A cross-IDC simulation method, device, system and related equipment

Abstract

This application discloses a cross-IDC simulation method, apparatus, system, and related equipment. The method includes: receiving a simulation request from a user; obtaining summary information of the drive test dataset based on keyword information of the drive test dataset contained in the simulation request; determining the local data center where the drive test dataset is located based on the summary information, and generating at least one simulation task based on the summary information; determining a target processing center for executing the simulation task based on the local data center; scheduling the simulation task to the target processing center, monitoring the execution status of the simulation task in the target processing center, and receiving simulation results returned by the target processing center. This application can perform simulation calculations on drive test datasets while avoiding the cross-regional transmission of large files, reducing long-distance transmission costs while improving the timeliness of simulation calculations, which is conducive to improving the iteration speed of autonomous driving algorithms.

Description

Technical Field

[0001] The present application relates to the field of autonomous driving technology, and more specifically, to a cross-IDC simulation method, apparatus, system, and related equipment. Background Art

[0002] During the development of autonomous driving systems, extensive road testing is required to evaluate the algorithm's overall performance under various road conditions. This process generates a large amount of data, such as incidents involving driver takeovers and driver behavior. This data is aggregated into a bag file (a vehicle information data package recorded in a format customized by the ROS system or the autonomous driving company). This bag file contains the message flows of each software module in the autonomous driving system, and each file can be several GB in size.

[0003] The bag files generated by vehicle road tests usually need to be centrally managed in a system called an accident warehouse to generate regression test sets so that the modified algorithms can be evaluated using historical data in the simulation system, ultimately achieving rapid algorithm iteration.

[0004] To quickly evaluate the effectiveness of algorithms, large-scale simulations of tens of thousands of scenarios in the accident repository are typically required, with evaluation reports generated within a few hours. However, as the scale of autonomous driving services grows, fleets may be distributed across different cities and regions. The timely collection of dispersed road test bag files and their use in simulation calculations has become a pressing issue. Summary of the Invention

[0005] In view of this, the present application provides a cross-IDC simulation method, apparatus, system and related equipment to implement simulation calculation of cross-IDC drive test data sets.

[0006] To achieve the above objectives, the first aspect of the present application provides a cross-IDC simulation method, which is applied to a central simulation platform and includes:

[0007] receiving a simulation request from a user, wherein the simulation request includes keyword information of a drive test data set;

[0008] acquiring summary information of the drive test dataset based on keyword information of the drive test dataset;

[0009] determining a local data center where the drive test data set is located based on the summary information, and generating at least one simulation task based on the summary information;

[0010] Determining a target processing center for executing the simulation task based on the local data center;

[0011] The simulation task is dispatched to the target processing center, and the execution status of the simulation task in the target processing center is monitored, and the simulation result returned by the target processing center is received.

[0012] Preferably, the process of generating at least one simulation task based on the summary information includes:

[0013] Determine the autonomous driving mode, map version number, and scenario configuration of each data packet in the drive test dataset based on the summary information;

[0014] Determine the data packets that have the same autonomous driving mode, the same map version number, and compatible scene configurations as data packets of the same category;

[0015] Generate a simulation task based on data packets of the same category.

[0016] Preferably, the process of determining a target processing center for executing the simulation task based on the local data center includes:

[0017] The local processing center corresponding to the local data center is determined as the target processing center.

[0018] Preferably, the process of determining a target processing center for executing the simulation task based on the local data center includes:

[0019] Determining whether the available computing power of the local processing center corresponding to the local data center meets the computing power requirement of the simulation task;

[0020] If so, determining the local processing center corresponding to the local data center as the target processing center;

[0021] If not, the local processing center whose available computing power meets the computing power requirements of the simulation task and has the lowest cost will be determined as the target processing center.

[0022] Preferably, before dispatching the simulation task to the target processing center, the method further includes:

[0023] Determine the simulation task as a task to be executed, and save the task to be executed in a database;

[0024] The process of dispatching the simulation task to the target processing center includes:

[0025] Regularly query the pending tasks stored in the database;

[0026] When it is determined that there is a new task to be executed, the new task to be executed is dispatched to the target processing center.

[0027] A second aspect of the present application provides a cross-IDC simulation method, which is applied to a local processing center and includes:

[0028] Uploading computing power information and cost information of the local processing center to the central simulation platform;

[0029] Determining a target working node for executing the simulation task based on the received simulation task;

[0030] Obtaining a data packet in a drive test data set used in the simulation task;

[0031] Based on the data packet, the preset simulation software is run using the target working node, and the simulation results are returned to the central simulation platform in the form of a message queue.

[0032] A third aspect of the present application provides a cross-IDC simulation system, comprising: a central simulation platform and at least one local processing center;

[0033] The central simulation platform is provided with an accident warehouse for implementing the various steps of the cross-IDC simulation method applied to the central simulation platform as described above; wherein the accident warehouse is used to store summary information of the drive test data set;

[0034] The local processing center is provided with a local data center for implementing the various steps of the above-mentioned cross-IDC simulation method applied to the local processing center; wherein the local data center is used to store the drive test data set.

[0035] Preferably, the local processing center is further provided with an uploader, and the uploader is used to upload the summary information of the drive test set to the accident warehouse, and upload the drive test data set to the local data center.

[0036] A fourth aspect of the present application provides a cross-IDC simulation device applied to a central simulation platform, comprising:

[0037] A simulation request receiving unit, configured to receive a simulation request from a user, wherein the simulation request includes keyword information of a drive test data set;

[0038] a summary information acquiring unit, configured to acquire summary information of the drive test dataset based on keyword information of the drive test dataset;

[0039] a task parsing and generating unit, configured to determine a local data center where the drive test data set is located based on the summary information, and to generate at least one simulation task based on the summary information;

[0040] a task acceptance determination unit, configured to determine a target processing center for executing the simulation task based on the local data center;

[0041] The task scheduling and monitoring unit is used to schedule the simulation task to the target processing center, monitor the execution status of the simulation task in the target processing center, and receive the simulation result returned by the target processing center.

[0042] A fifth aspect of the present application provides a cross-IDC simulation device applied to a local processing center, including:

[0043] Information reporting unit, used to upload computing power information and cost information to the central simulation platform;

[0044] A node allocation unit, configured to determine a target working node for executing the simulation task based on the received simulation task;

[0045] A data acquisition unit, configured to acquire data packets from a drive test data set used in the simulation task;

[0046] A simulation computing unit is used to run the preset simulation software using the target working node based on the data packet, and return the simulation results to the central simulation platform in the form of a message queue

[0047] A sixth aspect of the present application provides a cross-IDC simulation device, including: a memory and a processor;

[0048] The memory is used to store programs;

[0049] The processor is used to execute the program to implement the various steps of the above-mentioned cross-IDC simulation method applied to the central simulation platform or the cross-IDC simulation method applied to the local processing center.

[0050] In a seventh aspect, the present application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the computer program implements the various steps of the above-mentioned cross-IDC simulation method applied to the central simulation platform or the cross-IDC simulation method applied to the local processing center.

[0051] As can be seen from the above technical solution, the present application first receives a simulation request from a user, the simulation request including keyword information of a drive test dataset. Then, based on the keyword information of the drive test dataset, summary information of the drive test dataset is obtained. Summary information occupies a small amount of byte space, making it convenient to collect summary information of drive test datasets from various locations. Next, the local data center where the drive test dataset is located is determined based on the summary information, and at least one simulation task is generated based on the summary information. Then, based on the local data center, a target processing center for executing the simulation task is determined, and the simulation task is dispatched to the target processing center. It is understood that each local data center corresponds to a local processing center. Using the local processing center where the drive test dataset is located to execute the simulation task avoids the cross-regional transmission of large files for the drive test dataset, saving bandwidth resources and transmission time associated with cross-regional transmission. During this process, the execution status of the simulation task at the target processing center is monitored, and simulation results returned by the target processing center are received. The present application can perform simulation calculations on drive test datasets while avoiding the cross-regional transmission of large files, reducing long-distance transmission costs while improving the timeliness of simulation calculations and facilitating the iteration speed of autonomous driving algorithms. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.

[0053] Figure 1 A schematic diagram of a cross-IDC simulation system disclosed in an embodiment of the present application;

[0054] Figure 2 A schematic diagram of a cross-IDC simulation method applied to a central simulation platform according to an embodiment of the present application;

[0055] Figure 3 A schematic diagram of a cross-IDC simulation method applied to a local processing center according to an embodiment of the present application;

[0056] Figure 4 Another schematic diagram of the cross-IDC simulation system disclosed in an embodiment of the present application;

[0057] Figure 5 Another schematic diagram of the cross-IDC simulation system disclosed in an embodiment of the present application;

[0058] Figure 6A schematic diagram of a cross-IDC simulation device applied to a central simulation platform disclosed in an embodiment of the present application;

[0059] Figure 7 A schematic diagram of a cross-IDC simulation device disclosed in an embodiment of the present application and applied to a local processing center;

[0060] Figure 8 This is a schematic diagram of a cross-IDC simulation device disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0062] The inventors of this application have discovered that in the existing technical architecture, the simulation platform cluster and the data warehouse need to be deployed in the same computer room. Since a huge amount of accident data needs to be read during the simulation process, this architecture has the following defects:

[0063] Because the massive amount of data generated by the fleet needs to be transmitted back to a data center (IDC, Internet Data Center), when the fleet is large and when each team is scattered over a long distance, on the one hand, there is a large delay, which prevents engineers from reviewing the data in time and recreating the scene where the problem occurred; on the other hand, it is necessary to rent a very large bandwidth for the data center, which costs a lot of money.

[0064] In addition, deploying the simulation system in only one data center makes it difficult to fully utilize the computing power in different regions.

[0065] Based on this, this application proposes a cross-regional data management and simulation technology solution to overcome the above technical defects. The cross-IDC simulation method provided in the embodiment of this application can be run in the cross-IDC simulation system provided in the embodiment of this application. Figure 1 The cross-IDC simulation system provided by the embodiment of the present application may include a central simulation platform 10 and at least one local processing center 20.

[0066] The central simulation platform 10 includes an incident repository 11, which is used to execute the various steps of the cross-IDC simulation method provided in the embodiments of this application, dispatch simulation tasks to the local processing center 20, monitor the execution status of simulation tasks in the local processing center 20, and receive execution results returned by the local processing center 20. Incident repository 11 is used to store summary information of drive test datasets.

[0067] The local processing center 20 is provided with a local data center 21, which is used to respond to the task scheduling of the central simulation platform 10, execute simulation tasks, and return simulation results to the central simulation platform 10. The local data center 21 is used to store drive test data sets.

[0068] The following describes the cross-IDC simulation method applied to the central simulation platform 10 provided in the embodiment of the present application. Figure 2 The cross-IDC simulation method provided in the embodiment of the present application may include the following steps:

[0069] Step S101: receiving a simulation request from a user.

[0070] Among them, the simulation request includes keyword information of the road test data set. It can be understood that the keyword information of the road test data set is necessary information for applying the cross-IDC simulation method provided by this embodiment, and may include information such as safety score, scenario configuration, driving mode, etc., which is used to filter out the road test data set that needs to be simulated. In addition, the simulation request may also include information such as the priority of the task, the configuration of the autonomous driving software, and the algorithm derivation mechanism (such as GPU or Inference Server). This information is the information that is usually carried when submitting a simulation task in this field and will not be repeated here.

[0071] Step S102: Obtain summary information of the drive test dataset based on keyword information of the drive test dataset.

[0072] For example, summary information of the drive test data set may be stored in the accident repository in a table format, and the keyword of the drive test data set may be set as an index of the summary information. Then, the corresponding summary information may be found through the keyword information of the drive test data set.

[0073] Among them, the summary information may include vehicle information, autonomous driving software version, vehicle safety officer, road test date, road test city, road test route branch and other metadata information with smaller data volume.

[0074] Step S103: determining the local data center where the drive test data set is located based on the summary information, and generating at least one simulation task.

[0075] It is understandable that the summary information of the drive test data set and the drive test data set itself are stored separately. On the one hand, the drive test data set contains video files and bag files generated during the vehicle road test, which occupy a relatively large storage space. These files can be stored in a decentralized manner in the local data center where the vehicle road test is performed; on the other hand, the summary information part that occupies a smaller storage space can be uploaded to the central simulation platform in a unified manner.

[0076] Step S104: determining a target processing center for executing the simulation task based on the local data center.

[0077] It can be understood that each local data center is equipped with a local processing center, and the local data center and its supporting local processing center are set up in the same local area. For example, they can be set up in the same computer room, which makes it convenient for the local processing center to read the data of the local data center nearby, avoiding cross-regional and high-cost network transmission.

[0078] In this step, the local processing center corresponding to the local data center can be set as the target processing center. Of course, when it is found that the local processing center is not competent for the simulation task, other nearby local processing centers with lower costs can also be set as the target processing center.

[0079] Step S105 , dispatching the simulation task to the target processing center, monitoring the execution status of the simulation task in the target processing center, and receiving the simulation result returned by the target processing center.

[0080] This application first receives a simulation request from a user, which includes keyword information for a drive test dataset. Then, based on the keyword information in the drive test dataset, it obtains summary information for the drive test dataset. Summary information occupies a small amount of byte space, making it convenient to collect summary information for drive test datasets from various locations. Next, based on the summary information, it determines the local data center where the drive test dataset is located, and generates at least one simulation task based on the summary information. Then, based on the local data center, it determines a target processing center for executing the simulation task, and dispatches the simulation task to the target processing center. It will be appreciated that each local data center corresponds to a local processing center. Using the local processing center where the drive test dataset is located to execute the simulation task avoids the cross-regional transmission of large files for the drive test dataset, saving bandwidth resources and transmission time associated with cross-regional transmission. During this process, the execution status of the simulation task at the target processing center is monitored, and simulation results returned by the target processing center are received. This application allows simulation calculations on drive test datasets to be performed without the need for large files to be transferred across regions, reducing long-distance transmission costs while improving the timeliness of simulation calculations and facilitating the iteration speed of autonomous driving algorithms.

[0081] In some embodiments of the present application, the process of generating at least one simulation task based on the summary information in step S103 may include:

[0082] S1, based on the summary information, determines the autonomous driving mode, map version number and scenario configuration of each data packet in the road test dataset.

[0083] Among them, autonomous driving modes can include fully autonomous driving modes that are completely controlled by driving software, as well as driving modes that require manual driver participation. As maps are updated and iterated, road test datasets may involve different versions of maps, and maps with different version numbers may not be compatible in a single simulation operation of the same simulation software. Similarly, a road test dataset may involve thousands of scenarios, and not all scenarios can be compatible in a single simulation operation of the same simulation software. By analyzing the scenario configuration and combining it with the preset scenario rules, it is possible to determine which scenarios can be compatible in a single simulation operation of the same simulation software.

[0084] S2: Determine the data packets in each data packet that have the same autonomous driving mode and map version number and compatible scene configuration as data packets of the same category.

[0085] S3,generates a simulation task based on the data packets of the same category.

[0086] Since data packets of the same category have the same autonomous driving mode and the same map version number, and have compatible scenario configurations, simulation tasks generated by the same category can run in parallel in the simulation software in a fused simulation scenario, eliminating the need to start the simulation software multiple times and saving the total simulation calculation time.

[0087] In some embodiments of the present application, the process of determining the target processing center for executing the simulation task based on the local data center in step S104 may include:

[0088] The local processing center corresponding to the local data center is determined as the target processing center.

[0089] It is understandable that using a local processing center corresponding to the local data center to execute simulation tasks can avoid cross-regional and high-cost network transmission and complete simulation tasks more efficiently. The simulation results generated by the simulation tasks usually occupy a small amount of byte space and can be returned to the central simulation platform via the network.

[0090] When the local processing center corresponding to the local data center is unable to perform the simulation task, the execution failure status and specific information of the execution failure can be learned during the monitoring of the execution status in step S105.

[0091] To solve this problem, in some embodiments of the present application, the process of determining the target processing center for executing the simulation task based on the local data center in step S104 may include:

[0092] S1: Determine whether the available computing power of the local processing center corresponding to the local data center meets the computing power requirements of the simulation task. If so, execute S2; if not, execute S3.

[0093] The available computing power of a local processing center is the total computing power of the local processing center minus the occupied computing power.

[0094] S2, determining the local processing center corresponding to the local data center as the target processing center.

[0095] S3,determine the local processing center whose available computing power meets the computing power requirements of the simulation task and has the lowest cost as the target processing center.

[0096] By fully considering the available computing power and costs of each local processing center, we can fully utilize the cost advantages of multiple local processing centers, reduce the resource constraints of a single local processing center, and expand computing resources. In addition, distributed storage and computing reduce the risks of a single data center or a single processing center.

[0097] In some embodiments of the present application, before the above step S105 dispatches the simulation task to the target processing center, the following steps may also be included:

[0098] The simulation task is determined as a task to be executed, and the task to be executed is saved in a database.

[0099] Based on this, the process of dispatching the simulation task to the target processing center in step S105 may include:

[0100] S1, regularly query the pending tasks stored in the database.

[0101] S2: When it is determined that there is a new task to be executed, the new task to be executed is dispatched to the target processing center.

[0102] The following describes the cross-IDC simulation method provided by the embodiment of the present application and applied to the local processing center 20. Figure 3 The cross-IDC simulation method provided in the embodiment of the present application may include the following steps:

[0103] Step S201: Upload computing power information and cost information to the central simulation platform.

[0104] The computing power information may include the processing capacity and health status of each working node under the local processing center 20, and the cost information may include the computing cost of each working node. A working node is a computing node responsible for executing a simulation task.

[0105] Step S202: Based on the received simulation task, determine a target working node for executing the simulation task.

[0106] Generally, one simulation task can be assigned to one work node. When multiple simulation tasks are received, one work node is allocated to each simulation task one by one.

[0107] Step S203: obtaining data packets in the drive test data set used in the simulation task.

[0108] It is understandable that in addition to the data packets in the drive test dataset, configuration information such as the simulation scenario of the simulation task also needs to be obtained. This configuration information can be included in the description of the simulation task, while the data packets in the drive test dataset are obtained in the local data center that stores the drive test dataset.

[0109] Step S204 : Based on the data packet, the preset simulation software is run using the target working node, and the simulation result is returned to the central simulation platform 10 in the form of a message queue.

[0110] For example, a connection to a preset simulation software may be made through an HTTP interface, the simulation software may be used to perform simulation calculations, and the simulation results may be returned to the central simulation platform 10 .

[0111] Specifically, various indicators may be aggregated to form aggregated simulation results, and the aggregated simulation results may be uploaded to a database of a central simulation platform.

[0112] The above describes the cross-IDC simulation method from the perspectives of the central simulation platform 10 and the local processing center 20. Based on the cross-IDC simulation method described in the above embodiments, the cross-IDC simulation system provided by this application can be further described.

[0113] See also Figure 4 In some embodiments of the present application, in addition to the accident warehouse 11 , the central simulation platform 10 may also be provided with a task generator 12 , a scheduler 13 and a database 14 .

[0114] The task generator 12 is configured to generate at least one simulation task based on the summary information in step S103. The scheduler 13 is configured to determine the target processing center for executing the simulation task based on the local data center storing the target drive test dataset and schedule the simulation task to the target processing center. The scheduler 13 is also configured to receive computing power and cost information reported by each local processing center. The database 14 is configured to store pending simulation tasks and simulation results after metrics aggregation.

[0115] like Figure 4 As shown, in addition to the local data center 21 , the local processing center 20 may also be provided with a scheduler agent 22 and a plurality of working nodes 23 .

[0116] The scheduler agent 22 is used to report the processing capacity and health status of each work node 23 to the scheduler 13, accept task scheduling from the scheduler 13, and assign simulation tasks to be executed to corresponding work nodes 23. The work nodes 23 are used to execute simulation tasks and return simulation results to the database 14 of the central simulation platform 10.

[0117] The core idea of ​​the cross-IDC simulation method or cross-IDC simulation system provided in the embodiments of the present application is to perform simulation calculations on-site on the road test data set generated during vehicle road testing, so as to reduce data transmission costs and improve the failure rate of simulation calculations.

[0118] Based on this, in some embodiments of this application, please refer to Figure 5 The local processing center 20 is further provided with an uploader 24 , which is used to upload the summary information of the drive test set to the accident warehouse 11 , and upload the drive test data set to the local data center 21 .

[0119] It can be understood that the summary information of the drive test set includes the path information of the drive test set, and the path information indicates the local data center where a certain drive test data set is located.

[0120] The following describes a cross-IDC simulation device applied to the central simulation platform 10 provided in an embodiment of the present application. The cross-IDC simulation device described below and the cross-IDC simulation method applied to the central simulation platform 10 described above can refer to each other.

[0121] See Figure 6 The cross-IDC simulation device provided in the embodiment of the present application and applied to the central simulation platform 10 may include:

[0122] A simulation request receiving unit 41 is configured to receive a simulation request from a user, wherein the simulation request includes keyword information of a drive test data set;

[0123] A summary information obtaining unit 42 is configured to obtain summary information of the drive test dataset based on keyword information of the drive test dataset;

[0124] a task parsing and generating unit 43, configured to determine a local data center where the drive test dataset is located based on the summary information, and to generate at least one simulation task based on the summary information;

[0125] A task acceptance determination unit 44 is configured to determine a target processing center for executing the simulation task based on the local data center;

[0126] The task scheduling monitoring unit 45 is used to schedule the simulation task to the target processing center, monitor the execution status of the simulation task in the target processing center, and receive the simulation result returned by the target processing center.

[0127] In some embodiments of the present application, the process of the task parsing and generating unit 43 generating at least one simulation task based on the summary information may include:

[0128] Determine the autonomous driving mode, map version number, and scenario configuration of each data packet in the drive test dataset based on the summary information;

[0129] Determine the data packets that have the same autonomous driving mode, the same map version number, and compatible scene configurations as data packets of the same category;

[0130] Generate a simulation task based on data packets of the same category.

[0131] In some embodiments of the present application, the process of the task acceptance determination unit 44 determining the target processing center for executing the simulation task based on the local data center may include:

[0132] The local processing center corresponding to the local data center is determined as the target processing center.

[0133] In some embodiments of the present application, the process of the task acceptance determination unit 44 determining the target processing center for executing the simulation task based on the local data center may include:

[0134] Determining whether the available computing power of the local processing center corresponding to the local data center meets the computing power requirement of the simulation task;

[0135] If so, determining the local processing center corresponding to the local data center as the target processing center;

[0136] If not, the local processing center whose available computing power meets the computing power requirements of the simulation task and has the lowest cost will be determined as the target processing center.

[0137] In some embodiments of the present application, before the task scheduling monitoring unit 45 schedules the simulation task to the target processing center, the following steps may be further included:

[0138] Determine the simulation task as a task to be executed, and save the task to be executed in a database;

[0139] The process of the task scheduling monitoring unit 45 scheduling the simulation task to the target processing center may include:

[0140] Regularly query the pending tasks stored in the database;

[0141] When it is determined that there is a new task to be executed, the new task to be executed is dispatched to the target processing center.

[0142] The cross-IDC simulation device provided in the embodiment of the present application and applied to the central simulation platform 10 can be applied to cross-IDC simulation devices, such as computers. Figure 7 The hardware structure diagram of the cross-IDC simulation device is shown in FIG. Figure 7 The hardware structure of the cross-IDC simulation device may include: at least one processor 31 , at least one communication interface 32 , at least one memory 33 and at least one communication bus 34 .

[0143] In the embodiment of the present application, the number of the processor 31, the communication interface 32, the memory 33, and the communication bus 34 is at least one, and the processor 31, the communication interface 32, and the memory 33 communicate with each other through the communication bus 34;

[0144] The processor 31 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application;

[0145] The memory 32 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory;

[0146] The memory 33 stores a program, and the processor 31 can call the program stored in the memory 33, wherein the program is used to:

[0147] receiving a simulation request from a user, wherein the simulation request includes keyword information of a drive test data set;

[0148] acquiring summary information of the drive test dataset based on keyword information of the drive test dataset;

[0149] determining a local data center where the drive test data set is located based on the summary information, and generating at least one simulation task based on the summary information;

[0150] Determining a target processing center for executing the simulation task based on the local data center;

[0151] The simulation task is dispatched to the target processing center, and the execution status of the simulation task in the target processing center is monitored, and the simulation result returned by the target processing center is received.

[0152] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0153] An embodiment of the present application further provides a storage medium, which may store a program suitable for execution by a processor, wherein the program is used to:

[0154] receiving a simulation request from a user, wherein the simulation request includes keyword information of a drive test data set;

[0155] acquiring summary information of the drive test dataset based on keyword information of the drive test dataset;

[0156] determining a local data center where the drive test data set is located based on the summary information, and generating at least one simulation task based on the summary information;

[0157] Determining a target processing center for executing the simulation task based on the local data center;

[0158] The simulation task is dispatched to the target processing center, and the execution status of the simulation task in the target processing center is monitored, and the simulation result returned by the target processing center is received.

[0159] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0160] The following describes a cross-IDC simulation device applied to the local processing center 20 provided in an embodiment of the present application. The cross-IDC simulation device described below and the cross-IDC simulation method applied to the local processing center 20 described above can refer to each other.

[0161] See Figure 8 The cross-IDC simulation device provided in the embodiment of the present application and applied to the local processing center 20 may include:

[0162] An information reporting unit 51 is used to upload the computing power information and cost information of the local processing center to the central simulation platform;

[0163] The node allocation unit 52 is configured to determine a target working node for executing the simulation task based on the received simulation task;

[0164] A data acquisition unit 53 is configured to acquire data packets from a drive test data set used in the simulation task;

[0165] The simulation calculation unit 54 is used to run the preset simulation software using the target working node based on the data packet, and return the simulation results to the central simulation platform in the form of a message queue.

[0166] The cross-IDC simulation device provided in the embodiment of the present application and applied to the local processing center 20 can be applied to cross-IDC simulation devices, such as computers. Figure 7 The hardware structure diagram of the cross-IDC simulation device is shown in FIG. Figure 7 The hardware structure of the cross-IDC simulation device may include: at least one processor 31 , at least one communication interface 32 , at least one memory 33 and at least one communication bus 34 .

[0167] In the embodiment of the present application, the number of the processor 31, the communication interface 32, the memory 33, and the communication bus 34 is at least one, and the processor 31, the communication interface 32, and the memory 33 communicate with each other through the communication bus 34;

[0168] The processor 31 may be a central processing unit (CPU), or an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application;

[0169] The memory 32 may include a high-speed RAM memory, and may also include a non-volatile memory, such as at least one disk memory;

[0170] The memory 33 stores a program, and the processor 31 can call the program stored in the memory 33, wherein the program is used to:

[0171] Uploading computing power information and cost information of the local processing center to the central simulation platform;

[0172] Determining a target working node for executing the simulation task based on the received simulation task;

[0173] Obtaining a data packet in a drive test data set used in the simulation task;

[0174] Based on the data packet, the preset simulation software is run using the target working node, and the simulation results are returned to the central simulation platform in the form of a message queue.

[0175] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0176] An embodiment of the present application further provides a storage medium, which may store a program suitable for execution by a processor, wherein the program is used to:

[0177] Uploading computing power information and cost information of the local processing center to the central simulation platform;

[0178] Determining a target working node for executing the simulation task based on the received simulation task;

[0179] Obtaining a data packet in a drive test data set used in the simulation task;

[0180] Based on the data packet, the preset simulation software is run using the target working node, and the simulation results are returned to the central simulation platform in the form of a message queue.

[0181] Optionally, the detailed functions and extended functions of the program may refer to the above description.

[0182] In summary:

[0183] This application first receives a simulation request from a user, which includes keyword information for a drive test dataset. Then, based on the keyword information in the drive test dataset, it obtains summary information for the drive test dataset. Summary information occupies a small amount of byte space, making it convenient to collect summary information for drive test datasets from various locations. Next, based on the summary information, it determines the local data center where the drive test dataset is located, and generates at least one simulation task based on the summary information. Then, based on the local data center, it determines a target processing center for executing the simulation task, and dispatches the simulation task to the target processing center. It will be appreciated that each local data center corresponds to a local processing center. Using the local processing center where the drive test dataset is located to execute the simulation task avoids the cross-regional transmission of large files for the drive test dataset, saving bandwidth resources and transmission time associated with cross-regional transmission. During this process, the execution status of the simulation task at the target processing center is monitored, and simulation results returned by the target processing center are received. This application allows simulation calculations on drive test datasets to be performed without the need for large files to be transferred across regions, reducing long-distance transmission costs while improving the timeliness of simulation calculations and facilitating the iteration speed of autonomous driving algorithms.

[0184] Finally, it should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or device comprising the element.

[0185] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referenced to each other.

[0186] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present application. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application is not limited to the embodiments shown herein, but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A cross-IDC simulation method, characterized in that: include: receiving a simulation request from a user, wherein the simulation request includes keyword information of a drive test data set; acquiring summary information of the drive test dataset based on keyword information of the drive test dataset; determining a local data center where the drive test data set is located based on the summary information, and generating at least one simulation task based on the summary information; Determining a target processing center for executing the simulation task based on the local data center; Dispatching the simulation task to the target processing center, monitoring the execution status of the simulation task in the target processing center, and receiving the simulation result returned by the target processing center; The keyword information includes safety score, scenario configuration and driving mode; The keyword information of the drive test data set is set as the index of the summary information, and the corresponding summary information can be found through the keyword information of the drive test data set; The summary information includes vehicle information, autonomous driving software version, vehicle safety officer, road test date, the city where the road test was conducted, and the route branch where the road test was conducted.

2. The method according to claim 1, characterized in that The process of generating at least one simulation task based on the summary information includes: Determine the autonomous driving mode, map version number, and scenario configuration of each data packet in the drive test dataset based on the summary information; Determine the data packets that have the same autonomous driving mode, the same map version number, and compatible scene configurations as data packets of the same category; Generate a simulation task based on data packets of the same category.

3. The method according to claim 1, characterized in that The process of determining a target processing center for executing the simulation task based on the local data center includes: The local processing center corresponding to the local data center is determined as the target processing center.

4. The method according to claim 1, wherein The process of determining a target processing center for executing the simulation task based on the local data center includes: Determining whether the available computing power of the local processing center corresponding to the local data center meets the computing power requirement of the simulation task; If so, determining the local processing center corresponding to the local data center as the target processing center; If not, the local processing center whose available computing power meets the computing power requirements of the simulation task and has the lowest computing power cost will be determined as the target processing center.

5. The method according to claim 1, wherein Before dispatching the simulation task to the target processing center, the method further includes: Determine the simulation task as a task to be executed, and save the task to be executed in a database; The process of dispatching the simulation task to the target processing center includes: Regularly query the pending tasks stored in the database; When it is determined that there is a new task to be executed, the new task to be executed is dispatched to the target processing center.

6. The method according to claim 1, wherein Each road test data set is uploaded to the local processing center where the tested vehicle is located through the uploader, and the summary information of each road test data set is uploaded to the central simulation platform through the uploader.

7. A cross-IDC simulation system, characterized in that: include: a central simulation platform and at least one local processing center; The central simulation platform is provided with an accident warehouse, which is used to implement the various steps of the cross-IDC simulation method according to any one of claims 1 to 6; wherein the accident warehouse is used to store summary information of the drive test data set; The local processing center is provided with a local data center for implementing the various steps of the following simulation method; wherein the local data center is used to store the drive test data set; The various steps of the simulation method include: Uploading computing power information and cost information of the local processing center to the central simulation platform; Determining a target working node for executing the simulation task based on the received simulation task; Obtaining a data packet in a drive test data set used in the simulation task; Based on the data packet, the preset simulation software is run using the target working node, and the simulation results are returned to the central simulation platform in the form of a message queue.

8. The system according to claim 7, characterized in that include: The local processing center is further provided with an uploader, which is used to upload the summary information of the drive test set to the accident warehouse, and upload the drive test data set to the local data center.

9. A cross-IDC simulation device, characterized in that: include: A simulation request receiving unit, configured to receive a simulation request from a user, wherein the simulation request includes keyword information of a drive test data set; a summary information acquiring unit, configured to acquire summary information of the drive test dataset based on keyword information of the drive test dataset; a task parsing and generating unit, configured to determine a local data center where the drive test data set is located based on the summary information, and to generate at least one simulation task based on the summary information; a task acceptance determination unit, configured to determine a target processing center for executing the simulation task based on the local data center; A task scheduling and monitoring unit, configured to schedule the simulation task to the target processing center, monitor the execution status of the simulation task in the target processing center, and receive the simulation result returned by the target processing center; The keyword information includes safety score, scenario configuration and driving mode; The keyword information of the drive test data set is set as the index of the summary information, and the corresponding summary information can be found through the keyword information of the drive test data set; The summary information includes vehicle information, autonomous driving software version, vehicle safety officer, road test date, the city where the road test was conducted, and the route branch where the road test was conducted.

10. A cross-IDC simulation device, characterized in that: include: memory and processor; The memory is used to store programs; The processor is configured to execute the program to implement the steps of the cross-IDC simulation method according to any one of claims 1 to 6.

11. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, each step of the cross-IDC simulation method according to any one of claims 1 to 6 is implemented.

Images