Method, device and equipment for allocating resources of vehicle machine and storage medium

Abstract

A vehicle machine resource allocation method and device, equipment and storage medium. Involve the field of automobile technology. The method comprises: acquiring collection data and historical data of the vehicle machine; based on the collection data, determining the scene mode in which the vehicle machine is located in the first time unit; based on the historical data and the collection data, generating a prediction result, the prediction result being used to indicate a first application program to be started by the vehicle machine in a second time unit; based on the scene mode and the prediction result, determining a resource allocation strategy of the vehicle machine in the first time unit. The above method solves the problem of high application program start delay caused by the static vehicle machine resource allocation mode in the related art by using the current scene mode and the prediction result as the decision basis for vehicle machine resource scheduling, ensures that the resource allocation strategy meets the current scene mode, and when the user switches to the first application program, the first application program can be quickly responded and started, thereby improving the application program start speed.

Description

Technical Field

[0001] This application relates to the field of automotive technology, and in particular to a method, apparatus, device, and storage medium for allocating vehicle infotainment resources. Background Technology

[0002] Vehicle infotainment system resources may include at least one of the following: processor resources, memory resources, storage resources, etc., but this application does not limit them.

[0003] In related technologies, vehicle system resources can be scheduled based on a static resource allocation method. Specifically, this method can include static priority-based resource scheduling. In this approach, different applications are pre-assigned priorities, and vehicle system resources are allocated according to these priorities. For example, navigation applications can be set to the highest priority, multimedia entertainment applications to the second highest priority, and other background applications to lower priorities. When the vehicle system starts, it prioritizes allocating resources to high-priority applications, even if those applications do not currently have an urgent need for resources. Low-priority applications, even if they have an urgent resource requirement, must wait for higher-priority applications to release resources before they can acquire them.

[0004] However, the above-mentioned static vehicle system resource allocation method will lead to high application startup latency. Summary of the Invention

[0005] This application provides a method, apparatus, device, and storage medium for allocating vehicle system resources. The technical solution provided by this application includes the following aspects.

[0006] According to one aspect of the embodiments of this application, a method for allocating vehicle system resources is provided, the method comprising: Acquire the vehicle's data collection data and historical data. The data collection data is used to indicate the vehicle's operating status and interaction status in a first time unit. The interaction status is used to indicate the operation behavior of interacting with the vehicle. The historical data is used to indicate the vehicle's usage of at least one application in a historical period. Based on the collected data, the scene mode of the vehicle system in the first time unit is determined; Based on the historical data and the collected data, a prediction result is generated. The prediction result is used to indicate the first application that the vehicle system will launch in the second time unit, which is the time unit after the first time unit. Based on the scenario pattern and the prediction results, a resource allocation strategy for the vehicle system in the first time unit is determined. The resource allocation strategy is used to indicate the resource allocation method for at least one task of the vehicle system.

[0007] In some embodiments, determining the scene mode of the vehicle system in the first time unit based on the collected data includes: Based on the condition information and the collected data, the scene mode is determined from at least one candidate scene mode; wherein the condition information is used to indicate a preset condition corresponding to at least one of the running state and the interaction state corresponding to the at least one candidate scene mode.

[0008] In some embodiments, the condition information includes at least one condition expression, which indicates a preset condition corresponding to at least one of the running state and the interaction state of a candidate scene mode.

[0009] In some embodiments, the method further includes: The configuration data corresponding to the first application is preloaded into the memory of the vehicle system; In response to the launch operation for the first application, the first application is launched in the vehicle system based on the configuration data corresponding to the first application.

[0010] In some embodiments, generating a prediction result based on the historical data and the collected data includes: Based on the historical data, a first probability value is determined for each of the at least one application. The first probability value of the application is used to indicate the probability that the vehicle system will launch the application when the current application is running. Based on the historical data, a second probability value is determined for each of the at least one application, and the second probability value of the application is used to indicate the probability of the vehicle system launching the application within a first time period; Based on the historical data, a third probability value is determined for each of the at least one application. The third probability value of the application is used to indicate the probability that the vehicle system will launch the application when it is in the first position. The prediction result is determined based on at least one of the first probability value, the second probability value, and the third probability value.

[0011] In some embodiments, the historical data includes at least one entry, each entry including at least one of the following: application identification information, application location information, and application time information; wherein the identification information is used to distinguish different applications among the at least one application, the location information is used to indicate the location of the vehicle system when running the application, and the time information is used to indicate the time period during which the vehicle system runs the application.

[0012] In some embodiments, the method further includes: After scheduling the vehicle system to execute the resource allocation scheme, the performance compliance information of the at least one task is determined, and the performance compliance information is used to indicate the degree of deviation between the performance of the at least one task and the expected performance. Based on the performance compliance information, the resource allocation strategy is adjusted to obtain an adjusted resource allocation strategy, which is used to redetermine the resource allocation method for the at least one task.

[0013] Preferably, the method further includes: After scheduling the vehicle system to execute the resource allocation scheme, the performance compliance information of the at least one task is determined, and the performance compliance information is used to indicate the degree of deviation between the performance of the at least one task and the expected performance. Based on the performance compliance information, the resource allocation strategy is adjusted to obtain an adjusted resource allocation strategy, which is used to redetermine the resource allocation method for the at least one task.

[0014] According to one aspect of the embodiments of this application, a vehicle system resource allocation device is provided, the device comprising: The acquisition module is used to acquire the vehicle's collected data and historical data. The collected data is used to indicate the vehicle's operating status and interaction status in a first time unit. The interaction status is used to indicate the operation behavior of interacting with the vehicle. The historical data is used to indicate the vehicle's usage of at least one application in a historical period. The first determining module is used to determine the scene mode of the vehicle system in the first time unit based on the collected data; The generation module is used to generate a prediction result based on the historical data and the collected data. The prediction result is used to indicate the first application that the vehicle system will start in the second time unit, which is the time unit after the first time unit. The second determining module is used to determine the resource allocation strategy of the vehicle system in the first time unit based on the scene mode and the prediction result. The resource allocation strategy is used to indicate the resource allocation method for at least one task of the vehicle system.

[0015] Preferably, the collected data includes at least one of the following: user interaction data, vehicle status data, and system resource data.

[0016] According to one aspect of the embodiments of this application, a computer device is provided, the computer device including a processor and a memory, the memory storing a computer program, the computer program being loaded and executed by the processor to implement the above-described method for allocating vehicle system resources.

[0017] According to one aspect of the embodiments of this application, a computer-readable storage medium is provided, wherein a computer program is stored in the computer program, which is loaded and executed by a processor to implement the above-described method for allocating vehicle system resources.

[0018] According to one aspect of the embodiments of this application, a computer program product is provided, the computer program product including a computer program stored in a computer-readable storage medium, and a processor reading from the computer-readable storage medium and executing the computer program to implement the above-described method for allocating vehicle system resources.

[0019] The technical solution provided in this application can bring the following beneficial effects: By combining the current scenario mode with the prediction results as the decision basis for vehicle system resource scheduling, the technical problem of high application startup latency caused by static vehicle system resource allocation in related technologies is solved. This ensures that the resource allocation strategy can quickly respond and launch the first application when the user switches to the first application, while meeting the current scenario mode, thereby achieving the technical effect of improving application startup speed. Attached Figure Description

[0020] Figure 1 This is a flowchart of a method for allocating vehicle system resources according to an embodiment of this application; Figure 2 This is a schematic diagram of multiple modules included in a vehicle infotainment system according to one embodiment of this application; Figure 3 This is a flowchart of a method for allocating vehicle system resources according to another embodiment of this application; Figure 4 This is a block diagram of a vehicle system resource allocation device provided in one embodiment of this application; Figure 5 This is a structural block diagram of a computer device provided in one embodiment of this application. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of this application clearer, the embodiments of this application will be described in further detail below with reference to the accompanying drawings.

[0022] The vehicle system resource allocation method provided in this application embodiment can be executed by a computer device, which can be any electronic device with computing and storage capabilities, such as a PC (Personal Computer), a server, or other electronic devices.

[0023] Please refer to Figure 1 The diagram illustrates a flowchart of a method for allocating vehicle system resources according to an embodiment of this application. The execution entity of each step of the method can be a computer device, and the method may include at least one of the following steps (110-140).

[0024] Step 110: Obtain the vehicle's data collection and historical data. The data collection is used to indicate the vehicle's operating status and interaction status in the first time unit. The interaction status is used to indicate the operation behavior of interacting with the vehicle. The historical data is used to indicate the vehicle's usage of at least one application in a historical period.

[0025] A car infotainment system refers to the in-vehicle infotainment system installed in a car. It integrates multiple functions, such as navigation, multimedia playback, vehicle control, and intelligent connectivity, to provide users with a convenient driving experience and information services.

[0026] Optionally, the collected data may include at least one of the following: user interaction data, vehicle status data, and system resource data.

[0027] User interaction data, used to indicate user interactions with the vehicle's infotainment system, may include at least one of the following: the currently running application's package name, application switching history and frequency, the user's touchscreen focus area, and voice assistant module wake-up and command interaction events. The currently running application's package name is a string uniquely identifying an application within the infotainment system, used to distinguish different applications. The application switching history and frequency indicate how frequently a user switches between different applications within a certain period, including at least one of the switching time, order, and frequency. Analyzing this data helps understand user habits and application usage popularity, providing a reference for resource allocation. The user's touchscreen focus area indicates the specific area the user focuses on when touching the infotainment screen, reflecting the user's operational intent and interests. For example, frequent touches of a specific area may indicate that the corresponding function or application is the user's current focus. Voice assistant module wake-up and command interaction events indicate when the user wakes up the voice assistant and interacts with it, including the wake-up time and the content of the commands issued. By analyzing this data, we can understand the frequency and needs of users using voice assistants, as well as the applications or functions involved in voice commands, and thus allocate resources rationally.

[0028] Vehicle status data indicates the vehicle's current operating status and may include at least one of the following: vehicle speed information, vehicle gear information, vehicle charging status, and vehicle time information. Vehicle speed information indicates the vehicle's current speed, reflecting its motion state. At different speeds, the resource requirements of the vehicle's infotainment system may vary. For example, at high speeds, navigation applications may require more real-time and accurate map data and location information, thus requiring more resources. Vehicle gear information indicates the vehicle's current gear, such as drive, reverse, or neutral. Different gears may correspond to different driving scenarios and infotainment system functionalities. For example, when reversing, the reversing camera application may need to prioritize system resources to ensure real-time and clear image display. Vehicle charging status indicates the vehicle's charging progress, such as whether it is charging and the charging progress. While the vehicle is charging, the infotainment system may perform background data updates or system maintenance operations, requiring reasonable resource allocation to ensure these operations proceed smoothly. Vehicle time information indicates the current time, including date, hour, and minute. Time factors may affect users' usage habits and the resource requirements of the vehicle's infotainment system. For example, during weekday morning and evening rush hours, navigation applications may be used more frequently, requiring more resources to be allocated.

[0029] Optionally, the vehicle status data can be obtained through a vehicle gateway or a CAN (Controller Area Network) bus interface. The vehicle gateway serves as a communication hub between different networks within the vehicle, while the CAN bus is a serial communication protocol widely used in automotive electronic systems. Both can reliably acquire various vehicle status information.

[0030] System resource data indicates the resource usage of the vehicle infotainment system and may include at least one of the following: GPU (Graphics Processing Unit) utilization, memory utilization, CPU (Central Processing Unit) utilization, and network interface throughput. GPU utilization may include the rendering frame rate (Frames Per Second, FPS) of at least one GPU. Memory utilization includes Resident Set Size (RSS) and Proportional Set Size (PSS). RSS represents the actual physical memory used by the application, while PSS considers shared memory and more accurately reflects the application's actual memory usage. The CPU is the core computing unit of the vehicle infotainment system; CPU utilization reflects its current workload, i.e., the proportion of time the CPU spends executing various tasks within a given timeframe. Network interface throughput indicates the rate at which the vehicle infotainment system transmits data through the network interface, reflecting network resource usage. In the vehicle infotainment system, many applications require data to access the network, such as online navigation and multimedia playback; therefore, network interface throughput is crucial to the performance of these applications.

[0031] Optionally, system resource data can be obtained from the / proc filesystem of the operating system kernel (such as the Linux Kernel) or from the performance monitoring interface (Perf Event). The / proc filesystem is a special virtual filesystem in the Linux operating system that provides a convenient way to obtain various system runtime information; the performance monitoring interface (PerfEvent) is a set of interfaces provided by the Linux kernel for performance monitoring, which can obtain more accurate information about system resource usage.

[0032] The above method, by acquiring collected data, facilitates the accurate determination of the scene mode in which the vehicle's infotainment system is located.

[0033] Step 120: Based on the collected data, determine the scene mode in which the vehicle-mounted system is located in the first time unit.

[0034] Scene modes refer to specific operating states of the vehicle's infotainment system under different usage environments and user needs. These modes are used to rationally allocate system resources and provide targeted functional services based on different scenarios, thereby improving user experience and operational efficiency. Scene modes may include at least one of the following: safe driving mode, entertainment mode, standby power-saving mode, and intelligent voice interaction mode; this application does not limit the specific mode.

[0035] The Safe Driving Mode prioritizes driving safety-related functions such as navigation, real-time traffic updates, and collision warnings when the vehicle is in motion. In this mode, the system reduces unnecessary entertainment features and restricts operations that might distract the driver, such as lowering multimedia volume and blocking unnecessary pop-up notifications, ensuring the driver can concentrate on driving and improving safety.

[0036] The Entertainment & Leisure Mode indicates that when the vehicle is parked or stationary, the infotainment system will focus on providing a wealth of entertainment features, such as high-definition video playback, music playback, and gaming. The system will allocate more system resources to these entertainment applications to ensure a smooth user experience, and may also adjust the interface display and sound settings to create a comfortable entertainment atmosphere.

[0037] The standby power-saving mode indicates that the vehicle's infotainment system will enter a low-power operating state when it is not used for an extended period or when the vehicle is turned off. In this mode, the infotainment system will disable most unnecessary functions and services to reduce system resource consumption, retaining only basic functions such as the clock and battery monitoring, in order to extend the battery's range and prevent excessive discharge of the vehicle battery.

[0038] The intelligent voice interaction mode is designed for scenarios where users primarily interact with the vehicle's infotainment system via voice commands. In this mode, the system will highly sensitively recognize the user's voice commands and execute corresponding operations quickly and accurately, such as querying information, controlling vehicle equipment (e.g., adjusting air conditioning temperature, opening / closing windows), and setting navigation. The system will also provide more intelligent and personalized services and feedback based on the content and context of the voice interaction.

[0039] It should be noted that the above scenario modes are merely illustrative and this application does not limit them.

[0040] Step 130: Based on historical data and collected data, generate prediction results. The prediction results are used to indicate the first application that the vehicle system will launch in the second time unit, which is the time unit following the first time unit.

[0041] The first application refers to the application that the vehicle's infotainment system is most likely to launch in the second time unit, based on analysis and prediction of historical and collected data. For example, if the vehicle is currently on its way home from work on a weekday (first time unit), the music playback application that is predicted to launch after arriving at the destination and parking (second time unit) is the first application.

[0042] In some embodiments, the historical data includes at least one entry, each entry including at least one of the following: application identification information, application location information, and application time information; wherein the identification information is used to distinguish different applications in at least one application, the location information is used to indicate the location of the vehicle system when the application is running, and the time information is used to indicate the time period during which the vehicle system runs the application.

[0043] The application's identification information is used to distinguish different applications. The application's location information may include the vehicle's geographical coordinates, city / region, and specific location (such as a parking lot, shopping mall, or highway service area) when the application is running on the vehicle's infotainment system. The application's time information may include the specific date (such as weekday, weekend, or holiday), specific time of day (such as morning, afternoon, or evening), and season, etc., which are not limited in this application.

[0044] The above method uses historical and collected data to predict the first application that the vehicle's infotainment system may launch in the future. This allows the system to prepare the first application that the user needs in advance, improving the convenience and smoothness of using the infotainment system, optimizing the resource allocation of the system, and improving the overall system performance and response speed.

[0045] In some embodiments, generating a prediction result based on historical data and collected data includes: determining a probability value for at least one application based on historical data and collected data, the probability value of the application being used to indicate the probability that the vehicle system will launch the application in a second time unit; wherein the collected data includes at least one of the following: the currently running application of the vehicle system, the current first location of the vehicle system, and the current first time of the vehicle system; and determining a probability result based on the probability value for at least one application.

[0046] The currently running application in the vehicle's infotainment system refers to the application that the system is currently using.

[0047] Preferably, a behavior pattern is determined based on historical data. The behavior pattern, also known as a historical sequence, is used to indicate the regularity and habit of the vehicle system launching applications at different times, locations, and in different application states. Based on the behavior pattern and the collected data, the probability value of at least one application is determined.

[0048] In some embodiments, behavioral patterns are determined based on historical data; and a probability value for each of at least one application is determined using a Markov chain based on the behavioral patterns and the collected data.

[0049] A Markov chain is a stochastic process that exhibits the Markov property, meaning that given the current state, the probability distribution of its future states depends only on the current state and is independent of past states. In the context of this application, the startup state of the in-vehicle application can be viewed as a state in a Markov chain. Behavioral patterns obtained by analyzing historical data can be used to determine the transition probabilities between states, thereby predicting the probability of the in-vehicle system launching each application in a future second time unit based on currently collected data (such as the currently running application, the current time, and location).

[0050] Optionally, the behavior pattern may include at least one of the following: time pattern, location pattern, application-related pattern, time-location pattern, time-application pattern, location-application pattern, time-location-application pattern, etc. This application does not limit the specific behavior pattern.

[0051] Time-based launch patterns indicate the frequency with which a vehicle's infotainment system launches specific applications at different times or during different time periods. For example, between 7 and 9 PM every evening, there's an 80% probability that a user will launch a video playback app to watch programs. Location-based launch patterns indicate the tendency of the infotainment system to launch applications in different locations. For example, when the vehicle is in a shopping mall parking lot, there's a 60% probability that a mall navigation or shopping discount information app will launch. App-related launch patterns indicate the probability of launching other applications while one application is running. For example, when a music playback app is running, there's a 30% probability that a lyrics search app will launch. Time-location patterns consider the combined impact of time and location factors on application launches. For example, around noon on weekdays, near the office, there's a 75% probability that a food delivery app will launch. Time-application patterns consider the combined impact of time and currently running applications on application launches. Location-application patterns consider the combined impact of location and currently running applications on application launches. Time-location-application patterns consider the combined impact of time, location, and currently running applications on application launches.

[0052] In some embodiments, the probability value corresponding to the application is used to indicate the probability that the vehicle system will launch the application when at least one of the following conditions is met: the vehicle system is running the application, the vehicle system is in a first time period, or the vehicle system is in a first location.

[0053] Optionally, the probability value may include a first probability value, a second probability value, and a third probability value; determining the probability value of at least one application based on historical data and collected data includes: determining a first probability value for each of the at least one application, the first probability value of the application indicating the probability of launching the application when the vehicle system is running the current application; determining a second probability value for each of the at least one application based on historical data, the second probability value of the application indicating the probability of launching the application within a first time period; determining a third probability value for each of the at least one application based on historical data, the third probability value of the application indicating the probability of launching the application when the vehicle system is located in a first position; and determining a prediction result based on at least one of the first, second, and third probability values.

[0054] For example, based on behavioral patterns and collected data, the probability of launching another application B from the current application A (i.e., the first probability value mentioned above) can be calculated and denoted as P(B|A); the probability of launching another application from the current time period (i.e., the first time) (i.e., the second probability value) can be denoted as P(App|Time). The probability of launching another application at the first location (i.e., the third probability value) is not limited in this application.

[0055] It should be noted that the above is merely an example. The first probability value, second probability value, and third probability value of the application can also be used to indicate the probability of the vehicle system launching the application under at least one condition: the current application is running, the application is in a first time, or the application is in a first position. This application does not limit this.

[0056] The above method, through the analysis of historical and collected data, can accurately grasp the habits and patterns of users launching applications in different scenarios, and provide users with personalized application launch prediction services.

[0057] Step 140: Based on the scenario pattern and prediction results, determine the resource allocation strategy of the vehicle-mounted system in the first time unit. The resource allocation strategy is used to indicate the resource allocation method for at least one task of the vehicle-mounted system.

[0058] At least one task refers to the various operations and functions that the vehicle infotainment system needs to handle during operation, including but not limited to the following: Application running tasks: such as the startup and operation of multimedia applications (music playback, video playback, etc.), navigation applications, communication applications (telephone, SMS, etc.), entertainment games, etc. System service tasks: such as system update checks, background data synchronization, sensor data acquisition and processing (such as GPS positioning, accelerometer data processing, etc.), security protection services (virus scanning, firewall monitoring, etc.). User interaction tasks: responding to touchscreen operations, voice command recognition and processing, key input, etc.

[0059] Resource allocation strategy refers to a series of rules and methods formulated to allocate vehicle system resources in a reasonable and efficient manner to ensure that each task can run stably and smoothly.

[0060] Resource allocation strategies may include at least one of the following: GPU resource allocation strategy, memory resource allocation strategy, and scheduling strategy, and this application does not limit them.

[0061] GPU resource allocation strategies may include allocating high-priority graphics queues to rendering threads of critical foreground applications (such as navigation); throttling or reducing the precision of GPU requests from background applications (e.g., reducing rendering precision from FP32 to FP16); and, in "safe driving mode," forcibly disabling or significantly reducing UI animation effects and blur effects for non-critical applications.

[0062] Memory resource allocation strategies can include reserving memory space (Memory Pool) for applications (App_X) that are expected to launch soon; employing more aggressive memory compaction (ZRAM) and background process reclamation (Kill) strategies, prioritizing the termination of processes that have been inactive for a long time and are irrelevant to the current scenario (such as residual background game service processes after a game ends); and adjusting the sensitivity of the system memory watermark to trigger the memory reclamation mechanism earlier.

[0063] Process scheduling strategies can include dynamically adjusting process scheduling priorities (Nice value) and CPU core binding (Affinity). For example, setting the navigation process's Nice value to the highest and binding it to a performance core (Big Core). Utilizing Linux cgroups (control groups) technology, CPU and I / O bandwidth limits can be set for different application groups. For example, creating a cgroup for all background applications and limiting their combined CPU usage to no more than 10%. In "voice interaction mode," temporarily increasing the real-time priority (SCHED_FIFO) of the audio service process and the speech recognition process.

[0064] like Figure 3 The diagram illustrates a flowchart of a method for allocating vehicle system resources according to another embodiment of this application. It may include the following steps S1-S5.

[0065] Step S1, multi-dimensional data collection and perception, which corresponds to step 110 above (acquiring the collected data and historical data of the vehicle system).

[0066] Step S2, scene pattern recognition and classification, also known as step 120 above (based on the collected data, determine the scene pattern of the vehicle system in the first time unit).

[0067] Step S3, user habit learning and prediction, also known as step 130 above (generating prediction results based on historical data and collected data).

[0068] Step S4, dynamic resource strategy generation and allocation, also known as step 140 above (based on the scenario mode and prediction results, determine the resource allocation strategy of the vehicle system in the first time unit).

[0069] Step S5: Real-time feedback and closed-loop optimization.

[0070] In some embodiments, real-time feedback and closed-loop optimization include: after the vehicle scheduling system executes the resource allocation scheme, determining the performance compliance information of at least one task, the performance compliance information being used to indicate the degree of deviation between the performance of at least one task and the expected performance; based on the performance compliance information, adjusting the resource allocation strategy to obtain the adjusted resource allocation strategy, the adjusted resource allocation strategy being used to redetermine the resource allocation method for at least one task.

[0071] Performance compliance information refers to relevant data used to measure the difference between the actual performance and expected performance of at least one task in the vehicle infotainment system. It may include at least one of the following: response time, throughput, accuracy, resource utilization, application launch time, interface rendering frame rate (FPS), touch response latency, and overall system power consumption, which are not limited in this application.

[0072] The deviation between the performance of at least one task and its expected performance refers to the difference or ratio between the actual performance metric and the pre-set expected performance metric. Understandably, a larger deviation indicates a greater gap between the actual and expected performance, suggesting that the task's execution may not meet user needs or system design requirements; a smaller deviation indicates that the actual performance is closer to the expected performance, and the task execution is effective.

[0073] In some embodiments, the resource allocation strategy is adjusted based on performance target information to obtain an adjusted resource allocation strategy, which may include at least one of the following: increasing or decreasing resource allocation, adjusting task priority, or optimizing the resource allocation algorithm. This application does not limit the scope of the adjustment.

[0074] Increasing or decreasing resource allocation means that if the performance of a task deviates significantly from expectations, such as having a long response time, the resource allocation for that task can be increased appropriately, such as by increasing its CPU priority or allocating more memory; conversely, if the performance of a task far exceeds expectations and consumes too many resources, its resource allocation can be decreased.

[0075] Adjusting task priorities means reassessing the importance and urgency of each task based on performance compliance information, and then adjusting the task priorities accordingly. For important tasks with significant performance deviations, their priorities are increased to ensure they receive more resources; for tasks that meet performance standards but are less important, their priorities are decreased.

[0076] Optimizing resource allocation algorithms involves analyzing performance metrics to identify problems in the current resource allocation strategy and then optimizing the algorithm. For example, adjusting cgroups resource limit parameters to better suit the actual needs of the task. For instance, if expectations are not met (e.g., navigation frame rate remains below 50 FPS), a feedback loop will trigger strategy adjustments (e.g., further limiting background GPU usage), thereby achieving adaptive optimization of the strategy parameters.

[0077] The above method, by identifying performance compliance information and adjusting resource allocation strategies accordingly, can promptly identify performance issues during task execution and take targeted optimization measures. For example, for tasks with excessively long response times, increasing resource allocation can significantly shorten the response time, enabling the task to complete faster and improving execution efficiency and performance.

[0078] In some embodiments, such as Figure 2 As shown, this diagram illustrates multiple modules included in a vehicle-mounted infotainment system according to an embodiment of this application. The system includes a data perception module, a scene recognition module, a habit learning module, an execution and feedback module, and a strategy management module. The data perception module acquires the collected data and historical data; the scene perception module determines the scene mode of the vehicle-mounted infotainment system in its first time unit; the habit learning module generates prediction results based on historical and collected data; the strategy management module determines the resource allocation strategy for the vehicle-mounted infotainment system in its first time unit based on the scene mode and prediction results; and the execution and feedback module adjusts the resource allocation strategy to obtain the adjusted strategy. The system also includes a vehicle CAN bus and sensor module and an operating system kernel. The vehicle CAN bus and sensor module acquires collected and historical data from the operating system kernel and forwards it to the data perception module. The above devices (such as...) Figure 2The device (as shown) can be implemented on a vehicle-mounted hardware platform that includes a processor (CPU), graphics processing unit (GPU), memory, and storage. Each module in the device can be a software program unit stored in memory, read and executed by the CPU. The data sensing module connects to the vehicle network via a vehicle bus (such as CAN or Ethernet) and interacts with the kernel through operating system system calls. The device can also be partially or entirely programmed using a hardware description language (such as Verilog) and fabricated as an application-specific integrated circuit (ASIC) or a field-programmable gate array (FPGA) for hardware acceleration, thereby providing higher execution efficiency and response speed.

[0079] In some embodiments, such as Figure 2 As shown, based on the resource batching strategy, a set of resource control commands is generated and sent to the operating system kernel through the execution and feedback module.

[0080] The control command may be associated with at least one of the following: GPU control, memory and process control, adjusting process priority, using Cgroups for isolation, or terminating a process, without limitation in this application.

[0081] GPU control is used to allocate and manage the graphics processing unit (GPU) resources in the vehicle infotainment system. Optionally, it may include one of the following: dynamically adjusting the GPU frequency, allocating GPU video memory, and controlling GPU thread scheduling; this application does not limit this to any of these.

[0082] Memory and process control is used to allocate and manage the memory resources of the vehicle system, as well as to control the creation, destruction, and execution of processes. Optionally, it may include at least one of the following: memory allocation and reclamation, process priority adjustment, and process scheduling optimization, which are not limited in this application.

[0083] Adjusting process priorities allows for the dynamic adjustment of the execution priority of various processes in the vehicle's infotainment system based on different scenario modes and task characteristics. This can include adjustments based on task importance, where safety-related tasks (such as collision warning systems) and tasks with high real-time requirements (such as voice interaction responses) are given higher priority to ensure timely access to system resources and normal task execution. It can also be dynamically adjusted based on scenario modes, meaning the priority of each process is adjusted according to different scenario modes. For example, in driving mode, the priority of navigation applications and safety warning systems is increased, while the priority of non-critical processes such as entertainment applications is decreased; in parking mode, the priority of entertainment applications and other processes can be appropriately increased.

[0084] Cgroups (Control Groups) are used for isolation to separate different tasks or process groups within a vehicle infotainment system, ensuring that tasks do not interfere with each other and improving system stability and security. This can include resource limiting, which restricts resource usage for each task or process group through Cgroups; and process isolation, which involves placing different types of tasks or process groups into different Cgroups to achieve process isolation. For example, system service processes and user application processes can be isolated to prevent abnormal behavior of user application processes from affecting the normal operation of system services.

[0085] The process termination function is used to stop the operation of one or more processes in the vehicle system when necessary, in order to release system resources or resolve process abnormalities.

[0086] For example, GPU control can be implemented in a Linux environment by manipulating files in the ` / sys / class / kgsl / kgsl-3d0 / ` directory (e.g., `max_gpuclk` limits the maximum frequency, `min_pwrlevel` sets the minimum power consumption level) to dynamically adjust GPU frequency and power consumption. GPU priority for specific processes can be set via the `ioctl` interface provided by the GPU driver. For example, memory and process control refer to core functions of the Linux kernel. For example, adjusting process priority can include using the `setpriority()` system call to dynamically modify the `nice` value of a target process. Isolation using cgroups can be a more modern and efficient approach, such as grouping processes for management. For instance, creating a cgroup named `background.slice` and limiting the CPU usage of all processes in the group to 5% by writing to the `cpu.max` file and the memory usage limit by `memory.high`. Then, writing the PIDs of all background entertainment application processes to the `background.slice / cgroup.procs` file achieves overall resource limitation. Terminating processes can include ending abnormal or unnecessary processes by sending SIGTERM or SIGKILL signals.

[0087] In some embodiments, this application can be implemented by a resident background service (which may be named ResourceOrchestrator) in the vehicle infotainment system. This service runs on top of the operating system and interacts closely with the application framework layer and the system kernel. Figure 2As shown, after the ResourceOrchestrator service starts, its internal data sensing module continuously collects data through the following interfaces: It obtains foreground application switching information by subscribing to the Android Automotive ActivityManagerService event; it reads vehicle speed, gear position, and charging status signals from the CAN bus by connecting to the vehicle's VHAL (Vehicle Hardware AbstractionLayer) service; and it parses the Linux system's / proc / ... <pid>Various files under / (such as / proc / ) <pid>The ` / status` command checks memory usage, and ` / proc / ` is also mentioned. <pid>The ` / cmdline` command queries process names, allowing real-time monitoring of resource usage for each process. The collected data is then sent to the scene recognition module.

[0088] In summary, the technical solution provided by this application embodiment solves the technical problem of high application startup latency caused by static vehicle system resource allocation in related technologies by combining the current scene mode with the prediction result as the decision basis for vehicle system resource scheduling. It ensures that the resource allocation strategy can quickly respond and start the first application when the user switches to the first application, under the premise of meeting the current scene mode, thereby achieving the technical effect of improving the application startup speed.

[0089] The following describes how to determine the scene mode of the vehicle's infotainment system in the first time unit. That is, step 120 above can be implemented as the following steps 121 (…). Figure 1 (Not shown in the image).

[0090] Step 121: Based on condition information and collected data, determine a scene mode from at least one candidate scene mode; wherein, the condition information is used to indicate the preset conditions corresponding to at least one of the running state and interaction state corresponding to the at least one candidate scene mode.

[0091] In some embodiments, at least one candidate scenario module includes at least one of the following: safe driving mode, entertainment and leisure mode, standby power saving mode, and intelligent voice interaction mode, which are not limited in this application.

[0092] Conditional information refers to a series of standards set to determine the vehicle's infotainment system in various candidate scenario modes. These standards are related to the system's operating status (such as vehicle speed, gear position, power status, etc.) and interaction status (such as whether there is user touch interaction, the status of the voice assistant, etc.). By judging whether the collected data meets these preset conditions, the current scenario mode of the vehicle's infotainment system is determined.

[0093] For example, the conditional information may include a safe driving mode, where the vehicle speed is greater than a threshold K1 and the gear is in Drive (D). In this mode, the system prioritizes applications directly related to driving safety. The conditional information may also include an entertainment mode, where the vehicle speed is approximately zero and the gear is in Park (P), while the foreground application is an entertainment application such as a video or game, or the user is frequently interacting with the device via touch. The conditional information may also include a standby power-saving mode, where the vehicle's power is off or the duration of no user interaction exceeds a threshold T1. The system's goal is to minimize power consumption. The conditional information may also include an intelligent voice interaction mode, where the voice assistant is active (IsActive) and is processing speech recognition or natural language understanding tasks. This mode requires low-latency audio processing capabilities.

[0094] The above method can determine the scene mode of the vehicle's infotainment system by using conditional information and collected data.

[0095] In some embodiments, the condition information includes at least one condition expression, which indicates a preset condition corresponding to at least one of the running state and the interaction state of a candidate scene mode.

[0096] In some embodiments, a conditional expression, also known as a rule, refers to an expression that uses a specific combination of logic and conditions to describe at least one of the preset conditions in the running state and interaction state corresponding to a candidate scene mode. It defines in a clear and executable way under what circumstances the vehicle system should enter a specific scene mode.

[0097] There is a one-to-one correspondence between the conditional expression and the candidate scenario, that is, one conditional expression corresponds to one candidate scenario. Different conditional expressions correspond to different candidate scenarios, and this application does not limit this.

[0098] For example, the conditional expression can be: IF (vehicle_speed>5 km / h) AND (gear == 'D') THEN scenario = 'Safe Driving Mode', which indicates that when the vehicle speed is greater than 5 km / h and the gear is in drive (D), the vehicle system determines that it is currently in safe driving mode. In this case, the system should prioritize functions and applications related to driving safety according to the requirements of safe driving mode. The conditional expression can be IF(vehicle_speed == 0) AND (gear == 'P') AND (foreground_app.package in ['com.netflix.niv', 'com.tencent.tvgame']) THEN scenario = 'Entertainment and Leisure Mode', which indicates that when the vehicle speed is 0 (parked), the gear is in park (P), and the currently running application package name belongs to the specified entertainment category, the vehicle system determines that it is currently in entertainment and leisure mode, and the system will provide corresponding support and optimization for the user's entertainment activities. The conditional expression can be IF (voice_assistant.state == 'LISTENING') THEN scenario = 'Intelligent Voice Interaction Mode', which indicates that when the voice assistant is in listening mode, that is, ready to receive and process the user's voice commands, the vehicle system determines that it is currently in intelligent voice interaction mode. The system must ensure that it has low-latency audio processing capabilities to ensure the smoothness and accuracy of voice interaction.

[0099] The above method, by using conditional expressions to clarify the triggering conditions of each scenario mode, enables the vehicle system to adapt more intelligently and flexibly to different usage situations, bringing a better user experience while ensuring driving safety and saving energy.

[0100] In some embodiments, step 120 above can also be implemented as follows: when the scene mode of the vehicle system in the first time unit cannot be determined based on the above condition information (e.g., the vehicle is stationary but the front-end application is unknown, or multiple signals are mixed), the probability value of at least one candidate scene mode is determined based on the collected data and historical data using a pre-trained Lightweight Gradient Boosting Tree (LightGBM) model; and the scene mode of the vehicle system in the first time unit is determined based on the probability value of at least one candidate scene mode.

[0101] In some embodiments, the method further includes: preloading configuration data corresponding to the first application into the memory of the vehicle system; and launching the first application in the vehicle system based on the configuration data corresponding to the first application in response to a launch operation for the first application.

[0102] The configuration data contains various parameters and settings required for the first application to start. Preloading the configuration data allows the application to complete some initialization work before startup. For example, the configuration data may include at least one of the following: database connection information, interface layout parameters, etc. The application can directly use this information for initialization at startup without parsing and loading it during the startup process, thereby speeding up the startup. For example, the application code segment (binary) and frequently used data of App_X (i.e., the first application) are loaded into memory; a low-priority network connection is established with the cloud server of App_X. This aims to achieve a near-instantaneous launch experience when the user actually triggers the first application.

[0103] The above method enables the rapid startup of the first application by preloading configuration data.

[0104] In some embodiments, the method further includes at least one of the following: generating a preload instruction to prepare for the startup of the first application in advance; triggering the initialization of some components of the first application in advance; and pre-caching the core files of the first application.

[0105] Generating a preload instruction means that when the policy management module predicts based on learning results that a user may launch a certain first application (e.g., com.example.video), it will generate a preload instruction. The purpose of this instruction is to prepare for the launch of the first application in advance, but it does not directly launch the first application.

[0106] Pre-initializing certain components of an application can be achieved by calling the `bindService()` method of the Android system's `ActivityManager` or by sending an implicit `Intent`. In the Android system, application components (such as services and activities) require a series of initialization operations during startup. Completing some initialization in advance can reduce the time overhead when the application actually starts.

[0107] Pre-caching the core files of the first application can include instructing the file system to pre-cachate the application's core library files (.so files) and asset files into the memory file system. Simultaneously, leveraging the Linux readahead mechanism, which allows file data to be read from disk into memory in advance, reduces disk I / O latency. Since disk I / O operations are typically time-consuming, pre-caching and the readahead mechanism enable applications to obtain the necessary file data more quickly at startup.

[0108] The above methods have several advantages. First, the strategy management module generates pre-loading instructions to prepare for application startup in advance, making the entire startup process more orderly. The system can plan resource allocation in the background in advance, such as reserving memory space and scheduling related processes. When the user actually starts the application, since the preliminary preparation work has been completed, the application can quickly enter the startup process, skipping the initial preparation stage, thereby effectively shortening the startup time. Second, early component initialization reduces overhead. Specifically, in the Android system, the initialization of application components is a critical and time-consuming step in the startup process. By calling the ActivityManager's bindService() method or sending implicit Intents to trigger the initialization of some components in advance, these operations can be completed during system idle periods. When the user initiates a startup request, the application only needs to complete the remaining necessary initialization steps, greatly reducing the time overhead during startup and speeding up the application startup. Third, core file caching reduces I / O latency. Specifically, disk I / O operations are one of the main bottlenecks in the application startup process. By pre-caching the core library files (.so files) and asset files of the first application to the memory file system and leveraging the Linux readahead mechanism to read the file data from disk into memory in advance, frequent disk read operations during application startup can be avoided. In this way, the application can directly obtain the necessary file data from memory at startup, significantly reducing disk I / O latency and thus speeding up application startup.

[0109] The following are embodiments of the apparatus described in this application, which can be used to execute the embodiments of the method described in this application. For details not disclosed in the apparatus embodiments of this application, please refer to the embodiments of the method described in this application.

[0110] Please refer to Figure 4 This diagram illustrates a block diagram of a vehicle-mounted system resource allocation device according to an embodiment of this application. The device has the functions described above, which can be implemented in hardware or by hardware executing corresponding software. The device can be the computer device described above, or it can be installed within a computer device. Figure 4 As shown, the device 400 may include an acquisition module 410, a first determination module 420, a generation module 430, and a second determination module 440.

[0111] The acquisition module 410 is used to acquire the vehicle system's collected data and historical data. The collected data is used to indicate the vehicle system's operating status and interaction status in a first time unit. The interaction status is used to indicate the operational behavior of interacting with the vehicle system. The historical data is used to indicate the vehicle system's usage of at least one application in a historical period.

[0112] The first determining module 420 is used to determine the scene mode of the vehicle system in the first time unit based on the collected data.

[0113] The generation module 430 is used to generate a prediction result based on the historical data and the collected data. The prediction result is used to indicate the first application that the vehicle system will start in a second time unit, which is a time unit after the first time unit.

[0114] The second determining module 440 is used to determine the resource allocation strategy of the vehicle system in the first time unit based on the scene mode and the prediction result. The resource allocation strategy is used to indicate the resource allocation method for at least one task of the vehicle system.

[0115] In some embodiments, the first determining module 420 is used to determine the scene mode from at least one candidate scene mode based on condition information and the collected data; wherein the condition information is used to indicate a preset condition corresponding to at least one of the running state and the interaction state corresponding to the at least one candidate scene mode.

[0116] In some embodiments, the condition information includes at least one condition expression, which indicates a preset condition corresponding to at least one of the running state and the interaction state of a candidate scene mode.

[0117] In some embodiments, the device 400 further includes: a startup module ( Figure 4 (Not shown in the image).

[0118] The startup module is used to preload the configuration data corresponding to the first application into the memory of the vehicle system; in response to the startup operation of the first application, the first application is launched in the vehicle system based on the configuration data corresponding to the first application.

[0119] In some embodiments, the generation module 430 is used to determine the probability value of each of the at least one application based on the historical data and the collected data, wherein the probability value of the application is used to indicate the probability that the vehicle system will launch the application in the second time unit; wherein the collected data includes at least one of the following: the currently running application of the vehicle system, the current first location of the vehicle system, and the current first time of the vehicle system; and the probability result is determined based on the probability value of each of the at least one application.

[0120] In some embodiments, the historical data includes at least one entry, each entry including at least one of the following: application identification information, application location information, and application time information; wherein the identification information is used to distinguish different applications among the at least one application, the location information is used to indicate the location of the vehicle system when running the application, and the time information is used to indicate the time period during which the vehicle system runs the application.

[0121] In some embodiments, the device 400 further includes: an adjustment module ( Figure 4 (Not shown).

[0122] The adjustment module is used to determine the performance compliance information of at least one task after scheduling the vehicle system to execute the resource allocation scheme. The performance compliance information is used to indicate the degree of deviation between the performance of at least one task and the expected performance. Based on the performance compliance information, the resource allocation strategy is adjusted to obtain an adjusted resource allocation strategy. The adjusted resource allocation strategy is used to redetermine the resource allocation method of at least one task.

[0123] In some embodiments, the collected data includes at least one of the following: user interaction data, vehicle status data, and system resource data.

[0124] It should be noted that the apparatus provided in the above embodiments is only illustrated by the division of the above functional modules when implementing its functions. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the content structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0125] Please refer to Figure 5 The diagram shows a structural block diagram of a computer device 500 provided in one embodiment of this application.

[0126] Typically, computer device 500 includes a processor 510 and a memory 520.

[0127] Processor 510 may include one or more processing cores, such as a quad-core processor, an octa-core processor, etc. Processor 510 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field Programmable Gate Array), and PLA (Programmable Logic Array). Processor 510 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 510 may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, processor 510 may also include an AI processor for handling computational operations related to machine learning.

[0128] The memory 520 may include one or more computer-readable storage media, which may be non-transitory. The memory 520 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 520 are used to store a computer program configured to be executed by one or more processors to implement the above-described method for allocating vehicle system resources.

[0129] Those skilled in the art will understand that Figure 5 The structure shown does not constitute a limitation on the computer device 500, and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0130] In an exemplary embodiment, a computer-readable storage medium is also provided, wherein a computer program is stored in the storage medium, and the computer program, when executed by a processor, implements the above-described method for allocating vehicle system resources. Optionally, the computer-readable storage medium may include: ROM (Read-Only Memory), RAM (Random Access Memory), SSD (Solid State Drives), or optical disc, etc. The random access memory may include ReRAM (Resistance Random Access Memory) and DRAM (Dynamic Random Access Memory).

[0131] In an exemplary embodiment, a computer program product is also provided, the computer program product including a computer program stored in a computer-readable storage medium. A processor of a computer device reads the computer program from the computer-readable storage medium, and the processor executes the computer program, causing the terminal device to perform the above-described vehicle resource allocation method.

[0132] It should be noted that the collection and processing of relevant data (including historical data, collected data, etc.) in this application should strictly comply with the requirements of relevant national laws and regulations, obtain the informed consent or separate consent of the personal information subject, and carry out subsequent data use and processing within the scope of laws and regulations and the authorization of the personal information subject.

[0133] It should be understood that "multiple" as used herein refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. Furthermore, the step numbers described herein are merely illustrative of one possible execution order. In some other embodiments, the steps may not be executed in the order shown in the figures, such as two steps with different numbers being executed simultaneously, or two steps with different numbers being executed in the reverse order of the figures. This application does not limit this.

[0134] The above description is merely an exemplary embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.< / pid> < / pid> < / pid>

Claims

1. A method for allocating vehicle infotainment system resources, characterized in that, The method includes: Acquire the vehicle's data collection data and historical data. The data collection data is used to indicate the vehicle's operating status and interaction status in a first time unit. The interaction status is used to indicate the operation behavior of interacting with the vehicle. The historical data is used to indicate the vehicle's usage of at least one application in a historical period. Based on the collected data, the scene mode of the vehicle system in the first time unit is determined; Based on the historical data and the collected data, a prediction result is generated. The prediction result is used to indicate the first application that the vehicle system will launch in the second time unit, which is the time unit after the first time unit. Based on the scenario pattern and the prediction results, a resource allocation strategy for the vehicle system in the first time unit is determined. The resource allocation strategy is used to indicate the resource allocation method for at least one task of the vehicle system.

2. The method according to claim 1, characterized in that, Determining the scene mode of the vehicle system in the first time unit based on the collected data includes: Based on the condition information and the collected data, the scene mode is determined from at least one candidate scene mode; wherein the condition information is used to indicate a preset condition corresponding to at least one of the running state and the interaction state corresponding to the at least one candidate scene mode.

3. The method according to claim 2, characterized in that, The condition information includes at least one condition expression, which indicates a preset condition corresponding to at least one of the running state and the interaction state of a candidate scene mode.

4. The method according to claim 1, characterized in that, The method further includes: The configuration data corresponding to the first application is preloaded into the memory of the vehicle system; In response to the launch operation for the first application, the first application is launched in the vehicle system based on the configuration data corresponding to the first application.

5. The method according to claim 1, characterized in that, The process of generating prediction results based on the historical data and the collected data includes: Based on the historical data and the collected data, a probability value for each of the at least one application is determined. The probability value of the application is used to indicate the probability that the vehicle system will launch the application in the second time unit. The collected data includes at least one of the following: the currently running application of the vehicle system, the current first location of the vehicle system, and the current first time of the vehicle system. The probability result is determined based on the probability values ​​of each of the at least one application.

6. The method according to claim 1, characterized in that, The historical data includes at least one entry, each entry including at least one of the following: application identification information, application location information, and application time information; wherein, the identification information is used to distinguish different applications among the at least one application, the location information is used to indicate the location of the vehicle system when running the application, and the time information is used to indicate the time period during which the vehicle system runs the application.

7. A device for allocating vehicle-mounted system resources, characterized in that, The device includes: The acquisition module is used to acquire the vehicle's collected data and historical data. The collected data is used to indicate the vehicle's operating status and interaction status in a first time unit. The interaction status is used to indicate the operation behavior of interacting with the vehicle. The historical data is used to indicate the vehicle's usage of at least one application in a historical period. The first determining module is used to determine the scene mode of the vehicle system in the first time unit based on the collected data; The generation module is used to generate a prediction result based on the historical data and the collected data. The prediction result is used to indicate the first application that the vehicle system will start in the second time unit, which is the time unit after the first time unit. The second determining module is used to determine the resource allocation strategy of the vehicle system in the first time unit based on the scene mode and the prediction result. The resource allocation strategy is used to indicate the resource allocation method for at least one task of the vehicle system.

8. A computer device, characterized in that, The computer device includes a processor and a memory, the memory storing a computer program, which is loaded and executed by the processor to implement the vehicle resource allocation method as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which is loaded and executed by a processor to implement the vehicle system resource allocation method as described in any one of claims 1 to 6.

10. A computer program product, characterized in that, The computer program product includes a computer program stored in a computer-readable storage medium, and a processor reads from and executes the computer program to implement the vehicle system resource allocation method as described in any one of claims 1 to 6.

Images