Automatic driving simulation dynamic running resource scheduling method and device

Abstract

The application relates to an automatic driving simulation dynamic running resource scheduling method and device, wherein the method comprises the following steps: acquiring a scheduling task input by a user; evaluating dynamic resources of a basic scene in each scheduling task of at least one scheduling task, determining reserved resources of the basic scene; and receiving a scheduling instruction of scene switching, and executing a task scheduling action corresponding to the scheduling task based on an optimal resource allocation strategy, wherein the optimal resource allocation strategy is obtained based on running resource evaluation of each simulation scene of at least one simulation scene. According to the application, the resources of the basic scene of the user can be reserved based on the scheduling task of the user, and the running resource scheduling and allocation of the simulation scene can be performed according to the scene switching, so that the simulation dynamic running resource scheduling of the automatic driving is realized, and the fast iteration of the resource scheduling simulation scene switching can be performed under the condition of limited resources, and the simulation execution efficiency is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving cloud simulation, in particular to an automatic driving simulation dynamic running resource scheduling method and device. BACKGROUND

[0002] With the rapid development of modern technologies such as artificial intelligence, automation and control technology, automatic driving car technology, especially full-scene automatic driving, has become a hot spot for innovation in the academic and industrial fields. An automatic driving system contains a large number of machine learning algorithms, which need a large amount of scene simulation verification when training and verifying to ensure the safety of the automatic driving system driving domain.

[0003] In actual application, a set of automatic driving system needs to be tested and verified by at least 11 billion miles of driving data to meet the conditions for mass production. Among them, 11 billion miles means that all the roads in China need to be driven 4000 times. With a fleet of 100 test cars, 7*24 hours of uninterrupted testing, it will take 500 years to complete the test. Therefore, an automatic driving simulation system has become an indispensable component for the online of the automatic driving system and the landing of the car.

[0004] The current driving simulation system is developing rapidly and mainly includes an automatic driving scene manager, a simulation engine, an algorithm evaluation system and several other parts, and realizes scene digitization. In related technologies, a scene can be injected into a simulation engine, and a scene execution can be completed through information interaction between an automatic driving algorithm and the simulation engine. The scene execution result is recorded as an evaluation index item for analysis by the algorithm evaluation system module.

[0005] However, from the perspective of physical execution of the automatic driving simulation system, the execution of a simulation scene and an algorithm is a process of executing a large number of artificial intelligence algorithms on corresponding computer resources. In related technologies, there are many real simulation scenes, which are difficult to meet the requirements of rapid iteration. Under the premise of limited resources, it is difficult to efficiently execute simulation capabilities, and improvement is needed. SUMMARY

[0006] The present application provides an automatic driving simulation dynamic running resource scheduling method and device to solve the technical problem of low simulation execution efficiency due to the large number of simulation scenes and limited resources in related technologies, which makes it difficult to meet the requirements of rapid iteration when the scene is converted.

[0007] The first aspect embodiment of the application provides an automatic driving simulation dynamic running resource scheduling method, comprising the following steps: obtaining a scheduling task input by a user; evaluating dynamic resources of a basic scene in each scheduling task of at least one scheduling task, determining reserved resources of the basic scene; and receiving a scheduling instruction of scene switching, and performing a task scheduling action corresponding to the scheduling task based on an optimal resource allocation strategy, wherein the optimal resource allocation strategy is obtained based on running resources of each simulation scene of at least one simulation scene.

[0008] According to the above technical means, the embodiment of the application can reserve resources of a basic scene based on a scheduling task of a user, and perform running resource scheduling and allocation of a simulation scene according to scene switching, so as to realize simulation dynamic running resource scheduling of automatic driving, and enable fast iteration when a resource scheduling simulation scene switches under the condition of limited resources, thereby improving simulation execution efficiency.

[0009] Optionally, in an embodiment of the application, before performing the task scheduling action corresponding to the scheduling task based on the optimal resource allocation strategy, the method further comprises: obtaining at least one simulation scene input by the user; and evaluating running resources of each simulation scene of the at least one simulation scene, to generate an optimal resource allocation strategy that is adjusted based on a running index in real time.

[0010] According to the above technical means, the embodiment of the application can dynamically adjust simulation scene resources during the process to ensure that a priority algorithm is preferentially executed.

[0011] Optionally, in an embodiment of the application, each simulation scene comprises at least one of a generic scene, a selection algorithm, an execution priority, and limited resources and the basic scene.

[0012] According to the above technical means, the embodiment of the application can implement resource allocation strategies for different scenes based on different simulation scenes.

[0013] Optionally, in an embodiment of the application, after receiving the scheduling instruction of scene switching, the method further comprises: recording a task state currently running in a current simulation scene and storing a context to a corresponding space during automatic driving simulation dynamic running; and restoring the current simulation scene according to the context and the task state after the task scheduling action corresponding to the scheduling task is performed.

[0014] According to the above technical means, the embodiment of the application can record a task state currently running in a current simulation scene, and restore the current simulation scene after a scheduling action is performed, without repeated allocation, to realize automatic switching of simulation scenes, reduce manual participation under the requirements of efficiency and flexibility, and improve the automation level of a vehicle.

[0015] Optionally, in an embodiment of the present application, the scheduling task includes at least one of an autonomous driving scene, an autonomous driving algorithm to be simulated, a priority, a resource scheduling mode, and a running time period.

[0016] According to the above technical means, embodiments of the present application can include multiple scheduling tasks to meet the dynamic running resource scheduling of algorithm business requirements.

[0017] The second aspect embodiment of the present application provides an autonomous driving simulation dynamic running resource scheduling device, including: a first acquisition module configured to acquire a scheduling task input by a user; a first evaluation module configured to evaluate dynamic resources of a basic scene in each scheduling task of at least one scheduling task, and determine reserved resources of the basic scene; and a scheduling module configured to receive a scheduling instruction of scene switching, and perform a task scheduling action corresponding to the scheduling task based on an optimal resource allocation strategy, wherein the optimal resource allocation strategy is obtained based on running resource evaluation of each simulation scene of at least one simulation scene.

[0018] Optionally, in an embodiment of the present application, the device further includes: a second acquisition module configured to acquire at least one simulation scene input by the user; and a second evaluation module configured to evaluate running resources of each simulation scene of the at least one simulation scene, and generate an optimal resource allocation strategy adjusted based on a running index in real time.

[0019] Optionally, in an embodiment of the present application, each simulation scene includes at least one of a generic scene, a selected algorithm, an execution priority, and limited resources, and the basic scene.

[0020] Optionally, in an embodiment of the present application, the device further includes: a recording module configured to record a task state currently running in a current simulation scene and store a context to a corresponding space during autonomous driving simulation in real time; and a recovery module configured to recover the current simulation scene according to the context and the task state after the task scheduling action corresponding to the scheduling task is performed.

[0021] Optionally, in an embodiment of the present application, the scheduling task includes at least one of an autonomous driving scene, an autonomous driving algorithm to be simulated, a priority, a resource scheduling mode, and a running time period.

[0022] The third aspect embodiment of the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the autonomous driving simulation dynamic running resource scheduling method as described in the above embodiments.

[0023] The fourth aspect of the present application provides a computer readable storage medium, the computer readable storage medium stores a computer program, and the program is executed by a processor to realize the automatic driving simulation dynamic running resource scheduling method.

[0024] The beneficial effects of the embodiments of the present application are as follows:

[0025] (1) The embodiments of the present application can reserve the resources of the basic scene based on the scheduling tasks of the user, and perform running resource scheduling and distribution of the simulation scene according to scene switching, so as to realize simulation dynamic running resource scheduling of automatic driving, and make the resource scheduling simulation scene switching can be fast iteration under the condition of limited resources, and improve the simulation execution efficiency.

[0026] (2) The embodiments of the present application can evaluate the resources of the simulation scene, and dynamically adjust in the process to ensure that the priority algorithm is preferentially executed.

[0027] (3) The embodiments of the present application can record the current running task state of the current simulation scene, and restore after the scheduling action is completed, without repeated distribution, realize automatic switching of the simulation scene, reduce manual participation under the requirements of efficiency and flexibility, and improve the automation level of the vehicle.

[0028] Additional aspects and advantages of the present application will be in part apparent and in part pointed out hereinafter. BRIEF DESCRIPTION OF DRAWINGS

[0029] The above and / or additional aspects and advantages of the present application will become apparent and be readily understood from the following description, taken in conjunction with the accompanying drawings, in which:

[0030] Figure 1 A flowchart of an automatic driving simulation dynamic running resource scheduling method according to an embodiment of the present application is provided.

[0031] Figure 2 A principle schematic diagram of an automatic driving simulation dynamic running resource scheduling method according to an embodiment of the present application is provided.

[0032] Figure 3 An application schematic diagram of an automatic driving simulation dynamic running resource scheduling method according to an embodiment of the present application is provided.

[0033] Figure 4 A structural schematic diagram of an automatic driving simulation dynamic running resource scheduling device according to an embodiment of the present application is provided.

[0034] Figure 5 A structural schematic diagram of an electronic device according to an embodiment of the present application is provided.

[0035] Wherein, 10 - autonomous driving simulation dynamic running resource scheduling device; 100 - first acquisition module, 200 - first evaluation module, 300 - scheduling module. DETAILED DESCRIPTION

[0036] The embodiments of the present application are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar notations represent the same or similar elements having the same or similar functions throughout. The embodiments described below by reference to the drawings are exemplary and are intended to explain the present application, and cannot be understood as a limitation of the present application.

[0037] The automatic driving simulation dynamic running resource scheduling method and device of the embodiments of the present application are described below with reference to the accompanying drawings. In view of the technical problems in the related art that the simulation scene is numerous and the resources are limited, and it is difficult to meet the rapid iteration requirement during scene conversion, and the simulation execution efficiency is low, the present application provides an automatic driving simulation dynamic running resource scheduling method. In the method, the dynamic resources of the basic scene in each scheduling task of the scheduling task can be evaluated based on the scheduling task input by the user, and the scheduling instruction of scene switching is received, and the task scheduling action corresponding to the scheduling task is executed based on the best resource allocation strategy, thereby realizing the simulation dynamic running resource scheduling of autonomous driving, so that the resource scheduling simulation scene switching can be fast iteration, and the simulation execution efficiency is improved. Thus, the technical problems in the related art that the simulation scene is numerous and the resources are limited, and it is difficult to meet the rapid iteration requirement during scene conversion, and the simulation execution efficiency is low are solved.

[0038] Specifically, Figure 1 A flowchart of an automatic driving simulation dynamic running resource scheduling method provided by the embodiments of the present application is shown in the figure.

[0039] As Figure 1 shown, the automatic driving simulation dynamic running resource scheduling method includes the following steps:

[0040] In step S101, the scheduling task input by the user is acquired.

[0041] In the actual execution process, the embodiments of the present application can acquire the scheduling task input by the user to realize the add, delete, modify and query of the task, and view the state of the scheduling task and the execution report of the scheduling task for manual scheduling of the task.

[0042] Optionally, in an embodiment of the present application, the scheduling task includes at least one of the autonomous driving scene, the autonomous driving algorithm to be simulated, the priority, the resource scheduling mode and the running time period.

[0043] As a possible implementation manner, the content of the scheduling task in the embodiment of the application can include an autonomous driving scene, an autonomous driving algorithm to be simulated, a priority, a resource scheduling mode (static, dynamic), a running time period, etc., wherein the autonomous driving scene can be divided into a basic scene and a generic scene.

[0044] In step S102, the dynamic resources of the basic scene in each scheduling task of the at least one scheduling task are evaluated to determine the reserved resources of the basic scene.

[0045] In some embodiments, the evaluation of the dynamic resources of the basic scene can provide a basis for the scheduling algorithm to reserve the resources of the basic scene. Specifically, the embodiment of the application can take the resource consumption of the basic scene and the simulation algorithm as a basis, take the unit CPU / GPU, unit memory, and disk as the scheduling calculation unit, provide a scene execution algorithm benchmark for the scheduling algorithm execution, evaluate the dynamic resources of the basic scene in each scheduling task of the at least one scheduling task, and determine the reserved resources of the basic scene.

[0046] For example, the embodiment of the application can perform simulation running according to the minimum resource configuration of the basic scene and the algorithm, and dynamically reduce the resources according to the running index. This cycle obtains the reserved resources of the basic scene and the algorithm, i.e., the optimal resource standard.

[0047] In step S103, the scheduling instruction of the scene switching is received, and the task scheduling action corresponding to the scheduling task is executed based on the optimal resource allocation strategy, wherein the optimal resource allocation strategy is obtained based on the running resources of each simulation scene of the at least one simulation scene.

[0048] In actual execution, after receiving the scheduling instruction of the scene switching, the embodiment of the application can convert each basic simulation scene of the basic scene based on the reserved resources of the basic scene obtained in the above steps, and then obtain the optimal resource allocation strategy based on the running resources of each simulation scene of the at least one simulation scene, and then execute the task scheduling action corresponding to the scheduling task based on the optimal resource allocation strategy, to realize the automatic allocation and scheduling of resources.

[0049] The scheduling instruction of the scene switching can be obtained from the scheduling task input by the user, or can be obtained from the automatic simulation scene scheduling of the autonomous driving.

[0050] Optionally, in an embodiment of the application, before executing the task scheduling action corresponding to the scheduling task based on the optimal resource allocation strategy, the method further includes: obtaining at least one simulation scene input by the user; and evaluating the running resources of each simulation scene of the at least one simulation scene to generate the optimal resource allocation strategy which is adjusted based on the running index in real time.

[0051] As a possible implementation manner, the embodiment of the application can evaluate the running resource of each simulation scene based on at least one simulation scene input by a user, and generate an optimal resource allocation strategy adjusted based on a running index during running based on the reserved resource of the basic scene, so as to realize optimal resource scheduling and improve the efficiency of resource scheduling under the premise of limited resources.

[0052] Optionally, in an embodiment of the application, each simulation scene includes at least one of a generic scene, a selection algorithm, an execution priority, and a limited resource and the basic scene.

[0053] Specifically, each simulation scene can be extended based on the basic scene, the generic scene can be extended based on the basic scene for generality, the selection algorithm can be a manually selected scene extension based on the basic scene, the execution priority can be a selection based on the urgency of the simulation result, for example, the priority of no time requirement can be general (such as input during the day and execution at night), the priority of completing as soon as possible can be intermediate, and the priority of requiring strong occupation and real-time updating can be the highest (such a strong occupation task needs to be audited by a senior manager), and the limited resource can limit the maximum resource of the simulation task (including GPU / CPU, memory, disk, etc.).

[0054] In addition, the simulation scene of the embodiment of the application can also include global maximum resource, that is, all resources are used for scheduling of the task.

[0055] In actual execution, the embodiment of the application can determine the execution priority based on each simulation scene, and then obtain the optimal resource allocation strategy.

[0056] For example, the basic scheduling action of the embodiment of the application is to execute the project with the highest current priority, when the priorities are the same, the FIFO strategy can be used for scheduling and execution, when a task with higher priority appears while a task with a certain priority is running, the simulation scene switching process is triggered, the current simulation task is switched to the latest task, and when the new task is executed, the resource is allocated according to the resource mode set by the task, if the global maximum is used, all resources are allocated to the task, the limited resource is divided into a certain amount of data resources in the system, if the remaining resources cannot meet the minimum resource requirement of the basic scene, the task will be suspended due to insufficient resources, and other simulation tasks are scheduled, if there is no execution task and all tasks cannot be executed, a system warning will be generated to inform the relevant resource user.

[0057] Optionally, in an embodiment of the present application, after receiving the scheduling instruction of the scene switching, the method further comprises: recording the current running task state of the current simulation scene and storing the context to the corresponding space during the dynamic running of the autonomous driving simulation; and resuming the current simulation scene according to the context and the task state after the execution of the task scheduling action corresponding to the scheduling task is completed.

[0058] As a possible implementation, the embodiment of the present application can record the current running task state of the current simulation scene when interrupting the current simulation scene, and store the context to the corresponding space, so as to resume the current simulation scene according to the context and the task state after the execution of the task scheduling action corresponding to the scheduling task is completed, wherein the user can realize the corresponding simulation scene recovery through automatic recovery or manual recovery.

[0059] For example, when the task switching occurs, the embodiment of the present application can save the running state data into the system, and use the running resource to execute other simulation tasks, and when the scheduling algorithm recovers the simulation scene, the task execution can be quickly recovered according to the context.

[0060] In combination with Figure 2 and Figure 3 , the working principle of the autonomous driving simulation dynamic running resource scheduling method of the embodiment of the present application is described in detail. As shown in Figure 2 , the embodiment of the present application adds a simulation task scheduling part to the simulation system in the related art.

[0061] The simulation task scheduling part can include four parts: scheduling task management, scene switching management, scheduling algorithm and running resource evaluation.

[0062] The scheduling task management part can be used to obtain the scheduling task input by the user, realize the add, delete, modify and query of the task, manually schedule the task, view the scheduling task state, and view the task execution report. The core content of the scheduling task includes the autonomous driving scene, the autonomous driving algorithm to be simulated, the priority, the resource scheduling mode (static, dynamic), the running time period, etc. The autonomous driving scene is divided into a basic scene and a generic scene.

[0063] The running resource evaluation part can perform dynamic resource evaluation of the basic scene in the scheduling task, and provide the basis for the reservation resource for the scheduling algorithm. The resource evaluation can take the resource consumption of the basic scene and the simulation algorithm as the basis, take the unit CPU / GPU, the unit memory and the disk as the scheduling calculation unit, and provide the scene execution computing power benchmark for the execution of the scheduling algorithm.

[0064] The scene switching management part can realize automatic and manual simulation scene scheduling task scene switching under the scheduling algorithm instruction. The switching management mode requires the simulation execution environment to interrupt the execution of the unit scene simulation task, and to interrupt the corresponding execution engine and scene context. After the user automatically recovers or manually recovers, the corresponding scene simulation process can be restored.

[0065] The scheduling algorithm part can convert the static management of the scene simulation task to scheduling execution, including scene task switching algorithm execution, simulation execution environment resource calculation, application, and management, and can automatically or manually execute the switching algorithm.

[0066] In the actual execution process, for example, as shown in Figure 3 The simulation scene task creation can be performed, and the simulation scheduling engineer can enter the simulation scene into the system. The simulation task entry includes selecting the basic scene (mandatory), generic scene, selecting the algorithm, determining the execution priority, and determining the resource limit.

[0067] Specifically, the automatic driving simulation engineer can divide the automatic driving simulation scene into categories, and distinguish the basic scene and the generic scene: the basic scene is the main body, and the generic scene is an extension of the basic scene. The basic scene can be used for running resource evaluation to obtain the benchmark for resource scheduling.

[0068] In the execution process, the execution priority can be determined, that is, the simulation result is selected according to the urgency, such as the priority of no time requirement is general (such as entering during the day and executing at night), the priority of completing as soon as possible is intermediate, and the priority of requiring strong occupation and real-time updating is the highest (such a strong occupation task needs to be audited by a senior manager).

[0069] The priority is mainly used to combine the simulation task priority and resource utilization, to find a balance between automatic adaptation and manual management, and to maximize resource utilization.

[0070] The determination of the resource limit rule can be divided into two categories: one is global maximum resource, and the other is limited resource. The global maximum resource means that the simulation task uses all resources for the scheduling of the task. The limited resource is to limit the maximum resource upper limit (including GPU / CPU, memory, disk, etc.) of the simulation task. The limited resource is mainly used for resource worker management and distribution.

[0071] In the running resource evaluation, the basic unit of the basic scene and the algorithm can be used for running resource verification and evaluation for resource allocation in formal running.

[0072] For example, when a simulation task is created, embodiments of the present application can start simulation running according to the lowest resource configuration of the basic scene and the algorithm, and dynamically reduce the resource according to the running index, so as to obtain the best resource standard of the basic scene and the algorithm, which can be used when the basic scene, the generic scene and the parallel resource application are executed, so as to improve the parallelism of the system.

[0073] When automatic or manual task adjustment occurs, embodiments of the present application can record the state of the task being run, and store the context in the corresponding space, so as to realize fast storage and fast recovery of the task breakpoint. When task switching occurs, embodiments of the present application can save the running state data, and use the running resource to execute other simulation tasks. When the scheduling algorithm recovers the simulation scene, embodiments of the present application can quickly recover the task execution according to the context.

[0074] Specifically, in the application process, embodiments of the present application can allow the user to manually adjust the task priority or the system to automatically adjust the task priority. The task priority manually adjusted by the user can be a high-priority task, and the high-priority task is preferentially executed. The user can also reduce the execution order of the high-priority task. When the priority of the task being executed is adjusted, simulation scene switching occurs, and embodiments of the present application can record the state of the task being executed, the scene context, the algorithm and the like, so as to facilitate continuous execution when the execution is scheduled again.

[0075] Embodiments of the present application can trigger task scheduling actions after a task is created or the user adjusts the priority of the task or a simulation task is executed. The priority of the task can be adjusted as follows: the basic scheduling action is to calculate the task with the highest priority to execute, and when the priorities are the same, the FIFO strategy is used to schedule and execute; when a task with higher priority is running, simulation scene switching process is triggered, and the current simulation task is switched to the latest task; when the new task is executed, resource allocation can be performed according to the resource mode set by the task. If the global maximum is used, all resources are allocated to the task; the limited resource is divided into a certain amount of data resources in the system; if the remaining resources cannot meet the minimum resource requirement of the basic scene, the task will be suspended due to insufficient resources, and other simulation tasks are scheduled; if there is no execution task and all tasks cannot be executed, a system warning is generated, and the relevant resource user is notified.

[0076] According to the automatic driving simulation dynamic running resource scheduling method provided in the embodiments of the present application, the dynamic resources of the basic scene in each scheduling task of the scheduling task input by the user can be evaluated, the scheduling instruction of scene switching is received, and the task scheduling action corresponding to the scheduling task is executed based on the optimal resource allocation strategy, so that the simulation dynamic running resource scheduling of automatic driving is realized, and under the condition of limited resources, the fast iteration of the resource scheduling simulation scene switching can be performed, and the simulation execution efficiency is improved. Therefore, the technical problem of low simulation execution efficiency due to the large number of simulation scenes and limited resources in the related art, and the difficulty in meeting the fast iteration requirement in scene conversion is solved.

[0077] Secondly, the automatic driving simulation dynamic running resource scheduling device provided in the embodiments of the present application is described with reference to the accompanying drawings.

[0078] Figure 2 is a block schematic diagram of the automatic driving simulation dynamic running resource scheduling device in the embodiments of the present application.

[0079] As shown in Figure 2 , the automatic driving simulation dynamic running resource scheduling device 10 comprises a first acquisition module 100, a first evaluation module 200 and a scheduling module 300.

[0080] Specifically, the first acquisition module 100 is configured to acquire the scheduling task input by the user.

[0081] The first evaluation module 200 is configured to evaluate the dynamic resources of the basic scene in each scheduling task of at least one scheduling task, and determine the reserved resources of the basic scene.

[0082] The scheduling module 300 is configured to receive the scheduling instruction of scene switching, and execute the task scheduling action corresponding to the scheduling task based on the optimal resource allocation strategy, wherein the optimal resource allocation strategy is obtained based on the running resource evaluation of each simulation scene of at least one simulation scene.

[0083] Optionally, in an embodiment of the present application, the automatic driving simulation dynamic running resource scheduling device 10 further comprises a second acquisition module and a second evaluation module.

[0084] The second acquisition module is configured to acquire at least one simulation scene input by the user.

[0085] The second evaluation module is configured to evaluate the running resources of each simulation scene of at least one simulation scene, and generate the optimal resource allocation strategy adjusted based on the running index in real time.

[0086] Optionally, in an embodiment of the present application, each simulation scene comprises at least one of a generic scene, a selection algorithm, an execution priority and a limited resource and a basic scene.

[0087] Optionally, in an embodiment of the present application, the automatic driving simulation dynamic running resource scheduling device 10 further comprises a recording module and a recovery module.

[0088] The recording module is configured to record the current running task state of the current simulation scene and store the context to the corresponding space during the automatic driving simulation dynamic running.

[0089] The recovery module is configured to recover the current simulation scene according to the context and the task state after the execution of the task scheduling action corresponding to the scheduling task ends.

[0090] Optionally, in an embodiment of the present application, the scheduling task comprises at least one of an automatic driving scene, an automatic driving algorithm to be simulated, a priority, a resource scheduling mode and a running time period.

[0091] It should be noted that the foregoing explanation and description of the embodiment of the automatic driving simulation dynamic running resource scheduling method also apply to the automatic driving simulation dynamic running resource scheduling device of this embodiment, which will not be described here.

[0092] The automatic driving simulation dynamic running resource scheduling device according to the embodiment of the present application can evaluate the dynamic resources of the basic scene in each scheduling task of the scheduling task based on the scheduling task input by the user, receive a scheduling instruction for scene switching, and execute a task scheduling action corresponding to the scheduling task based on the best resource allocation strategy, thereby realizing simulation dynamic running resource scheduling of automatic driving, and enabling fast iteration when the resource scheduling simulation scene switches under the condition of limited resources, and improving the simulation execution efficiency. Thus, the technical problem of low simulation execution efficiency due to the difficulty in meeting the fast iteration requirement when the scene is converted because of the large number of simulation scenes and limited resources in the related art is solved.

[0093] Figure 5 The structure schematic diagram of the electronic device provided by the embodiment of the present application is provided. The electronic device can include:

[0094] The memory 501, the processor 502 and the computer program stored in the memory 501 and executable on the processor 502.

[0095] The processor 502 implements the automatic driving simulation dynamic running resource scheduling method provided in the above embodiments when executing the program.

[0096] Further, the electronic device further includes:

[0097] The communication interface 503 is configured to communicate between the memory 501 and the processor 502.

[0098] The memory 501 is configured to store a computer program capable of being executed on the processor 502.

[0099] The memory 501 can include a high-speed RAM memory, and can further include a non-volatile memory, for example, at least one disk memory.

[0100] If the memory 501, the processor 502 and the communication interface 503 are independently implemented, the communication interface 503, the memory 501 and the processor 502 can be connected with each other through a bus and complete communication between each other. The bus can be an Industry Standard Architecture (ISA) bus, a Peripheral Component (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For the convenience of representation, Figure 5 In the figure, only one thick line is used to represent the bus, but it does not mean that there is only one bus or only one type of bus.

[0101] Optionally, in a specific implementation, if the memory 501, the processor 502 and the communication interface 503 are integrated on a chip, the memory 501, the processor 502 and the communication interface 503 can complete communication between each other through an internal interface.

[0102] The processor 502 can 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.

[0103] The embodiments of the present application further provide a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the automatic driving simulation dynamic running resource scheduling method.

[0104] In the description of the application, reference to "one embodiment", "some embodiments", "an example", "a specific example", or "some examples" means that a particular feature, structure, material, or characteristic being described is included in at least one embodiment or example of the application. The appearances of the phrase in various places in the specification are not necessarily all referring to the same embodiment or example. Furthermore, the described specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples. In addition, the usage of "N" means at least two, for example, two, three or the like, unless explicitly stated otherwise.

[0105] Furthermore, the terms "first", "second", or the like, are used merely as a designation of certain elements or features, and do not imply or connote relative importance or a specific order of categorization thereof. Accordingly, features described as "first" or "second" can implicitly or explicitly include at least one of the features. In the description of the application, the meaning of "N" is at least two, for example, two, three, etc., unless explicitly specified otherwise.

[0106] Any process or method descriptions or blocks in flow charts or otherwise described herein represent embodiments which can be managed as one or more modules, segments, or portions of code which include one or more executable instructions for implementing specific logic functions or steps, and alternate implementations are possible. In some embodiments, the processes and methods described can be executed by one or more apparatuses or devices, either directly or after conversion to another language. Alternate implementations are possible.

[0107] The logic and / or steps represented in the flowcharts and / or described herein, for example, can be considered as a sequence of instructions to implement logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, apparatus, or device, such as a computer-based system, processor- based system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For purposes of this specification, a "computer-readable medium" can be any apparatus that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. The computer-readable medium can be a computer- readable storage medium or a computer-readable signal medium. The computer- readable storage medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer-readable storage medium include the following: an electrical connection having one or more wires (electrical connections), a portable computer diskette (a magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium can even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via, for example, optical scanning of the paper or other medium, then compiled, interpreted, or otherwise processed in a suitable manner, if necessary, and then stored in a computer memory.

[0108] It should be understood that aspects of the application can be implemented in hardware, software, firmware or combinations thereof. In the above embodiments, the N steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. If implemented in hardware and in another embodiment, any of the following technologies, known in the art, or their combinations can be used: discrete logic circuitry having logic gates for implementing logic functions on data signals, application specific integrated circuits having appropriate combinational logic gates, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc.

[0109] Those skilled in the art can understand that all or part of the steps carried out by the above-mentioned embodiment methods can be completed by programs instructing related hardware, and the programs can be stored in a computer-readable storage medium. When the programs are executed, one or a combination of the steps of the method embodiments is included.

[0110] In addition, each of the functional units in the various embodiments of the present application can be integrated in one processing module, or each of the units can be physically present separately, or two or more units can be integrated in one module. The integrated module can be implemented in the form of hardware or in the form of a software functional module. When the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it can also be stored in a computer readable storage medium.

[0111] The storage medium mentioned above can be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it should be understood that the above embodiments are exemplary and should not be construed as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above embodiments within the scope of the present application.

Claims

1. A method for dynamic resource scheduling in autonomous driving simulation, characterized in that, Includes the following steps: Retrieve user-entered scheduling tasks; Evaluate the dynamic resources of the basic scenario in each of at least one scheduling task, and determine the reservation resources of the basic scenario; as well as Receive the scheduling instruction for scene switching, and execute the task scheduling action corresponding to the scheduling task based on the optimal resource allocation strategy, wherein the optimal resource allocation strategy is obtained by evaluating the running resources of each simulation scene in at least one simulation scene; Before executing the task scheduling action corresponding to the scheduled task based on the optimal resource allocation strategy, the method further includes: Obtain at least one simulation scenario entered by the user; Evaluate the runtime resources of each simulation scenario in the at least one simulation scenario, and generate the optimal resource allocation strategy that is adjusted at runtime based on runtime metrics; Each simulation scenario includes at least one of the following: a generic scenario, a selection algorithm, an execution priority, and limited resources, and the base scenario.

2. The method according to claim 1, characterized in that, After receiving the scheduling instruction for scene switching, the process also includes: During the dynamic execution of autonomous driving simulation, the current running task state of the current simulation scenario is recorded and the context is stored in the corresponding space; After the task scheduling action corresponding to the scheduled task is completed, the current simulation scenario is restored according to the context and the task status.

3. The method according to claim 1, characterized in that, The scheduling task includes at least one of the following: autonomous driving scenario, autonomous driving algorithm to be simulated, priority, resource scheduling mode, and running time period.

4. A dynamic operation resource scheduling device for autonomous driving simulation, characterized in that, include: The first acquisition module is used to acquire the scheduling tasks entered by the user; The first evaluation module is used to evaluate the dynamic resources of the basic scenario in each of at least one scheduling task and determine the reservation resources of the basic scenario. as well as The scheduling module is used to receive scheduling instructions for scene switching and execute the task scheduling action corresponding to the scheduling task based on the optimal resource allocation strategy, wherein the optimal resource allocation strategy is obtained by evaluating the running resources of each simulation scene in at least one simulation scene. Also includes: The second acquisition module is used to acquire at least one simulation scenario entered by the user; The second evaluation module is used to evaluate the running resources of each simulation scenario in the at least one simulation scenario and generate the best resource allocation strategy that is adjusted based on running indicators at runtime. Each simulation scenario includes at least one of the following: a generic scenario, a selection algorithm, an execution priority, and limited resources, and the base scenario.

5. The apparatus according to claim 4, characterized in that, Also includes: The recording module is used to record the current running task state of the current simulation scenario and store the context to the corresponding space during the dynamic running of the autonomous driving simulation. The recovery module is used to restore the current simulation scene based on the context and the task status after the task scheduling action corresponding to the scheduling task has been completed.

6. A vehicle, characterized in that, include: The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the dynamic operation resource scheduling method for autonomous driving simulation as described in any one of claims 1-3.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the dynamic operation resource scheduling method for autonomous driving simulation as described in any one of claims 1-3.

Images