Urban traffic high-speed simulation platform and method, electronic equipment and storage medium
By designing a high-speed urban traffic simulation platform, using data processing modules, traffic simulation modules and intelligent optimization modules, the problem that existing technology cannot achieve large-scale urban traffic simulation and intelligent optimization is solved, and efficient urban traffic simulation and optimization decisions are achieved. .
Patent Information
- Application Number
- CN202411245266.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-06
- Publication Date
- 2024-11-19
- Estimated Expiration
- 2044-09-06
AI Technical Summary
The existing urban traffic simulation platform cannot realize large-scale urban traffic simulation and diverse decision-making optimization applications, and cannot automatically realize intelligent optimization for arbitrary map data.
A high-speed urban traffic simulation platform is designed, including data processing module, traffic simulation module and intelligent optimization module. By obtaining map data and vehicle data of urban areas, pre-updating and formal updating of simulation objects is carried out according to preset time intervals. Generate traffic simulation data and intelligently optimize based on these data to meet preset optimization goals.
It has realized intelligent optimization of large-scale urban traffic simulation and automation, and can be optimized for any map data, improving traffic simulation efficiency and decision-making support capabilities.
Smart Images

Figure CN118762525B_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of traffic simulation, and in particular to a high-speed urban traffic simulation platform and method, electronic equipment and storage medium. Background Art
[0002] Since its birth, urban traffic simulation technology has been widely used to analyze, study and solve various decision-making optimization problems in urban traffic. Specifically, the urban traffic simulation platform can simulate the driving behavior of vehicles on urban roads and their mutual influence and correlation, reproduce the traffic conditions of specific sections or areas, and help policy makers optimize and test different decision-making plans. Typical decision-making optimization problems include signal light control, lane channelization setting, tidal lane control, road planning, traffic distribution, etc.
[0003] However, the existing urban traffic simulation platform is unable to realize large-scale traffic simulation and various decision optimization applications at the city level, and only supports some decision optimization tasks, and fails to realize the automated intelligent optimization of various decision optimization tasks. Summary of the invention
[0004] In view of this, the present disclosure proposes a city traffic high-speed simulation platform and method, electronic equipment and storage medium, which can realize large-scale city traffic simulation and automatically realize intelligent optimization of optimization tasks for any map data.
[0005] According to one aspect of the present disclosure, a city traffic high-speed simulation platform is provided, comprising: a data processing module, for acquiring map data and vehicle data of a city area of a designated map data to be optimized, the map data comprising: lane data, intersection data, road data, and area of interest data; the designated map data is at least one of the lane data, intersection data, road data, and area of interest data, the vehicle data comprising: vehicle routing data, vehicle attribute data, and initial driving state data, the vehicle routing data representing the vehicle driving trajectory, the vehicle attribute data representing the vehicle driving attributes, and the initial driving state data comprising the initial position information and speed information of the vehicle; a traffic simulation module, for: performing an update operation on a plurality of simulation objects based on the map data and the vehicle data at a preset time interval to obtain traffic simulation data at a plurality of moments, the plurality of simulation objects comprising vehicle objects, lane objects, intersection objects, and area of interest objects, the update operation comprising a pre-update operation and a formal update operation; wherein the pre-update operation comprises: updating the lane index data of the lane object at a previous moment to obtain the lane index data at a current moment, updating the signal light state of the intersection object at a previous moment to obtain the signal light state at a current moment, updating the area of interest object at a The vehicle array at the previous moment is used to obtain the vehicle array at the current moment, and the driving status data of each vehicle object at the previous moment is copied to obtain the backup driving status data at the current moment; the lane index data is used to record the relative position relationship between the vehicle object contained in the lane object and the adjacent vehicle object, and the vehicle array is used to indicate the vehicle object staying in the interest area object; the formal update operation includes: based on the vehicle data of the vehicle object, the lane index data of the lane object at the current moment and the signal light state of the intersection object at the current moment, determining the driving behavior strategy of the vehicle object at the current moment; according to the vehicle object at The driving behavior strategy at the current moment and the backup driving status data of the vehicle object at the previous moment are used to update the driving status data of the vehicle object at the previous moment to obtain the driving status data of the vehicle object at the current moment; wherein the traffic simulation data at the current moment includes the lane index data of the lane object at the current moment, the traffic light status of the intersection object at the current moment, the vehicle array of the area of interest object at the current moment, and the driving status data of the vehicle object at the current moment; an intelligent optimization module is used to optimize the specified map data based on the traffic simulation data at the multiple moments to obtain the target map data that meets the preset optimization goals.
[0006] In one possible implementation, the traffic simulation module includes a pre-update unit and a formal update unit; the pre-update unit is used to use a graphics processor to perform pre-update operations on each simulation object in parallel, and after completing the pre-update operations on each simulation object, synchronize the pre-update completion results to the formal update unit; the formal update unit is used to use the graphics processor to perform formal update operations on each vehicle object in parallel upon receiving the pre-update completion results synchronized by the pre-update unit, and after completing the formal update operations on each vehicle object, synchronize the formal update completion results to the pre-update unit so as to perform the update operation at the next moment.
[0007] In a possible implementation, the obtaining of the map data and vehicle data of the urban area of the designated map data to be optimized includes: obtaining the original map data and original vehicle data of the urban area, converting the original map data and the original vehicle data into map data and vehicle data of a preset data structure and storing them; wherein the preset data structure includes an array structure; the lane data includes: lane number, speed limit information, left lane pointer, right lane pointer, front lane array, rear lane array, vehicle pointer closest to the starting point of the lane, lane index data; the front lane array includes the lane pointer array of the front lane, and the rear lane array includes the lane pointer array of the rear lane; the intersection data includes: the intersection number, the lane array of the lanes connected to the intersection, and the signal light control information, the signal light control information includes the phase array of the intersection signal light, and the phase array includes the state of each phase of the intersection signal light and the duration of each phase; the road data includes: the road number, the lane array of the lanes included in the road, and whether it is one-way; the area of interest data includes: The area number, the array of vehicles including the vehicle in the area of interest, the array of entrances and exits of the area of interest, the lane pointer of the lane connected to the entrance and exit of the area of interest, and the entrance and exit position, wherein the entrance and exit position includes the distance between the entrance and exit and the lane starting point; the area of interest includes a geographical area outside the road; the vehicle routing data includes: the vehicle starting point, the vehicle end point, the vehicle departure time, and the vehicle path; the vehicle path includes a road array of the road passed from the vehicle starting point to the vehicle end point, the vehicle starting point includes the entrance and exit pointer of the area of interest as the starting point, and the vehicle end point includes the entrance and exit pointer of the area of interest as the end point; the vehicle attribute data includes: the vehicle number, the maximum speed of the vehicle itself, the headway, the maximum acceleration, and the maximum braking speed; the initial position information includes the lane pointer of the lane where the vehicle is initially located, the initial vehicle position, and the vehicle index data; wherein the initial vehicle position includes the distance between the vehicle and the lane starting point of the initial lane, and the vehicle index data includes the vehicle pointer located in at least one direction of the left front, front, right front, left rear, rear, and right rear of the vehicle.
[0008] In a possible implementation, the updating of the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment includes: updating the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment according to the vehicle position of the vehicle object contained in the lane object at the current moment; the updating of the traffic light status of the intersection object at the previous moment to obtain the traffic light status at the current moment includes: updating the traffic light status of the intersection object at the previous moment to obtain the traffic light status at the current moment according to the traffic light control information of the intersection object; the updating of the vehicle array of the area of interest object at the previous moment to obtain the vehicle array at the current moment includes: updating the vehicle array of the area of interest object at the previous moment to obtain the vehicle array at the current moment according to the vehicle intersection data of the vehicle object.
[0009] In a possible implementation, the determining of the driving behavior strategy of the vehicle object at the current moment based on the vehicle data of the vehicle object, the lane index data of the lane object at the current moment, and the traffic light state of the intersection object at the current moment includes at least one of the following: determining a lane change strategy according to the vehicle routing data of the vehicle object and the driving status data of the surrounding vehicle objects, the lane change strategy including whether to change lanes, whether to accelerate or decelerate after changing lanes; determining a speed limit strategy of the vehicle object according to the maximum speed limit on the lane object where the vehicle object is located, the maximum speed of the vehicle object itself, and the maximum speed limit of the next lane object when approaching the next lane object, the speed limit strategy being used to indicate the maximum driving speed of the vehicle object at the current moment; ... lane object at the current moment The method comprises the following steps: determining the relative position and relative speed of the vehicle object and the vehicle object in front of the vehicle object according to the relative position and relative speed of the vehicle object and the vehicle object in front of the vehicle object, wherein the following strategy includes the driving speed required to avoid collision with the vehicle object in front of the vehicle object; determining the intersection strategy of each vehicle object according to the traffic light status of the intersection object to be passed by the vehicle object at the current moment, wherein the intersection strategy includes: slowing down and stopping when the intersection object cannot be passed, and the driving speed determined according to the intersection traffic flow when the intersection object is passable; determining the stopping strategy according to the vehicle routing data and the vehicle attribute data of the vehicle object, wherein the stopping strategy includes: reducing the driving speed when the vehicle object approaches the vehicle end point, and entering the interest area of the vehicle end point.
[0010] In a possible implementation, the designated map data is optimized based on the traffic simulation data at the multiple moments to obtain target map data that meets a preset optimization goal, including: determining statistical data required to optimize the designated map data based on the traffic simulation data at the multiple moments, and judging whether the statistical data meets the preset optimization goal; in the case where the statistical data does not meet the preset optimization goal, optimizing the designated map data based on the statistical data using a preset optimization algorithm to obtain optimized designated map data; the preset optimization algorithm includes a black box optimization algorithm and / or a reinforcement learning algorithm; controlling the traffic simulation module to redetermine the traffic simulation data at multiple moments and redetermine the statistical data based on the optimized designated map data until the redetermined statistical data meets the preset optimization goal, thereby obtaining target map data that meets the preset optimization goal.
[0011] In one possible implementation, the simulation platform also provides a data reading interface and a simulation control interface; the data reading interface is used to provide at least one of the map data, vehicle data, traffic simulation data and statistical data within the simulation platform; the simulation control interface is used to modify at least one of the map data and vehicle data within the simulation platform, so that the simulation platform can optimize the specified map data based on at least one of the modified map data and vehicle data; the simulation control interface is also used to control the update progress of the traffic simulation module, and the control of the update progress of the traffic simulation module includes at least one of the following: performing update operations for multiple time periods forward, pausing update operations, and resuming update operations to any historical time period.
[0012] According to another aspect of the present disclosure, a method for simulating urban traffic high-speed is provided, comprising: obtaining map data and vehicle data of an urban area of designated map data to be optimized, the map data comprising: lane data, intersection data, road data, and area of interest data; the designated map data is at least one of the lane data, intersection data, road data, and area of interest data, the vehicle data comprising: vehicle routing data, vehicle attribute data, and initial driving state data, the vehicle routing data representing the vehicle driving trajectory, the vehicle attribute data representing the vehicle driving attributes, and the initial driving state data comprising the initial position information and speed information of the vehicle; at a preset time interval, based on the map data and the vehicle data, performing an update operation on a plurality of simulation objects to obtain traffic simulation data at a plurality of moments, the plurality of simulation objects comprising vehicle objects, lane objects, intersection objects, and area of interest objects, the update operation comprising a pre-update operation and a formal update operation; wherein the pre-update operation comprises: updating the lane index data of the lane object at a previous moment to obtain the lane index data at a current moment, updating the signal light state of the intersection object at a previous moment to obtain the signal light state at a current moment, updating the signal light state of the area of interest object at a previous moment to obtain the signal light state at a current moment, and updating the signal light state of the area of interest object at a previous moment to obtain the signal light state at a current moment. The vehicle array at the previous moment is obtained by copying the driving state data of each vehicle object at the previous moment to obtain the backup driving state data at the current moment; the lane index data is used to record the relative position relationship between the vehicle object contained in the lane object and the adjacent vehicle object, and the vehicle array is used to indicate the vehicle object staying in the interest area object; the formal update operation includes: determining the driving behavior strategy of the vehicle object at the current moment based on the vehicle data of the vehicle object, the lane index data of the lane object at the current moment and the signal light state of the intersection object at the current moment; updating the driving state data of the vehicle object at the previous moment according to the driving behavior strategy of the vehicle object at the current moment and the backup driving state data of the vehicle object at the previous moment to obtain the driving state data of the vehicle object at the current moment; wherein the traffic simulation data at the current moment includes the lane index data of the lane object at the current moment, the signal light state of the intersection object at the current moment, the vehicle array of the interest area object at the current moment and the driving state data of the vehicle object at the current moment; based on the traffic simulation data at the multiple moments, optimizing the specified map data to obtain the target map data that meets the preset optimization target.
[0013] According to another aspect of the present disclosure, an electronic device is provided, comprising: a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to implement the above method when executing the instructions stored in the memory.
[0014] According to another aspect of the present disclosure, a non-volatile computer-readable storage medium is provided, on which computer program instructions are stored, wherein the computer program instructions implement the above method when executed by a processor.
[0015] According to another aspect of the present disclosure, a computer program product is provided, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above method.
[0016] According to various aspects of the present disclosure, by acquiring map data and vehicle data of an urban area, and then pre-updating and formally updating each simulation object based on the map data and vehicle data at a preset time interval, traffic simulation data is obtained, and the specified map data to be optimized is optimized based on the traffic simulation data, thereby realizing a large-scale urban traffic simulation and automatically realizing intelligent optimization of optimization problems for arbitrary map data in an urban area.
[0017] Further features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0018] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments, features, and aspects of the disclosure and, together with the description, serve to explain the principles of the disclosure.
[0019] Figure 1 A block diagram of a high-speed urban traffic simulation platform according to an embodiment of the present disclosure is shown.
[0020] Figure 2 A schematic diagram showing the relationship between various data objects in physical space according to an embodiment of the present disclosure.
[0021] Figure 3 A schematic diagram showing the logical relationship between various data objects according to an embodiment of the present disclosure.
[0022] Figure 4 A schematic diagram showing the workflow of the traffic simulation module 102 according to an embodiment of the present disclosure.
[0023] Figure 5 A schematic diagram of the structure of lane index data according to an embodiment of the present disclosure is shown.
[0024] Figure 6 A schematic diagram showing the interaction relationship between an automatic optimization algorithm, a data reading interface, and a simulation control interface according to an embodiment of the present disclosure is shown.
[0025] Figure 7 A schematic diagram showing a platform workflow according to an embodiment of the present disclosure is shown.
[0026] Figure 8a A road network schematic diagram of a certain urban area according to an embodiment of the present disclosure is shown.
[0027] Figure 8b A schematic diagram of a process for optimizing a traffic light control strategy for multiple intersections in a city area using a model simulation platform according to an embodiment of the present disclosure is shown.
[0028] Figure 9a A schematic diagram of a road network in an urban area with one-way streets to be optimized according to an embodiment of the present disclosure is shown.
[0029] Figure 9b A schematic diagram of a process for completing one-way transformation of 54 roads in a city area using a model simulation platform according to an embodiment of the present disclosure is shown.
[0030] Fig.10 A flow chart of a method for simulating urban traffic high-speed according to an embodiment of the present disclosure is shown.
[0031] Fig.11 A block diagram of an electronic device 1900 according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0032] Various exemplary embodiments, features and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the accompanying drawings represent elements with the same or similar functions. Although various aspects of the embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless otherwise specified.
[0033] The word “exemplary” is used exclusively herein to mean “serving as an example, example, or illustration.” Any embodiment described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.
[0034] The term "and / or" herein is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C can represent including any one or more elements selected from the set consisting of A, B, and C.
[0035] It should be understood that the terms “include” and “comprising” used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.
[0036] In addition, in order to better illustrate the present disclosure, numerous specific details are given in the following specific embodiments. It should be understood by those skilled in the art that the present disclosure can also be implemented without certain specific details. In some examples, methods, means, components and circuits well known to those skilled in the art are not described in detail in order to highlight the subject matter of the present disclosure.
[0037] It can be seen that in order to support large-scale traffic simulation and various decision-making optimization applications at the city level, the urban traffic simulation platform needs to have a coordinated acceleration design of software and hardware to efficiently complete the simulation calculation tasks, and on the other hand, it also needs to provide a variety of convenient data access, modification and simulation control interfaces to support various applications. However, the current urban traffic simulation platforms cannot meet the above requirements. In terms of performance, the existing simulation platform adopts a single-threaded or multi-threaded simulation operation on the central processing unit CPU. Due to the number of CPU cores, it cannot support the parallel simulation calculation of vehicles at the city scale, which means that the simulation platform cannot efficiently complete the city-level traffic simulation calculation, and the application scale is limited; in terms of application, the existing simulation platform usually focuses on the optimization of signal light control, and only provides related functions of obtaining road traffic status and modifying signal light phase, and cannot complete intelligent decision-making applications such as lane optimization control and road planning; and the data access and platform interaction interface of the existing simulation platform are single, which cannot support various application scenarios and intelligent decision-making applications in actual use.
[0038] In recent years, with the development of deep learning technology, the computing acceleration hardware technology system represented by graphics processing units (GPUs) has become increasingly mature, and it has become possible to achieve large-scale accelerated computing on GPUs. How to make good use of the hardware characteristics of GPUs to achieve high-speed computing for large-scale urban traffic simulations and better play the role of simulation platforms in supporting various intelligent decision-making problems is an important direction that needs to be explored.
[0039] In view of this, the embodiments of the present disclosure provide a city traffic high-speed simulation platform, which can fully utilize the hardware-level parallel computing capabilities of the GPU that far exceeds the CPU by building a system architecture designed specifically for the hardware characteristics of the GPU, and realize large-scale high-speed traffic simulation at the city level. It also provides a variety of data reading interfaces, simulation control interfaces, and preset optimization algorithms to effectively support the intelligent solution of various decision-making optimization problems.
[0040] In practical applications, the urban traffic high-speed simulation platform of the embodiment of the present disclosure can be deployed on various terminal devices through software or hardware modification. The terminal device involved in the embodiment of the present disclosure may refer to a device with a wireless connection function and / or a wired connection function. The wireless connection function refers to the ability to connect to other devices through wireless connection methods such as wifi and Bluetooth. The terminal device involved in the embodiment of the present disclosure may also communicate with other devices through a wired connection function. The terminal device involved in the embodiment of the present disclosure may be a touch screen, a non-touch screen, or a screenless terminal device. The touch screen terminal device may be controlled by clicking and sliding on the display screen with a finger or a stylus. The non-touch screen device may be connected to an input device such as a mouse, a keyboard, a touch panel, and the terminal device may be controlled through the input device. For example, the device without a screen may be a Bluetooth speaker without a screen. For example, the terminal device of the present application may include but is not limited to a user equipment (UE), a mobile device, a user terminal, a terminal, a handheld device, a tablet computer, a laptop, a PDA, a computing device, etc.
[0041] The urban traffic high-speed simulation platform of the embodiment of the present disclosure can also be deployed on a server, which can be located in the cloud or locally, and can be a physical device or a virtual device, such as a virtual machine, a container, etc., and has a wireless communication function, wherein the wireless communication function can be set in the chip (system) or other parts or components of the server. It can refer to a device with a wireless connection function, and the wireless connection function means that it can be connected to other servers or terminal devices through wireless connection methods such as Wi-Fi and Bluetooth. The server involved in the embodiment of the present disclosure may also have the function of communicating through a wired connection. For example, the server can receive the map data and vehicle data of the urban area sent by the above-mentioned terminal device, and use the urban traffic high-speed simulation platform deployed on the server to output an optimization plan (that is, target map data that meets the preset optimization target) based on the map data and vehicle data, and return the optimization plan to the terminal device, so as to display the optimization plan to the user through the terminal device, and the embodiment of the present disclosure is not limited to this.
[0042] Figure 1 FIG. 1 is a block diagram of a high-speed urban traffic simulation platform according to an embodiment of the present disclosure. Figure 1 As shown, the simulation platform includes:
[0043] The data processing module 101 is used to obtain map data and vehicle data of the urban area of the designated map data to be optimized, the map data including: lane data, intersection data, road data, and area of interest data; the designated map data is at least one of the lane data, intersection data, road data, and area of interest data, the vehicle data including: vehicle routing data, vehicle attribute data, and initial driving state data, the vehicle routing data represents the vehicle driving trajectory, the vehicle attribute data represents the vehicle driving attributes, and the initial driving state data includes the vehicle's initial position information and speed information;
[0044] The traffic simulation module 102 is used to: perform an update operation on multiple simulation objects based on map data and vehicle data at a preset time interval to obtain traffic simulation data at multiple times, wherein the multiple simulation objects include vehicle objects, lane objects, intersection objects, and interest area objects, and the update operation includes a pre-update operation and a formal update operation;
[0045] The pre-update operation includes: updating the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment, updating the signal light state of the intersection object at the previous moment to obtain the signal light state at the current moment, updating the vehicle array of the interest area object at the previous moment to obtain the vehicle array at the current moment, and copying the driving state data of each vehicle object at the previous moment to obtain the backup driving state data at the current moment; the lane index data is used to record the relative position relationship between the vehicle object contained in the lane object and the adjacent vehicle object, and the vehicle array is used to indicate the vehicle object staying in the interest area object;
[0046] The formal update operation includes: determining the driving behavior strategy of the vehicle object at the current moment based on the vehicle data of the vehicle object, the lane index data of the lane object at the current moment, the signal light state of the intersection object at the current moment, and the vehicle array of the region of interest object at the current moment; updating the driving state data of the vehicle object at the previous moment according to the driving behavior strategy of the vehicle object at the current moment and the backup driving state data of the vehicle object at the previous moment, and obtaining the driving state data of the vehicle object at the current moment;
[0047] The traffic simulation data at the current moment includes lane index data of the lane object at the current moment, the signal light state of the intersection object at the current moment, the vehicle array of the interest area object at the current moment, and the driving state data of the vehicle object at the current moment;
[0048] The intelligent optimization module 103 is used to optimize the designated map data based on the traffic simulation data at multiple moments to obtain target map data that meets the preset optimization target.
[0049] The data objects involved in the simulation platform of the embodiment of the present disclosure are divided into two categories: maps and vehicles. Map objects can be subdivided into areas of interest (AOI), roads, lanes, intersections, traffic lights, speed limit and traffic restriction indicators, etc. Among them, traffic lights and speed limit and traffic restriction indicators are logical objects and do not have physical space entities. For example, Figure 2 Shows the relationship between various data objects in physical space, Figure 3 Shows the logical relationship between various data objects, such as Figure 2 and Figure 3 As shown, the area of interest is used to represent the geographical area outside the road, such as buildings, communities, and schools on the map. Each area of interest will have one or more entrances and exits connected to the road, and vehicles can enter the road network from the area of interest or enter the area of interest from the road network through these entrances and exits; each road is composed of several parallel adjacent lanes in the same direction; an intersection can connect multiple roads, and the lanes on the roads connected by the intersection will connect the lanes of related roads according to the road traffic relationship. Each intersection has at most one logical traffic light to control the traffic status of all roads connected to the intersection; each lane has at most one logical speed limit indication, which is used to indicate the maximum speed limit of the lane, or to mark whether the lane is restricted or not.
[0050] In practical applications, in order to better utilize the single instruction multiple data memory access mode of the GPU, the data items of the data objects of vehicles, AOIs, intersections, roads, and lanes can be stored using an array structure, for example. Specifically, each data object can be stored using its own array structure, where a traffic light can be stored as additional data of an intersection, and a speed limit indication can be stored as additional data of a lane. Thus, the data items of each data object shown in Table 1 can be obtained.
[0051] Table 1 Data items of data objects
[0052]
[0053] It should be understood that the data items contained in the various data objects shown in the above Table 1 are some possible implementation methods provided by the embodiments of the present disclosure. In fact, those skilled in the art can set the data contained in each data object according to actual needs. For example, the road data can also include a data item of whether it is a one-way road, which is used to indicate whether the road is a one-way road. The lane data can also include a lane index data for orderly recording the vehicles on the lane. The embodiments of the present disclosure do not limit this.
[0054] Considering that the data structures of the original map data and the original vehicle data obtained in actual situations may not conform to the data structure shown in Table 1 above, in a possible implementation, the step of obtaining the map data and the vehicle data of the urban area of the designated map data to be optimized includes:
[0055] Acquire original map data and original vehicle data of the urban area, convert the original map data and original vehicle data into map data and vehicle data of a preset data structure and store them; wherein the preset data structure includes an array structure;
[0056] The lane data includes: lane number, speed limit information (i.e. speed limit indication, including maximum speed limit, whether it is restricted, etc.), left lane pointer (used to indicate the left lane), right lane pointer (used to indicate the right lane), front lane array, rear lane array, vehicle pointer closest to the lane starting point (used to indicate the vehicle closest to the lane starting point), lane index data; the front lane array includes a lane pointer array of the front lane (used to indicate the lane in front of the lane, or the lane connected to the lane end), and the rear lane array includes a lane pointer array of the rear lane (used to indicate the lane behind the lane, or the lane connected to the lane starting point); wherein, through the lane index data, a linked list structure is formed between the vehicles on the lane, so that the vehicle passing the lane starting point can access the information of subsequent vehicles in sequence;
[0057] The intersection data includes: the intersection number, the lane array of the lanes connected to the intersection, and the signal light control information, the signal light control information includes the phase array of the intersection signal light, the phase array includes the state of each phase of the intersection signal light (such as the state of the red light, green light, and yellow light, including on and off) and the duration of each phase (such as the duration of the red light, the duration of the green light, and the duration of the yellow light);
[0058] The road data includes: road number, lane array of lanes included in the road, and whether it is one-way;
[0059] The area of interest data includes: an area number, a vehicle array containing vehicles in the area of interest, an entrance and exit array of the area of interest, a lane pointer of a lane connected to the entrance and exit of the area of interest, and an entrance and exit position, wherein the entrance and exit position includes a distance between the entrance and exit and a lane start point; the area of interest includes a geographical area outside the road, such as a building, a community, a school, etc., and each entrance and exit can be represented by the lane number and the entrance and exit position;
[0060] The vehicle routing data includes: a vehicle starting point, a vehicle end point, a vehicle departure time, and a vehicle path; the vehicle path includes a road array of roads passed from the vehicle starting point to the vehicle end point, the vehicle starting point includes an entrance and exit pointer of the interest area as the starting point, and the vehicle end point includes an entrance and exit pointer of the interest area as the end point; wherein the entrance and exit pointer is used to indicate the entrance and exit of the interest area;
[0061] The vehicle attribute data includes: vehicle number, maximum speed of the vehicle itself (i.e., the maximum speed that the vehicle can reach), headway, maximum acceleration (i.e., the maximum acceleration that the vehicle can reach), and maximum braking speed (i.e., the maximum braking speed that the vehicle can reach);
[0062] The initial position information includes a lane pointer of the lane where the vehicle is initially located, an initial vehicle position, and vehicle index data; wherein the initial vehicle position includes the distance between the vehicle and the lane starting point of the lane where the vehicle is initially located, and the vehicle index data mainly includes a vehicle pointer initially located in at least one direction of the left front, front, right front, left rear, rear, and right rear of the vehicle. The vehicle position on the lane can be initialized based on the vehicle index data, and then the lane index data can be used to maintain the relative position relationship of the vehicle objects on the lane objects during subsequent updates.
[0063] It should be noted that the vehicle array in the embodiment of the present disclosure may be an array including vehicle pointers, the lane array may be an array including lane pointers, and the road data may be an array including road pointers. The vehicle pointer may be used to point to a vehicle object, the lane pointer may be used to point to a lane object, and the road pointer may be used to indicate a road object.
[0064] It should be understood that those skilled in the art can adopt data format conversion methods known in the art to convert original map data and original vehicle data into map data and vehicle data in an array structure and store them, and the above array structure is a possible preset data structure provided by the embodiment of the present disclosure. In fact, those skilled in the art can adopt any data structure known in the art as the above preset data structure, and the embodiment of the present disclosure is not limited to this.
[0065] In actual applications, the preset time interval can be customized according to actual needs, and the embodiments of the present disclosure do not limit this. After obtaining the map data and vehicle data, multiple simulation objects can be constructed, that is, vehicle objects, lane objects, intersection objects and area of interest objects are constructed. It can be understood that since the road includes lanes, constructing a lane object is equivalent to constructing a road object. Among them, performing update operations on multiple simulation objects can be understood as updating the status of each simulation object. In order to maximize the parallelization capabilities of the GPU and reduce the delay caused by data access conflicts, copies are created for the data of simulation objects that will have read-write conflicts, and each update is divided into two steps: a pre-update operation and a formal update operation. Synchronization can be performed between each step, and the operations within each step are completed in parallel.
[0066] Thus, the traffic simulation module 102 may include a pre-update unit and a formal update unit; wherein the pre-update unit is used to use the graphics processor to perform the pre-update operation on each simulation object in parallel, and after completing the pre-update operation on each simulation object, synchronize the pre-update completion result to the formal update unit; the formal update unit is used to use the graphics processor to perform the formal update operation on each vehicle object in parallel when receiving the pre-update completion result synchronized by the pre-update unit, and after completing the formal update operation on each vehicle object, synchronize the formal update completion result to the pre-update unit, so as to perform the update operation at the next moment. In this way, the parallelization capability of the GPU can be fully utilized and the read-write conflict between data can be avoided, which is conducive to realizing large-scale urban traffic simulation calculations and improving the efficiency of urban traffic simulation.
[0067] Figure 4 A schematic diagram showing the workflow of the traffic simulation module 102 according to an embodiment of the present disclosure is shown as follows: Figure 4 As shown, in the pre-update operation, for the vehicle object, a data copy is created. Specifically, a copy of the driving status data of the vehicle object can be created to avoid read-write conflicts. For the lane object, the lane index data is maintained, that is, an ordered vehicle index data can be maintained according to the position of the vehicle object contained in the lane object. Each lane object has a vehicle index data, so that the vehicles on the lane can find the vehicles in the six directions of the front, rear, left front, right front, left rear, and right rear according to the vehicle index data. Specifically, the vehicle index data is empty in the initial case. When a vehicle enters the lane, the vehicle needs to be inserted into the appropriate position in the vehicle index data to maintain the orderliness of the vehicle index data. When the vehicle leaves the lane, it also needs to be deleted from the vehicle index data, that is, the vehicle leaving the lane can be deleted from the lane index data, and the vehicle entering the lane can be added to the lane index data to maintain the orderliness of the lane index data. The lane index data can adopt a linked list structure. Figure 5 A schematic diagram of the structure of lane index data is shown, such as Figure 5 As shown, there are three lanes A, B and C. For vehicle B0 on lane B, vehicle B1 in front of vehicle B0, vehicle B2 behind it, vehicle A1 in front of the left, vehicle A2 in the rear left, vehicle C1 in front of the right and vehicle C2 in the rear right can be recorded; for intersection objects, the state of the signal light is updated. Specifically, the state of the signal light is updated according to the signal light control strategy. For example, the countdown of the current signal light phase can be updated. If the countdown ends, switch to the next phase; for the area of interest AOI, the vehicle array is updated. Specifically, since there may be vehicles that leave the AOI and enter the lane in the last formal update, and there may also be vehicles that leave the lane and enter the AOI, the vehicles that have reached the departure time can be deleted from the vehicle array in the AOI and put into the lane where the exit is located, and the vehicles that enter the area of interest are added to the vehicle array in the AOI.
[0068] Based on this, the above-mentioned updating of the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment includes: updating the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment according to the vehicle position of the vehicle object contained in the lane object at the current moment; it should be understood that the vehicle routing data and vehicle attribute data of the known vehicle object can simulate the driving process of the vehicle object on the lane object, that is, the vehicle position of each vehicle object on each lane object at each moment can be calculated, and then the lane index data of the lane object at the previous moment can be updated according to the vehicle position of the vehicle object contained in the lane object at the current moment, wherein the vehicle object entering the lane object is inserted into the vehicle index data, and the vehicle object leaving the lane object is deleted from the vehicle index data.
[0069] The above-mentioned updating of the signal light status of the intersection object at the previous moment to obtain the signal light status at the current moment includes: updating the signal light status of the intersection object at the previous moment to obtain the signal light status at the current moment according to the signal light control information of the intersection object; as mentioned above, the signal light control information includes: the status of each phase of the intersection signal light and the duration of each phase, therefore, the phase change of the signal light of each intersection object can be simulated, thereby calculating the signal light status of the signal light of each intersection object at each moment (i.e. the status of the signal light phase, the remaining time of the lit phase, etc.), thereby realizing the updating of the signal light status of the intersection object, for example, if it is calculated that the signal light status of a certain intersection object at the current moment is that the red light is on, then the signal light status of the intersection object at the current moment is updated to the red light being on and the remaining time of the red light.
[0070] The above-mentioned updating of the vehicle array of the object of the region of interest at the previous moment to obtain the vehicle array of the current moment includes: updating the vehicle array of the object of the region of interest at the previous moment to obtain the vehicle array of the current moment according to the vehicle objects contained in the object of the region of interest at the current moment. It should be understood that the vehicle intersection data of the known vehicle objects can know the departure time of each vehicle object, thereby the vehicle object arriving at the departure time at the current moment can be deleted from the vehicle array of the region of interest, and by simulating the driving process of each vehicle object, the vehicle object arriving at the vehicle end point at the current moment can be determined, and the vehicle object arriving at the vehicle end point can be added to the vehicle array of the region of interest as the vehicle end point, thereby updating the vehicle array of the region of interest.
[0071] like Figure 4 As shown, after the pre-update operation is completed, synchronization is performed, and then the formal update operation is entered; in the formal update operation, driving decisions and driving status updates are performed for the vehicle object. Specifically, the current driving speed can be obtained according to the lane change strategy, speed limit strategy, following strategy, intersection strategy and terminal strategy, and the driving status data such as the position and speed of the vehicle object are updated. If the current lane is completed, the next lane will be entered; after the formal update operation is completed, synchronization is performed, and the update operation at the next moment is entered, and the cycle is repeated in sequence. Among them, the lane change strategy may include: changing lanes according to the current route navigation requirements, or entering a faster lane from a slower lane, and making decisions on whether to change lanes and accelerating or decelerating according to the surrounding vehicle conditions; the speed limit strategy may include: the vehicle speed cannot exceed the maximum speed limit of the current lane, as well as its own maximum speed, and when approaching the next lane, the speed limit of the next lane must also be considered; the following strategy may include considering the relative position and speed relationship of the vehicle in front, and using a suitable following model to calculate the speed required to avoid collision; the intersection strategy may include: observing the intersection signal lights, slowing down and stopping when it is estimated that it is impossible to cross the stop line before the red light, entering the intersection when it is estimated that it can pass, and choosing a suitable speed to drive considering the intersecting traffic in other directions; the stop decision may include: when the vehicle approaches the end point, reducing the driving speed, and entering the area of interest at the entrance and exit at the end point to end the driving. Then, after deciding on various strategies, the acceleration or speed can be determined according to various strategies, and the driving status data such as the speed and position at the current moment can be determined according to the acceleration or speed decided and the speed and vehicle position at the previous moment.
[0072] Therefore, the above-mentioned determination of the driving behavior strategy of the vehicle object at the current moment based on the vehicle data of the vehicle object, the lane index data of the lane object at the current moment, and the traffic light state of the intersection object at the current moment includes at least one of the following:
[0073] Determine a lane change strategy based on the vehicle routing data of the vehicle object and the driving status data of the surrounding vehicle objects, the lane change strategy including whether to change lanes, whether to accelerate or decelerate after changing lanes;
[0074] Determine a speed limit strategy for the vehicle object according to the maximum speed limit of the lane object where the vehicle object is located, the maximum speed of the vehicle object itself, and the maximum speed limit of the next lane object when approaching the next lane object, wherein the speed limit strategy is used to indicate the maximum driving speed of the vehicle object at the current moment;
[0075] Determine the relative position and relative speed of the vehicle object and the front vehicle object according to the lane index data of the lane object at the current moment, and determine the following strategy of each vehicle object according to the phase position and phase speed of the lane object and the front vehicle object, wherein the following strategy includes the driving speed required to avoid collision with the front vehicle object;
[0076] Determine the intersection strategy of each vehicle object according to the current signal light state of the intersection object that the vehicle object is to pass, wherein the intersection strategy includes: slowing down and stopping when the intersection object cannot be passed, and determining the driving speed according to the intersection traffic flow when the intersection object is passable;
[0077] A stopping strategy is determined based on the vehicle routing data and the vehicle attribute data of the vehicle object, wherein the stopping strategy includes: reducing the driving speed when the vehicle object approaches the vehicle end point and entering the interest area of the vehicle end point.
[0078] It should be understood that the vehicle routing data of a known vehicle object can simulate the driving trajectory of the vehicle object from the vehicle starting point to the vehicle end point, and it can also be known whether the vehicle object currently needs to change lanes to meet the driving trajectory. For example, it may be necessary to change from a straight lane to a left turn lane, from a slow lane to a fast lane, from a fast lane to a slow lane, etc.; it can be understood that the driving status of surrounding vehicle objects needs to be considered when changing lanes. For example, when changing lanes to the left, the speed and position of the vehicle objects to the left rear and left front need to be considered. Therefore, it can be decided whether to accelerate or decelerate after changing lanes based on the vehicle routing data and the driving status data of the surrounding vehicle objects; among them, the surrounding vehicle objects of any vehicle object on the lane object can be determined by the lane index data.
[0079] It should be understood that the vehicle's driving speed should not exceed the maximum speed limit of the lane and the maximum speed of the vehicle itself. At the same time, the maximum speed limit of the next lane object should also be considered when approaching the next lane object. Therefore, the maximum driving speed of the vehicle object at the current moment can be determined based on the maximum speed limit of the lane object where the vehicle object is located, the maximum speed of the vehicle object itself, and the maximum speed limit of the next lane object when approaching the next lane object to ensure that the normal driving process of the vehicle object is simulated.
[0080] It should be understood that the vehicle object in front of any vehicle object can be obtained based on the lane index data of the lane object. Thereby, the relative position and relative speed of the vehicle object and the vehicle object in front can be calculated based on the position and speed of each of the vehicle object and the vehicle object in front at the previous moment. In practical applications, those skilled in the art can use any following model known in the art to realize the determination of the driving speed required to avoid a collision with the vehicle object in front based on the relative position and relative speed of the vehicle object and the vehicle object in front, and the embodiments of the present disclosure are not limited to this.
[0081] It should be understood that if the signal light state of the intersection object that the vehicle object is to pass at the current moment is known, the phase of the signal light at the current moment and the remaining time are also known. Therefore, it can be determined whether the vehicle object can pass through the intersection object, that is, whether the intersection object is passable. For example, when the current light is red or the red light is about to come on, it can be considered that the vehicle object cannot pass through the intersection object. In the case where the red light is about to come on, it can be estimated whether the vehicle object can cross the stop line of the intersection object before the red light comes on. If it cannot cross the stop line, it is considered that it cannot pass through the intersection object. Otherwise, it is considered that it can pass through the intersection object. Then, when it is determined that the intersection object cannot be passed, the vehicle object can be slowed down and stopped. Alternatively, when it is determined that the intersection object is passable, the driving speed can be determined according to the intersection traffic flow. The intersection traffic flow is the vehicle object that intersects at the intersection object and enters the same lane as the current vehicle object. It can be known that the speed of the intersection traffic flow is usually reduced. Therefore, in order to avoid collision, it is necessary to determine the driving speed of the vehicle object that is about to pass through the intersection object in combination with the speed of the intersection traffic flow.
[0082] As described above, the vehicle routing data and vehicle attribute data of the known vehicle object can simulate the driving trajectory of the vehicle object from the vehicle starting point to the vehicle end point. When the vehicle object is close to the vehicle end point (for example, when the distance between the vehicle object and the vehicle end point is less than a specified threshold), it is necessary to reduce the driving speed and enter the area of interest from the entrance of the area of interest at the vehicle end point. The reduced driving speed can be determined in combination with the vehicle attribute data of the vehicle object (such as the maximum braking speed), so as to effectively simulate the behavior of the vehicle object arriving at the end point.
[0083] In actual applications, by cyclically executing the above-mentioned pre-update operation and formal update operation based on the above-mentioned vehicle data and map data at preset time intervals, traffic simulation data at multiple moments can be obtained. The traffic simulation data at each moment may include the lane index data of the above-mentioned lane object at each moment, the traffic light status of the intersection object at each moment, the vehicle array of the area of interest object at each moment, and the driving status data of the vehicle object at each moment; of course, it may also include other status data that can be obtained from the above-mentioned update process, for example, it may also include the driving status data of the vehicle object at each moment. It should be understood that the driving status data at the first moment may be the initial driving status data of the vehicle object.
[0084] After the traffic simulation module 102 obtains the traffic simulation data at multiple times, the intelligent optimization module 103 optimizes the specified map data based on the traffic simulation data at multiple times to obtain target map data that meets the preset optimization target, including:
[0085] Based on the traffic simulation data at multiple times, determine the statistical data required to optimize the specified map data, and judge whether the statistical data meets the preset optimization goals;
[0086] When the statistical data does not meet the preset optimization target, a preset optimization algorithm is used to optimize the specified map data based on the statistical data to obtain optimized specified map data; the preset optimization algorithm includes a black box optimization algorithm and / or a reinforcement learning algorithm;
[0087] The control traffic simulation module redetermines the traffic simulation data at multiple times and redetermines the statistical data based on the optimized designated map data until the redetermined statistical data meets the preset optimization target, thereby obtaining the target map data that meets the preset optimization target.
[0088] As described above, the designated map data may be at least one of lane data, intersection data, road data, and area of interest data. For example, if a user wishes to optimize the traffic light control strategy at an intersection to reduce congestion in an urban area, the map data to be optimized may be the signal control information of the relevant intersection object. If a user wishes to transform roads in an urban area into one-way streets to improve traffic operation efficiency, the map data to be optimized may be the road data of the relevant roads.
[0089] Among them, statistical data may include, but are not limited to: the number of vehicles in the lane, the average speed of the lane, the average speed of the road, the traffic volume at the intersection, the length of the vehicle queue, the average vehicle travel time and other data. It should be understood that those skilled in the art can determine the required statistical data according to actual needs in order to optimize the specified map data, and the embodiments of the present disclosure are not limited to this.
[0090] Among them, the preset optimization target can be customized according to actual needs. For example, if the user expects to optimize the traffic light control strategy at the intersection to reduce congestion in the urban area, the preset optimization target can be minimizing the length of the vehicle queue, wherein the length of the vehicle queue can be understood as the number of vehicles queuing at the intersection, wherein the vehicle queue length can be obtained by calculating the number of vehicle objects with a vehicle speed of 0 between the intersection and the starting point of the lane connecting the intersection; if the user hopes to transform the roads in the urban area into one-way streets to improve traffic operation efficiency, then the preset optimization target can be minimizing the average vehicle travel time, wherein the average vehicle travel time can be understood as the average time it takes for the vehicle object to enter the simulation area and leave the simulation data, wherein the simulation area can be the entire urban area or a part of the urban area, and the embodiments of the present disclosure are not limited to this.
[0091] It should be understood that when the statistical data does not meet the preset optimization target, for example, the length of the vehicle queue is still greater than a certain length threshold, and the average vehicle travel time is still greater than a certain time threshold, the preset preset optimization algorithm can be used to optimize the specified map data based on the statistical data to obtain the optimized specified map data, wherein the preset optimization algorithm may include but is not limited to any known black box optimization algorithm and reinforcement learning algorithm in the art. Black box optimization is to regard the object to be optimized as a black box with an opaque internal mechanism, and to find the optimal parameter combination by constantly trying various parameters. It is suitable for optimization scenarios where parameters do not need to change according to real-time traffic conditions, such as finding the most suitable road repair plan, setting speed limit and traffic restriction instructions, etc. The goal of reinforcement learning is to obtain an optimal interactive strategy, or an intelligent agent, which can respond to changes in real-time traffic conditions to maximize long-term benefits. Related use scenarios include finding the optimal intersection signal control strategy to maximize throughput, finding the optimal tidal lane switching strategy to minimize the average vehicle travel time and other control strategies. It is understandable that those skilled in the art may also use other known optimization algorithms in the art, and the embodiments of the present disclosure are not limited to this.
[0092] After obtaining the optimized map data using the preset optimization algorithm, the traffic simulation module 102 can be controlled to redetermine the traffic simulation data at multiple times and redetermine the statistical data based on the optimized specified map data until the redetermined statistical data meets the preset optimization target, thereby obtaining the target map data that meets the preset optimization target.
[0093] In actual applications, in order to facilitate the interaction between the platform and the outside and meet the user's data reading and simulation control needs, in a possible implementation method, the simulation platform also provides a data reading interface and a simulation control interface; wherein the data reading interface is used to provide at least one of the map data, vehicle data, traffic simulation data and statistical data within the simulation platform, for example, that is, the internal information of the simulation platform can be provided to the outside, such as but not limited to map data such as lanes, roads, intersections, AOIs, vehicle position and speed, vehicle information around the vehicle, vehicle data such as vehicles in intersections, and traffic simulation data, as well as statistical data obtained by statistical aggregation based on the current and historical states of the simulation platform, such as the number of vehicles in the lane, the average speed of the lane, the average speed of the road, the traffic volume at the intersection, etc., which is not limited to the embodiments of the present disclosure.
[0094] The simulation control interface is used to modify at least one of the map data and vehicle data in the simulation platform, so that the simulation platform can optimize the specified map data based on at least one of the modified map data and vehicle data; that is, the simulation control interface provides a way to modify the map data and vehicle data, wherein modifying the map data and vehicle data of the data object can be understood as modifying the data items of the data object shown in Table 1 above, such as modifying the vehicle path, modifying the road speed limit and traffic restriction status, modifying the intersection signal light status, etc., which is not limited by the embodiment of the present disclosure. The simulation control interface can also be used to control the simulation progress, and controlling the simulation progress can be understood as controlling the update progress of the traffic module, that is, the simulation control interface is also used to control the update progress of the traffic simulation module, wherein controlling the update progress of the traffic simulation module includes at least one of the following: performing update operations at multiple moments forward (that is, simulating several steps forward), pausing the update operation (that is, pausing the simulation), and restoring the update operation to any historical moment (that is, restoring the simulation to a certain historical moment).
[0095] In practical applications, the intelligent optimization module 103 can rely on the simulation capability of the traffic simulation module 102 for real traffic, and interact with the data reading interface and simulation control interface provided by the platform, so as to use the automatic optimization algorithm to find the optimal strategy, such as Figure 6The interactive relationship between the automatic optimization algorithm and the data reading interface and the simulation control interface is shown. The automatic optimization algorithm can obtain object data such as vehicle data, lane data, intersection data, and obtain lane vehicle data, average speed, traffic flow and other statistical data through the data reading interface; it can also modify object data such as vehicle path, road restrictions and speed limits, traffic light status, etc. through the simulation control interface, and can control the simulation progress, that is, control forward simulation, pause simulation, restore simulation to historical moments, etc., and can also include: archiving traffic simulation data, restoring the simulation to the archived state at a certain historical moment and other control functions.
[0096] Based on the above-mentioned urban traffic high-speed simulation platform, the embodiment of the present disclosure also provides the following Figure 7 A schematic diagram of a simulation platform workflow is shown in FIG. Figure 7 As shown, the original data (i.e., the original map data and the original vehicle data) and the optimization problem can be input, that is, the original map data, the original vehicle data and the optimization problem (used to indicate the specified map data to be optimized) of the studied urban area can be input into the simulation platform, and the data processing module 101 of the simulation platform will convert the data format of the input original data, that is, convert the input original data into an efficient data structure storage suitable for the hardware characteristics of the GPU; the traffic simulation module 102 will complete the efficient simulation calculation of large-scale vehicle behaviors in parallel on the GPU based on the built-in vehicle driving simulation algorithm, and obtain the fine-grained vehicle trajectory information and the overall traffic operation situation of the urban area under the specified road network and traffic facilities settings; next, the intelligent optimization module 103 can automatically optimize according to the simulation results of the traffic simulation module 102 (such as the above-mentioned traffic simulation data), the optimization objective function (i.e., the preset optimization algorithm), the optimization object (i.e., the specified map data to be optimized) and the parameter range (i.e., the preset optimization range) set by the user, and obtain the optimization solution (such as the target map data). Among them, thanks to the platform's efficient traffic simulation, within a limited time, the traffic simulation module 102 and the intelligent optimization module 103 can iteratively complete multiple rounds of optimization and continuously improve the effect of the optimization solution.
[0097] According to the embodiments of the present disclosure, by acquiring map data and vehicle data of an urban area, and then pre-updating and formally updating each simulation object based on the map data and vehicle data at preset time intervals, traffic simulation data is obtained, and then an intelligent optimization module is used to optimize the specified map data to be optimized based on the traffic simulation data, thereby realizing a large-scale urban traffic simulation and automatically realizing intelligent optimization of optimization problems for arbitrary map data in an urban area.
[0098] For example, for Figure 8a The schematic diagram of the road network in a certain urban area is shown, and the embodiment of the present disclosure provides Figure 8b The figure shows a flow chart of using a model simulation platform to complete the optimization of the signal light control strategy for multiple intersections in a city area, so that the signal lights at these intersections can intelligently switch phases according to road condition information such as the length of the vehicle queue at the current intersection, so as to minimize the length of the vehicle queue in the city area. In actual applications, users can obtain the original map data of a city area, including road data, lane data, AOI data, intersection data, and obtain the original vehicle data in the city area. Users can use C++ and CUDA (Compute Unified Device Architecture) on the GPU to deploy and build a simulation platform. The data structure of the above-mentioned vehicle, AOI, intersection, road, lane and other data objects can be implemented using a struct structure, and each data object can be stored using a structure array. Then the above-mentioned original map data and original vehicle data can be imported into the structure array of the corresponding data object to realize data structure conversion. Figure 8a As shown, the user can import map data and vehicle data into the platform, and the platform can use the traffic simulation module to perform simulation calculations, wherein the pre-update and formal update steps of the simulation calculations can be implemented as two functions on the GPU, which are called cyclically during the simulation process, and the synchronization function is used for global synchronization after each call. Pybind11 can be used to implement the data access interface and simulation control interface of the simulation platform, so that the intelligent optimization module can interact with the traffic simulation module to obtain the vehicle queue length information of the lanes connected to the specified intersection, that is, to count the length of the vehicle queue at the intersection, and then use the agent decision signal phase to modify the signal light state of the specified intersection. Python can be used to implement a multi-layer perceptron neural network agent model in the intelligent optimization module, taking the vehicle queue length of the lanes connected to the intersection as input and outputting the signal light state of the intersection. The agent model reads the required input data through the data access interface of the simulation platform, and changes the signal light state of the intersection in the traffic simulation module to its own output value through the simulation control interface. After a large number of interactions between the intelligent agent and the traffic simulation module, the intelligent agent behavior and simulation result data can be collected (that is, the interaction data is collected cyclically, such as the above-mentioned traffic simulation data, the statistical vehicle queue length, etc.), and then the reinforcement learning algorithm can be used to use the interaction data to train the intelligent agent model with the goal of minimizing the vehicle queue length, and the Adam optimizer is used to update the neural network parameters of the intelligent agent model. After several rounds of training (that is, after reaching the preset time limit), an intelligent agent model that can intelligently switch phases according to road conditions such as the vehicle queue length at the current intersection is obtained, thereby realizing the optimization of traffic light control strategies for multiple intersections using the simulation platform.
[0099] For example, for Figure 9a The schematic diagram of the road network of a certain urban area to be optimized for one-way travel is shown in FIG. Figure 9b The figure shows a flow chart of using a model simulation platform to complete the one-way transformation of 54 roads in a certain urban area, wherein the one-way transformation of multiple roads can be understood as selecting the most suitable several roads from multiple candidate transformation roads to transform them into one-way roads, so that the overall traffic operation efficiency in the area is improved and the average vehicle travel time is reduced. In practical applications, users can obtain the urban map data of the urban area, including road data, lane data, AOI data, intersection data, and obtain vehicle data on the corresponding map of the urban area. Users can set multiple roads to be transformed into two types of roads, one is a one-way road and the other is a two-way road, that is, the two-way road can be released, the one-way road is restricted, or the two-way road is restricted and the one-way road is released, so as to achieve the effect of not transforming the one-way road and transforming the one-way road equivalently. Users can use C++ and CUDA technology on GPU to deploy and build simulation platforms. The data structures of the above-mentioned vehicles, AOIs, intersections, roads, lanes and other data objects can be implemented using struct structures, and each data object can be stored in a structure array. Then, the above-mentioned original map data and original vehicle data can be imported into the structure array of the corresponding data object to realize data structure conversion. Figure 9b As shown, users can import map data and vehicle data into the platform and initialize optimization variables , each optimization variable represents a road object to be optimized, and the restriction status is set according to the optimization variable, that is, the restriction status of two-way roads is set to be released, one-way roads are restricted, or two-way roads are restricted, one-way roads are released, and then the traffic simulation module is used to simulate and calculate the average travel time in the urban area (that is, the average vehicle travel time). Among them, the pre-update and formal update steps of the simulation calculation are implemented as two functions on the GPU. These two functions are called cyclically during the simulation process, and the preset synchronization function is used for global synchronization after each call. Among them, pybind11 can be used to implement the data access interface and simulation control interface of the simulation platform, so that the intelligent optimization module can interact with the traffic simulation module, obtain the average vehicle travel time in the simulation area, and modify the restriction status of the specified road. Among them, the one-way street transformation problem can be modeled as a black box optimization problem, using 54 0 / 1 variables Indicates whether each road should be converted into a one-way street. For each variable, when its value is 0, it means that the corresponding road will not be converted into a one-way street, and the road status of the road can be set to two-way road with no restrictions, one-way road with restrictions, etc. through the simulation control interface. When its value is 1, it means that the corresponding road will be converted into a one-way street, and the road status of the road can be set to two-way road with restrictions, one-way road with no restrictions, etc. through the simulation control interface. After setting the road status for different variables, the average vehicle travel time under the road status can be obtained through simulation calculation, and this average vehicle travel time can be used as the optimization target of the black box optimization problem, that is, the existing black box optimizer can be used to update , that is, to find The optimal parameter combination, where the average travel time can be recalculated based on the optimized variables using the traffic simulation module before the time limit is reached. After several rounds of iterations (i.e., after the time limit is reached), the optimal result is selected to obtain the final road reconstruction plan, which is the output Corresponding transformation plan.
[0100] It should be understood that the above Figure 8b and Figure 9b The two optimization scenarios used are some possible optimization scenarios provided by the embodiments of the present disclosure. In fact, those skilled in the art can use the simulation platform provided by the embodiments of the present disclosure to implement intelligent optimization for various optimization problems for map data, and the embodiments of the present disclosure do not limit this.
[0101] Based on the above simulation platform, the embodiment of the present disclosure also provides Fig.10 A flow chart of a method for simulating high-speed urban traffic is shown in FIG. Fig.10 As shown, the method includes:
[0102] Step S11, obtaining map data and vehicle data of the urban area of the designated map data to be optimized, wherein the map data includes: lane data, intersection data, road data, and area of interest data; the designated map data is at least one of the lane data, intersection data, road data, and area of interest data; the vehicle data includes: vehicle routing data, vehicle attribute data, and initial driving state data, wherein the vehicle routing data represents the vehicle driving trajectory, the vehicle attribute data represents the vehicle driving attributes, and the initial driving state data includes the vehicle's initial position information and speed information;
[0103] Step S12, performing an update operation on multiple simulation objects based on the map data and the vehicle data at a preset time interval to obtain traffic simulation data at multiple times, wherein the multiple simulation objects include vehicle objects, lane objects, intersection objects, and area of interest objects, and the update operation includes a pre-update operation and a formal update operation;
[0104] The pre-update operation includes: updating the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment, updating the signal light state of the intersection object at the previous moment to obtain the signal light state at the current moment, updating the vehicle array of the interest area object at the previous moment to obtain the vehicle array at the current moment, and copying the driving state data of each vehicle object at the previous moment to obtain the backup driving state data at the current moment; the lane index data is used to record the relative position relationship between the vehicle object contained in the lane object and the adjacent vehicle object, and the vehicle array is used to indicate the vehicle object staying in the interest area object;
[0105] The formal update operation includes: determining the driving behavior strategy of the vehicle object at the current moment based on the vehicle data of the vehicle object, the lane index data of the lane object at the current moment, and the signal light state of the intersection object at the current moment; updating the driving state data of the vehicle object at the previous moment according to the driving behavior strategy of the vehicle object at the current moment and the backup driving state data of the vehicle object at the previous moment, so as to obtain the driving state data of the vehicle object at the current moment;
[0106] The traffic simulation data at the current moment includes lane index data of the lane object at the current moment, the signal light state of the intersection object at the current moment, the vehicle array of the interest area object at the current moment, and the driving state data of the vehicle object at the current moment;
[0107] Step S13, optimizing the designated map data based on the traffic simulation data at the multiple moments to obtain target map data that meets a preset optimization goal.
[0108] In a possible implementation, the method also includes: using a graphics processor to perform pre-update operations on each simulation object in parallel, and synchronizing the pre-update completion results after completing the pre-update operations on each simulation object; after synchronizing the pre-update completion results, using the graphics processor to perform formal update operations on each vehicle object in parallel, and after completing the formal update operations on each vehicle object, synchronizing the formal update completion results to the pre-update unit to perform the update operation at the next moment.
[0109] In a possible implementation, the obtaining of the map data and vehicle data of the urban area of the designated map data to be optimized includes: obtaining the original map data and original vehicle data of the urban area, converting the original map data and the original vehicle data into map data and vehicle data of a preset data structure and storing them; wherein the preset data structure includes an array structure; the lane data includes: lane number, speed limit information, left lane pointer, right lane pointer, front lane array, rear lane array, vehicle pointer closest to the starting point of the lane, lane index data; the front lane array includes the lane pointer array of the front lane, and the rear lane array includes the lane pointer array of the rear lane; the intersection data includes: the intersection number, the lane array of the lanes connected to the intersection, and the signal light control information, the signal light control information includes the phase array of the intersection signal light, and the phase array includes the state of each phase of the intersection signal light and the duration of each phase; the road data includes: the road number, the lane array of the lanes included in the road, and whether it is one-way; the area of interest data includes: The area number, the array of vehicles including the vehicle in the area of interest, the array of entrances and exits of the area of interest, the lane pointer of the lane connected to the entrance and exit of the area of interest, and the entrance and exit position, wherein the entrance and exit position includes the distance between the entrance and exit and the lane starting point; the area of interest includes a geographical area outside the road; the vehicle routing data includes: the vehicle starting point, the vehicle end point, the vehicle departure time, and the vehicle path; the vehicle path includes a road array of the road passed from the vehicle starting point to the vehicle end point, the vehicle starting point includes the entrance and exit pointer of the area of interest as the starting point, and the vehicle end point includes the entrance and exit pointer of the area of interest as the end point; the vehicle attribute data includes: the vehicle number, the maximum speed of the vehicle itself, the headway, the maximum acceleration, and the maximum braking speed; the initial position information includes the lane pointer of the lane where the vehicle is initially located, the initial vehicle position, and the vehicle index data; wherein the initial vehicle position includes the distance between the vehicle and the lane starting point of the initial lane, and the vehicle index data includes the vehicle pointer located in at least one direction of the left front, front, right front, left rear, rear, and right rear of the vehicle.
[0110] In a possible implementation, the updating of the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment includes: updating the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment according to the vehicle position of the vehicle object contained in the lane object at the current moment; the updating of the traffic light status of the intersection object at the previous moment to obtain the traffic light status at the current moment includes: updating the traffic light status of the intersection object at the previous moment to obtain the traffic light status at the current moment according to the traffic light control information of the intersection object; the updating of the vehicle array of the area of interest object at the previous moment to obtain the vehicle array at the current moment includes: updating the vehicle array of the area of interest object at the previous moment to obtain the vehicle array at the current moment according to the vehicle intersection data of the vehicle object.
[0111] In a possible implementation, the determining of the driving behavior strategy of the vehicle object at the current moment based on the vehicle data of the vehicle object, the lane index data of the lane object at the current moment, and the traffic light state of the intersection object at the current moment includes at least one of the following: determining a lane change strategy according to the vehicle routing data of the vehicle object and the driving status data of the surrounding vehicle objects, the lane change strategy including whether to change lanes, whether to accelerate or decelerate after changing lanes; determining a speed limit strategy of the vehicle object according to the maximum speed limit on the lane object where the vehicle object is located, the maximum speed of the vehicle object itself, and the maximum speed limit of the next lane object when approaching the next lane object, the speed limit strategy being used to indicate the maximum driving speed of the vehicle object at the current moment; ... lane object at the current moment The method comprises the following steps: determining the relative position and relative speed of the vehicle object and the vehicle object in front of the vehicle object according to the relative position and relative speed of the vehicle object and the vehicle object in front of the vehicle object, wherein the following strategy includes the driving speed required to avoid collision with the vehicle object in front of the vehicle object; determining the intersection strategy of each vehicle object according to the traffic light status of the intersection object to be passed by the vehicle object at the current moment, wherein the intersection strategy includes: slowing down and stopping when the intersection object cannot be passed, and the driving speed determined according to the intersection traffic flow when the intersection object is passable; determining the stopping strategy according to the vehicle routing data and the vehicle attribute data of the vehicle object, wherein the stopping strategy includes: reducing the driving speed when the vehicle object approaches the vehicle end point, and entering the interest area of the vehicle end point.
[0112] In a possible implementation, the designated map data is optimized based on the traffic simulation data at the multiple moments to obtain target map data that meets a preset optimization goal, including: determining statistical data required to optimize the designated map data based on the traffic simulation data at the multiple moments, and judging whether the statistical data meets the preset optimization goal; in the case where the statistical data does not meet the preset optimization goal, optimizing the designated map data based on the statistical data using a preset optimization algorithm to obtain optimized designated map data; the preset optimization algorithm includes a black box optimization algorithm and / or a reinforcement learning algorithm; controlling the traffic simulation module to redetermine the traffic simulation data at multiple moments and redetermine the statistical data based on the optimized designated map data until the redetermined statistical data meets the preset optimization goal, thereby obtaining target map data that meets the preset optimization goal.
[0113] In a possible implementation, the method further includes: using a data reading interface to provide at least one of map data, vehicle data, traffic simulation data and statistical data within the simulation platform; using a simulation control interface to modify at least one of the map data and vehicle data within the simulation platform, so that the simulation platform completes the optimization of the specified map data based on at least one of the modified map data and vehicle data; using the simulation control interface to control the update progress, and the control of the update progress includes at least one of the following: performing update operations at multiple time points forward, pausing the update operation, and resuming the update operation to any historical time point.
[0114] According to the method of the embodiment of the present disclosure, by acquiring map data and vehicle data of the urban area, and then pre-updating and formally updating each simulation object based on the map data and vehicle data at a preset time interval, traffic simulation data is obtained, and the specified map data to be optimized is optimized based on the traffic simulation data, so as to realize a large-scale urban traffic simulation and automatically realize the intelligent optimization of the optimization problem of any map data in the urban area.
[0115] In some embodiments, the specific implementation of the method provided by the embodiments of the present disclosure can refer to the description of the embodiments of the simulation platform above, and for the sake of brevity, it will not be repeated here.
[0116] The present disclosure also provides a computer-readable storage medium on which computer program instructions are stored, and when the computer program instructions are executed by a processor, the above platform or method is implemented. The computer-readable storage medium may be a volatile or non-volatile computer-readable storage medium.
[0117] An embodiment of the present disclosure further proposes an electronic device, comprising: a processor; and a memory for storing instructions executable by the processor; wherein the processor is configured to implement the above-mentioned platform or method when executing the instructions stored in the memory.
[0118] The embodiments of the present disclosure also provide a computer program product, including a computer-readable code, or a non-volatile computer-readable storage medium carrying the computer-readable code. When the computer-readable code runs in a processor of an electronic device, the processor in the electronic device executes the above platform or method.
[0119] Fig.11 1 is a block diagram of an electronic device 1900 according to an embodiment of the present disclosure. For example, the electronic device 1900 may be provided as a server or a terminal device. Fig.11 , the electronic device 1900 includes a processing component 1922, which further includes one or more processors, and a memory resource represented by a memory 1932 for storing instructions executable by the processing component 1922, such as an application. The application stored in the memory 1932 may include one or more modules, each corresponding to a set of instructions. In addition, the processing component 1922 is configured to execute instructions to perform the above platform or method.
[0120] The electronic device 1900 may also include a power supply component 1926 configured to perform power management of the electronic device 1900, a wired or wireless network interface 1950 configured to connect the electronic device 1900 to a network, and an input / output interface 1958 (I / O interface). The electronic device 1900 may operate based on an operating system stored in the memory 1932, such as Windows Server 2000. TM , Mac OS X TM , Unix TM , Linux TM , FreeBSD TM or similar.
[0121] In an exemplary embodiment, a non-volatile computer-readable storage medium is also provided, such as a memory 1932 including computer program instructions, which can be executed by the processing component 1922 of the electronic device 1900 to implement the above platform or method.
[0122] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.
[0123] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the above. More specific examples (a non-exhaustive list) of computer-readable storage media include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the above. The computer-readable storage medium used herein is not to be interpreted as a transient signal itself, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.
[0124] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.
[0125] The computer program instructions for performing the operations of the present disclosure may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages, such as Smalltalk, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer, partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., using an Internet service provider to connect through the Internet). In some embodiments, by using the state information of the computer-readable program instructions to personalize an electronic circuit, such as a programmable logic circuit, a field programmable gate array (FPGA), or a programmable logic array (PLA), the electronic circuit may execute the computer-readable program instructions, thereby implementing various aspects of the present disclosure.
[0126] Various aspects of the present disclosure are described herein with reference to flowcharts and / or block diagrams of methods, platforms (systems) and computer program products according to embodiments of the present disclosure. It should be understood that each box in the flowchart and / or block diagram and the combination of boxes in the flowchart and / or block diagram can be implemented by computer-readable program instructions.
[0127] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, thereby producing a machine, so that when these instructions are executed by the processor of the computer or other programmable data processing device, a device that implements the functions / actions specified in one or more boxes in the flowchart and / or block diagram is generated. These computer-readable program instructions can also be stored in a computer-readable storage medium, and these instructions cause the computer, programmable data processing device, and / or other equipment to work in a specific manner, so that the computer-readable medium storing the instructions includes a manufactured product, which includes instructions for implementing various aspects of the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0128] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device so that a series of operating steps are performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to implement the functions / actions specified in one or more boxes in the flowchart and / or block diagram.
[0129] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system, method and computer program product according to multiple embodiments of the present disclosure. In this regard, each square box in the flow chart or block diagram can represent a part of a module, program segment or instruction, and a part of the module, program segment or instruction includes one or more executable instructions for realizing the specified logical function. In some alternative implementations, the function marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or action, or can be implemented with a combination of special hardware and computer instructions.
[0130] The embodiments of the present disclosure have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.
Claims
1. A high-speed urban traffic simulation platform, characterized in that: include: a data processing module, for acquiring map data and vehicle data of a city area of designated map data to be optimized, wherein the map data includes lane data, intersection data, road data, and area of interest data; the designated map data is at least one of the lane data, intersection data, road data, and area of interest data; the vehicle data includes vehicle routing data, vehicle attribute data, and initial driving state data, wherein the vehicle routing data represents a vehicle driving trajectory, the vehicle attribute data represents a vehicle driving attribute, and the initial driving state data includes initial position information and speed information of the vehicle; A traffic simulation module is used to: perform an update operation on a plurality of simulation objects based on the map data and the vehicle data at a preset time interval to obtain traffic simulation data at a plurality of times, wherein the plurality of simulation objects include vehicle objects, lane objects, intersection objects, and area of interest objects, and the update operation includes a pre-update operation and a formal update operation; The pre-update operation includes: updating the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment, updating the signal light state of the intersection object at the previous moment to obtain the signal light state at the current moment, updating the vehicle array of the interest area object at the previous moment to obtain the vehicle array at the current moment, and copying the driving state data of each vehicle object at the previous moment to obtain the backup driving state data at the current moment; the lane index data is used to record the relative position relationship between the vehicle object contained in the lane object and the adjacent vehicle object, and the vehicle array is used to indicate the vehicle object staying in the interest area object; The formal update operation includes: determining the driving behavior strategy of the vehicle object at the current moment based on the vehicle data of the vehicle object, the lane index data of the lane object at the current moment, and the signal light state of the intersection object at the current moment; updating the driving state data of the vehicle object at the previous moment according to the driving behavior strategy of the vehicle object at the current moment and the backup driving state data of the vehicle object at the previous moment to obtain the driving state data of the vehicle object at the current moment; The traffic simulation data at the current moment includes lane index data of the lane object at the current moment, the signal light state of the intersection object at the current moment, the vehicle array of the interest area object at the current moment, and the driving state data of the vehicle object at the current moment; The intelligent optimization module is used to optimize the specified map data based on the traffic simulation data at the multiple moments to obtain target map data that meets the preset optimization target.
2. The platform according to claim 1, characterized in that The traffic simulation module includes a pre-update unit and a formal update unit; The pre-update unit is used to use the graphics processor to perform pre-update operations on each simulation object in parallel, and after completing the pre-update operations on each simulation object, synchronize the pre-update completion results to the formal update unit; The formal update unit is used to use the graphics processor to perform formal update operations on each vehicle object in parallel upon receiving the pre-update completion result synchronized by the pre-update unit, and after completing the formal update operation on each vehicle object, synchronize the formal update completion result to the pre-update unit to perform the update operation at the next moment.
3. The platform according to claim 1, characterized in that: The step of obtaining the map data and vehicle data of the urban area of the designated map data to be optimized includes: Acquire original map data and original vehicle data of the urban area, convert the original map data and the original vehicle data into map data and vehicle data of a preset data structure and store them; wherein the preset data structure includes an array structure; The lane data includes: lane number, speed limit information, left lane pointer, right lane pointer, front lane array, rear lane array, vehicle pointer closest to the lane start point, lane index data; the front lane array includes a lane pointer array of the front lane, and the rear lane array includes a lane pointer array of the rear lane; The intersection data includes: an intersection number, a lane array of lanes connected to the intersection, and signal light control information, wherein the signal light control information includes a phase array of the intersection signal light, and the phase array includes the state of each phase of the intersection signal light and the duration of each phase; The road data includes: road number, lane array of lanes included in the road, and whether it is one-way; The area of interest data includes: an area number, a vehicle array containing vehicles in the area of interest, an entrance and exit array of the area of interest, a lane pointer of a lane connected to the entrance and exit of the area of interest, and an entrance and exit position, wherein the entrance and exit position includes a distance between the entrance and exit and a lane start point; the area of interest includes a geographical area outside a road; The vehicle routing data includes: a vehicle starting point, a vehicle end point, a vehicle departure time, and a vehicle path; the vehicle path includes a road array of roads passed from the vehicle starting point to the vehicle end point, the vehicle starting point includes an entrance and exit pointer of an interest area as a starting point, and the vehicle end point includes an entrance and exit pointer of an interest area as a end point; The vehicle attribute data includes: vehicle number, maximum speed of the vehicle itself, headway, maximum acceleration and maximum braking speed; The initial position information includes a lane pointer of the vehicle's initial lane, an initial vehicle position, and vehicle index data; wherein the initial vehicle position includes the distance between the vehicle and the lane starting point of the initial lane, and the vehicle index data includes a vehicle pointer located in at least one direction of the vehicle's left front, front, right front, left rear, rear, and right rear.
4. The platform according to claim 3, characterized in that: The updating of the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment includes: updating the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment according to the vehicle position of the vehicle object included in the lane object at the current moment; The updating of the signal light state of the intersection object at the previous moment to obtain the signal light state at the current moment includes: updating the signal light state of the intersection object at the previous moment to obtain the signal light state at the current moment according to the signal light control information of the intersection object; The updating of the vehicle array of the interest area object at the previous moment to obtain the vehicle array of the current moment includes: updating the vehicle array of the interest area object at the previous moment to obtain the vehicle array of the current moment according to the vehicle intersection data of the vehicle object.
5. The platform according to claim 3 or 4, characterized in that: The determining of the driving behavior strategy of the vehicle object at the current moment based on the vehicle data of the vehicle object, the lane index data of the lane object at the current moment, and the traffic light state of the intersection object at the current moment includes at least one of the following: Determine a lane change strategy according to the vehicle routing data of the vehicle object and the driving status data of the surrounding vehicle objects, wherein the lane change strategy includes whether to change lanes, whether to accelerate or decelerate after changing lanes; Determine a speed limit strategy for the vehicle object according to the maximum speed limit of the lane object where the vehicle object is located, the maximum speed of the vehicle object itself, and the maximum speed limit of the next lane object when approaching the next lane object, wherein the speed limit strategy is used to indicate the maximum driving speed of the vehicle object at the current moment; Determine the relative position and relative speed of the vehicle object and the vehicle object in front according to the lane index data of the lane object at the current moment, and determine the following strategy of each vehicle object according to the relative position and relative speed of the vehicle object and the vehicle object in front, wherein the following strategy includes the required driving speed to avoid collision with the vehicle object in front; Determine the intersection strategy of each vehicle object according to the current signal light state of the intersection object that the vehicle object is to pass, wherein the intersection strategy includes: slowing down and stopping when the intersection object cannot be passed, and the driving speed determined according to the intersection traffic flow when the intersection object is passable; A stopping strategy is determined based on the vehicle routing data and the vehicle attribute data of the vehicle object, wherein the stopping strategy includes: reducing the driving speed when the vehicle object approaches the vehicle end point and entering the interest area of the vehicle end point.
6. The platform according to claim 1, characterized in that: The optimizing the designated map data based on the traffic simulation data at the plurality of moments to obtain target map data that meets a preset optimization goal includes: Determining statistical data required for optimizing the specified map data based on the traffic simulation data at the multiple moments, and judging whether the statistical data meets the preset optimization goal; When the statistical data do not meet the preset optimization target, optimizing the designated map data based on the statistical data using a preset optimization algorithm to obtain optimized designated map data; the preset optimization algorithm includes a black box optimization algorithm and / or a reinforcement learning algorithm; The traffic simulation module is controlled to redetermine the traffic simulation data at multiple moments and redetermine the statistical data based on the optimized designated map data until the redetermined statistical data meets the preset optimization target, thereby obtaining the target map data meeting the preset optimization target.
7. The platform according to claim 6, characterized in that The simulation platform also provides a data reading interface and a simulation control interface; The data reading interface is used to provide at least one of map data, vehicle data, traffic simulation data and statistical data within the simulation platform; The simulation control interface is used to modify at least one of the map data and the vehicle data in the simulation platform, so that the simulation platform optimizes the specified map data based on the modified at least one of the map data and the vehicle data; The simulation control interface is also used to control the update progress of the traffic simulation module, and the control of the update progress of the traffic simulation module includes at least one of the following: updating operations at multiple time points forward, pausing update operations, and resuming update operations to any historical time point.
8. A method for simulating high-speed urban traffic, characterized in that: include: Acquire map data and vehicle data of a city area of designated map data to be optimized, the map data comprising: lane data, intersection data, road data, and area of interest data; the designated map data is at least one of the lane data, intersection data, road data, and area of interest data, the vehicle data comprising: vehicle routing data, vehicle attribute data, and initial driving state data, the vehicle routing data representing a vehicle driving trajectory, the vehicle attribute data representing a vehicle driving attribute, and the initial driving state data comprising initial position information and speed information of the vehicle; At a preset time interval, based on the map data and the vehicle data, an update operation is performed on a plurality of simulation objects to obtain traffic simulation data at a plurality of times, wherein the plurality of simulation objects include vehicle objects, lane objects, intersection objects, and area of interest objects, and the update operation includes a pre-update operation and a formal update operation; The pre-update operation includes: updating the lane index data of the lane object at the previous moment to obtain the lane index data at the current moment, updating the signal light state of the intersection object at the previous moment to obtain the signal light state at the current moment, updating the vehicle array of the interest area object at the previous moment to obtain the vehicle array at the current moment, and copying the driving state data of each vehicle object at the previous moment to obtain the backup driving state data at the current moment; the lane index data is used to record the relative position relationship between the vehicle object contained in the lane object and the adjacent vehicle object, and the vehicle array is used to indicate the vehicle object staying in the interest area object; The formal update operation includes: determining the driving behavior strategy of the vehicle object at the current moment based on the vehicle data of the vehicle object, the lane index data of the lane object at the current moment, and the signal light state of the intersection object at the current moment; updating the driving state data of the vehicle object at the previous moment according to the driving behavior strategy of the vehicle object at the current moment and the backup driving state data of the vehicle object at the previous moment to obtain the driving state data of the vehicle object at the current moment; The traffic simulation data at the current moment includes lane index data of the lane object at the current moment, the signal light state of the intersection object at the current moment, the vehicle array of the interest area object at the current moment, and the driving state data of the vehicle object at the current moment; Based on the traffic simulation data at the multiple moments, the designated map data is optimized to obtain target map data that meets a preset optimization target.
9. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to implement the platform described in any one of claims 1 to 7 when executing the instructions stored in the memory.
10. A non-volatile computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the platform according to any one of claims 1 to 7 is implemented.
Citation Information
Patent Citations
Storage device for micro traffic simulation based on GPU (Graphics Processing Unit)
CN117332559A
Road change detection method, computing device, and storage medium
EP4303834A1
Information processing system, information processing server, information processing method and program
JP2013210913A
Road traffic control system
JP2020067743A
Interactive vehicle information mapping system
US10460602B1
