculation, it writes the result to the global result set.

7. The method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism according to claim 1, characterized in that, The metrics include: road occupancy rate, traffic density, average speed, traffic flow, queue length, and delay time.

8. A vehicle simulation evaluation index calculation system based on spatial partitioning and data parallelism, characterized in that, A method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism as described in any one of claims 1-7 includes: The grid division module is configured to acquire the current map data and divide it into subdivided grids, and then map the acquired target road data into the subdivided grids. The dictionary building module is configured to obtain vehicle coordinates, locate the target subdivision grid based on the vehicle coordinates, determine the road where the current vehicle is located in the target subdivision grid, and build a vehicle-road dictionary; The indicator calculation module is configured to read the vehicle set of each road, obtain various indicators of the current road based on the vehicle set, and output the indicators to the global result set.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the vehicle simulation evaluation index calculation method based on spatial partitioning and data parallelism as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements a method for calculating vehicle simulation evaluation indicators based on spatial partitioning and data parallelism as described in any one of claims 1-7.