A method, device, equipment and storage medium for loading autonomous driving simulation data

Abstract

The present application discloses a method, device, equipment and storage medium for loading autonomous driving simulation data. The method comprises: obtaining autonomous driving simulation data from a cloud storage system through a high-speed cache system; and storing the autonomous driving simulation data in a simulation task executor according to the type of the autonomous driving simulation data. The present application obtains autonomous driving simulation data from a cloud storage system through a high-speed cache system, and then stores the autonomous driving simulation data in a simulation task executor according to the type of the autonomous driving simulation data, thereby greatly improving the overall data download rate, thereby achieving the purpose of improving simulation efficiency and reducing time cost.

Description

Technical Field

[0001] The present invention relates to the field of autonomous driving technology, and in particular to an autonomous driving simulation data loading method, device, equipment and storage medium. Background Art

[0002] At present, large-scale simulation technology based on cloud platforms plays an important role in the field of autonomous driving. It can reduce the time cost and safety hazards of algorithm testing in test vehicles, while improving the efficiency of algorithm testing and the coverage of algorithm testing for various scenarios.

[0003] In order to comprehensively evaluate the current version of the autonomous driving algorithm, each single test will contain hundreds of kilometers of road test data, including high-precision map data, scene data, and simulation code files of the current version of the autonomous driving system. Hundreds of thousands of kilometers of simulation mileage every day will use a huge amount of data. The relevant existing technology is to download the required relevant data directly from the storage file system to the simulation container used to run the simulation algorithm.

[0004] It can be seen that the related existing technologies are relatively simple to implement, and the storage method is cheap and relatively simple, but the slow file download speed leads to low simulation efficiency and increases a lot of time costs. Compared with the simulation time of several minutes, the time cost of downloading data is very large. Summary of the invention

[0005] To this end, the technical problem solved by the embodiments of the present application is to provide an autonomous driving simulation data loading method, device, equipment and storage medium, which can increase the file download speed, thereby improving simulation efficiency and reducing time cost.

[0006] In order to solve the above technical problems, the technical solutions adopted in this application are as follows:

[0007] On the one hand, an embodiment of the present application provides a method for loading autonomous driving simulation data, the method comprising:

[0008] Obtain autonomous driving simulation data from a cloud storage system through a high-speed cache system;

[0009] The autonomous driving simulation data are classified and stored in a simulation task executor according to the type of the autonomous driving simulation data.

[0010] Furthermore, the step of classifying and storing the autonomous driving simulation data into the simulation task executor according to the type of the autonomous driving simulation data specifically includes:

[0011] Different loading methods are used according to the type of autonomous driving simulation data to classify and store the autonomous driving simulation data in the simulation task executor.

[0012] Furthermore, when the autonomous driving simulation data is high-precision map data, the high-precision map data is loaded into the simulation task executor by local loading;

[0013] When the autonomous driving simulation data is an autonomous driving simulation code package, loading the autonomous driving simulation code package into the simulation task executor in a one-time loading manner;

[0014] When the autonomous driving simulation data is a scene data packet, the scene data packet is loaded into the simulation task executor in a simulation and loading asynchronous manner.

[0015] Preferably, when the autonomous driving simulation data is high-precision map data, the method includes:

[0016] Mounting the high-precision map data from the cloud storage system once through a high-speed cache system;

[0017] Calculate the map range required for the current simulation task according to the trajectory information of the current simulation task;

[0018] The local high-precision map data corresponding to the map range is loaded into the simulation container of the simulation task executor through the cache system.

[0019] More preferably, the number of the simulation containers is several; the step of loading the local high-precision map data corresponding to the map range into the simulation container of the simulation task executor through the cache system specifically includes:

[0020] Different local high-precision map data corresponding to different map ranges are loaded into different simulation containers through the cache system.

[0021] Still further preferably, the trajectory information includes starting point information, path information and end point information of the simulated vehicle travel.

[0022] Preferably, when the autonomous driving simulation data is an autonomous driving simulation code package, the method includes:

[0023] Acquire the autonomous driving simulation code package from the cloud storage system through a high-speed cache system;

[0024] The autonomous driving simulation code package is loaded into the simulation container of the simulation task executor at one time.

[0025] More preferably, the number of the simulation containers is several; and the step of loading the autonomous driving simulation code package into the simulation container of the simulation task executor at one time specifically includes:

[0026] According to the version of the autonomous driving simulation code package, different versions of the autonomous driving simulation code packages are loaded into different simulation containers at one time.

[0027] Preferably, when the autonomous driving simulation data is a scene data packet, the method includes:

[0028] Mounting the scene data packet from the cloud storage system through a high-speed cache system;

[0029] Reading the current scene data packet into the simulation container of the simulation executor through the cache system;

[0030] While starting the simulation of the current scene data packet in the simulation container, the next scene data packet is read into the simulation container in parallel through the cache system.

[0031] On the other hand, an embodiment of the present application provides an automatic driving simulation data loading device, the device comprising:

[0032] A data acquisition module, used to acquire autonomous driving simulation data from a cloud storage system through a high-speed cache system;

[0033] The data processing module is used to classify and store the autonomous driving simulation data into a simulation task executor according to the type of the autonomous driving simulation data.

[0034] On the other hand, an embodiment of the present application provides a device, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor implements any one of the above-mentioned steps of the method for loading autonomous driving simulation data when executing the computer program.

[0035] On the other hand, an embodiment of the present application provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, the steps of any one of the above-mentioned methods for loading autonomous driving simulation data are performed.

[0036] In summary, compared with the prior art, the technical solution provided in the embodiment of the present application has at least the following beneficial effects:

[0037] 1. The embodiment of the present application obtains autonomous driving simulation data from a cloud storage system by adopting a high-speed cache system, and then classifies and stores the autonomous driving simulation data into a simulation task executor according to the type of the autonomous driving simulation data, thereby greatly improving the overall data download rate, thereby achieving the purpose of improving simulation efficiency and reducing time cost.

[0038] 2. The embodiments of the present application classify and store the autonomous driving simulation data into the simulation task executor by using different loading methods according to the type of the autonomous driving simulation data, thereby laying a foundation for implementing fast and efficient large-scale autonomous driving simulation when the cache system traffic is limited.

[0039] 3. When the autonomous driving simulation data is high-precision map data, the embodiment of the present application uses a high-speed cache system to load only the local high-precision map data corresponding to the map range required for the current simulation task into the simulation container. Not only can the local high-precision map data required for the current simulation task be accurately loaded into the simulation container for immediate use, but there is no need to download all high-precision map data for each current simulation task. On the one hand, the overall download speed of the data is further improved, thereby further improving the simulation efficiency and reducing the time cost. On the other hand, it also greatly reduces the workload of the cache system.

[0040] 4. When the autonomous driving simulation data is an autonomous driving simulation code package, the embodiment of the present application uses a cache system to load the autonomous driving simulation code package into the simulation container once. After the autonomous driving simulation code package is loaded once, for the subsequent re-simulation of the same simulation task, a strategy of reusing the autonomous driving simulation code package can be adopted to minimize repeated loading, improve simulation efficiency, and reduce the workload of the cache system. Moreover, simulation tasks using the same version of the autonomous driving simulation code package can be collected in the simulation container to form a simulation task group. For the simulation of each simulation task in the simulation task group, the autonomous driving simulation code package can be reused, thereby significantly reducing the workload of the cache system.

[0041] 5. When the autonomous driving simulation data is a scene data packet, the embodiment of the present application adopts a high-speed cache system to start simulating the current scene data packet in the simulation container and read the next scene data packet into the simulation container in parallel through the asynchronous simulation and loading method. Thus, without affecting the simulation of the current scene data packet, the data reading time of the high-speed cache system is prolonged, reducing the workload of the high-speed cache system. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 It is a flowchart of the method for loading autonomous driving simulation data provided by the first exemplary embodiment of the present application.

[0043] Figure 2 It is a flow chart of the method for loading autonomous driving simulation data provided by the second exemplary embodiment of the present application.

[0044] Figure 3It is a flowchart of the method for loading autonomous driving simulation data provided by the third exemplary embodiment of the present application.

[0045] Figure 4 It is a flow chart of the method for loading autonomous driving simulation data provided by the fourth exemplary embodiment of the present application.

[0046] Figure 5 It is a flowchart of the method for loading autonomous driving simulation data provided by the fifth exemplary embodiment of the present application.

[0047] Figure 6 It is a flow chart of the method for loading autonomous driving simulation data provided by the sixth exemplary embodiment of the present application.

[0048] Figure 7 It is a structural diagram of an autonomous driving simulation data loading device provided by the seventh exemplary embodiment of the present application.

[0049] Figure 8 It is a schematic diagram of the structure of the device provided by the eighth exemplary embodiment of the present application. DETAILED DESCRIPTION

[0050] This specific embodiment is merely an explanation of the present application and is not a limitation of the present application. After reading this specification, those skilled in the art may make modifications to the present embodiment without any creative contribution as needed, but such modifications are protected by the patent law as long as they are within the scope of the claims of the present application.

[0051] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0052] The term "comprise" and any variations thereof in the specification and claims of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products or apparatuses.

[0053] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0054] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.

[0055] Figure 1 The first exemplary embodiment of the present application provides a method for loading autonomous driving simulation data. The main steps of the method are described as follows:

[0056] Obtain autonomous driving simulation data from a cloud storage system through a high-speed cache system;

[0057] The autonomous driving simulation data are classified and stored in a simulation task executor according to the type of the autonomous driving simulation data.

[0058] This embodiment uses a high-speed cache system to obtain autonomous driving simulation data from a cloud storage system, and then classifies and stores the autonomous driving simulation data into a simulation task executor according to the type of the autonomous driving simulation data, thereby greatly improving the overall data download rate, thereby achieving the purpose of improving simulation efficiency and reducing time costs.

[0059] The inventors also found in the process of implementing this application that since the cost of using the cache system is relatively high, an upper limit is generally set for the overall flow of the cache system. If the same loading method is used for different types of autonomous driving simulation data, the overall flow of the cache system will not be reasonably configured, making it difficult to achieve fast and efficient large-scale autonomous driving simulation.

[0060] In order to solve the technical problem of reasonably configuring the overall traffic of the cache system to achieve faster and more efficient large-scale autonomous driving simulation, Figure 2 A method for loading autonomous driving simulation data is provided by a second exemplary embodiment of the present application. Figure 1 The first exemplary embodiment shown is further improved, and the main steps of the method are described as follows:

[0061] Obtain autonomous driving simulation data from a cloud storage system through a high-speed cache system;

[0062] Different loading methods are used according to the type of autonomous driving simulation data to classify and store the autonomous driving simulation data in the simulation task executor.

[0063] This embodiment adopts different loading methods according to the type of autonomous driving simulation data to classify and store the autonomous driving simulation data into the simulation task executor, so as to reasonably configure the overall traffic of the cache system when the traffic of the cache system is limited, alleviate the working pressure of the cache system, and lay the implementation foundation for realizing fast and efficient large-scale autonomous driving simulation.

[0064] The types of the autonomous driving simulation data include, but are not limited to: high-precision map data, autonomous driving simulation code packages, and scenario data packages.

[0065] The high-precision map data is collected by manually driving a collection vehicle, and then a high-precision map is generated based on the collected map data. The collection vehicle is a mobile collection system that is precisely integrated with a variety of advanced measurement sensors, generally including Lidar, inertial navigation, cameras and other equipment, and is equipped with different types of sensor equipment according to different collection scenarios. After the sensor equipment collects the precision data from the external real world, it must go through image recognition, precision processing, manual processing and other steps to become usable high-precision map data.

[0066] The autonomous driving simulation code package includes the simulation executable file and the data files that the simulation executable file depends on. Most simulation tasks are to evaluate different versions of autonomous driving algorithms, so the simulation executable files of the autonomous driving algorithms required for simulation are generated in real time during the entire simulation workflow. Therefore, the simulation code package used for a specific simulation task is fixed.

[0067] The scene data packet includes a plurality of parameterized initial scene data and drive test sampling data; the parameterized initial scene data includes the initial position information and orientation information of the autonomous driving vehicle, and also includes road information and obstacle information, wherein the obstacle information includes at least one of the position information, speed and acceleration of the obstacle. The drive test sampling data includes the initial scene data and the change state information of at least one of the position, speed and acceleration of the obstacle at multiple time points within the time interval.

[0068] In order to further realize the use of different loading methods to load different types of autonomous driving simulation data into the simulation task executor, Figure 3 A method for loading autonomous driving simulation data is provided by the third exemplary embodiment of the present application. Figure 2 Further improvements are made on the basis of the second exemplary embodiment shown, and the main steps of the method are described as follows:

[0069] When the autonomous driving simulation data is high-precision map data, the high-precision map data is loaded into the simulation task executor by local loading;

[0070] When the autonomous driving simulation data is an autonomous driving simulation code package, loading the autonomous driving simulation code package into the simulation task executor in a one-time loading manner;

[0071] When the autonomous driving simulation data is a scene data packet, the scene data packet is loaded into the simulation task executor in a simulation and loading asynchronous manner.

[0072] Among them, high-precision map data records the map data necessary for autonomous driving simulation. For simulation tasks performed in the same city, the map data is the same and relatively fixed, but at the same time the map data is also very large. Loading map data for multiple simulation tasks at the same time can easily put tremendous pressure on the cache system. Moreover, not all map data information is needed in a single simulation task. Therefore, for high-precision map data, the inventors adopt a precise partial loading solution to find out the required map range for different simulation tasks, and only load the local high-precision map data corresponding to the map range, without loading all the high-precision map data. In this way, on the one hand, the data download rate is further improved and the simulation rate is increased, reducing the time cost, and on the other hand, the workload of the cache system is reduced.

[0073] Most simulation tasks are to evaluate different versions of autonomous driving algorithms, so the autonomous driving algorithm executable files required for simulation are generated in real time throughout the simulation workflow. The simulation code package used for a specific autonomous driving is fixed. Therefore, for the autonomous driving simulation code package, a strategy of loading it into the simulation executor once or once and then reusing it is adopted to minimize repeated loading, further improve simulation efficiency, and reduce the workload of the cache system.

[0074] Each specific simulation task will evaluate a large number of different scenarios. All types of data related to the scenarios are unique to each scenario. The simulation task executor can only obtain the scenario data to be performed after initialization, so the scenario data must be loaded. However, if a large amount of scenario data is loaded at the same time, it is easy to put a lot of pressure on the cache system. Considering that in the simulation task executor, multiple simulation tasks are performed serially. Therefore, for the scene data packet, its loading and the execution of the simulation task are performed asynchronously. When the current simulation task is executed, the scene data packet required for the next simulation task is loaded, thereby achieving the purpose of reducing the workload of the cache system.

[0075] The simulation executor is provided with a simulation container, and the simulation container is used to execute at least one simulation task.

[0076] Figure 4 A method for loading autonomous driving simulation data is provided by the fourth exemplary embodiment of the present application. Figure 3 Further improvements are made on the basis of the third exemplary embodiment shown, and the main steps of the method are described as follows:

[0077] When the autonomous driving simulation data is high-precision map data, the method includes:

[0078] Mounting the high-precision map data from the cloud storage system once through a high-speed cache system;

[0079] Calculate the map range required for the current simulation task according to the trajectory information of the current simulation task;

[0080] The local high-precision map data corresponding to the map range is loaded into the simulation container of the simulation task executor through the cache system.

[0081] This embodiment uses a high-speed cache system to mount the high-precision map data from the cloud storage system at one time, thereby ensuring that the current simulation task executor can obtain the latest high-precision map data; then the map range required for the current simulation task is calculated based on the trajectory information of the current simulation task; finally, the high-speed cache system is used to load only the local high-precision map data corresponding to the required map range into the simulation container. Not only can the local high-precision map data required for the current simulation task be accurately loaded into the simulation container for immediate use, but there is no need to download all the high-precision map data for each current simulation task.

[0082] The trajectory information includes the starting point information, path information and end point information of the simulated vehicle. The calculation method of the map range is as follows:

[0083] For the current simulation task, we will set a starting point (x1, y1) and multiple end points ((x2, y2), ..., (xn, yn)). For these multiple end points, we take the min, max, and add padding (i.e. path), thus obtaining the following data:

[0084] Xmin=MIN(x1,…,xn)–padding_size

[0085] Xmax=MAX((x1,…,xn)+padding_size

[0086] Ymin=MIN(y1,…,yn)-padding_size

[0087] Ymax=MAX((y1,…,yn)+padding_size) Draw a rectangular range through the four points Xmin, Xmax, Ymin, and Ymax, which is the map range required for the current simulation task.

[0088] As a further improvement of this embodiment, the number of the simulation containers is several; the loading of the local high-precision map data corresponding to the map range into the simulation container of the simulation task executor through the cache system specifically includes:

[0089] Different local high-precision map data corresponding to different map ranges are loaded into different simulation containers through the cache system.

[0090] By setting up multiple simulation containers to execute multiple different simulation tasks in parallel, the simulation efficiency is further improved without increasing the workload of the cache system.

[0091] Figure 5 A method for loading autonomous driving simulation data is provided in the fifth exemplary embodiment of the present application. Figure 3 Further improvements are made on the basis of the third exemplary embodiment shown, and the main steps of the method are described as follows:

[0092] When the autonomous driving simulation data is an autonomous driving simulation code package, the method includes:

[0093] Acquire the autonomous driving simulation code package from the cloud storage system through a high-speed cache system;

[0094] The autonomous driving simulation code package is loaded into the simulation container of the simulation task executor at one time.

[0095] The embodiment of the present application uses a cache system to obtain the autonomous driving simulation code package from the cloud storage system; and then loads the autonomous driving simulation code package into the simulation container once. After the autonomous driving simulation code package is loaded once, for subsequent re-simulations of the same simulation task, a strategy of reusing the autonomous driving simulation code package can be adopted to minimize repeated loading, improve simulation efficiency, and reduce the workload of the cache system. Moreover, simulation tasks using the same version of the autonomous driving simulation code package can be collected in the simulation container to form a simulation task group. For the simulation of each simulation task in the simulation task group, the autonomous driving simulation code package can be reused, thereby significantly reducing the workload of the cache system.

[0096] As a further improvement of this embodiment, the number of the simulation containers is several; the step of loading the autonomous driving simulation code package into the simulation container of the simulation task executor at one time specifically includes:

[0097] According to the version of the autonomous driving simulation code package, different versions of the autonomous driving simulation code packages are loaded into different simulation containers at one time.

[0098] By setting up multiple simulation containers to obtain different versions of autonomous driving simulation code packages, simulation tasks using the same version of the autonomous driving simulation code package are collected in each simulation container to form a simulation task group. For the simulation of each simulation task in the simulation task group, the same version of the autonomous driving simulation code package can be reused, thereby significantly reducing the workload of the cache system. Moreover, multiple simulation containers work in parallel, which further improves the simulation efficiency without increasing the workload of the cache system.

[0099] Figure 6 A method for loading autonomous driving simulation data is provided in the sixth exemplary embodiment of the present application. Figure 3 Further improvements are made on the basis of the third exemplary embodiment shown, and the main steps of the method are described as follows:

[0100] When the autonomous driving simulation data is a scene data packet, the method includes:

[0101] Mounting the scene data packet from the cloud storage system through a high-speed cache system;

[0102] Reading the current scene data packet into the simulation container of the simulation executor through the cache system;

[0103] While starting the simulation of the current scene data packet in the simulation container, the next scene data packet is read into the simulation container in parallel through the cache system.

[0104] This embodiment uses a cache system to first mount the scene data packet from the cloud storage system; then read the current scene data packet into the simulation container through the cache system; then, while starting the simulation of the current scene data packet in the simulation container, the next scene data packet is read into the simulation container through the cache system in parallel. Thus, without affecting the simulation of the current scene data packet, the data reading time of the cache system is prolonged, reducing the working pressure of the cache system.

[0105] Figure 7The seventh exemplary embodiment of the present application provides an automatic driving simulation data loading device, which corresponds one-to-one to the loading method in the above embodiment, and the loading device includes:

[0106] A data acquisition module, used to acquire autonomous driving simulation data from a cloud storage system through a high-speed cache system;

[0107] The data processing module is used to classify and store the autonomous driving simulation data into a simulation task executor according to the type of the autonomous driving simulation data.

[0108] Each module of the above-mentioned loading device can be implemented in whole or in part by software, hardware or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute the operations corresponding to each module.

[0109] Figure 8 It is a device provided by the eighth exemplary embodiment of the present application, and the device may be a server. The device includes a processor, a memory and a communication interface connected via a system bus. Among them, the processor of the device is used to provide computing and control capabilities. The memory of the device can be implemented by any type of volatile or non-volatile storage device or a combination thereof, and the volatile or non-volatile storage device includes but is not limited to: a disk, an optical disk, an EEPROM, an EPROM, an SRAM, a ROM, a magnetic storage device, a flash memory, and a PROM. The memory of the device provides an environment for the operation of the operating system and computer programs stored therein. The communication interface of the device is a network interface, and the network interface is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the loading method steps described in the above embodiment are implemented.

[0110] In another embodiment of the present application, a storage medium is provided, wherein the storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the loading method described in the above embodiment are implemented. The storage medium includes but is not limited to: ROM, RAM, CD-ROM, magnetic disk, and floppy disk.

[0111] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device described in the present application is divided into different functional units or modules to complete all or part of the functions described above.

[0112] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of the present application, ordinary technicians in this field can also make many forms without departing from the purpose of the present application and the scope of protection of the claims, all of which are within the protection of the present application.

Claims

1. A method for loading autonomous driving simulation data, characterized in that: The method comprises: Obtain autonomous driving simulation data from a cloud storage system through a high-speed cache system; According to the type of the autonomous driving simulation data, different loading methods are used to classify and store the autonomous driving simulation data into the simulation task executor; When the autonomous driving simulation data is high-precision map data, the high-precision map data is loaded into the simulation task executor by local loading; When the autonomous driving simulation data is an autonomous driving simulation code package, the autonomous driving simulation code package is obtained from the cloud storage system through a high-speed cache system, and according to the version of the autonomous driving simulation code package, different versions of the autonomous driving simulation code packages are loaded into different simulation containers of the simulation task executor at one time, and the number of the simulation containers is several; When the autonomous driving simulation data is a scene data packet, the scene data packet is loaded into the simulation task executor in a simulation and loading asynchronous manner.

2. The method for loading autonomous driving simulation data according to claim 1, characterized in that: When the autonomous driving simulation data is high-precision map data, the method includes: Mounting the high-precision map data from the cloud storage system once through a high-speed cache system; Calculate the map range required for the current simulation task according to the trajectory information of the current simulation task; The local high-precision map data corresponding to the map range is loaded into the simulation container of the simulation task executor through the cache system.

3. The method for loading automatic driving simulation data according to claim 2, characterized in that: The number of the simulation containers is several; the local high-precision map data corresponding to the map range is loaded into the simulation container of the simulation task executor through the high-speed cache system, specifically including: Different local high-precision map data corresponding to different map ranges are loaded into different simulation containers through the cache system.

4. The method for loading automatic driving simulation data according to claim 2 or 3, characterized in that: The trajectory information includes the starting point information, path information and end point information of the simulated vehicle travel.

5. The method for loading autonomous driving simulation data according to claim 1, characterized in that: When the autonomous driving simulation data is a scene data packet, the method includes: Mounting the scene data packet from the cloud storage system through a high-speed cache system; Reading the current scene data packet into the simulation container of the simulation executor through the cache system; While starting the simulation of the current scene data packet in the simulation container, the next scene data packet is read into the simulation container in parallel through the cache system.

6. An automatic driving simulation data loading device, characterized in that: The device comprises: A data acquisition module, used to acquire autonomous driving simulation data from a cloud storage system through a high-speed cache system; A data processing module, used to classify and store the autonomous driving simulation data into a simulation task executor by using different loading methods according to the type of the autonomous driving simulation data; A classification loading module is used for loading the high-precision map data into the simulation task executor by local loading when the autonomous driving simulation data is high-precision map data; for obtaining the autonomous driving simulation code package from the cloud storage system by a high-speed cache system when the autonomous driving simulation data is an autonomous driving simulation code package, and for loading different versions of the autonomous driving simulation code package into different simulation containers of the simulation task executor at one time according to the version of the autonomous driving simulation code package, where the number of the simulation containers is several; and for loading the scene data package into the simulation task executor by asynchronous simulation and loading when the autonomous driving simulation data is a scene data package.

7. A device, characterized in that: It includes a memory, a processor, and a computer program stored in the memory and capable of running on the processor, and when the processor executes the computer program, the steps of the autonomous driving simulation data loading method as described in any one of claims 1 to 5 are implemented.

8. A storage medium, characterized in that: The storage medium stores a computer program, which, when executed by a processor, implements the steps of the method for loading autonomous driving simulation data as described in any one of claims 1 to 5.

Images