Resource scheduling method and apparatus, as well as chip systems, electronic devices and vehicles
The integrated resource scheduling method optimizes resource allocation between cockpit and intelligent driving systems in intelligent connected vehicles, enhancing utilization and stability by sharing resources based on user scenes, thus improving the vehicle's performance.
Patent Information
- Application Number
- JP2025059234
- Authority / Receiving Office
- JP · JP
- Patent Type
- Patents
- Current Assignee / Owner
- Priority Date
- 2024-08-20
- Filing Date
- 2025-03-31
- Publication Date
- 2026-11-19
- Estimated Expiration
- 2045-03-31
AI Technical Summary
Existing resource scheduling in intelligent connected vehicles is isolated between operating systems, leading to unreasonable resource allocation and low utilization rates, as seen in the independent operation of cockpit and intelligent driving systems.
A resource scheduling method and apparatus that integrates the cockpit and intelligent driving systems, allowing shared resource allocation based on user scenes, utilizing a chip system with a hardware layer, system layer, and resource management layer to optimize resource distribution.
Enhances resource utilization and stability by enabling inter-system resource scheduling, ensuring rational allocation and meeting the resource requirements of each operating system in different user scenarios, improving the overall performance of the vehicle's functions.
Smart Images

Figure 0007953864000002 
Figure 0007953864000003 
Figure 0007953864000004
Abstract
Description
TECHNICAL FIELD
[0001] The present disclosure relates to the technical field of intelligent driving, and in particular, relates to a resource scheduling method and apparatus, as well as a chip system, an electronic device and a vehicle.
BACKGROUND ART
[0002] Multiple operating systems, such as an intelligent driving system and a cockpit system, can operate on an intelligent connected vehicle. Among them, the intelligent driving system is mainly responsible for autonomous driving and safety of the vehicle, including environment sensing, path planning, decision making and control, etc. The cockpit system is mainly responsible for the driving and riding experience inside the vehicle and human-computer interaction, including infotainment, meter display, media player and environment control, etc.
[0003] At present, the resource scheduling of each operating system is isolated, which causes problems such as unreasonable resource allocation and low resource utilization. For example, when the intelligent driving system requires more resources to perform complex path planning, the cockpit system may just have free resources, however, these free resources cannot be utilized by the intelligent driving system, which reduces the utilization rate of system resources.
SUMMARY OF THE INVENTION
PROBLEMS TO BE SOLVED BY THE INVENTION
[0004] In order to improve the resource utilization rate among multiple operating systems, the embodiments of the present disclosure provide a resource scheduling method, apparatus, chip system, electronic device and vehicle.
MEANS FOR SOLVING THE PROBLEMS
[0005] A first aspect of the present invention provides a resource scheduling method that can be used in an integrated circuit running at least a first operating system and a second operating system, the method comprising: determining a first user scene of the first operating system; and determining a resource scheduling policy for scheduling first resources between the first operating system and the second operating system based on the first user scene, wherein the first resources include resources shared by the first operating system and the second operating system.
[0006] For example, the first operating system is a cockpit system and the second operating system is an intelligent driving system, or the first operating system is an intelligent driving system and the second operating system is a cockpit system.
[0007] A second aspect of the present invention provides a resource scheduling device that can be used in an integrated circuit running at least a first operating system and a second operating system, the device comprising: a scene recognition unit configured to determine a first user scene of the first operating system; and a scheduling management unit configured to determine a resource scheduling policy for scheduling first resources between the first operating system and the second operating system based on the first user scene, wherein the first resources include resources shared by the first operating system and the second operating system.
[0008] A third aspect of the present disclosure provides a chip system comprising a processor and a memory, the memory storing computer program instructions that, when executed by the processor, cause the chip system to execute steps of the resource scheduling method relating to the first aspect of the present disclosure and each implementation thereof.
[0009] A fourth aspect of this disclosure provides electronic equipment that may include a chip system provided in a third aspect of this disclosure.
[0010] A fifth aspect of this disclosure provides a vehicle that may include a chip system provided in a third aspect of this disclosure, or electronic equipment provided in a fourth aspect of this disclosure. [Effects of the Invention]
[0011] The technical proposals relating to the embodiments of this disclosure enable resource scheduling between different operating systems based on user scenes of each operating system. In other words, they realize inter-system resource scheduling based on user scenes, satisfy the resource requirements of each operating system in different user scenes, improve the execution stability of the operating systems, and further make resource allocation more rational and improve resource utilization. [Brief explanation of the drawing]
[0012] The above and other purposes, features and advantages of this disclosure will become clearer by describing the embodiments of this disclosure in more detail with reference to the drawings. The drawings are used to further illustrate the embodiments of this disclosure, constitute part of the specification, and are used to describe this disclosure together with the embodiments of this disclosure, and do not limit this disclosure. In the drawings, the same reference numerals generally represent the same members or steps. [Figure 1] This is an architectural diagram of the software and hardware of a cockpit driving integration system according to one exemplary embodiment of the present disclosure. [Figure 2] This is a flowchart of a resource scheduling method according to one exemplary embodiment of the present disclosure. [Figure 3] This is a flowchart of step S200 of a resource scheduling method according to one exemplary embodiment of the present disclosure. [Figure 4] This is another flowchart of a resource scheduling method relating to one exemplary embodiment of the present disclosure. [Figure 5] This is a flowchart of step S250 of a resource scheduling method according to one exemplary embodiment of the present disclosure. [Figure 6] This is a flowchart of the decision-making process for a first scheduling policy relating to one exemplary embodiment of the present disclosure. [Figure 7] This is a schematic diagram illustrating the execution of a first scheduling policy relating to one exemplary embodiment of the present disclosure. [Figure 8] This is another schematic diagram illustrating the execution of the first scheduling policy relating to one exemplary embodiment of the present disclosure. [Figure 9] This is a schematic diagram illustrating the execution of a second scheduling policy relating to one exemplary embodiment of the present disclosure. [Figure 10] This is a structural block diagram of a resource scheduling device according to one exemplary embodiment of the present disclosure. [Modes for carrying out the invention]
[0013] Hereinafter, exemplary embodiments relating to this disclosure will be described in detail with reference to the drawings. Clearly, the embodiments described are only a part of the embodiments of this disclosure, not all embodiments, and this disclosure is not limited to the exemplary embodiments described herein.
[0014] Unless otherwise specifically stated, the relative arrangements of the components and steps, the formulas and numerical values described in these embodiments do not limit the scope of this disclosure.
[0015] Summary of the Application Intelligent connected vehicles are automobiles integrated with advanced sensors, controllers, actuators and communication systems, and can implement functions such as autonomous driving, Internet of Vehicles, and in-vehicle infotainment. They collect and analyze large volumes of data to achieve a safe, efficient, comfortable and entertaining driving experience.
[0016] Intelligent connected vehicles include abundant functions such as driving and entertainment, and to implement these functions, a plurality of operating systems, for example, an intelligent driving system and a cockpit system, are running on the intelligent connected vehicles.
[0017] Among them, the intelligent driving system can implement functions related to autonomous driving and safety of a vehicle (for example, a motor vehicle). The functions of the intelligent driving system include, but are not limited to, environment sensing, path planning and / or decision-making control, thereby implementing related autonomous driving functions of the vehicle such as autonomous driving in urban areas / on expressways, navigate on autopilot (NOA) in urban areas / on expressways, valet parking, automatic parking, adaptive cruise control (ACC), lane centering control (LCC), and also implementing related functions of autonomous emergency braking (AEB).
[0018] The cockpit system can implement functions related to the driving and riding experience inside the vehicle, and human-computer interaction. The functions of the cockpit system include, but are not limited to, infotainment, meter display, and / or human-computer interaction.
[0019] In some implementations, the intelligent driving system and the cockpit system can each operate on different chip systems. For example, an intelligent connected vehicle may include two chip systems, where one chip system executes the intelligent driving system, and the other chip system executes the cockpit system. Each operating system operates independently and can individually schedule resources provided by the chip system on which it is located, that is, the system scheduling of the intelligent driving system and the cockpit system are isolated from each other, and the respective operating systems cannot share resources with each other, that is, resource scheduling between the systems cannot be implemented, which results in unreasonable resource allocation and low utilization rate. For example, when the intelligent driving system requires more resources to perform complex path planning, the cockpit system may just have idle resources, however, these idle resources cannot be utilized by the intelligent driving system, which reduces the resource utilization rate.
[0020] Exemplary System Embodiments of the present disclosure provide a cockpit-driving integrated system.
[0021] Figure 1 is a software and hardware architecture diagram of a cockpit-driving integrated system according to one exemplary embodiment of the present disclosure.
[0022] In the embodiments of the present disclosure, the cockpit-driving integrated system can be applied to vehicles. Here, the vehicles may include land vehicles, aerial vehicles, surface water vehicles, underwater vehicles, space vehicles and the like.
[0023] Land vehicles may include, for example, automobiles (e.g., intelligent connected vehicles, trolleybuses, buses, etc.), railway vehicles (e.g., trains, subways, trams, etc.), various mobile robots (e.g., service robots, transport robots, automated guided vehicles (AGVs), unmanned ground vehicles (UGVs), bionic robots (e.g., robot dogs), etc.), various construction machinery, and other mobile land devices. Air vehicles may include, for example, airplanes, low-altitude aircraft, high-altitude aircraft, flying cars, and drones. Water vehicles may include ships, motorboats, etc. Underwater vehicles may include submarines, diving equipment, and underwater detectors, etc. Space vehicles may include spacecraft, spacecraft, and satellites, etc.
[0024] In some examples, a vehicle may be a vehicle with a driver, i.e., a manned vehicle, or it may be a vehicle without a driver, i.e., an unmanned vehicle.
[0025] In the embodiments of this disclosure, the cockpit-driving integrated system is an integrated system that combines a cockpit system and an intelligent driving system into a single unit. Compared to separate cockpit systems and intelligent driving systems, the cockpit-driving integrated system offers a higher degree of integration and intelligence, which is advantageous in improving the driving safety, comfort, and driving / riding experience of a mobile vehicle. In one embodiment, the cockpit-driving integrated system may include a hardware layer, a resource management layer, and a system layer.
[0026] Of these, the hardware layer includes the hardware portion of the cockpit driving integrated system, and for example, includes an integrated circuit, which may be a chip system, for example, a system on a chip (SoC) 110. Exemplarily, the SoC 110 may include one or more central processing units (CPUs) and / or one or more graphics processing units (GPUs) and / or one or more neural processing units (NPUs), digital signal processors (DSPs), and image signal processors (ISPs). These processors can be used to provide resources to the system layer, for example, to provide computing resources to the cockpit system and the intelligent driving system to support the implementation of the functions of the cockpit system and the intelligent driving system.
[0027] In embodiments of the present invention, "resources" is a general term for various hardware used in an integrated circuit (e.g., SoC) to perform tasks, run programs, and provide computing power. Exemplary examples of resources may include processor resources (e.g., CPU resources, GPU resources), memory resources, storage resources, network resources, input / output (I / O) resources, and the like.
[0028] In one implementation, each processor (e.g., CPU, GPU, NPU, etc.) may be an independent device. That is, different processors may be located on different chips. Alternatively, at least one processor may be integrated on one chip, and at least one other processor may be integrated on another chip, and the embodiments of this disclosure are not limited thereto.
[0029] The hardware layer may include other hardware in addition to the SoC110, such as memory 120 and communication interface 130.
[0030] Of these, memory 120 can be used for system files, application files, and data files (e.g., cache data, high-precision map data) of the intelligent driving system and cockpit system. Exemplaryly, memory 120 may include volatile memory such as dynamic random access memory (DRAM) and static random access memory (SRAM), or it may include non-volatile memory such as read-only memory (ROM) and flash memory (NVM). In some examples, memory 120 may be a discrete device or integrated with at least some devices of a chip system (e.g., at least some processors), for example, integrated into a single SoC. Some memory 120 (e.g., cache) may be integrated inside the processor and become part of the processor.
[0031] In the embodiments of this disclosure, the memory 120 further stores computer program instructions, and when the computer program instructions are executed by the processor, the chip system is made to execute the steps of the resource scheduling method according to the embodiments of this disclosure.
[0032] The communication interface 130 is used for communication between each piece of hardware in the hardware layer, and between the hardware layer and other devices such as vehicle sensors, display screens, microphones, and speakers. Exemplary examples of communication interfaces 130 may include, for example, a controller area network (CAN) interface, an inter-integrated circuit (I2C) interface, an inter-integrated circuit sound (I2S) interface, a pulse code modulation (PCM) interface, a digital video port (DVP), a peripheral component interconnect express (PCI-E) interface, a serial peripheral interface bus (SPI), a universal asynchronous receiver / transmitter (UART) interface, a mobile industry processor interface (MIPI), a general-purpose input / output (GPIO) interface, a local interconnect network (LIN) interface, a FlexRay interface, a media-oriented systems transport (MOST) interface, an Ethernet interface, and the like. In one implementation, the communication interface 130 can be integrated into the SoC and become part of the SoC in order to improve the system's integration density.
[0033] The system layer may include multiple operating systems, for example, operating system 1, operating system 2, ... operating system N. Different operating systems can be used to implement different functions of the cockpit driving integrated system. Multiple operating systems can run within an integrated circuit; in other words, multiple operating systems run within a chip system.
[0034] Here, the multiple operating systems may be operating systems built on the same system kernel. This is advantageous for exchanging and sharing data between multiple operating systems, and naturally, to facilitate the selection of a more appropriate operating system depending on the functions to be actually implemented, the multiple operating systems may be operating systems built on different system kernels. Here, the system kernel may be, for example, a microkernel, monolithic kernel, hybrid kernel, or exokernel. Exemplarily, let the system layer include operating system 1 and operating system 2, and operating system 1 and operating system 2 may be the same or different operating systems from among, for example, a real-time operating system (RTOS), a Unix real-time operating system (quick unix, QUX), a Linux operating system, or an Android operating system. For example, operating system 1 is a Linux operating system, and operating system 2 is an Android operating system.
[0035] In one implementation, operating system 1 may be a cockpit system for realizing functions related to the driving and riding experience and human-computer interaction inside the vehicle. Operating system 2 may be an intelligent driving system for realizing functions related to the autonomous driving and safety of the vehicle. Alternatively, operating system 1 may be an intelligent driving system and operating system 2 may be a cockpit system; that is, if the system layer includes multiple operating systems, the embodiments of this disclosure are not limited to the implementation of each operating system.
[0036] In one implementation, at least two operating systems may have an interconnection interface, such as an inter-process communication (IPC) interface. Thus, different operating systems can communicate via the IPC interface to send and receive messages and share information. Exemplarily, different operating systems can share load information, temperature information, computing power requests, etc., via the IPC interface, and the embodiments of this disclosure are not specifically limited thereto.
[0037] In one implementation, at least one operating system may include multiple subsystems. Taking a cockpit system as an example, the cockpit system may include multiple subsystems, and the number and function of subsystems in the cockpit system can be determined based on the configuration of the cockpit. Exemplarily, taking as an example that the cockpit includes an instrument panel, a central control screen, and a passenger-side screen, the cockpit system may include, for example, an instrument panel subsystem, a central control screen subsystem, and a passenger-side screen subsystem. The instrument panel subsystem can be used to display important vehicle information on the vehicle's instrument panel, such as information such as speed, rotational speed, fuel consumption, power consumption, temperature, warning information, navigation instructions, driver assistance system status, and vehicle malfunction warnings. The central control screen subsystem can be used to provide vehicle infotainment functions such as audio playback, video playback, navigation system, Bluetooth® connectivity, and smartphone mirroring; vehicle installation functions such as adjustment of air conditioning, seats, lights, driving modes, and power output configuration; vehicle driving assistance functions such as displaying rear view images, 360-degree panoramic images, and relevant information obtained from other operating systems (e.g., intelligent driving systems); and further, to provide vehicle human-computer interaction functions such as voice control and gesture control. The passenger-side screen subsystem can be used to provide entertainment and information services to the passenger, for example, by playing music and videos, providing game content, and displaying navigation information.
[0038] In some cases, the cockpit system can allocate acquired resources to each subsystem and schedule resources for each subsystem based on the task execution status, load, and / or resource requests of each subsystem. For example, if the cockpit system detects that the driver is viewing navigation information using the instrument panel, it can allocate more resources to the instrument panel subsystem. Similarly, if the cockpit system detects that the front passenger is watching video using the front passenger screen, it can allocate more resources to the front passenger subsystem.
[0039] The resource management layer is an intermediate layer located between the hardware layer and the system layer. The resource management layer may be implemented by software, by hardware, or by a combination of software and hardware. The resource management layer is used to allocate and use the resources of the hardware layer to the system layer. For example, some resources may be allocated to operating system 1 for use, and other resources may be allocated to operating system 2 for use.
[0040] In one implementation, at least a portion of the hardware layer resources may be allocated to and used by operating system 1 or by operating system 2. In other words, at least a portion of the hardware layer resources may be resources shared by operating system 1 and operating system 2, and these shared resources can be scheduled between operating system 1 and operating system 2. For the sake of explanation, the resources shared by operating system 1 and operating system 2 may hereafter be referred to as the first resources.
[0041] In one example, the first resource may include all resources in the hardware layer, such as all CPU resources, all GPU resources, all NPU resources, and all I / O resources. In other words, all resources in the hardware layer are resources shared by operating system 1 and operating system 2. Taking CPU resources in an SoC as an example, and assuming the CPU has 8 cores, these 8 cores may be allocated to operating system 1 or to operating system 2.
[0042] In another example, the first resource may include some resources in the hardware layer, such as some CPU resources, some GPU resources, some NPU resources, and some I / O resources. That is, some resources in the hardware layer are not shared resources and are allocated to operating system 1 only, or to operating system 2 only, while other resources are shared resources and may be allocated to operating system 1 or to operating system 2.
[0043] Let's take CPU resources in an SoC as an example, and assume that the CPU includes eight cores, for example, cores 1 through 8. Of these, cores 1 and 2 are unshared resources; core 1 can be allocated to operating system 1, core 2 can be allocated to operating system 2, and cores 3 through 8 are shared resources, i.e., the first resources, which may be allocated to operating system 1 or operating system 2. Here, the first resources may be fixedly set or dynamically adjusted; for example, cores 1 through 4 may be adjusted to be unshared resources, and cores 5 through 8 may be adjusted to be shared resources.
[0044] Alternatively, 20% of the CPU resources are unshared resources, of which 10% are allocated to operating system 1 only, and another 10% are allocated to operating system 2 only. The remaining 80% of the CPU resources are shared resources, i.e., the first resources, and may be allocated to operating system 1 or operating system 2. Here, the 80% of the CPU resources may be virtualized CPU resources, that is, these 80% of the CPU resources do not correspond to any or any core of the CPU. Of course, the 80% of the CPU resources may also be resources provided by designated cores of the CPU, for example, resources provided by cores 3 through 8.
[0045] In one implementation, the resource management layer can virtualize the resources of the hardware layer and schedule the virtualized resources for use by the system layer. In other words, the resource management layer can establish and manage multiple virtual machine instances based on the resources of the hardware layer, allocate the hardware layer resources to each virtual machine instance, and multiple operating systems can run on different virtual machine instances.
[0046] For example, the resource management layer may include, for instance, a virtual machine manager hypervisor210, which manages hardware layer resources such as processor, memory, and network resources, and can allocate these resources to multiple virtual machine instances. Thus, the operating systems running on each virtual machine instance can share these resources and operate in isolated virtual environments, ensuring that the operating systems do not interfere with each other, thereby improving the safety and reliability of the cockpit driving integration system. Furthermore, hypervisor210 can coordinate the requests of each operating system, map these requests to the physical hardware in the hardware layer, and return the processing results to the corresponding operating system. Additionally, hypervisor210 can schedule resources among virtual machine instances to ensure highly efficient resource utilization.
[0047] The structures shown in the embodiments of this disclosure do not specifically limit the cockpit driving integration system. In other embodiments of the present invention, the cockpit driving integration system may include more or fewer components than those shown in Figure 1, or may be a combination of components, a division of components, or have a different component configuration. The components in Figure 1 may be realized by hardware, software, or a combination of software and hardware.
[0048] Exemplary Method Embodiments of this disclosure provide a resource scheduling method which may be implemented in the cockpit driving integrated system shown in Figure 1, or in other software and hardware systems. For example, the method may be used in an integrated circuit which runs at least a first operating system and a second operating system.
[0049] In each of the following embodiments, a first operating system and a second operating system are referred to. Here, the first operating system and the second operating system may be any two systems in the cockpit driving integrated system. For example, the first operating system may be the cockpit system and the second operating system may be the intelligent driving system, or the first operating system may be the intelligent driving system and the second operating system may be the cockpit system.
[0050] Accordingly, the first operating system, the second operating system, and / or the resource management layer may invoke the processor to perform the steps of each method relating to the embodiments of the Disclosure. For example, the steps of some methods may be performed by the first operating system invoking the processor, some steps may be performed by the second operating system invoking the processor, and some steps may be performed by the resource management layer invoking the processor, and the embodiments of the Disclosure can be designed flexibly when implemented, and the embodiments of the Disclosure are not limited thereto.
[0051] Here, the invocation of a processor by a first operating system, a second operating system, and / or a resource management layer to perform a step of each method according to an embodiment of the disclosure may be described as the first operating system, a second operating system, and / or a resource management layer performing a step of each method according to an embodiment of the disclosure.
[0052] Figure 2 is a flowchart of a resource scheduling method according to one exemplary embodiment of the present disclosure.
[0053] As shown in Figure 2, in some embodiments, the resource scheduling method may include steps S100 and S200.
[0054] In step S100, the first user scene of the first operating system is determined.
[0055] Here, the first user scene may be the current user scene of the first operating system.
[0056] In the embodiments of this disclosure, there may be various methods for setting up the user scene.
[0057] In the first implementation, a user scene can be set based on the state of the vehicle, where the state of the vehicle may include, for example, a driving state, a parked state, etc. Exemplary, a user scene includes, but is not limited to, at least one of the following: Driving scene: The state in which a vehicle is in motion. Parking scene: A vehicle is in a parked state. Intelligent Driving Scene: The vehicle is in motion and the intelligent driving function is turned on. Manual driving scenario: The vehicle is in motion and the intelligent driving function is turned off.
[0058] In the second implementation, user scenes can be divided based on the system state of the operating system, where the system state may include, for example, standby state, screen on state, screen off state, active state, charging state, etc. Exemplarily, a user scene includes, but is not limited to, at least one of the following: Standby scene: The operating system is in a standby state. Screen-on scene: The operating system is in a standby state with the screen on. Screen off scene: The operating system is in standby mode + screen off state. Active scene: When the operating system is playing music, playing videos, or running game applications.
[0059] In the third implementation, user scenes can be divided based on the input state of the operating system, where the input state may include, for example, a touch input state, a voice input state, a visual input state, etc. Exemplarily, a user scene includes, but is not limited to, at least one of the following: Touch scene: When a user performs a touch operation on the operating system's user interface. Audio scene: The operating system is performing speech recognition. Visual scene: The operating system is performing visual recognition.
[0060] In the fourth implementation, user scenes can be divided based on the operating system usage state. For example, an operating system user scene includes, but is not limited to, at least one of the following: Game scene: The operating system is running the game application. Video scene: The operating system is running a video application. Music scene: The operating system is running a music application. Navigation scene: The operating system is running a map application. Download scene: An operating system application downloading content.
[0061] In several other implementations, the methods for dividing user scenes described above can be used in combination to create a richer range of user scenes. For example, by combining the division methods in the first and fourth implementations described above, the user scenes can further include intelligent driving + video scenes, parking + game scenes, and so on.
[0062] In several other implementations, user scenes can be further subdivided to obtain more user scenarios. For example, an intelligent driving scene can be further subdivided into intelligent driving scenes on highways, intelligent driving scenes in urban areas, automatic parking scenes, and so on.
[0063] In several other implementations, multiple user scenes can be integrated to simplify the number of user scenes. For example, touch scenes, audio scenes, and visual scenes can be integrated into an interaction scene; intelligent driving scenes and manual driving scenes can be integrated into a driving scene; and game scenes, video scenes, music scenes, navigation scenes, and download scenes can be integrated into an active scene.
[0064] The embodiments of this disclosure do not specifically limit the method of dividing user scenes.
[0065] The above provides an illustrative explanation of how to configure user scenes, but since different operating systems have different functionalities, the user scenes included in different operating systems may differ. Furthermore, an operating system may include more user scenes, fewer user scenes, or divide user scenes using other methods. Naturally, users can customize the configuration of user scenes, and the operating system can update user scenes using methods such as over-the-air (OTA) downloads, thereby continuously enriching the user scenes. None of these actions exceed the scope of protection of the embodiments of this disclosure.
[0066] In one embodiment, step S100 allows the first operating system to invoke and execute the processor.
[0067] The first operating system determining the first user scene can be achieved in various forms.
[0068] In one implementation, application information running on a first operating system can be obtained, and a first user scene can be determined based on the obtained application information.
[0069] Here, the application information may include the name of the application running on the first operating system, the package name, and / or the process name. In this way, the first operating system can determine which application is running based on the acquired application information, and further determine the first user scene. For example, the application name for WeChat® may be "WeChat", the package name may be "com.tencent.xin", and the process name may be "WeChat.exe".
[0070] For example, by obtaining application information for an intelligent driving application, it can be determined that the intelligent driving application is running, and then it can be determined that the first usage scenario is an intelligent driving scenario.
[0071] Generally, a running application may include applications running in the foreground and applications running in the background. Here, an application running in the foreground may also be called a foreground application, where "running in the foreground" means that the application is running and its content is displayed on the screen. For example, a foreground application may specifically include a foreground focus application and a foreground focus-out application, of which a foreground focus application includes an application whose content is displayed on the screen and is being interacted with by the user, and a foreground focus-out application includes an application that is displayed on the screen but is not being interacted with by the user. Accordingly, an application running in the background may also be called a background application, where "running in the background" means that the application is running but its content is not displayed on the screen.
[0072] When an application is running in the foreground, it generally means that the application is currently being used by the user. For example, if a user is watching a video (e.g., a video scene), the video application will run in the foreground and be able to play video content, etc. When an application is running in the background, it generally means that the application is not currently being used by the user. For example, if a user is not watching a video, the video application can be switched to the background, and if a user is not using navigation, the map application can be switched to the background.
[0073] As can be seen from the above, the foreground application is useful in accurately determining the first user scene of the first operating system. Thus, in one implementation, foreground application information in the first operating system can be obtained, and the first user scene can be determined based on the obtained foreground application information.
[0074] Furthermore, the first operating system can run multiple applications in the foreground. For example, the first operating system can simultaneously display the interfaces of multiple applications on the display screen using split screens, card windows, or floating windows, for instance, displaying the interface of a map application and the interface of a music application simultaneously. In this case, the foreground application information acquired by the first operating system can include application information for multiple applications, for example, including application information for a map application and application information for a music application. In this case, the first user scene determined by the first operating system can include multiple scenes, for example, the first user scene can include a navigation scene and a music scene.
[0075] In one implementation, a first user scene can be determined based on the state of the vehicle. For example, if the vehicle is in a driving state, for instance, if the vehicle is operating in D shift, the first user scene can be determined to be a driving scene. If the vehicle is in a parked state, for instance, if the vehicle is operating in P shift, the first user scene can be determined to be a parking scene.
[0076] In one implementation, a first user scene can be determined based on the system state of a first operating system. For example, if the first operating system is in a standby state, the first user scene can be determined to be a standby scene. If the first operating system is in a screen-on state, the first user scene can be determined to be a screen-on scene.
[0077] In one implementation, a first user scene can be determined based on the input state of a first operating system. For example, if the first operating system is in a touch input state, for instance, responding to a user's touch operation, the first user scene can be determined to be a touch scene. If the first operating system is in a voice input state, for instance, receiving or recognizing a user's voice, the first user scene can be determined to be a voice scene.
[0078] In one implementation, each of the above-described methods for determining the first user scene can be used in combination, and in this way, the first user scene can be further subdivided in order to perform resource scheduling at a finer granularity. For example, the first user scene can be determined based on the state and usage state of the vehicle. For example, if the vehicle is in motion and the intelligent driving application is running, the first user scene can be determined to be an intelligent driving scene. Alternatively, for example, if the vehicle is in motion and the intelligent driving application is not running, the first user scene can be determined to be a manual driving scene.
[0079] The above provides an illustrative explanation of the first method of determining the user scene. The above example is not an exhaustive representation of all methods for determining the first user scene, but rather only a part of them. In actual use, other methods for determining the first user scene can be set based on the usage scenarios, methods for dividing user scenes, and operating system functions of the embodiments of this disclosure. None of these methods exceed the scope of protection of the embodiments of this disclosure.
[0080] In step S200, a resource scheduling policy is determined for scheduling a first resource between a first operating system and a second operating system, based on the first user scene.
[0081] In one embodiment, step S200 allows the first operating system to invoke and execute the processor.
[0082] Here, the first resource includes resources shared by the first operating system and the second operating system.
[0083] Generally, during the initial operation of a chip system, for example, when the chip system is powered on, the resource management layer can initially allocate the first resource to the first operating system and / or the second operating system. During past operation of the chip system, scheduling may have been performed on the first resource, and the allocation state of the first resource may have been changed. Therefore, before executing step S200, there may be multiple allocation states for the first resource, which are illustrated below.
[0084] In the first allocation state, all of the first resources are allocated to the first operating system and / or the second operating system. For example, 50% of the first resources are allocated to the first operating system and another 50% are allocated to the second operating system. Or, 40% of the first resources are allocated to the first operating system and another 60% are allocated to the second operating system. Or, all of the first resources are allocated to the first operating system and the second operating system does not acquire any of the first resources. Or, all of the first resources are allocated to the second operating system and the first operating system does not acquire any of the first resources.
[0085] In the second allocation state, only a portion of the first resources are allocated to the first and / or second operating systems. For example, 40% of the first resources are allocated to the first operating system, another 50% are allocated to the second operating system, and the remaining 10% of the first resources are unallocated.
[0086] There may be multiple resource scheduling policies based on the various allocation states of the first resource. For example, scheduling the first resource between the cockpit system and the intelligent driving system may include the following resource scheduling policies: Policy 1: Allocate at least a portion of the primary resources already allocated to the cockpit system to the intelligent driving system. Policy 2: Allocate at least a portion of the first resource already allocated to the intelligent driving system to the cockpit system. Policy 3: Schedule at least some of the unallocated primary resources to the cockpit system. Policy 4: Schedule at least some of the unallocated primary resources to the intelligent driving system.
[0087] Since the resource requirements for the cockpit system and the intelligent driving system differ in different user scenarios, a resource scheduling policy can be determined based on the first user scenario, and the first resources can be scheduled between the cockpit system and the intelligent driving system using that resource scheduling policy, thereby satisfying the resource requirements for the cockpit system and the intelligent driving system in different user scenarios.
[0088] The following provides an illustrative example of how resource scheduling policies are determined based on the first user scenario.
[0089] When the first user scenario is a driving scenario, an intelligent driving scenario, or a manual driving scenario, it is necessary to prioritize ensuring that the intelligent driving system acquires more resources in order to ensure that functions related to intelligent driving or vehicle safety, such as autonomous driving, constant speed driving / inter-vehicle distance control, lane centering (lane keeping) assist, and automatic emergency braking, can be reliably provided. In this case, the first resources can be scheduled from the cockpit system to the intelligent driving system. For this purpose, policy 1 and / or policy 4 can be selected.
[0090] If the first user scenario is a parking scenario, the intelligent driving system will turn off functions related to intelligent driving, enter a standby state or another state other than intelligent driving, and will have low resource demands. In this case, the first resource can be scheduled from the intelligent driving system to the cockpit system. For this reason, policy 2 and / or policy 3 can be selected.
[0091] The resource scheduling method according to an embodiment of the present invention provides a resource scheduling mechanism between systems, enabling inter-system resource scheduling between a first operating system and a second operating system based on a first user scene. This not only ensures that the resources acquired by each operating system meet the requirements of the first user scene, but also makes resource allocation more rational, improves resource utilization, and enhances the efficiency and stability of the operating system's operation.
[0092] Figure 3 is a flowchart of step S200 of a resource scheduling method according to one exemplary embodiment of the present disclosure.
[0093] As shown in Figure 3, in one embodiment, step S200 can be realized by the following steps S210 and S220.
[0094] In step S210, the first resource state of the first operating system is determined.
[0095] Here, step S210 allows the first operating system to call and execute the processor. After determining the first resource state, the first operating system can send the first resource state to the second operating system via the IPC interface.
[0096] In one implementation, the first resource state includes the load state of the first operating system, for example, the current load. In specific implementation, a performance management tool or load monitoring tool can be executed within the first operating system, and the load state of the first operating system can be monitored and obtained through the performance management tool or load monitoring tool. Exemplaryly, the load state of the first operating system may include the utilization rate of resources already acquired by the first operating system, for example, CPU utilization, GPU utilization, NPU utilization, etc.
[0097] In one implementation, the first resource state may include the load changes of the first operating system over a certain period in the future. Here, the certain period in the future may be, for example, a few seconds or a few minutes in the future, and the embodiments of this disclosure are not limited thereto. In specific implementation, in order to predict and obtain the load changes of the first operating system over a certain period in the future, the load changes of the first operating system can be estimated based on a first user scene.
[0098] Taking the example that the first operating system is a cockpit system and the first user scene is a video scene, when a video application plays a video at a specific resolution, frame rate, and bitrate, the resources consumed by the video application are predictable. Based on this, it is possible to predict the load changes of the cockpit system within a certain period in the future (for example, within the time remaining until video playback ends). Here, the resources consumed when playing videos at different resolutions, frame rates, and bitrates can be obtained by statistically analyzing historical data on the local or cloud side, and are not specifically limited here.
[0099] As an example, if the first operating system is an intelligent driving system and the first user scenario is an intelligent driving scenario, and the intelligent driving application performs an intelligent driving task on a highway, the resources consumed by the intelligent driving application are predictable. Based on this, it is possible to predict the load changes of the intelligent driving system within a certain period in the future (for example, within the time remaining until the current intelligent driving is completed). Here, the expected resource consumption for the intelligent driving task on the highway can be obtained by statistically analyzing historical data on the local or cloud side, and is not specifically limited here.
[0100] As an example, if the first operating system is an intelligent driving system and the first user scenario is an intelligent driving scenario, when an intelligent driving application performs an intelligent driving task in an urban area, the resources consumed by the intelligent driving application may fluctuate. For example, in sections / times with few vehicles and simple road conditions (e.g., elevated roads in urban areas / outside of morning and evening peak hours), the intelligent driving application will consume fewer resources, while in sections / times with many vehicles and complex road conditions (e.g., intersections / morning and evening peak hours), the intelligent driving application will consume more resources. Based on this, it is possible to estimate the time periods during which a vehicle will pass through each section based on the intelligent driving route, thereby predicting the resource consumption of the intelligent driving application at each time period, and further predicting the load changes of the intelligent driving system within a certain period in the future (e.g., within the time remaining until the current intelligent driving is completed). Here, the expected resource consumption for an intelligent driving task in an urban area can be obtained by statistically analyzing historical data on the local or cloud side, and is not specifically limited here.
[0101] In this way, based on the first resource state, the first operating system can determine the allocation of resources that are free and the proportion of resources that are used, thereby providing a basis for determining a resource scheduling policy.
[0102] In step S220, a resource scheduling policy is determined based on the first user scene and the first resource state.
[0103] Here, step S220 allows the first operating system to call the processor and execute.
[0104] For example, a resource scheduling policy may include scheduling a first resource from a first operating system to a second operating system, or scheduling a first resource from a second operating system to a first operating system.
[0105] Let's take the example of a system where the first operating system is the cockpit system and the second operating system is the intelligent driving system.
[0106] If the first user scene is a driving scene, it is necessary to prioritize the supply of resources to the second operating system. In this case, it is possible to determine whether or not resource scheduling needs to be performed based on the state of the first resource, and if resource scheduling needs to be performed, the resource scheduling policy may include scheduling the first resources from the first operating system to the second operating system.
[0107] If the first user scene is a parking scene, the second operating system can turn off functions related to intelligent driving and, in this case, determine whether resource scheduling can be performed based on the first resource state. If resource scheduling can be performed, the resource scheduling policy may include scheduling the first resource from the second operating system to the first operating system.
[0108] The method according to the embodiment of this disclosure determines a resource scheduling policy based on the user scene and resource state of the first operating system, and matches the resources acquired by the first operating system with resource requests in each user scene. This ensures that each application in the first operating system can operate efficiently and stably, thereby improving the reliability of the system.
[0109] Figure 4 is another flowchart of a resource scheduling method according to one exemplary embodiment of the present disclosure.
[0110] As shown in Figure 4, in one embodiment, the resource scheduling method may include steps S100, S210, S230, S240 and S250.
[0111] In step S100, the first user scene of the first operating system is determined.
[0112] In step S210, the first resource state of the first operating system is determined.
[0113] In step S230, the second user scene of the second operating system is determined.
[0114] The second user scene may be the current user scene of the second operating system.
[0115] Here, step S230 allows the second operating system to call the processor and execute. After determining the second user scene, the second operating system can send the second user scene to the first operating system via the IPC interface.
[0116] The determination of a second user scene by a second operating system can be achieved in various ways. For example, the second user scene can be determined by acquiring application information running in the second operating system; by acquiring foreground application information in the second operating system; by acquiring foreground application information; by determining the second user scene based on the state of the vehicle; by determining the system state of the second operating system; or by determining the input state of the second operating system.
[0117] Any aspects of the determination of the second user scene by the second operating system that are not elaborated upon here are implemented by referring to the form in which the first operating system determines the first user scene, and redundant explanations are omitted here.
[0118] Because the functions of the first and second operating systems differ and the applications they run are different, the first and second user scenes may or may not be the same at the same time or within the same time period.
[0119] Let's take the example of a system where the first operating system is the cockpit system and the second operating system is the intelligent driving system.
[0120] If the vehicle is in motion and a video is being played on the first operating system, the first user scene determined by the first operating system may be a video scene, and the second user scene determined by the second operating system may be a driving scene.
[0121] If the vehicle is parked and a game application is running on the first operating system, the first user scene determined by the first operating system may be a game scene, and the second user scene determined by the second operating system may be a parking scene.
[0122] In step S240, the second resource state of the second operating system is determined.
[0123] In one implementation, step S240 allows a second operating system to call and execute a processor. After determining a second resource state, the second operating system can send the second resource state to the first operating system via the IPC interface.
[0124] In one implementation, the second resource state includes the load state of the second operating system, for example, the current load. In specific implementation, a performance management tool or load monitoring tool can be run within the second operating system, and the load state of the second operating system can be monitored and obtained through the performance management tool or load monitoring tool. Exemplaryly, the load state of the second operating system may include, for example, the CPU usage, GPU usage, NPU usage, memory usage, magnetic device usage, network usage, etc.
[0125] In one implementation, the second resource state may include the load changes of the second operating system over a certain period in the future. Here, the certain period in the future may be, for example, a few seconds or a few minutes in the future, and the embodiments of this disclosure are not limited thereto. In specific implementation, in order to predict and obtain the load changes of the second operating system over a certain period in the future, the load changes of the second operating system can be estimated based on the second user scene of the second operating system.
[0126] Any aspects of the determination of the second resource state by the second operating system that are not elaborated upon here are implemented by referring to the form in which the first operating system determines the first resource state, and redundant explanations are omitted here.
[0127] In step S250, a resource scheduling policy is determined based on the first user scene, the second user scene, the first resource state, and the second resource state.
[0128] Here, step S250 may be performed by the first operating system calling the processor and executing, or by the second operating system calling the processor and executing, or by the first operating system and the second operating system calling the processor and executing in parallel, that is, the two operating systems can each determine one resource scheduling policy.
[0129] Figure 5 is a flowchart of step S250 of a resource scheduling method according to one exemplary embodiment of the present disclosure.
[0130] As shown in Figure 5, in one embodiment, step S250 may include the following steps S251 and S252.
[0131] In step S251, the priority of the first operating system is determined based on the first user scene, and the priority of the second operating system is determined based on the second user scene.
[0132] In one implementation, priority rules can be set for user scenes, and these priority rules may include the priority of the cockpit system and / or intelligent driving system in each user scene. Herein, the priority rules can be flexibly set by the engineer, and the embodiments of this disclosure are not limited thereto.
[0133] Taking the example that priority includes high priority and low priority, the priority rules may include, for example, the following:
[0134] When the intelligent driving system and / or cockpit system are in a driving scene, an intelligent driving scene, or a manual driving scene, the intelligent driving system is given high priority and the cockpit system is given low priority. In other words, the priority of the intelligent driving system is higher than the priority of the cockpit system.
[0135] If the Intelligent Driving System is in a parking scene and the Cockpit System is in a touch, audio, visual, game, video, music, or download scene, the Intelligent Driving System will be given lower priority and the Cockpit System will be given higher priority.
[0136] When the Intelligent Driving System is in a manual driving scene and the Cockpit System is in a navigation scene, the Intelligent Driving System will be given lower priority and the Cockpit System will be given higher priority.
[0137] If the cockpit system is in a standby scene or a screen-off scene, the cockpit system will be given low priority.
[0138] When the Intelligent Driving System is in standby mode or screen-off mode, the Intelligent Driving System will be given lower priority.
[0139] In this way, the priority of the first operating system and the priority of the second operating system can be determined based on the priority rules.
[0140] For example, if the first operating system is the cockpit system and the second operating system is the intelligent driving system, and the first user scene is a video scene and the second user scene is a driving scene, then the first operating system has a low priority and the second operating system has a high priority.
[0141] In one implementation, user scenes can be prioritized and sorted in order of priority, either from highest to lowest.
[0142] For example, the priority of user scenarios is, in descending order: The sequence may include: driving scene > interaction scene > active scene > parking scene > screen on scene > screen off scene.
[0143] Among these, driving scenes may include, for example, intelligent driving scenes and manual driving scenes; interaction scenes may include, for example, touch scenes, audio scenes and visual scenes; and active scenes may include, for example, game scenes, video scenes, music scenes, navigation scenes and download scenes.
[0144] Thus, based on the sorting described above, the priority of the first user scene can be set as the priority of the first operating system, and the priority of the second user scene can be set as the priority of the second operating system.
[0145] For example, if the first operating system is a cockpit system and the second operating system is an intelligent driving system, and the first user scene is a video scene and the second user scene is a driving scene, then the priority of the first user scene is lower than the priority of the second user scene, and accordingly, the priority of the first operating system is lower than the priority of the second operating system.
[0146] In one embodiment, priority scores can be pre-set for multiple user scenes.
[0147] Table 1 shows, as an example, the priority scores for user scenes.
[0148] [Table 1]
[0149] The priority scores mentioned above are for illustrative purposes only and do not specifically limit the embodiments of this disclosure. Those skilled in the art can adjust the priority scores for each user scene based on the embodiments of this disclosure, and none of these adjustments will exceed the scope of protection of the embodiments of this disclosure.
[0150] In this way, the priority relationship between the first operating system and the second operating system can be determined based on the priority scores of the first and second user scenes. In other words, the priority score of the first user scene is used as the priority score for the first operating system, and the priority score of the second user scene is used as the priority score for the second operating system.
[0151] For example, if the first operating system is a cockpit system and the second operating system is an intelligent driving system, and the first user scene is a video scene with a priority score of 50, and the second user scene is a driving scene with a priority score of 90, then the priority of the first user scene is lower than the priority of the second user scene, and accordingly, the priority of the first operating system is lower than the priority of the second operating system.
[0152] In step S252, a resource scheduling policy is determined based on the priority of the first operating system, the priority of the second operating system, the first resource state, and the second resource state.
[0153] In one implementation, step S252 can be implemented by the following steps S2521 to S2523.
[0154] In step S2521, it is determined whether the first operating system has free resources based on the first resource state, and it is also determined whether the second operating system has free resources based on the second resource state.
[0155] For example, the load state of a first operating system can be determined by setting a first load threshold. If the current load of the first operating system is greater than (or equal to) the first load threshold, it can be determined that the first operating system has no free resources. If the current load of the first operating system is less than the first load threshold, it can be determined that the first operating system has free resources.
[0156] For example, the first load threshold may include utilization thresholds for processor resources already acquired by the first operating system, such as a CPU utilization threshold, a GPU utilization threshold, an NPU utilization threshold, etc. The value of the first load threshold may be, for example, 80%, 85%, 88%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, etc. Here, the utilization thresholds corresponding to different processors may be the same or different.
[0157] For example, if the CPU usage threshold is 90%, then if the CPU usage is greater than 90%, it can be determined that the first operating system has no available CPU resources. If the CPU usage is less than 90%, it can be determined that the first operating system has available CPU resources.
[0158] Similarly, a second load threshold can be set for the second operating system, and the second load threshold and the first load threshold may be the same or different. If the current load of the second operating system is greater than (or equal to) the second load threshold, it can be determined that the second operating system has no free resources. If the current load of the second operating system is less than the second load threshold, it can be determined that the second operating system has free resources.
[0159] Any details regarding the determination of whether or not the second operating system has available resources are not specifically elaborated upon and will be carried out by referring to the method used to determine whether or not the first operating system has available resources; therefore, redundant explanations will be omitted here.
[0160] In step S2522, if the second operating system has no available resources and the priority of the first operating system is lower than the priority of the second operating system, a first resource scheduling policy is determined, which includes scheduling the first resources from the first operating system to the second operating system.
[0161] Here, the lack of available resources in the second operating system can include the following two cases:
[0162] One possible scenario is when neither the first nor the second operating system has any available resources. In this case, it is necessary to prioritize ensuring that the higher-priority operating system can operate stably. Therefore, if the priority of the first operating system is lower than that of the second operating system, the first operating system can schedule its resources to be transferred to the second operating system. This allows the second operating system to acquire more resources, avoiding a resource shortage and ensuring the stable operation of the second operating system.
[0163] Another scenario is when the first operating system has available resources, but the second operating system does not. In this case, if the priority of the first operating system is lower than that of the second operating system, the first operating system can schedule its resources to be transferred to the second operating system. This allows the second operating system to acquire more resources, avoiding resource shortages and ensuring the stable operation of the second operating system. It also allows the second operating system to utilize the available resources in the first operating system, thereby improving resource utilization.
[0164] In step S2523, if the first operating system has no available resources and the priority of the first operating system is higher than the priority of the second operating system, a second resource scheduling policy is determined, which includes scheduling the first resources from the second operating system to the first operating system.
[0165] Here, the lack of available resources in the first operating system can include the following two cases:
[0166] One possible scenario is when neither the first nor the second operating system has any available resources. In this case, it is essential to prioritize ensuring that the higher-priority operating system can operate stably. Therefore, if the priority of the first operating system is higher than that of the second operating system, the second operating system can schedule resources from the first operating system to the first operating system. This allows the first operating system to acquire more resources, avoids resource shortages, and ensures the stable operation of the first operating system.
[0167] Another scenario is when the second operating system has available resources, but the first operating system does not. In this case, if the priority of the first operating system is higher than that of the second operating system, the second operating system can schedule resources from the first operating system to the first operating system. This allows the first operating system to acquire more resources, avoids resource shortages, and ensures the stable operation of the first operating system. It also allows the first operating system to utilize any available resources in the second operating system, thereby improving resource utilization.
[0168] Figure 6 is a flowchart of the decision-making process for a first scheduling policy according to one exemplary embodiment of the present disclosure.
[0169] In one implementation, the first scheduling policy can be determined by the first operating system, which includes, for example, the following steps S310 to S360.
[0170] In step S310, the first resource state of the first operating system is determined.
[0171] Here, step S310 can be specifically implemented by referring to step S210, and a redundant explanation is omitted here. After determining the first resource state, the first operating system can send the first resource state to the second operating system via the IPC interface.
[0172] In step S320, the second resource state of the second operating system is determined.
[0173] For example, a first operating system can receive a second resource state transmitted from a second operating system via an IPC interface.
[0174] In step S330, it is determined whether the first operating system has available resources based on the first resource state.
[0175] Here, step S330 can be specifically implemented by referring to step S2521, and redundant explanations are omitted here.
[0176] After the first operating system determines whether or not there are free resources, it can jump to step S350.
[0177] In step S340, based on the second resource state, it is determined whether or not there are free resources in the second operating system.
[0178] Here, step S340 can be specifically implemented by referring to step S2521, and a redundant explanation is omitted here. If there are no free resources in the second operating system, the process can jump to step S350. If there are free resources in the second operating system, the process can jump to step S320.
[0179] In step S350, it is determined whether the priority of the first operating system is higher than the priority of the second operating system.
[0180] Here, if the priority of the first operating system is higher than the priority of the second operating system, the process can jump to step S320. That is, if the priority of the first operating system is higher than the priority of the second operating system, the first operating system will not determine the first resource scheduling policy, even if the second operating system has no free resources, and will not schedule the first resources from the higher-priority operating system to the lower-priority operating system.
[0181] If the priority of the first operating system is lower than the priority of the second operating system, the process can jump to step S360.
[0182] In step S360, the first resource scheduling policy is determined.
[0183] In other words, if the priority of the first operating system is lower than the priority of the second operating system, the first operating system can determine a first scheduling policy regardless of whether it has available resources, thereby scheduling the first resources from the lower-priority first operating system to the higher-priority second operating system.
[0184] Next, after determining the first resource scheduling policy, the resource scheduling can be completed by executing the first resource scheduling policy. The following provides an illustrative description of how the first resource scheduling policy can be executed.
[0185] In one implementation, the first resource scheduling policy can be executed by the first operating system and resource management layer calling the processor. As shown in Figure 6, the execution of the first resource scheduling policy may include steps S370 and S380.
[0186] In step S370, based on the resource request of the second operating system, the first operating system releases the first target resource from the first resource it has already acquired, and the first target resource matches the resource request of the second operating system.
[0187] Here, step S370 allows the first operating system to invoke and execute the processor.
[0188] In one example, if a second operating system has a resource request, the second operating system can send the resource request to the first operating system via the IPC interface. That is, the first operating system can receive the resource request from the second operating system via the IPC interface. Here, the resource request from the second operating system may include the type of resource requested by the second operating system and the amount of resources requested.
[0189] Among these, resource types may include, for example, CPU resources, GPU resources, NPU resources, and memory resources. Resource requests may include, for example, the amount of CPU resources requested, the amount of GPU resources requested, and the amount of NPU resources requested.
[0190] In one implementation, the second operating system can predict the resources it will need within a certain period in the future based on a second user scene in order to determine the resource requirements of the second operating system, and then send the resource requests of the second operating system to the first operating system.
[0191] Let's take the example of a second operating system being an intelligent driving system and a second user scene being an intelligent driving scene. Assume that the second operating system currently has 200 TOPS (Tera Operations Per Second, meaning 1 trillion operations per second) of NPU resources, the intelligent driving application requires 150 TOPS of NPU resources in elevated urban sections, and 250 TOPS of NPU resources in pedestrian-vehicle sections of urban areas. Then, when the vehicle travels from an elevated urban section to a pedestrian-vehicle section of urban areas, the second operating system will need to acquire at least an additional 50 TOPS of NPU resources to meet the computing power requirements of the intelligent driving application. In other words, within a certain period in the future, for example, within the time remaining until the current intelligent driving reaches its destination, the resource requirements of the second operating system will include at least 50 TOPS of NPU resources.
[0192] After obtaining a resource request from the second operating system, the first operating system can determine whether the free resources already acquired by the first operating system in the first resource (hereinafter referred to as the first free resources) match the resource request from the second operating system.
[0193] Exemplary, it can be determined whether the type of the first free resource matches the resource type requested by the second operating system. For example, if the type of the first free resource is the same as the resource type requested by the second operating system, it is shown that the type of the first free resource matches the resource type requested by the second operating system. If the type of the first free resource is different from the resource type requested by the second operating system, it is shown that the type of the first free resource does not match the resource type requested by the second operating system.
[0194] For example, if the first free resource includes a CPU resource and the second operating system requests an NPU resource, the type of the first free resource does not match the resource type requested by the second operating system. If the first free resource includes a CPU resource and the second operating system requests a CPU resource, the type of the first free resource does match the resource type requested by the second operating system.
[0195] For example, it is also possible to determine whether the resource amount of the first free resource matches the resource amount requested by the second operating system. For instance, if the resource amount of the first free resource is greater than or equal to the resource request for the same type of resource by the second operating system, it is shown that the resource amount of the first free resource matches the resource request for the second operating system. If the resource amount of the first free resource is less than the resource request for the same type of resource by the second operating system, it is shown that the resource amount of the first free resource does not match the resource request for the second operating system.
[0196] For example, if the first free resource includes 200 TOPS of NPU resources and the second operating system requests 100 TOPS of NPU resources, the amount of resources in the first free resource matches the resource request of the second operating system. However, if the first free resource includes 200 TOPS of NPU resources and the second operating system requests 300 TOPS of NPU resources, the amount of resources in the first free resource does not match the resource request of the second operating system.
[0197] Based on the above, if the resource type and resource quantity of the first free resource match the resource type and resource quantity requested by the second operating system, it can be determined that the first free resource matches the resource request of the second operating system; otherwise, it can be determined that the first free resource does not match the resource request of the second operating system.
[0198] In one example, a resource request from the second operating system may include the resource request time of the second operating system. The resource request time may be a predetermined time or the predicted remaining duration of the current user scene of the second operating system.
[0199] For example, it is also possible to determine whether the remaining free time of the first free resource matches the resource request time of the second operating system. If the remaining free time of the first free resource is greater than or equal to the resource request time of the second operating system, it is shown that the remaining free time of the first free resource matches the resource request time of the second operating system. If the remaining free time of the first free resource is less than the resource request time of the second operating system, it is shown that the remaining free time of the first free resource does not match the resource request time of the second operating system.
[0200] In this case, if the resource type, resource amount, and remaining free time of the first free resource match the resource type, resource request amount, and resource request time requested by the second operating system, it can be determined that the first free resource matches the resource request of the second operating system; otherwise, it can be determined that the first free resource does not match the resource request of the second operating system.
[0201] Next, if the first free resource matches the resource request of the second operating system, the first operating system releases the first target resource from the first free resource, leaving the first target resource unallocated. That is, the released first target resource belongs neither to the first operating system nor to the second operating system. In some implementations, the release of the first target resource may be performed by the resource management layer, and this is not particularly limited here.
[0202] For example, as shown in Figure 7, if the first free resource includes 200 TOPS of NPU resources and the second operating system requests 100 TOPS of NPU resources, the first operating system can release 100 TOPS of NPU resources from the first free resource. Naturally, the first operating system may release more NPU resources from the first free resource, for example, 150 TOPS of NPU resources, and in this way, more NPU resources can subsequently be scheduled for the second operating system.
[0203] Furthermore, if the available resources of the first operating system do not match the resource request of the second operating system, the first operating system can perform internal resource scheduling to acquire an available first target resource and release the first target resource.
[0204] For example, a first operating system can perform internal resource scheduling based on a first user scene. As shown in Figure 8, let's take the example where the first operating system is a cockpit system and the first user scene is a video scene. If the first free resources include 10% of the total CPU resources and the second operating system requests 20% of the total CPU resources, the first operating system can reduce the CPU resource utilization by the video application by reducing parameters such as the video bitrate, resolution, and frame rate, thereby freeing up more CPU resources. For example, it can free up another 10% of the total CPU resources, and in this way, free up 20% of the total CPU resources in total, which can then be designated as the first target resource and released.
[0205] Similarly, if the first user scene is a game scene, the game's image quality can be reduced by, for example, lowering the game's frame rate, lowering the resolution, lowering the game's graphics quality, or reducing or turning off motion in the game. This reduces the resources occupied by the game application, freeing up more resources. If the first user scene is a download scene, the application's download speed can be reduced or the download task can be paused. This reduces the resources occupied by the download task, freeing up more resources.
[0206] Furthermore, the first operating system can manage and control background tasks, for example, by turning off background tasks or reducing the resources allocated to them to free up more resources. Turning off background tasks may include, for example, turning off background applications that are not used by the user for a long time, and turning off background update tasks and download tasks.
[0207] The embodiments of this disclosure do not specifically limit the forms in which the operating system frees up more resources.
[0208] In step S380, the first target resource is scheduled to the second operating system.
[0209] In one implementation, step S380 can be executed by the resource management layer calling the processor.
[0210] In one example, after releasing a first target resource, the first operating system may send a first notification message to the resource management layer, instructing the resource management layer to execute a first resource scheduling policy. Thus, upon receiving the first notification message, the resource management layer can schedule the first target resource to the second operating system. Here, the first notification message may include, for example, a first resource scheduling policy and the object to be scheduled (e.g., the first target resource), and is used to instruct the resource management layer to schedule the first target resource to the second operating system.
[0211] For example, the first operating system may send a first notification message to a virtual machine manager hypervisor, and after receiving the first notification message, the virtual machine manager hypervisor schedules the released first target resource to the second operating system.
[0212] Exemplary, as shown in Figure 7, the first operating system may, after releasing 100 TOPS of NPU resources, send a first notification message to the virtual machine manager hypervisor, instructing the virtual machine manager hypervisor to schedule 100 TOPS of NPU resources to the second operating system. The virtual machine manager hypervisor receives the first notification message, performs resource scheduling, and schedules 100 TOPS of NPU resources to the second operating system.
[0213] For example, as shown in Figure 8, the first operating system may, after releasing 20% of its total CPU resources, send a first notification message to the virtual machine manager hypervisor, instructing the virtual machine manager hypervisor to schedule 20% of its total CPU resources to the second operating system. The virtual machine manager hypervisor receives the first notification message, performs resource scheduling, and schedules 20% of its total CPU resources to the second operating system.
[0214] In one implementation, the second scheduling policy can be determined by the second operating system, and the second resource scheduling policy can be executed by the second operating system and the resource management layer. The determination of the second scheduling policy by the second operating system can be achieved by reference to the determination of the first scheduling policy by the first operating system, and the execution of the second scheduling policy by the second operating system and the resource management layer can be achieved by reference to the execution of the first scheduling policy by the first operating system and the resource management layer. Hereinafter, redundant explanations will be omitted.
[0215] As an example, as shown in Figure 9, the first operating system is a cockpit system, the first user scene is a voice scene, the second operating system is an intelligent driving system, and the second user scene is a parking scene.
[0216] Generally, intelligent driving tasks require execution on a second operating system, and therefore, by default, a large or all of the NPU resources are allocated to the second operating system, while the first operating system receives only a small number of NPU resources, or no NPU resources at all. This means that when the second user scene is a parking scene, the second operating system will have a large amount of available NPU resources.
[0217] At the same time, if the first user scene is a speech scene, the speech recognition application of the first operating system may need to understand the speech content input by the user through a local large-scale model and then provide a corresponding response. This means that the first operating system will need a large amount of NPU resources to satisfy the computational power requirements of the local large-scale model. In this case, the available NPU resources in the second operating system (i.e., the second available resources) can be freed up and scheduled for the first operating system.
[0218] For example, if the second free resource includes 400 TOPS of NPU resources and the first operating system's resource request is for 300 TOPS of NPU resources, the second operating system can release 300 TOPS of NPU resources from the second free resource. Naturally, the second operating system may release more NPU resources from the second free resource, for example, 350 TOPS of NPU resources, or 400 TOPS of NPU resources, i.e., all free NPU resources.
[0219] After the second operating system releases 300 TOPS of NPU resources, it can send a second notification message to the virtual machine manager hypervisor, instructing the virtual machine manager hypervisor to schedule 300 TOPS of NPU resources to the first operating system. The virtual machine manager hypervisor receives the second notification message, performs resource scheduling, and schedules 300 TOPS of NPU resources to the first operating system.
[0220] As can be seen from the above technical proposals, the method according to the embodiment of this disclosure can perform resource scheduling between different operating systems based on resource requirements in different user scenes of each operating system. In other words, it realizes inter-system resource scheduling based on user scenes, satisfies the resource requirements of each operating system in different user scenes, improves the execution stability of the operating systems, and further makes resource allocation more rational and improves resource utilization.
[0221] Exemplary device The resource scheduling method according to the embodiments of this disclosure has been described above. The cockpit driving integrated system (for example, the first operating system, the second operating system, and the resource management layer in the cockpit driving integrated system) includes software modules corresponding to the execution of each function in order to realize the above functions.
[0222] Naturally, those skilled in the art will readily conceive that the Disclosure can be implemented in hardware or in combination of hardware and computer software, in accordance with the steps of the resource scheduling methods described in each embodiment of the Disclosure. Whether a function is performed by hardware or by software driving hardware will depend on the specific application and design constraints of the proposed technology. Those skilled in the art may implement the functions described using different methods for each specific application, but such implementations should not be considered beyond the scope of the Disclosure.
[0223] Figure 10 is a structural block diagram of a resource scheduling device according to one exemplary embodiment of the present disclosure.
[0224] Embodiments of this disclosure provide resource scheduling devices used in integrated circuits running at least a first operating system and a second operating system.
[0225] In one embodiment, as shown in Figure 10, the resource scheduling device for implementing each step of the resource scheduling method may include a scene recognition unit 710, a scheduling management unit 720, and a scheduling execution unit 730. Of these, the functions of each unit can be implemented in a first operating system and / or a second operating system. For example, the functions of each unit are implemented in the first operating system.
[0226] The scene recognition unit 710 is configured to determine a first user scene in the first operating system and to transmit the first user scene to the scheduling management unit 720. For example, the scene recognition unit 710 can determine the first user scene based on application information running in the first operating system.
[0227] The scheduling management unit 720 is configured to determine a resource scheduling policy based on a first user scenario.
[0228] In one implementation, the scene recognition unit 710 is configured to determine a first resource state of a first operating system and transmit the first resource state to the scheduling management unit 720. The scheduling management unit 720 is configured to determine a resource scheduling policy based on the first user scene and the first resource state.
[0229] In one implementation, the scheduling management unit 720 is configured to communicate with the second operating system to obtain the second user scene and the second resource state of the second operating system, for example, by communicating with the second operating system via an IPC interface, and to determine a resource scheduling policy based on the first user scene, the second user scene, the first resource state, and the second resource state.
[0230] In one implementation, the scheduling management unit 720 is configured to determine the priority of a first operating system based on a first user scene, the priority of a second operating system based on a second user scene, and to determine a resource scheduling policy based on the priority of the first operating system, the priority of the second operating system, the first resource state, and the second resource state.
[0231] In one implementation, the scheduling management unit 720 is configured to determine whether there are available resources in the first operating system based on a first resource state, whether there are available resources in the second operating system based on a second resource state, and if there are no available resources in the second operating system and the priority of the first operating system is lower than the priority of the second operating system, it determines a first resource scheduling policy. The first resource scheduling policy includes scheduling the first resources from the first operating system to the second operating system.
[0232] In one implementation, the scheduling management unit 720 is configured to receive resource requests from the second operating system, determine the internal scheduling policy of the first operating system if the available resources in the first resources already acquired by the first operating system do not match the resource requests of the second operating system, and send the internal scheduling policy of the first operating system to the scheduling execution unit 730. The scheduling execution unit 730 is configured to execute the internal scheduling policy of the first operating system, perform internal resource scheduling for the first operating system, and acquire available first target resources.
[0233] In one implementation, the scheduling execution unit 730 is further configured to release the first target resource and leave the first target resource unallocated.
[0234] In one implementation, the resource scheduling device further includes a resource management unit 810.
[0235] The scheduling management unit 720 is configured to send a resource scheduling policy (for example, a first resource scheduling policy) to the resource management unit 810. The resource management unit 810 can be configured to execute the resource scheduling policy and perform resource scheduling between systems. For example, it may execute the first resource scheduling policy and schedule a first target resource to a second operating system.
[0236] In one implementation, the resource management unit 810 is configured to release the first target resource and leave it unallocated. That is, the release of the first target resource may be performed by the scheduling execution unit 730 or by the resource management unit 810.
[0237] In one implementation, the scheduling management unit 720 is configured to predict the resources required by the first operating system within a predetermined time in the future based on a first user scene, and to determine the resource requests of the first operating system.
[0238] In one implementation, the scheduling management unit 720 is configured to send the first user scene, the first resource state, and resource requests of the first operating system to the second operating system.
[0239] In one implementation, the function of any unit in the resource scheduling device can be realized by the cooperation of multiple units; that is, the resource scheduling device may include more units. For example, the resource scheduling device may include a cross-domain scheduling unit, and at least some of the functions of the scheduling management unit 720 can be realized by the cross-domain scheduling unit.
[0240] For example, the scheduling management unit 720 may be configured to transmit a first user scene and a first resource state to the cross-domain scheduling unit, or the scene recognition unit 710 may transmit a first user scene and a first resource state to the cross-domain scheduling unit. The cross-domain scheduling unit is configured to determine a resource scheduling policy based on the first user scene and the first resource state.
[0241] Exemplary, the cross-domain scheduling unit is configured to communicate with a second operating system to obtain a second user scene and a second resource state of the second operating system, and to determine a resource scheduling policy based on the first user scene, the second user scene, the first resource state, and the second resource state.
[0242] For example, a cross-domain scheduling unit is configured to receive resource requests from a second operating system and send the resource scheduling policy and the resource requests from the second operating system to a scheduling management unit 720. In this way, the scheduling management unit 720 can determine the internal scheduling policy of the first operating system based on the resource scheduling policy and the resource requests from the second operating system.
[0243] For example, a cross-domain scheduling unit may be configured to send a resource scheduling policy (e.g., a first resource scheduling policy) to a resource management unit 810, causing the resource management unit 810 to execute the resource scheduling policy.
[0244] In one implementation, the functions of multiple units in a resource scheduling device may be implemented in a single unit, for example, by integrating at least two of the scene recognition unit 710, scheduling management unit 720, and scheduling execution unit 730 into one unit. In other words, the resource scheduling device may include fewer units, none of which exceed the scope of protection of the embodiments of this disclosure.
[0245] Furthermore, in the first or second operating system, each unit of the resource scheduling device can be implemented in various forms, and the embodiments of this disclosure are not limited thereto. For example, the first operating system may be the Android system. The scene recognition unit 710 may be located in the application framework layer of the Android system, the scheduling management unit 720 may be located in the native layer of the Android system, the scheduling execution unit 730 may be located in the kernel layer of the Android system, and the resource management unit 810 may be located in the resource management layer of the cockpit driving integration system. For example, the resource management unit 810 may be the virtual machine manager hypervisor of the resource management layer.
[0246] The resource scheduling device according to the embodiment of this disclosure can perform resource scheduling between different operating systems based on resource requests in different user scenes of each operating system. In other words, it realizes inter-system resource scheduling based on user scenes, satisfies the resource requests of each operating system in different user scenes, improves the execution stability of the operating systems, and further makes resource allocation more rational and improves resource utilization.
[0247] Exemplary electronic device Embodiments of this disclosure provide electronic devices that can be used in vehicles, for example, in intelligent connected vehicles. These electronic devices may be, for example, in-vehicle computing devices, in-vehicle computing platforms, in-vehicle computing units, intelligent driving computing platforms, and the like.
[0248] In one embodiment, the electronic device may include a chip system according to an embodiment of the present disclosure, which can run at least one operating system, for example, an intelligent driving system and a cockpit system, to realize related functions of a cockpit driving integrated system.
[0249] In one embodiment, the electronic device may include a housing, a power supply module, a heat dissipation module, and an external interface.
[0250] Enclosures are used to house and mount the various hardware units of electronic equipment, protecting them from dust and water ingress, and improving the reliability of the electronic equipment.
[0251] Power modules are used to supply power to each hardware unit of an electronic device.
[0252] Heat dissipation modules, such as cooling fans and liquid cooling devices, are used to dissipate heat from hardware units of electronic devices, thereby lowering the operating temperature of each hardware unit and improving the operational stability of each hardware unit.
[0253] External interfaces are used to connect to various sensors in the vehicle, such as cameras, LiDAR, millimeter-wave radar, and ultrasonic radar, to send commands to each sensor or receive data returned from the sensors. Furthermore, external interfaces can connect to various actuators in the vehicle, such as the drive system, braking system, and steering system, thereby sending control commands to each system and receiving feedback from each system.
[0254] Exemplary vehicles The embodiments of this disclosure provide vehicles, which may include land vehicles, air vehicles, water vehicles, underwater vehicles, space vehicles, and the like. Specific details of the vehicles can be found in the embodiments described above, and redundant explanations are omitted here.
[0255] The vehicle may include a chip system or electronic device according to an embodiment of the disclosure in order to implement the resource scheduling method according to an embodiment of the disclosure. In this way, when a user uses the vehicle, for example, by driving the vehicle or by using the vehicle to watch videos, listen to music, or play games, the vehicle can perform inter-system resource scheduling between the cockpit system and the intelligent driving system based on the user scene, thereby providing the user with a better user experience.
[0256] Exemplary computer program products and computer-readable storage media In addition to the methods and apparatus described above, embodiments of the present disclosure may include a computer program product that, when executed by a processor, causes the processor to perform steps in the resource scheduling methods relating to various embodiments of the present disclosure as described in the “Exemplary Methods” portion of this specification.
[0257] Computer program products can be created using any combination of one or more programming languages to produce program code for performing the operations of the embodiments of this disclosure, and the programming languages include object-oriented programming languages such as Java and C++, and conventional procedural programming languages such as the C language or similar programming languages. The program code may run entirely on the user's computing device, partially on the user's device, as separate software packages, partly on the user's computing device and partly on a remote computing device, or entirely on a remote computing device or server.
[0258] Furthermore, embodiments of the present disclosure may be computer-readable storage media on which computer program instructions are stored, and when executed by a processor, the computer program instructions cause the processor to perform steps in the resource scheduling method according to various embodiments of the present disclosure as described in the “Exemplary Methods” section above.
[0259] Computer-readable storage media may be any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable medium may include, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples (non-exclusive enumeration) of readable storage media include electrical connections with one or more reads, portable disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, compact disk read-only memory (CD-ROM), optical memory elements, magnetic memory elements, or any suitable combination of the above.
[0260] While the basic principles of this disclosure have been explained above with reference to specific examples, the advantages, advantages, and effects mentioned herein are not limited to those mentioned above, but are merely illustrative, and these advantages, advantages, and effects are not necessarily present in every example of this disclosure. Furthermore, the specific details disclosed above are not limited to those mentioned above, but are merely illustrative and intended to facilitate understanding, and these details do not necessarily limit this disclosure to being realized by those specific details.
[0261] Those skilled in the art can make various modifications and alterations to this disclosure without departing from the spirit and scope of this disclosure. Thus, if such modifications and alterations of this disclosure fall within the scope of the claims of this disclosure and the equivalent art, this disclosure is intended to include such modifications and alterations.
Claims
1. A resource scheduling method used in an integrated circuit that runs at least a first operating system and a second operating system, wherein each step is performed by a resource scheduling device, A step of determining a first user scene set based on the state of the vehicle and / or the state of the first operating system, wherein the state of the first operating system includes at least one of the system state, input state and usage state of the first operating system; A resource scheduling method comprising: determining a resource scheduling policy for scheduling a first resource between a first operating system and a second operating system based on a first user scene, wherein the first resource includes a resource shared by the first operating system and the second operating system.
2. The resource scheduling method according to claim 1, wherein the first operating system is a cockpit system and the second operating system is an intelligent driving system.
3. The step of determining the resource scheduling policy for scheduling the first resource between the first operating system and the second operating system based on the first user scene is: The steps include determining the first resource state of the first operating system, The resource scheduling method according to claim 1, comprising the step of determining the resource scheduling policy based on the first user scene and the first resource state.
4. A step of determining a second user scene set based on the vehicle state and / or the state of the second operating system, wherein the state of the second operating system includes at least one of the system state, input state and usage state of the second operating system. The steps include determining the second resource state of the second operating system, The resource scheduling method according to claim 3, further comprising the step of determining the resource scheduling policy based on the first user scene, the second user scene, the first resource state, and the second resource state.
5. The step of determining the resource scheduling policy based on the first user scene, the second user scene, the first resource state, and the second resource state is: The steps include determining the priority of the first operating system based on the first user scene, and determining the priority of the second operating system based on the second user scene, The resource scheduling method according to claim 4, comprising the step of determining the resource scheduling policy based on the priority of the first operating system, the priority of the second operating system, the first resource state, and the second resource state.
6. The step of determining the resource scheduling policy based on the priority of the first operating system, the priority of the second operating system, the first resource state, and the second resource state is: A step of determining whether the second operating system has available resources based on the second resource state, The resource scheduling method according to claim 5, comprising the step of determining a first resource scheduling policy if the second operating system has no available resources and the priority of the first operating system is lower than the priority of the second operating system, wherein the first resource scheduling policy includes scheduling the first resources from the first operating system to the second operating system.
7. A step of releasing a first target resource from the first resource already acquired by the first operating system based on a resource request from the second operating system, wherein the first target resource matches the resource request from the second operating system. The resource scheduling method according to claim 6, further comprising the step of scheduling the first target resource to the second operating system.
8. The resource scheduling method according to claim 7, further comprising the step of performing internal resource scheduling on the first operating system to acquire an available first target resource if any available resources in the first resource already acquired by the first operating system do not match the resource request of the second operating system.
9. The resource scheduling method according to claim 7, further comprising the step of predicting the resources required by the second operating system within a predetermined future time based on the second user scene, in order to determine the resource requests of the second operating system.
10. A step of determining whether the first operating system has available resources based on the first resource state, The resource scheduling method according to claim 6, comprising the step of determining a second resource scheduling policy if the first operating system has no available resources and the priority of the first operating system is higher than the priority of the second operating system, wherein the second resource scheduling policy includes scheduling the first resources from the second operating system to the first operating system.
11. A step in which the second operating system releases a second target resource from the first resource already acquired by the second operating system based on the resource request of the first operating system, wherein the amount of the second target resource matches the resource request of the first operating system. The resource scheduling method according to claim 10, further comprising the step of scheduling the second target resource to the first operating system.
12. The resource scheduling method according to claim 11, further comprising the step of performing internal resource scheduling on the second operating system to acquire the available second target resource if the amount of free resources in the first resource already acquired by the second operating system is less than the amount of resources in the second target resource.
13. The resource scheduling method according to claim 11, further comprising the step of predicting the resources required by the first operating system within a predetermined future time based on the first user scene, in order to determine the resource requests of the first operating system.
14. The step of determining the first user scene of the first operating system is: The process includes the step of determining the first user scene based on application information running in the first operating system, The step of determining the second user scene of the second operating system is: A resource scheduling method according to any one of claims 4 to 13, comprising the step of determining a second user scene based on application information running in the second operating system.
15. The resource scheduling method according to claim 1, wherein the first operating system and the second operating system communicate via an inter-process communication IPC interface.
16. A resource scheduling device used in an integrated circuit that runs at least a first operating system and a second operating system, A scene recognition unit configured to determine a first user scene set based on the state of the vehicle and / or the state of the first operating system, The system includes a scheduling management unit configured to determine a resource scheduling policy for scheduling a first resource between the first operating system and the second operating system based on the first user scene, A resource scheduling device wherein the first resource includes resources shared by the first operating system and the second operating system, and the state of the first operating system includes at least one of the system state, input state and utilization state of the first operating system.
17. The resource scheduling device according to claim 16, wherein the first operating system is an intelligent driving system, and the second operating system is a cockpit system.
18. The scene recognition unit is configured to determine the first resource state of the first operating system, The resource scheduling device according to claim 16 or 17, wherein the scheduling management unit is configured to determine the resource scheduling policy based on the first user scene and the first resource state.
19. The resource scheduling device according to claim 18, wherein the scheduling management unit is configured to acquire a second user scene and a second resource state set based on the vehicle state and / or the state of the second operating system, and to determine the resource scheduling policy based on the first user scene, the second user scene, the first resource state and the second resource state, the state of the second operating system includes at least one of the system state, input state and utilization state of the second operating system.
20. A chip system comprising a processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, cause the chip system to perform steps according to any one of claims 1 to 13.
21. An electronic device comprising the chip system described in claim 20.
22. A vehicle comprising the chip system described in claim 20.
Citation Information
Patent Citations
Operation protection mode scheduling
JP2011526040A
Power control method, computer system, and program
JP2012074084A
Computer system, kernel scheduling system, resource allocation method, and process execution sharing method
JP2015038757A
Behavioral model based malware protection system and method
US20170230389A1
