Vehicle machine performance adjusting method and system, storage medium and electronic equipment

Abstract

The invention discloses a vehicle machine performance adjusting method and system, a storage medium and electronic equipment, and the method comprises the steps: collecting the operation parameters of a plurality of dimensions of a vehicle machine, and carrying out the data cleaning of the operation parameters, and obtaining effective data; the valid data are sent to the trained neural network for deduction, and the performance state of the vehicle machine is judged and output; or / and performing multi-dimension comprehensive analysis on the effective data, and obtaining the performance state of the vehicle machine based on the analysis data; based on the performance state feedback processing measures of the vehicle machine, the performance of the vehicle machine is adjusted; the condition of the vehicle machine system can be dynamically displayed in real time, the evaluation level is output through the neural network, the performance condition of the whole vehicle machine can be known according to the state judged by the broken line graph and the neural network, and evaluation is more comprehensive.

Description

Technical Field

[0001] The present invention relates to the field of vehicle system testing, and in particular to a vehicle performance adjustment method, system, storage medium and electronic equipment. Background Art

[0002] In the vehicle system testing work, the performance evaluation of the vehicle computer (in-vehicle infotainment system) involves multiple aspects to ensure that its functions, stability and user experience meet the requirements. In terms of stability evaluation, it is generally evaluated based on the performance occupied by the app in the system, and only one aspect is evaluated. For example, the CPU load is used for evaluation, or the top command can roughly know the range of computing power. However, this data evaluation method is relatively one-sided, and the stability evaluation is not accurate. Summary of the invention

[0003] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a vehicle computer performance adjustment method, system, storage medium and electronic device.

[0004] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: a vehicle computer performance adjustment method, comprising:

[0005] S1: Collecting operating parameters of several dimensions of the vehicle computer, and performing data cleaning on the operating parameters to obtain valid data;

[0006] S2: sending the valid data to a trained neural network for deduction, and determining and outputting the performance status of the vehicle computer;

[0007] or / and performing comprehensive analysis of the valid data in several dimensions, and obtaining the performance status of the vehicle computer based on the analysis data;

[0008] S3: Adjusting the performance of the vehicle computer based on the performance status feedback processing measures of the vehicle computer.

[0009] As a further description of the above technical solution: in the step S1, the operating parameters are counted to obtain the data distribution of the operating parameters, a first threshold is set, the data distribution is compared with the first threshold, and the operating parameters greater than the first threshold are cleaned.

[0010] As a further description of the above technical solution: the operating parameters include first monitoring data of the control end and second monitoring data of the software end; the first monitoring data and the second monitoring data include at least two data types.

[0011] As a further description of the above technical solution: the training method of the neural network includes:

[0012] According to the dimensions of the operating parameters, corresponding different evaluation intervals and evaluation levels are set;

[0013] Based on different evaluation intervals and evaluation levels, the evaluation results under several dimensions are obtained;

[0014] Based on the evaluation results under the several dimensions, the balance between the several dimensions is judged, and the performance status of the vehicle computer is determined;

[0015] Iterative training is performed based on the above steps.

[0016] As a further description of the above technical solution: the operating parameters of the vehicle computer in several dimensions include CPU idle, CPU occupancy, free memory, GPU usage, video memory occupancy, I / O waiting time ratio, frame number or memory occupancy operating parameters.

[0017] As a further description of the above technical solution: the step of performing a comprehensive analysis of the valid data in several dimensions includes: selecting data information at consecutive time points in the valid data, and further obtaining the worst value of the corresponding dimension, and making these data into a line graph to complete the analysis.

[0018] As a further description of the above technical solution: the feedback processing measures include: adjusting the operating parameters according to the evaluation results and the line chart.

[0019] Also included is a vehicle computer performance evaluation system, which is applicable to the evaluation method in any one of the above technical solutions, including:

[0020] The acquisition module collects the operating parameters of the vehicle computer;

[0021] A data cleaning module receives the operating parameters sent by the acquisition module and removes abnormal data to obtain valid data;

[0022] A training module receives the valid data sent by the data cleaning module and performs training through a neural network;

[0023] The evaluation module determines the operating status of the vehicle computer according to the valid data and outputs an evaluation level.

[0024] It also includes a computer-readable storage medium storing a computer program for running the evaluation method, wherein the computer program enables a computer to execute the evaluation method as described in any one of the above technical solutions.

[0025] Also included is an electronic device, comprising:

[0026] one or more processors; memory; and

[0027] One or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the programs include a method for executing the evaluation method described in any one of the above technical solutions.

[0028] The above technical solution has the following advantages or beneficial effects:

[0029] After reading the operating parameters of the vehicle computer, data cleaning is performed to remove abnormal data, and the data is sorted according to the reading time. A line graph is generated for display. At the same time, it is determined whether there is a worst value. The neural network is iteratively trained to adjust the operating parameters and dynamically display the status of the vehicle computer system in real time. The evaluation level is output through the neural network. The performance of the entire vehicle computer can be known based on the status determined by the line graph and the neural network, making the evaluation more comprehensive. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 The process of the evaluation method proposed for the present invention Figure 1 ;

[0031] Figure 2 The process of the evaluation method proposed for the present invention Figure 2 ;

[0032] Figure 3 The process of the evaluation method proposed for the present invention Figure 3 ;

[0033] Figure 4 It is a structural schematic diagram of the evaluation system proposed by the present invention;

[0034] Figure 5 A line graph of the first monitoring data of the control terminal in the present invention;

[0035] Figure 6 It is a line chart of the second monitoring data of the software end in the present invention.

[0036] Legend:

[0037] 1. Collection module; 2. Data cleaning module; 3. Training module; 4. Evaluation module. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0039] During the research, the applicant found that in the evaluation of vehicle computer stability, special tools and software are used to conduct benchmark tests to evaluate the performance of the processor, memory and storage, and then stress tests and long-term operation tests are conducted to detect the stability of the system under high load and after long-term use. Generally, the evaluation is based on the performance occupied by the app in the system, and only one aspect is evaluated. However, this data evaluation method is relatively one-sided and the assessment of stability is not accurate.

[0040] Reference Figure 1 , an embodiment of the present invention provides: a vehicle computer performance adjustment method, comprising:

[0041] S1: Collect the operating parameters of several dimensions of the vehicle computer, clean the operating parameters, and obtain valid data.

[0042] The operation of the car computer involves many aspects. It is necessary to determine which dimensions of operating parameters to collect, including: performance dimension, network dimension, network dimension or hardware status dimension. You can use the system monitoring tool that comes with the car computer, or develop special collection software to obtain the required operating parameters, and determine the frequency of collecting operating parameters according to actual needs. If you want to understand the operating changes of the car computer in a short period of time in detail, you can set a higher collection frequency, such as collecting every few seconds; if you only focus on the general operating trend, the collection frequency can be appropriately reduced, such as collecting every few minutes, and the collected operating parameter data will be stored in time for subsequent processing. You can choose to store the data on the local storage device of the car computer (such as a built-in memory card or hard disk). The storage format can be a common text format (such as CSV) or a binary format, depending on the convenience of subsequent data processing and storage efficiency requirements. Abnormal data can be identified and screened by observing data distribution, setting thresholds, and other methods to obtain valid data.

[0043] S2: Send valid data to the trained neural network for deduction, and determine the performance status of the output vehicle computer;

[0044] Or / and conduct comprehensive analysis of valid data in several dimensions, and obtain the performance status of the vehicle computer based on the analyzed data.

[0045] Before sending valid data to the trained neural network, the neural network needs to be trained first. A training data set is set. The training data set contains a large amount of data on the vehicle computer operating parameters and the corresponding accurately labeled vehicle computer performance status. According to the characteristics and complexity of the vehicle computer performance status discrimination task, different types of neural network architectures can be selected, such as multi-layer perceptron (MLP), convolutional neural network (CNN), recurrent neural network (RNN) and its variants, etc. The selected neural network is trained using the prepared training data set. During the training process, by adjusting the weights and biases of the neural network and other parameters, the neural network can predict the corresponding vehicle computer performance status as accurately as possible according to the input vehicle computer operating parameters, continuously optimize the performance of the neural network, obtain the trained neural network, and then receive valid data for deduction. After receiving the input data, the neural network will perform a series of calculations and processing based on the weights and other parameters obtained by its training. This process is called deduction. After deduction, the neural network will eventually output a result about the vehicle computer performance status.

[0046] The operating parameters are sorted and a line graph is generated to determine whether there is a worst value; if so, the operating parameters are adjusted according to the worst value; if there is no worst value, the operating parameters are sent to the neural network for training.

[0047] S3: Based on the performance status feedback processing measures of the vehicle computer, the performance of the vehicle computer is adjusted.

[0048] Continuously collect the operating parameters of various dimensions of the vehicle computer, perform data cleaning and analysis according to the above methods, or accurately judge the performance status of the vehicle computer through neural network deduction, so as to timely capture the changes in the performance of the vehicle computer and feedback to the user end. The vehicle computer can also provide some operation prompts to guide users to perform correct operations to improve the performance of the vehicle computer. Users can make adjustments according to the performance status information fed back by the vehicle computer and the above operation prompts.

[0049] In step S1, statistics are collected on the operating parameters to obtain data distribution of the operating parameters, a first threshold is set, the data distribution is compared with the first threshold, and data cleaning is performed on the operating parameters that are greater than the first threshold.

[0050] In this embodiment, the operating parameters are counted, and abnormal data is identified and eliminated based on the data distribution of the operating parameters; eliminating abnormal data is a step in data cleaning, which is used to remove data points that do not conform to the expected pattern or are obviously incorrect.

[0051] Data distribution includes: normal distribution: use mean and standard deviation to identify abnormal data; uniform distribution: screen based on the range and distribution characteristics of the data; exponential distribution: eliminate abnormal data by setting thresholds or using statistical tests; lognormal distribution: apply normal distribution methods after logarithmic transformation of the data; Poisson distribution: detect abnormal data based on the counting characteristics of the data; different abnormal data detection methods can be selected according to different data distributions.

[0052] Reference Figure 2 , this application preferably adopts normal distribution, the method is as follows:

[0053] S11: Calculate mean and standard deviation: First calculate the mean (μ) and standard deviation (σ) of the data set;

[0054] S12: Determine the first threshold: Set the first threshold, such as the 3σ principle, to remove data points that exceed the mean ±3σ standard deviation;

[0055] S13: Filter data: remove data points that deviate from the normal range according to the first threshold.

[0056] It can effectively remove significantly abnormal data and retain most of the data that conforms to the normal distribution. 3σ is a statistical principle based on the normal distribution. It assumes that a set of data contains only random errors and obtains the standard deviation (σ) through calculation and processing to determine a probability interval. Errors outside this interval are considered to be gross errors rather than random errors, and data containing such errors should be eliminated.

[0057] In the normal distribution, σ represents the standard deviation, μ represents the mean, and x=μ is the axis of symmetry of the graph. The 3σ principle states that the probability of a value being distributed in (μ-σ,μ+σ) is 0.6826, while the probability of being distributed in (μ-2σ,μ+2σ) and (μ-3σ,μ+3σ) is 0.9544 and 0.9974 respectively. This means that almost all values ​​are concentrated in the interval (μ-3σ,μ+3σ).

[0058] The 3σ principle is based on repeated measurements of equal precision. If the absolute value of the residual error of a measurement value in a set of measurement data is greater than 3σ, the measurement value is considered a bad value and should be eliminated. Usually, errors equal to ±3σ are considered limit errors. For normal distribution random errors, the probability of falling outside ±3σ is only 0.27%, which is very unlikely to occur in a limited number of measurements. Therefore, the 3σ principle is the most commonly used and simplest criterion for gross error discrimination. It is usually applied to situations where the number of repeated measurements is large (n≥30) or when n>10 is used for rough discrimination.

[0059] Data cleaning methods include: statistical methods: using mean and standard deviation to identify and remove abnormal data that exceeds the range; box plots: identifying and removing extreme values ​​through quartile ranges; Z-score: calculating the Z-score of the data and removing data points that deviate too far from the mean; IQR (interquartile range) method: identifying and removing abnormal data based on the interquartile range (IQR) of the data; machine learning methods: using algorithms such as isolation forests and local anomaly factors to detect and process abnormal data. It helps to improve the quality of the data set and the reliability of the analysis results. This application preferably uses statistical methods to obtain the data distribution of the operating parameters.

[0060] Reference Figure 5-Figure 6 The operating parameters include first monitoring data from the control end and second monitoring data from the software end; the first monitoring data and the second monitoring data include at least two data types.

[0061] The operating parameters of the vehicle computer in several dimensions include CPU idle, CPU occupancy, free memory, GPU usage, video memory occupancy, I / O waiting time ratio, frame rate or memory occupancy.

[0062] In this embodiment, the data type of the first monitoring data includes: CPU idle (CPU idle), free memory (FrMm), GPU utilization (GPU), video memory occupancy (GMem) or I / O waiting time ratio (IOW); the data type of the second monitoring data includes: CPU occupancy (CPU), memory occupancy (Mem), GPU utilization (GPU), video memory occupancy (GMem) or frame rate (fps).

[0063] There is a corresponding relationship between data types. For example, if the I / O waiting time ratio is too high, the CPU occupancy is too high, and the GPU usage rate is too high, the frame rate (fps) will be too low. The cause of the low fps can be quickly located. When the computer is performing a large number of input and output operations, if the performance of the storage device cannot keep up with the demand, the I / O waiting time will increase. During the time waiting for the I / O operation to complete, other processes may be blocked, resulting in a slower overall screen update speed, that is, a lower fps. If the CPU is occupied too much by a large number of other non-rendering related tasks, the computing resources allocated to screen rendering to maintain normal fps will be relatively insufficient, which will lead to a drop in fps.

[0064] Neural network training methods include:

[0065] Set corresponding different evaluation intervals and evaluation levels according to the dimensions of operating parameters;

[0066] Based on different evaluation intervals and evaluation levels, the evaluation results under several dimensions are obtained;

[0067] Based on the evaluation results under several dimensions, the balance between the dimensions is judged, and the performance status of the vehicle computer is determined;

[0068] Iterative training is performed based on the above steps.

[0069] The neural network receives operating parameters, sets evaluation intervals according to data types, divides them into several evaluation levels, determines the balance between several data types, determines the operating status of the vehicle computer, and outputs the corresponding evaluation level.

[0070] In this embodiment, the neural network receives the operating parameters and then performs training. The evaluation interval is set according to the data type, and the evaluation is divided into several evaluation levels. Different data types are evaluated respectively. In a specific embodiment, for example, a CPU value less than 50% is excellent, greater than 50% and less than 60% is good, greater than 60% and less than 70% is fair, greater than 70% and less than 80% is poor, and greater than 80% is extremely poor.

[0071] Free memory (FrMm) greater than 80% is excellent, greater than 70% and less than 80% is good, greater than 60% and less than 70% is fair, greater than 50% and less than 60% is poor, and less than 50% is very poor.

[0072] According to actual needs, the evaluation interval corresponding to each data type is set, and the corresponding evaluation level is determined. The more balanced the evaluation level is, for example, they are all close to good, the more stable the vehicle computer state is. After monitoring the above-mentioned data types of the vehicle computer and determining their respective evaluation levels, the stability of the vehicle computer state can be evaluated by calculating the average evaluation level or observing the distribution of the evaluation level. Specifically, the distribution of the evaluation level is observed according to the distribution of the evaluation level of the data type. If the evaluation levels of each data type are relatively close (for example, they are all in the good or medium range), then the vehicle computer state is relatively stable; on the contrary, if the evaluation level of a certain data type is very different from that of other data types (for example, one is excellent and the other is extremely poor), then it means that there is a large imbalance in the vehicle computer state, and further attention and analysis of the vehicle computer function corresponding to the data type may be required.

[0073] The steps of conducting comprehensive analysis of valid data in several dimensions include: selecting data information at consecutive time points in the valid data, and further obtaining the worst value of the corresponding dimension, and making these data into a line graph to complete the analysis.

[0074] The operating parameters are sorted according to the reading time and divided into several reading time points. A reading cycle and a second threshold are set. The operating parameters at the reading time points are identified in turn within the reading cycle, and it is determined whether there are at least two consecutive time points exceeding the second threshold. If so, the reading time point is determined to be the worst value, and the operating parameters are adjusted. If not, the operating parameters are sent to the neural network.

[0075] Reference Figure 3 In this embodiment, the method for determining the worst value is as follows:

[0076] S21: setting a reading cycle, the reading time is 1 second, reading the first monitoring data or the second monitoring data every 1 second, and sorting them according to the reading time to form a plurality of line graphs;

[0077] S22: setting a second threshold value, judging the operating parameters, and taking the value exceeding the second threshold value as the worst value;

[0078] S23: Determine whether there are at least two consecutive time points exceeding the second threshold value. If so, determine that the read time point is the worst value and adjust the operating parameters. If not, send the operating parameters to the neural network.

[0079] This embodiment is preferably a data line graph to reflect the changing trend of the operating parameters over time. The method for judging whether there is a worst value for the first monitoring data and the second monitoring data is as follows:

[0080] Judge the first monitoring data:

[0081] If IOW is greater than 50% and CPU is less than 60%, it means that the current disk reading and writing consumes a lot of performance;

[0082] If the CPU is less than 60% and the GPU is more than 90%, it means that the GPU computing power of the vehicle is insufficient;

[0083] If the CPU is greater than 60% and the GPU is less than 60%, it means that the computing power of the vehicle computer is unevenly distributed and part of the computing power needs to be transferred to the GPU;

[0084] If IOW is greater than 50%, and the GMem occupied by the GPU and video memory is less than 100M, it means that the video memory utilization of the vehicle is low and some art materials need to be cached.

[0085] Judge the second monitoring data:

[0086] If the CPU is larger than the GPU, it means that many functions are concentrated on the CPU, and some functions need to be transferred to the GPU to complete.

[0087] If the fps is lower than 30 and the CPU is greater than 60%, it means that the CPU computing power is insufficient.

[0088] If the fps is lower than 30 and the GPU is greater than 90%, it means that the GPU computing power is insufficient.

[0089] If the fps is lower than 30, the CPU is greater than 60%, and the GPU is greater than 90%, it means that the art materials are used in large quantities and need to be streamlined.

[0090] The second threshold range for judgment can be set according to actual needs. By judging whether there is a worst value between the first monitoring data and the second monitoring data, the performance of the vehicle computer is determined. If the broken lines of the data types of the first monitoring data and the second monitoring data are relatively even (the fluctuation range is less than 5-10%, determined according to actual needs), the vehicle computer is considered to be healthy.

[0091] Feedback processing measures include: adjusting operating parameters based on evaluation results and line graphs.

[0092] In this embodiment, it is determined whether the operating parameters need to be adjusted based on the evaluation results. If necessary, the operating parameters are adjusted and the neural network is controlled to iteratively train the operating parameters. The performance of the entire vehicle computer is known based on the state determined by the line graph and the neural network, and the evaluation is more comprehensive.

[0093] The neural network outputs an evaluation level, reflecting that the current vehicle computer status is unstable, and sends a warning message to remind that the operating parameters need to be adjusted, conduct iterative training, update the operating parameters, and adjust the resources on the control and software ends.

[0094] Furthermore, it can prompt the possible reasons for the status of the control end or software end, improvement suggestions, and potential risks, etc. Specifically, for the analysis results corresponding to multiple data types, it can remind you to do corresponding processing; or according to the analysis results corresponding to multiple data types, combined with the status of the vehicle computer, it can give corresponding processing solutions, for example:

[0095] If the fps is low (less than 30) and the CPU is high (greater than 60%), it means that the computing power is tight, and it is suggested whether the GPU utilization rate can be increased. The current vehicle computer status is extremely poor, indicating that the GPU utilization rate needs to be increased. Set the first threshold or the second threshold range for judgment according to the actual situation;

[0096] If the evaluation levels of the data types are all reasonable, but the analysis results show that the vehicle computer status is extremely poor, the developer will be notified that the hardware computing power needs to be improved.

[0097] Example 2

[0098] Reference Figure 4, a vehicle computer performance evaluation system, the evaluation system is applicable to any evaluation method in the above technical solutions, including:

[0099] Collection module 1, collects the operating parameters of the vehicle computer;

[0100] Data cleaning module 2 receives the operating parameters sent by the acquisition module and removes abnormal data to obtain valid data;

[0101] Training module 3, receiving valid data sent by the data cleaning module, and performing training through the neural network;

[0102] The evaluation module 4 determines the operating status of the vehicle computer according to the valid data and outputs the evaluation level.

[0103] In this embodiment, the acquisition module 1 collects the operating parameters of the vehicle computer and sends them to the data cleaning module 2. The data cleaning module 2 reads and sorts the operating parameters and removes abnormal data according to the normal distribution. After completion, it is sent to the training module 3, and the operating parameters are updated through iterative training of the neural network. The evaluation module 4 judges the operating status of the vehicle computer according to the operating parameters and outputs an evaluation level.

[0104] It is understandable that the software evaluation system provided in the embodiment of the present invention corresponds to the above-mentioned adjustment method, and the explanations, examples, beneficial effects and other parts of its relevant contents can refer to the corresponding contents in the adjustment method, which will not be repeated here.

[0105] Example 3

[0106] Also included is a computer-readable storage medium storing a computer program for executing the selection method, wherein the computer program causes a computer to execute the following steps:

[0107] S1: Collect the operating parameters of several dimensions of the vehicle computer, clean the operating parameters, and obtain valid data;

[0108] S2: Send valid data to the trained neural network for deduction, and determine the performance status of the output vehicle computer;

[0109] S3: Based on the performance status feedback processing measures of the vehicle computer, the performance of the vehicle computer is adjusted.

[0110] Among them, the computer-readable storage medium can be a computer storage medium or a communication medium. The communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The computer storage medium can be any available medium that can be accessed by a general-purpose or special-purpose computer. For example, a computer-readable storage medium is coupled to a processor so that the processor can read information from the computer-readable storage medium and write information to the computer-readable storage medium. Of course, the computer-readable storage medium can also be a component of the processor. The processor and the computer-readable storage medium can be located in an application-specific integrated circuit (ASIC). In addition, the ASIC can be located in a user device. Of course, the processor and the computer-readable storage medium can also exist in a communication device as discrete components.

[0111] Specifically, the computer-readable storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The storage medium can be any available medium that can be accessed by a general or special-purpose computer.

[0112] Example 4

[0113] Also included is an electronic device, comprising:

[0114] one or more processors; memory; and

[0115] One or more programs, wherein the one or more programs are stored in a memory and configured to be executed by one or more processors, the programs comprising steps for performing the following steps:

[0116] S1: Collect the operating parameters of several dimensions of the vehicle computer, clean the operating parameters, and obtain valid data;

[0117] S2: Send valid data to the trained neural network for deduction, and determine the performance status of the output vehicle computer;

[0118] S3: Based on the performance status feedback processing measures of the vehicle computer, the performance of the vehicle computer is adjusted.

[0119] Memory is used to store computer programs. The memory may include high-speed random access memory (RAM) and may also include non-volatile memory (NVM), such as at least one disk memory, and may also be a USB flash drive, a mobile hard disk, a read-only memory, a disk or an optical disk.

[0120] The processor is used to execute the computer program stored in the memory. The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. The general-purpose processor can be a microprocessor or the processor can be any conventional processor, etc. The steps of the method disclosed in the invention can be directly embodied as being executed by a hardware processor, or can be executed by a combination of hardware and software modules in the processor.

[0121] Optionally, the memory can be independent or integrated with the processor.

[0122] When the memory is a device independent of the processor, the electronic device may further include a bus. The bus is used to connect the memory and the processor. The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus. The bus may be divided into an address bus, a data bus, a control bus, and the like.

[0123] It should be noted that, through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform. Based on such an understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiment. In this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the term "include", "comprise" or any other variant thereof is intended to cover non-exclusive inclusion, so that the process, method, article or device including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, article or device. Without more constraints, an element defined by the phrase "comprising a..." does not exclude the existence of other identical elements in the process, method, article or apparatus comprising the element.

[0124] Finally, it should be noted that the above is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A vehicle computer performance adjustment method, characterized in that: include: S1: Collecting operating parameters of several dimensions of the vehicle computer, and performing data cleaning on the operating parameters to obtain valid data; S2: sending the valid data to a trained neural network for deduction, and determining and outputting the performance status of the vehicle computer; or / and performing comprehensive analysis of the valid data in several dimensions, and obtaining the performance status of the vehicle computer based on the analysis data; S3: Adjusting the performance of the vehicle computer based on the performance status feedback processing measures of the vehicle computer.

2. The adjustment method according to claim 1, characterized in that: In the step S1, the operating parameters are counted to obtain data distribution of the operating parameters, a first threshold is set, the data distribution is compared with the first threshold, and data cleaning is performed on the operating parameters that are greater than the first threshold.

3. The adjustment method according to claim 1, characterized in that: The operating parameters include first monitoring data of the control end and second monitoring data of the software end; the first monitoring data and the second monitoring data include at least two data types.

4. The adjustment method according to claim 2, characterized in that: The training method of the neural network includes: According to the dimensions of the operating parameters, corresponding different evaluation intervals and evaluation levels are set; Based on different evaluation intervals and evaluation levels, the evaluation results under several dimensions are obtained; Based on the evaluation results under the several dimensions, the balance between the several dimensions is judged, and the performance status of the vehicle computer is determined; Iterative training is performed based on the above steps.

5. The adjustment method according to claim 4, characterized in that: The operating parameters of the vehicle computer in several dimensions include CPU idle, CPU occupancy, free memory, GPU usage, video memory occupancy, I / O waiting time ratio, frame number or memory occupancy.

6. The adjustment method according to claim 5, characterized in that: The step of performing comprehensive analysis of the valid data in several dimensions includes: selecting data information at consecutive time points in the valid data, and further obtaining the worst value of the corresponding dimension, and making these data into a line graph to complete the analysis.

7. The adjustment method according to claim 6, characterized in that: The feedback processing measures include: adjusting the operating parameters according to the evaluation results and the line graph.

8. A vehicle performance adjustment system, characterized in that: The evaluation system is applicable to the evaluation method according to any one of claims 1 to 7, including: The acquisition module collects the operating parameters of the vehicle computer; A data cleaning module receives the operating parameters sent by the acquisition module and removes abnormal data to obtain valid data; A training module receives the valid data sent by the data cleaning module and performs training through a neural network; The evaluation module determines the operating status of the vehicle computer according to the valid data and outputs an evaluation level.

9. A computer-readable storage medium, characterized in that: It stores a computer program for running the adjustment method, wherein the computer program enables a computer to execute the adjustment method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: one or more processors; Memory; as well as One or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, the programs comprising instructions for executing the adjustment method according to any one of claims 1 to 7.

Images