Vehicle-mounted process optimization method and device, electronic equipment and computer storage medium

Abstract

The invention provides a vehicle-mounted process optimization method and device, electronic equipment and a computer storage medium, relates to the technical field of automobiles, and is applied to a vehicle control system. The method comprises the following steps: determining a vehicle panoramic data vector based on multi-dimensional vehicle running state panoramic data; inputting the vehicle panoramic data vector into an intelligent scene recognition model, outputting a vehicle operation activation scene and querying a corresponding target vehicle-mounted resource allocation strategy to schedule vehicle operation resources and process operation modes; by fusing multi-dimensional vehicle running state panoramic data and utilizing an intelligent model to actively identify a driving scene, resource scheduling is converted from static and universal to dynamic and accurate; according to the method, the system resources are finely distributed according to the real-time scene, the real-time performance and the fluency of the key tasks are effectively guaranteed, the processing capacity of the system in the multi-task concurrent scene is optimized, and therefore the response speed, the resource utilization efficiency and the overall user experience of the vehicle-mounted system are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of automotive technology, and in particular to an in-vehicle process optimization method, apparatus, electronic device, and computer storage medium. Background Technology

[0002] Currently, in-vehicle systems based on the Android platform typically employ a common process lifecycle management strategy to allocate system resources. This strategy primarily relies on the static running state of processes (such as foreground, background, and service) and their importance, configuring default CPU priorities and cgroup (resource container) affiliations for them.

[0003] However, in intelligent cockpit environments with stringent requirements for real-time performance, smoothness, and multi-task concurrent processing capabilities, this static and universal scheduling mechanism has significant limitations. First, its rigid priority configuration fails to finely differentiate between critical tasks in the foreground (such as navigation and voice interaction), potentially leading to response delays caused by resource contention. Second, existing resource guarantee mechanisms are insufficient; many latency-sensitive background tasks (such as voice wake-up word detection) are easily subject to resource constraints under high loads, resulting in function interruptions. Finally, the general strategy cannot perceive dynamic changes in the driving situation and lacks the ability to anticipate and respond to sudden resource demands, thus impacting user experience and driving safety. Summary of the Invention

[0004] The purpose of this invention is to provide a method, device, electronic device, and computer storage medium for optimizing vehicle processes. By integrating multi-dimensional panoramic data of vehicle operation status and actively identifying driving scenarios using intelligent models, it realizes the transformation of resource scheduling from "static and universal" to "dynamic and precise". It allocates system resources in a refined manner according to real-time scenarios, effectively ensuring the real-time performance and smoothness of critical tasks, optimizing the system's processing capabilities in multi-task concurrent scenarios, thereby significantly improving the response speed, resource utilization efficiency, and overall user experience of the vehicle system.

[0005] In a first aspect, the present invention provides an in-vehicle process optimization method, applied to a vehicle control system, the method comprising: Acquire multi-dimensional panoramic data of vehicle operating status; The multi-dimensional panoramic data of vehicle operation status is preprocessed to obtain a vehicle panoramic data vector. The vehicle panoramic data vector is input into a pre-trained intelligent scene recognition model, which outputs the vehicle operation activation scene. Query the target vehicle resource allocation strategy corresponding to the vehicle operation activation scenario in the pre-set strategy mapping table; The vehicle operation resources and process operation modes are scheduled based on the target on-board resource allocation strategy.

[0006] In some preferred embodiments of the present invention, the multidimensional vehicle operation status panoramic data includes: vehicle status data, user status data, and environmental status data; the vehicle status data represents the real-time dynamic performance and operating status of the vehicle itself; the user status data represents the user's physiological state, behavioral intentions, and interaction needs; and the environmental status data represents the external environment and traffic situation in which the vehicle is located.

[0007] In some preferred embodiments of the present invention, preprocessing includes: standardization processing, noise reduction processing, and synchronization processing; the step of preprocessing the multi-dimensional vehicle operating state panoramic data to obtain a vehicle panoramic data vector includes: Standardize the multi-dimensional panoramic data of vehicle operation status to eliminate the influence of dimensions; The standardized multi-dimensional panoramic data of vehicle operation status is denoised based on filtering algorithms and outlier detection. The noise-reduced multi-dimensional panoramic data of vehicle operation status is synchronized based on timestamps to obtain a vehicle panoramic data vector.

[0008] In some preferred embodiments of the present invention, the intelligent scene recognition model includes: a vehicle operation scene inference sub-model and a confidence determination sub-model; The vehicle operation scenario inference sub-model is used to infer the current vehicle operation scenario and the corresponding confidence score based on the vehicle panoramic data vector; The confidence score determination sub-model is used to determine the current vehicle operation scenario corresponding to the confidence score as the vehicle operation activation scenario if the confidence score is greater than a preset threshold.

[0009] In some preferred embodiments of the present invention, the method further includes: training a vehicle operation scenario inference sub-model through the following steps: Historical multidimensional panoramic data of vehicle operation status is acquired, and preprocessed and labeled to obtain historical vehicle panoramic data vectors; among them, the labeling process is used to map the historical vehicle panoramic data vectors to preset operation scene labels. The training set containing historical vehicle panoramic data vectors is input into the pre-established basic vehicle operation scenario inference sub-model. The parameters of the basic vehicle operation scenario inference sub-model are iteratively optimized until the preset training completion conditions are met. The validation set containing historical vehicle panoramic data vectors is input into the trained basic vehicle operation scenario inference sub-model for model performance validation. The basic vehicle operation scenario inference sub-model that passes the validation is determined as the vehicle operation scenario inference sub-model.

[0010] In some preferred embodiments of the present invention, the strategy mapping table includes: a one-to-one correspondence between vehicle operation scenarios and on-board resource allocation strategies; the method further includes: External data is received via wired or wireless means to update the policy mapping table.

[0011] In some preferred embodiments of the present invention, vehicle operating resources include: task process priority, resource containers, and threads; different modes correspond to different vehicle screen display interfaces; the steps of scheduling vehicle operating resources and process running modes based on the target vehicle resource allocation strategy include: Prioritize critical task processes to a higher priority than background task processes; higher-priority task processes can obtain computing resources first. Resource containers are allocated according to the priority of the task processes; where the resource quota of the resource container corresponding to the task process with higher priority is greater than the resource quota of the resource container corresponding to the task process with lower priority. Thread scheduling is based on task priority and corresponding resource containers; Adjust the display interface of the vehicle screen.

[0012] Secondly, the present invention provides an in-vehicle process optimization device, applied to a vehicle control system, wherein the device includes: The data acquisition module is used to acquire multi-dimensional panoramic data of vehicle operating status; The data processing module is used to preprocess the multi-dimensional panoramic data of vehicle operation status to obtain a vehicle panoramic data vector. The scene judgment module is used to input the vehicle panoramic data vector into the pre-trained intelligent scene recognition model and output the vehicle running activation scene. The resource allocation strategy determination module is used to query the target vehicle resource allocation strategy corresponding to the vehicle operation activation scenario in a pre-set strategy mapping table. The resource scheduling module is used to schedule vehicle operating resources and process operation modes based on the target on-board resource allocation strategy.

[0013] Thirdly, the present invention provides an electronic device including a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the vehicle process optimization method provided in the first aspect above.

[0014] Fourthly, the present invention provides a computer storage medium storing computer-executable instructions, which, when called and executed by a processor, cause the processor to implement the vehicle process optimization method provided in the first aspect.

[0015] The invention has brought about the following beneficial effects: This invention provides a method, apparatus, electronic device, and computer storage medium for optimizing vehicle processes, applied to a vehicle control system. The method includes: acquiring multi-dimensional panoramic data of vehicle operating status; preprocessing the multi-dimensional panoramic data of vehicle operating status to obtain a vehicle panoramic data vector; inputting the vehicle panoramic data vector into a pre-trained intelligent scene recognition model to output a vehicle operating activation scene; querying a target vehicle resource allocation strategy corresponding to the vehicle operating activation scene in a pre-set strategy mapping table; scheduling vehicle operating resources and process operation modes based on the target vehicle resource allocation strategy; by fusing multi-dimensional panoramic data of vehicle operating status and actively recognizing driving scenes using an intelligent model, the resource scheduling is transformed from "static and universal" to "dynamic and precise"; the system resources are allocated in a refined manner according to the real-time scene, effectively ensuring the real-time performance and smoothness of key tasks, optimizing the system's processing capability in multi-task concurrent scenarios, thereby significantly improving the response speed, resource utilization efficiency, and overall user experience of the vehicle system. Attached Figure Description

[0016] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 A flowchart of an in-vehicle process optimization method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an in-vehicle process optimization device provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0018] Icons: 310 - Data Acquisition Module; 320 - Data Processing Module; 330 - Scene Judgment Module; 340 - Resource Allocation Strategy Determination Module; 350 - Resource Scheduling Module; 400 - Memory; 401 - Processor; 402 - Bus; 403 - Communication Interface. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0020] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0021] It should be noted that similar labels and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0022] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings, or the orientation or positional relationship commonly used when the product of this invention is in use. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. In addition, the terms "first," "second," "third," etc., are only used to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0023] Furthermore, terms such as "horizontal," "vertical," and "sag" do not imply that components must be absolutely horizontal or suspended, but rather that they can be slightly tilted. For example, "horizontal" simply means that its direction is more horizontal relative to "vertical," and does not mean that the structure must be completely horizontal, but can be slightly tilted.

[0024] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0025] The following detailed description of some embodiments of the present invention is provided in conjunction with the accompanying drawings. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0026] Example 1 This invention provides an in-vehicle process optimization method, applied to a vehicle control system. (See also...) Figure 1 The flowchart shown in this embodiment of the invention provides a method for optimizing in-vehicle processes, the method including: Step S102: Obtain multi-dimensional panoramic data of vehicle operating status.

[0027] Specifically, the vehicle collects three main categories of data in real time, forming a comprehensive picture of its operational status, through various sensors and data interfaces integrated within the vehicle. Vehicle status data is acquired at millisecond-level frequencies via the CAN bus interface, directly representing the vehicle's real-time dynamic performance and operational status. This includes, but is not limited to, vehicle speed, engine speed, throttle / brake depth, gear position, headlight status, and wiper status. User status data is acquired through in-vehicle cameras and microphone arrays, combined with sensor data (such as heart rate wristbands), to deeply characterize the driver's (or passenger's) physiological state, behavioral intentions, and interaction needs. This includes: driver facial expressions analyzed through facial recognition models (for assessing fatigue, distraction, etc.), voice commands, eye-tracking information, and historical operating habits. Environmental status data originates from the GPS module, the vehicle networking unit (T-Box), and environmental sensors, comprehensively depicting the external physical world and traffic conditions in which the vehicle operates, such as: geographical location, traffic conditions (congestion / smooth traffic), weather information, external light intensity, and temperature. By integrating these three types of heterogeneous data, the system constructs a panoramic information view necessary for understanding the current driving situation, providing a data foundation for subsequent intelligent scene recognition.

[0028] Furthermore, in some preferred embodiments of the present invention, the multidimensional vehicle operation status panoramic data includes: vehicle status data, user status data, and environmental status data; the vehicle status data represents the real-time dynamic performance and operating status of the vehicle itself; the user status data represents the user's physiological state, behavioral intentions, and interaction needs; and the environmental status data represents the external environment and traffic situation in which the vehicle is located.

[0029] Specifically, these three types of data, each with its own characteristics and complementing each other, together form the cornerstone of the system's perception of the driving environment. Vehicle status data is acquired through the CAN bus interface; its high frequency (millisecond level) and real-time nature accurately reflect the vehicle's instantaneous operation and dynamic performance, serving as a direct basis for determining "what the vehicle is doing." User status data is acquired through in-vehicle cameras, microphone arrays, and optional wearable devices, transforming the driver from an external controller into a dynamically perceptible and understandable variable within the system, which is key to achieving personalized services and human-machine co-driving. Environmental status data is acquired through GPS, T-Box, and environmental sensors, constructing a contextual information field for the vehicle to perceive and make decisions, defining "under what conditions the vehicle is operating." The organic combination of these three types of data enables the system to comprehensively and accurately grasp the overall state of the "human-vehicle-environment" closed-loop system.

[0030] Step S104: Preprocess the multi-dimensional vehicle operation status panoramic data to obtain the vehicle panoramic data vector.

[0031] Specifically, due to the diverse sources, varying formats, and noise inherent in the raw data, meticulous preprocessing is essential to create unified features suitable for efficient model learning. This step involves the initial integration and cleaning of the collected multi-source heterogeneous data, providing high-quality, well-organized input for subsequent scene recognition models. For example, vehicle data from the CAN bus may have a frequency as high as 100Hz, while user status data collected by cameras has a frequency of 30Hz, and GPS data may only have a frequency of 1Hz. These frequency differences and potential transmission delays must be addressed through preprocessing to ensure the accuracy and reliability of subsequent analysis.

[0032] Furthermore, in some preferred embodiments of the present invention, the preprocessing includes: standardization processing, noise reduction processing, and synchronization processing; the step of preprocessing the multi-dimensional vehicle operation state panoramic data to obtain a vehicle panoramic data vector includes: standardizing the multi-dimensional vehicle operation state panoramic data to eliminate the influence of dimensions; performing noise reduction processing on the standardized multi-dimensional vehicle operation state panoramic data based on filtering algorithms and outlier detection; and performing synchronization processing on the noise-reduced multi-dimensional vehicle operation state panoramic data based on timestamps to obtain a vehicle panoramic data vector.

[0033] Specifically, each of the three stages in the preprocessing process has its unique role and technical details. Standardization aims to transform data with different dimensions and numerical ranges (such as vehicle speed 0-200 km / h and steering wheel angle 0-360 degrees) to the same reference scale to eliminate the influence of dimensions and lay the foundation for subsequent fusion calculations. Noise reduction, based on filtering algorithms (such as Kalman filtering) and outlier detection technology, cleans the standardized data, identifying and eliminating unreliable data points and noise caused by sensor errors or transient interference, thereby extracting the effective components that truly reflect the physical signals. Synchronization processing is the final crucial step. It uses precise timestamps on each data point to solve the time delay problem in the acquisition and transmission of data streams from different sensors. Through interpolation and alignment methods, it aligns all features to a unified time reference, ultimately forming a high-dimensional vehicle panoramic data vector that is consistent in scale, quality, and time. This vector completely describes the overall state of the system at a specific moment.

[0034] Step S106: Input the vehicle panoramic data vector into the pre-trained intelligent scene recognition model and output the vehicle operation activation scene.

[0035] Specifically, this step is the core of achieving intelligent scheduling. The intelligent scene recognition model is a pre-trained machine learning model that receives standardized vehicle panoramic data vectors as input and outputs the most likely driving scenario. This model can learn the complex correlations between multi-dimensional data. For example, when features such as "high speed," "late at night," and "facial expression showing fatigue" occur simultaneously, the model can identify the complex scenario of "high-speed nighttime fatigued cruising." This model-based reasoning approach overcomes the limitations of traditional rule-based methods in covering all complex scenario combinations, achieving proactive and accurate perception of driving situations.

[0036] Furthermore, in some preferred embodiments of the present invention, the intelligent scene recognition model includes: a vehicle operation scene inference sub-model and a confidence score determination sub-model; the vehicle operation scene inference sub-model is used to infer the current vehicle operation scene and the confidence score corresponding to the current vehicle operation scene based on the vehicle panoramic data vector; the confidence score determination sub-model is used to determine the current vehicle operation scene corresponding to the confidence score as the vehicle operation active scene if the confidence score is greater than a preset threshold.

[0037] Specifically, this dual-sub-model design ensures both accurate identification and a flexible decision-making mechanism, which is key to improving system reliability. The vehicle operation scenario inference sub-model (e.g., a pre-trained deep neural network (DNN) or recurrent neural network (RNN)) is responsible for extracting patterns from high-dimensional feature vectors and calculating the confidence score for each preset scenario. The confidence score determination sub-model acts as a decision filter, setting a preset threshold. Only when the confidence score of the inference result exceeds this threshold will the scenario be recognized as an active vehicle operation scenario and subsequent scheduling be triggered. For example, the model might calculate a confidence score of 92% for the "urban congestion following" scenario (exceeding the 85% threshold), in which case the system would determine that scenario is active. This mechanism effectively prevents false identifications with low confidence scores from causing unnecessary or even erroneous scheduling behavior, thus improving system robustness.

[0038] Furthermore, in some preferred embodiments of the present invention, the method further includes: training a vehicle operation scenario inference sub-model through the following steps: acquiring historical multi-dimensional vehicle operation state panoramic data, and performing preprocessing and labeling to obtain historical vehicle panoramic data vectors; wherein, the labeling process is used to map the historical vehicle panoramic data vectors to preset operation scenario labels; inputting the training set containing the historical vehicle panoramic data vectors into a pre-established basic vehicle operation scenario inference sub-model, and iteratively optimizing the parameters of the basic vehicle operation scenario inference sub-model until the preset training completion conditions are met; inputting the validation set containing the historical vehicle panoramic data vectors into the trained basic vehicle operation scenario inference sub-model for model performance validation, and determining the basic vehicle operation scenario inference sub-model that passes the validation as the vehicle operation scenario inference sub-model.

[0039] Specifically, model training is an offline, systematic, and rigorous process. First, massive amounts of historical multi-dimensional vehicle operational status panoramic data need to be collected and preprocessed using the methods described above. Next, labeling is performed—a laborious but crucial step—where human experts or rules are used to label each processed data segment (historical vehicle panoramic data vector) with a pre-defined operational scenario label (such as "high-speed cruising," "urban congestion," "fatigue driving," etc.), thus forming a high-quality labeled dataset. This dataset is divided into training, validation, and test sets. The training set is used to iteratively optimize the parameters of the base model (such as an initialized DNN or RNN) using algorithms like backpropagation and gradient descent, aiming to minimize the loss between the model's predictions and the true labels (such as cross-entropy loss). The validation set is used to monitor the model's performance on unseen data during training (i.e., model performance validation), and to adjust hyperparameters or implement early stopping techniques to prevent overfitting based on performance. Finally, a one-time evaluation is performed on the test set. Once the performance meets the standards, the model is considered trained and can be deployed to the vehicle. This rigorous training process ensures that the model can accurately learn the mapping relationship from complex data to specific scenarios.

[0040] Step S108: Query the target vehicle resource allocation strategy corresponding to the vehicle operation activation scenario in the pre-set strategy mapping table.

[0041] Specifically, the policy mapping table is a predefined, static "condition-action" query hub, serving as a bridge connecting scenario recognition and scheduling execution. It takes the vehicle operation activation scenario output in step S106 as its sole input and, through a table lookup, maps the abstract driving scenario into a specific, executable set of target vehicle resource allocation policies. For example, when the activation scenario is "high-speed nighttime fatigue cruise," the retrieved policies might include: "increase the CPU priority of the navigation application and alarm service to RT_PRIO," or "place the relevant processes in a cgroup with higher resource quotas." This design makes the decision-making process transparent, efficient, and deterministic.

[0042] Furthermore, in some preferred embodiments of the present invention, the strategy mapping table includes: a one-to-one correspondence between vehicle operation scenarios and on-board resource allocation strategies; the method further includes: receiving external data via wired or wireless means to update the strategy mapping table.

[0043] Specifically, the core content of the policy mapping table is a one-to-one correspondence between vehicle operation scenarios and onboard resource allocation strategies. A key advantage and feature of this policy library lies in its updability. The system can receive external data via wired or wireless means (such as OTA over-the-air download), thereby enabling online updates and optimizations of the policy mapping table. This allows the system to flexibly adapt to new business requirements, user experience feedback, or the discovery of better scheduling strategies without modifying the underlying onboard system software or recalling vehicles, greatly improving the system's flexibility and continuous optimization capabilities throughout its lifecycle.

[0044] Step S110: Schedule vehicle operating resources and process operation modes based on the target vehicle resource allocation strategy.

[0045] Specifically, this step is the final execution phase of the strategy. It transforms abstract scheduling instructions into actual system behaviors, directly impacting the in-vehicle computing platform. Based on the queried target in-vehicle resource allocation strategy, the system dynamically and finely adjusts vehicle operating resources (including CPU / GPU priority, cgroup affiliation, and thread scheduling) and process execution modes (often manifested as different vehicle screen display interfaces and functional services). Its fundamental purpose is to ensure that, in specific scenarios, critical tasks always obtain the necessary computing resources, guaranteeing the system's real-time performance, smoothness, and security.

[0046] Furthermore, in some preferred embodiments of the present invention, vehicle operating resources include: task process priorities, resource containers, and threads; different modes correspond to different vehicle screen display interfaces; the steps of scheduling vehicle operating resources and process operating modes based on the target vehicle resource allocation strategy include: increasing the priority of critical task processes to be greater than the priority of background task processes; wherein, task processes with higher priority can obtain computing resources first; allocating resource containers according to the priority order of task processes; wherein, the resource quota of the resource container corresponding to a higher priority task process is greater than the resource quota of the resource container corresponding to a lower priority task process; performing thread scheduling based on the priority of the task and the corresponding resource container; and adjusting the display interface of the vehicle screen.

[0047] Specifically, resource scheduling involves multi-layered collaborative work, primarily implemented through Android kernel interfaces or APIs. Adjusting scheduling priorities: For example, when a "voice interaction" scenario is activated, the thread priority of voice recognition and processing services is immediately elevated to a real-time scheduling strategy (such as SCHED_FIFO) to ensure that its computational requests can be responded to by the CPU immediately, avoiding stuttering and latency caused by resource contention. Dividing into control groups (cgroups): For example, high-priority tasks (such as navigation path rendering and sensor fusion algorithms) are placed in cgroups with more abundant resource quotas to guarantee their CPU time slices, memory bandwidth, and I / O weight; at the same time, background tasks (such as app store updates and data synchronization) are restricted to cgroups with limited resources to fundamentally prevent them from excessively consuming system resources and affecting foreground critical tasks. Optimizing thread scheduling: The system scheduler intelligently performs thread migration between CPU cores, core binding (such as binding critical threads to high-performance large cores), and load balancing based on task process priority and cgroup settings, while strictly enforcing priority-based preemption rules to ensure that high-priority threads can immediately preempt low-priority tasks. Adjust the display interface of the vehicle screen: trigger corresponding functions and services according to the strategy, such as calling the audio interface to switch the audio source, or notifying the UI framework to adjust the interface layout and color mode through system-level broadcast (such as switching to dark mode in night driving scenarios).

[0048] This invention provides an in-vehicle process optimization method applied to a vehicle control system. The method includes: acquiring multi-dimensional panoramic data of vehicle operation status; preprocessing the multi-dimensional panoramic data of vehicle operation status to obtain a vehicle panoramic data vector; inputting the vehicle panoramic data vector into a pre-trained intelligent scene recognition model to output a vehicle operation activation scene; querying a target in-vehicle resource allocation strategy corresponding to the vehicle operation activation scene in a pre-set strategy mapping table; scheduling vehicle operation resources and process operation modes based on the target in-vehicle resource allocation strategy; by fusing multi-dimensional panoramic data of vehicle operation status and actively recognizing driving scenes using an intelligent model, the method achieves a transformation of resource scheduling from "static and universal" to "dynamic and precise"; and by finely allocating system resources according to real-time scenes, it effectively ensures the real-time performance and smoothness of key tasks, optimizes the system's processing capability in multi-task concurrent scenarios, and thus significantly improves the response speed, resource utilization efficiency, and overall user experience of the in-vehicle system.

[0049] Example 2 Based on the above embodiments, this invention provides an in-vehicle process optimization device, see [link to previous embodiment]. Figure 2 The diagram shown is a structural schematic of an in-vehicle process optimization device provided in an embodiment of the present invention. The device includes: Data acquisition module 310 is used to acquire multi-dimensional panoramic data of vehicle operating status; Data processing module 320 is used to preprocess multi-dimensional vehicle operation status panoramic data to obtain vehicle panoramic data vector; The scene judgment module 330 is used to input the vehicle panoramic data vector into the pre-trained intelligent scene recognition model and output the vehicle running activation scene. The resource allocation strategy determination module 340 is used to query the target vehicle resource allocation strategy corresponding to the vehicle operation activation scenario in a pre-set strategy mapping table. The resource scheduling module 350 is used to schedule vehicle operating resources and process operation modes based on the target on-board resource allocation strategy.

[0050] Furthermore, in some preferred embodiments of the present invention, the multidimensional vehicle operation status panoramic data includes: vehicle status data, user status data, and environmental status data; the vehicle status data represents the real-time dynamic performance and operating status of the vehicle itself; the user status data represents the user's physiological state, behavioral intentions, and interaction needs; and the environmental status data represents the external environment and traffic situation in which the vehicle is located.

[0051] Furthermore, in some preferred embodiments of the present invention, the preprocessing includes: standardization processing, noise reduction processing, and synchronization processing; the data processing module 320 is used to standardize the multidimensional vehicle operation status panoramic data to eliminate the influence of dimensions; to perform noise reduction processing on the standardized multidimensional vehicle operation status panoramic data based on filtering algorithms and outlier detection; and to perform synchronization processing on the noise-reduced multidimensional vehicle operation status panoramic data based on timestamps to obtain a vehicle panoramic data vector.

[0052] Furthermore, in some preferred embodiments of the present invention, the intelligent scene recognition model includes: a vehicle operation scene inference sub-model and a confidence score determination sub-model; the vehicle operation scene inference sub-model is used to infer the current vehicle operation scene and the confidence score corresponding to the current vehicle operation scene based on the vehicle panoramic data vector; the confidence score determination sub-model is used to determine the current vehicle operation scene corresponding to the confidence score as the vehicle operation active scene if the confidence score is greater than a preset threshold.

[0053] Furthermore, in some preferred embodiments of the present invention, the apparatus further includes: a model training module, used to acquire historical multi-dimensional panoramic data of vehicle operation status, and perform preprocessing and labeling to obtain historical vehicle panoramic data vectors; wherein, the labeling process is used to map the historical vehicle panoramic data vectors to preset operation scene labels; the training set containing the historical vehicle panoramic data vectors is input into a pre-established basic vehicle operation scene inference sub-model, and the parameters of the basic vehicle operation scene inference sub-model are iteratively optimized until the preset training completion conditions are met; the validation set containing the historical vehicle panoramic data vectors is input into the trained basic vehicle operation scene inference sub-model for model performance validation, and the basic vehicle operation scene inference sub-model that passes the validation is determined as the vehicle operation scene inference sub-model.

[0054] Furthermore, in some preferred embodiments of the present invention, the strategy mapping table includes: a one-to-one correspondence between vehicle operation scenarios and vehicle resource allocation strategies; the device further includes: a strategy mapping table update module, used to receive external data via wired or wireless means to update the strategy mapping table.

[0055] Furthermore, in some preferred embodiments of the present invention, vehicle operating resources include: task process priorities, resource containers, and threads; different modes correspond to different vehicle screen display interfaces; a resource scheduling module 350 is used to increase the priority of critical task processes to be greater than the priority of background task processes; wherein, task processes with higher priority can obtain computing resources first; resource containers are allocated according to the priority order of task processes; wherein, the resource quota of the resource container corresponding to a higher priority task process is greater than the resource quota of the resource container corresponding to a lower priority task process; thread scheduling is performed based on the priority of the task and the corresponding resource container; and the display interface of the vehicle screen is adjusted.

[0056] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the vehicle process optimization device described above can be referred to the corresponding process in the embodiments of the aforementioned vehicle process optimization method, and will not be repeated here.

[0057] Example 3 This invention also provides an electronic device for running an in-vehicle process optimization method; see [link to related documentation]. Figure 3 The schematic diagram of an electronic device provided by the embodiment of the present invention shown above includes a memory 400 and a processor 401. The memory 400 is used to store one or more computer instructions, which are executed by the processor 401 to implement the above-mentioned vehicle process optimization method.

[0058] Furthermore, Figure 3 The electronic device shown also includes a bus 402 and a communication interface 403. The processor 401, the communication interface 403 and the memory 400 are connected via the bus 402.

[0059] The memory 400 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 403 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc. The bus 402 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 3 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0060] Processor 401 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 401 or by instructions in software form. Processor 401 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly manifested as execution by a hardware decoding processor, or execution by a combination of hardware and software modules in the decoding processor. The software module can reside in a readily available storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory 400, and processor 401 reads information from memory 400 and, in conjunction with its hardware, completes the steps of the method described in the foregoing embodiments.

[0061] This invention also provides a computer storage medium storing computer-executable instructions. When these computer-executable instructions are called and executed by a processor, they cause the processor to implement the above-described vehicle process optimization method. For specific implementation details, please refer to the method embodiments, which will not be repeated here.

[0062] The computer program products of the vehicle process optimization method, apparatus and electronic device provided in the embodiments of the present invention include a computer storage medium storing program code. The instructions included in the program code can be used to execute the methods in the preceding method embodiments. For specific implementation, please refer to the method embodiments, which will not be repeated here.

[0063] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system and / or device described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0064] Furthermore, in the description of the embodiments of the present invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in the present invention based on the specific circumstances.

[0065] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An in-vehicle process optimization method, characterized by, Applied to a vehicle control system, the method comprises: acquiring multi-dimensional vehicle running state panoramic data; preprocessing the multi-dimensional vehicle running state panoramic data to obtain a vehicle panoramic data vector; inputting the vehicle panoramic data vector into a pre-trained intelligent scene recognition model to output a vehicle running active scene; querying a target vehicle-mounted resource allocation strategy corresponding to the vehicle running active scene in a pre-set strategy mapping table; scheduling vehicle running resources and process running modes based on the target vehicle-mounted resource allocation strategy.

2. The in-vehicle procedure optimization method according to claim 1, characterized by, The multi-dimensional vehicle running state panoramic data comprises vehicle state data, user state data and environment state data; the vehicle state data represents real-time dynamic performance and operating state of a vehicle body; the user state data represents physiological state, behavior intention and interaction demand of a user; and the environment state data represents an external environment and traffic situation in which the vehicle is located.

3. The in-vehicle procedure optimization method of claim 1, wherein, The preprocessing comprises normalization processing, noise reduction processing and synchronization processing; the step of preprocessing the multi-dimensional vehicle running state panoramic data to obtain a vehicle panoramic data vector comprises: performing normalization processing on the multi-dimensional vehicle running state panoramic data to eliminate dimensional influence; performing noise reduction processing on the multi-dimensional vehicle running state panoramic data based on a filtering algorithm and outlier detection after the normalization processing; performing synchronization processing on the multi-dimensional vehicle running state panoramic data based on a timestamp after the noise reduction processing to obtain the vehicle panoramic data vector.

4. The in-vehicle procedure optimization method of claim 1, wherein The intelligent scene recognition model comprises a vehicle running scene reasoning sub-model and a confidence determination sub-model; The vehicle running scene reasoning sub-model is configured to infer a vehicle running current scene and a confidence score corresponding to the vehicle running current scene based on the vehicle panoramic data vector; The confidence determination sub-model is configured to determine the vehicle running current scene corresponding to the confidence score as the vehicle running active scene if the confidence score is greater than a pre-set threshold.

5. The in-vehicle procedure optimization method according to claim 4, characterized by, The method further comprises training the vehicle running scene reasoning sub-model by the following steps: acquiring historical multi-dimensional vehicle running state panoramic data and preprocessing and labeling the historical multi-dimensional vehicle running state panoramic data to obtain historical vehicle panoramic data vectors; wherein the labeling processing is configured to one-to-one correspond the historical vehicle panoramic data vectors with pre-set running scene labels; inputting a training set containing the historical vehicle panoramic data vectors into a pre-established basic vehicle running scene reasoning sub-model, and iteratively optimizing parameters of the basic vehicle running scene reasoning sub-model until a pre-set training completion condition is reached; inputting a verification set containing the historical vehicle panoramic data vectors into the trained basic vehicle running scene reasoning sub-model to verify model performance, and determining the basic vehicle running scene reasoning sub-model that passes the verification as the vehicle running scene reasoning sub-model.

6. The in-vehicle procedure optimization method of claim 1, wherein, The strategy mapping table comprises one-to-one corresponding vehicle running scenes and vehicle-mounted resource allocation strategies; the method further comprises: receiving external data in a wired or wireless manner to update the strategy mapping table.

7. The in-vehicle procedure optimization method of claim 1, wherein, The vehicle operation resources include: priority of task processes, resource containers and threads; different modes correspond to different vehicle screen display interfaces; the step of scheduling vehicle operation resources and process operation modes based on the target vehicle resource allocation strategy includes: The priority of the key task process is raised to be greater than the priority of the background task process; wherein, the task process with high priority can acquire computing resources preferentially; The resource containers are allocated according to the priority order of the task processes; wherein, the resource quota of the resource container corresponding to the task process with high priority is greater than the resource quota of the resource container corresponding to the task process with low priority; Thread scheduling is performed based on the priority of the task process and the corresponding resource container; The display interface of the vehicle screen is adjusted.

8. An in-vehicle process optimization apparatus characterized by comprising: The device is applied to a vehicle control system, and the device comprises: a data acquisition module configured to acquire multi-dimensional vehicle operation state panoramic data; a data processing module configured to pre-process the multi-dimensional vehicle operation state panoramic data to obtain a vehicle panoramic data vector; a scene judgment module configured to input the vehicle panoramic data vector into a pre-trained intelligent scene recognition model to output a vehicle operation active scene; a resource allocation strategy determination module configured to query a target vehicle resource allocation strategy corresponding to the vehicle operation active scene in a pre-set strategy mapping table; a resource scheduling module configured to schedule vehicle operation resources and process operation modes based on the target vehicle resource allocation strategy.

9. An electronic device, comprising: The device comprises a processor and a memory, the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the vehicle process optimization method in any one of claims 1 to 7.

10. A computer storage medium, characterized in that, The computer storage medium stores computer executable instructions, and when the computer executable instructions are called and executed by the processor, the computer executable instructions cause the processor to implement the vehicle process optimization method in any one of claims 1 to 7.

Images