Data return method and device, vehicle and storage medium

Abstract

The invention discloses a data return method and device, a vehicle and a storage medium. The method comprises the steps of determining importance degree coefficients and measured data corresponding to M initial data types in to-be-returned data based on a current scene; n target data types are determined from the M initial data types, target return data are formed based on measured data corresponding to the N target data types, the total data volume of the target return data is smaller than the target data volume, and the sum of importance degree coefficients corresponding to the N target data types is maximum; sending the target return data to the cloud, so that the cloud analyzes the measured data corresponding to the N target data types, and determines feedback data; and updating the importance degree coefficients corresponding to the M initial data types in the to-be-returned data based on feedback data sent by the cloud. According to the method, efficient and accurate return of the target return data can be realized, so that the cloud can efficiently and accurately analyze the fault reason of the current scene according to the target return data.

Description

Technical Field

[0001] The present invention relates to the field of intelligent driving control technology, and in particular to a data feedback method, device, vehicle and storage medium. Background Art

[0002] As more and more sensor components are connected to the integrated driving domain controller, the amount of data collected by the vehicle side increases, which means that the amount of data that the vehicle side needs to transmit back to the cloud is increasing. When a large amount of data is transmitted back, problems such as low data transmission efficiency, transmission interruption and packet loss are prone to occur. It is impossible to guarantee that the cloud will receive the data used to locate the cause of the fault in time, resulting in low efficiency and accuracy of fault analysis in the cloud. In order to improve the efficiency and accuracy of fault location analysis in the cloud, the vehicle side needs to transmit the data required by the cloud more accurately and efficiently. Therefore, how to enable the vehicle side to accurately and efficiently transmit the data required by the cloud so that the cloud can accurately and efficiently locate the cause of the fault is a technical problem that needs to be solved urgently. Summary of the invention

[0003] The embodiments of the present invention provide a data feedback method, device, vehicle and storage medium to solve the problem of how to enable the vehicle to accurately and efficiently transmit the data required by the cloud.

[0004] A data return method includes the following steps performed by a vehicle: Based on the current scenario, determine the importance coefficients and measured data corresponding to the M initial data types in the data to be returned; Determine N target data types from the M initial data types, and form target return data based on the measured data corresponding to the N target data types, wherein the total data volume of the target return data is less than the target data volume, and the sum of the importance coefficients corresponding to the N target data types is the largest, 1≤N≤M; Sending the target return data to the cloud, so that the cloud analyzes the measured data corresponding to the N target data types to determine the feedback data; Based on the feedback data sent by the cloud, the importance coefficients corresponding to the M initial data types in the data to be returned are updated.

[0005] Preferably, the determining, based on the current scenario, the importance coefficients and measured data corresponding to the M initial data types in the data to be returned includes: Based on the current scenario, determine the data priorities corresponding to the M initial data types in the data to be returned; Based on the data priorities corresponding to the M initial data types in the data to be returned, importance coefficients and measured data of the M initial data types in the data to be returned are determined.

[0006] Preferably, the target data volume is the product of the system operation index and a preset calibration coefficient; The system operation index is determined based on system operation data.

[0007] Preferably, the system operation data includes network connection data and system load data; The system operation index is the difference between the network connection index and the system load index; The network connection index is determined based on the network connection data, and the system load index is determined based on the system load data; Among them, the network connection data is data related to the communication network between the vehicle end and the cloud end; the system load data is data related to the load between the vehicle end and the cloud end.

[0008] Preferably, the network connection data includes available bandwidth, network latency, and network latency variance; The network connection index is the sum value obtained by weighted processing of the available bandwidth, the network latency, and the network latency variance.

[0009] Preferably, the system load data includes CPU occupancy rate and encoding resource occupancy rate; The system load index is the sum value obtained by weighted processing of the CPU occupancy rate and the encoding resource occupancy rate.

[0010] Preferably, updating the importance degree coefficients corresponding to M initial data types in the data to be returned based on the feedback data sent by the cloud end includes: If the feedback data includes critical data types and non-critical data types, increase the importance degree coefficient corresponding to the critical data type, and / or decrease the importance degree coefficient corresponding to the non-critical data type, where the critical data type is a target data type whose failure cause correlation based on the target data to be returned is greater than a preset threshold, and the non-critical data type is a target data type whose failure cause correlation based on the target data to be returned is not greater than the preset threshold; If the feedback data does not include critical data types and non-critical data types, decrease the importance degree coefficient corresponding to the target data type, and / or increase the importance degree coefficient corresponding to the remaining data types; where the remaining data types are the data types in the initial data types other than the target data type.

[0011] A data return device, comprising: An importance degree coefficient determination module, which determines the importance degree coefficients and measured data corresponding to M initial data types in the data to be returned based on the current scenario; A target feedback data determination module, configured to determine N target data types from M initial data types, and form target feedback data based on the measured data corresponding to the N target data types, where the total data volume of the target feedback data is less than the target data volume, and the sum of the importance degree coefficients corresponding to the N target data types is the largest, and 1 ≤ N ≤ M; A target feedback data sending module, configured to send the target feedback data to the cloud, so that the cloud analyzes the measured data corresponding to the N target data types to determine feedback data; An importance degree coefficient updating module, configured to update the importance degree coefficients corresponding to the M initial data types in the data to be feedback based on the feedback data sent by the cloud.

[0012] A vehicle includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the above data feedback method is implemented.

[0013] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above data feedback method is implemented.

[0014] The above data feedback method, device, vehicle, and storage medium screen out the measured data corresponding to the N target data types from the M initial data types as the target feedback data, so that the total data volume of the measured data in the target feedback data is less than the target data volume, and the sum of the importance degree coefficients corresponding to the N target data types is the largest. It can ensure that the target feedback data with a higher importance degree coefficient is feedback to the cloud to the greatest extent under the constraints of the target data volume and the importance degree coefficient, so as to achieve the efficient and accurate feedback of the target feedback data. Dynamically adjust the importance degree coefficients of the M initial data types in the data to be feedback according to the feedback data, and establish a dynamic feedback adjustment mechanism between the cloud and the vehicle side, so that the vehicle side can timely correct the importance degree coefficients corresponding to the M initial data types in the data to be feedback according to the feedback data of the cloud, ensure the priority feedback of the target feedback data with a higher importance degree coefficient, and there is no need to feedback all the data to be feedback on the vehicle side, which can effectively avoid the system crash and data loss of the systems to which the vehicle side and the cloud belong caused by more feedback data, and achieve the purpose of the efficient and accurate feedback of the target feedback data, so that the cloud can efficiently and accurately locate and analyze the cause of the failure in the current scenario according to the efficiently and accurately feedback target feedback data. Description of the Drawings

[0015] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments of the present invention. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0016] Figure 1 is a flowchart of a data feedback method in an embodiment of the present invention; Figure 2 is another flowchart of a data feedback method in an embodiment of the present invention; Figure 3 is another flowchart of a data feedback method in an embodiment of the present invention; Figure 4 is a schematic diagram of a data feedback device in an embodiment of the present invention; Figure 5 is a schematic diagram of data transmission between the vehicle end and the cloud end in a system according to an embodiment of the present invention. Specific Embodiments

[0017] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some but not all of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0018] The data feedback method provided by the embodiments of the present invention can be applied to the vehicle end installed in a vehicle. As Figure 5 shown, it is a schematic diagram of data transmission between the vehicle end and the cloud end in a system. The vehicle end adopts the data feedback method provided by the present invention to determine the target feedback data from the data to be feedback, transmit the target feedback data to the cloud end, and update the importance degree coefficient corresponding to each initial data type in the data to be feedback based on the received feedback data from the cloud end, so that the vehicle end can accurately and efficiently feedback the target feedback data required by the cloud end according to the updated importance degree coefficient. In the embodiments of the present invention, the vehicle end communicates with the cloud end through a network. The vehicle end is also called the vehicle end controller, which is installed on the vehicle and is used to obtain the data to be feedback of different data types collected by the sensors installed on the vehicle, and determine the target feedback data from the data to be feedback of different data types collected, and transmit the target feedback data to the cloud end. The cloud end is also called the cloud server, which is used to receive the target feedback data transmitted by the vehicle end and perform fault cause location analysis on the feedback scenario corresponding to the target feedback data according to the target feedback data transmitted by the vehicle end.

[0019] In one embodiment, as Figure 1As shown in the figure, a data feedback method is provided. Taking the vehicle terminal in Figure 5 as an example, the following steps executed by the vehicle terminal are included: S101: Based on the current scenario, determine the importance degree coefficients and measured data corresponding to M initial data types in the data to be feedback. S102: Determine N target data types from the M initial data types, and form target feedback data based on the measured data corresponding to the N target data types. The total data volume of the target feedback data is less than the target data volume, and the sum of the importance degree coefficients corresponding to the N target data types is the largest, where 1 ≤ N ≤ M. S103: Send the target feedback data to the cloud so that the cloud can analyze the measured data corresponding to the N target data types to determine the feedback data. S104: Update the importance degree coefficients corresponding to the M initial data types in the data to be feedback based on the feedback data sent by the cloud.

[0020] Herein, the current scenario refers to the scenario when the vehicle breaks down, including but not limited to scenarios such as abnormal exit of the ACC (Adaptive Cruise Control) function, abnormal exit of the ICA (Integrated Cruise Assist) function, abnormal exit of the APA (Automatic Park Assist) function, and occurrence of a forward collision. The data to be feedback includes M initial data types generated during the vehicle operation and the measured data corresponding to the M initial data types collected in real time in the current scenario, which are used to be feedback to the cloud so that the cloud can perform fault cause location analysis on the vehicle in the current scenario. Herein, M ≥ 1. The measured data refers to the data collected in real time in the current scenario. The initial data type refers to the data type of the measured data in the data to be feedback. The initial data types include but not limited to data types such as front camera data, rear camera data, control data, vehicle environment data, and vehicle position. The measured data is the specific data corresponding to the above initial data types. It can be understood that during the vehicle driving process, the vehicle may be in different fault scenarios. At this time, it is necessary to collect the data to be feedback corresponding to the current scenario, determine the target feedback data based on the data to be feedback, and feedback the target feedback data to the cloud so that the cloud can receive the target feedback data feedback by the vehicle terminal and perform fault cause location analysis. It can be understood that during the vehicle operation, the measured data of different initial data types are collected by different types of sensors, and these measured data need to be feedback by the vehicle terminal to the cloud so that the cloud can perform fault cause analysis on the current scenario according to the received measured data. The initial importance degree coefficient refers to the coefficient used to represent the importance degree of the measured data of different initial data types in the data to be feedback for fault analysis of the current scenario.

[0021] As an example, in step S101, the vehicle end obtains the measured data of M initial data types collected by M types of sensors, performs a failure analysis on the measured data of the M initial data types, determines the fault scenario in which the vehicle is located, and determines the fault scenario in which the vehicle is located as the current scenario. In this example, the vehicle end can compare and analyze the measured data of each initial data type collected in real time with the standard data of the same initial data type in the system operation database, determine the abnormal data in the measured data, determine the fault scenario in which the vehicle is located according to the abnormal data, and determine this fault scenario as the current scenario to complete the failure analysis. For example, if the fault scenario in which the vehicle is located is the abnormal exit of the ACC function, then the abnormal exit of the ACC function is determined as the current scenario; if the fault scenario in which the vehicle is located is a forward collision, then the occurrence of a forward collision is determined as the current scenario. Among them, the standard data refers to the normal vehicle data used for comparing with the measured data for fault analysis. After the vehicle end determines the current scenario, it determines the measured data of the M initial data types collected in real time and the M initial data types as the data to be transmitted back under the current scenario.

[0022] In this example, after the vehicle end determines the current scenario and the data to be transmitted back, it determines the importance degree coefficients corresponding to the M initial data types in the data to be transmitted back according to the relevance between the measured data of the M initial data types in the data to be transmitted back and the current scenario. It can be understood that the higher the relevance between the measured data of the initial data type and the current scenario, the higher the importance of the measured data of this initial data type for locating and analyzing the fault cause of the current scenario, and then the larger the importance degree coefficient corresponding to this initial data type. Therefore, the initial data type with a higher relevance to the current scenario has a larger importance degree coefficient, and the initial data type with a larger importance degree coefficient has a higher priority for transmitting the corresponding measured data to the cloud, so that the cloud can efficiently and accurately locate and analyze the fault cause of the current scenario of the vehicle end according to the measured data with a larger importance degree coefficient.

[0023] Among them, the target data type refers to the data type corresponding to the data to be transmitted back to the cloud. The target data volume refers to the maximum data volume allowed to be transmitted between the vehicle end and the cloud, and this target data volume is determined based on factors such as the network connection status between the vehicle end and the cloud and the system load situation of the system to which the vehicle end and the cloud belong. The target data to be transmitted back includes the target data type to be transmitted back to the cloud and the measured data of the target data type. The network connection status is used to reflect the connection status between the vehicle end and the cloud. The system load situation is used to characterize the load occupancy situation of the system to which the vehicle end and the cloud belong. When the network connection status is good and the load occupancy in the system load situation is low, the transmitted target data volume is large, and as many target data to be transmitted back as possible can be transmitted; otherwise, the transmitted target data volume is small.

[0024] As an example, in step S102, the vehicle end screens the M initial data types according to the total data volume of the measured data corresponding to the M initial data types, determines N target data types to be transmitted back to the cloud, and determines the measured data corresponding to the N target data types as the target data to be transmitted back, ensuring that the total data volume of the measured data corresponding to the N target data types is less than the target data volume, and the sum of the importance degree coefficients corresponding to the N target data types is the largest, where 1 ≤ N ≤ M. It can be understood that due to the limitations of the systems to which the vehicle end and the cloud belong, if all the measured data corresponding to the M initial data types are transmitted to the cloud, not only the data transmission efficiency will be reduced, but also the systems to which the vehicle end and the cloud belong may crash and data may be lost due to the large number of target data to be transmitted back. Therefore, it is necessary to reasonably select the target data to be transmitted back from the data to be transmitted back and transmit it to the cloud to ensure efficient data transmission. Analyze the importance degree coefficient corresponding to each initial data type in the data to be transmitted back, so that the target data to be transmitted back transmitted to the cloud has a high correlation with the current scenario, ensuring the accuracy of data transmission, so as to facilitate the cloud to accurately analyze the target data to be transmitted back and improve the efficiency of fault cause analysis and location.

[0025] In this example, the data to be transmitted back obtained by the vehicle end in the current scenario contains M initial data types. For the measured data of the i-th initial data type, its corresponding real-time data volume is . Among them, the real-time data volume refers to the data volume in the measured data of each initial data type monitored in real time. The vehicle end obtains the total data volume corresponding to the measured data of the M initial data types in all the data to be transmitted back , where is the representation value corresponding to the measured data of the i-th initial data type. This representation value is a numerical value used to represent whether the measured data of the i-th initial data type is transmitted back to the cloud, and can be 1 or 0; if the representation value is 1, it indicates that the measured data of this initial data type is the target data to be transmitted back and needs to be transmitted back to the cloud; if the representation value is 0, it indicates that the measured data of this initial data type is not the target data to be transmitted back and does not need to be transmitted back to the cloud. The vehicle end determines the data volume constraint condition as ≤ D, where D is the target data volume, and determines the objective function as , where is the importance degree coefficient corresponding to the i-th initial data type in the j-th current scenario. The vehicle end solves the objective function according to the data volume constraint condition. When the objective function takes the maximum value and the data volume constraint condition is satisfied, it determines the representation value corresponding to the measured data of each initial data type in the current scenario. When determining that the representation value is 1, it determines that the measured data of the i-th initial data type is the target data to be transmitted back. When determining that the representation value When it is 0, it is determined that the measured data of the i-th initial data type is not the target data to be transmitted back. When the vehicle terminal determines that ≤ D and the objective function is is the maximum, among the M initial data types, each initial data type with a representation value of 1 is determined as the target data type, and the measured data corresponding to the target data type is determined as the target data to be transmitted back.

[0026] In this example, when ensuring that the total data volume of the measured data in the target data to be transmitted back is less than the target data volume and the sum of the importance coefficient corresponding to the N target data types is the largest, the measured data corresponding to the N target data types selected from the measured data corresponding to the M initial data types is obtained, and the target data to be transmitted back is obtained. This method can, under the constraint of the target data volume, maximize the guarantee that the target data to be transmitted back with a higher importance coefficient is transmitted back to the cloud to achieve the efficient and accurate transmission of the target data to be transmitted back.

[0027] Among them, the feedback data is the data fed back by the cloud to the vehicle terminal, and this feedback data is the data determined after analyzing the measured data corresponding to the N target data types.

[0028] As an example, in step S103, the vehicle terminal sends the target data to be transmitted back to the cloud, so that the cloud locates and analyzes the cause of the fault corresponding to the current scenario according to the received target data to be transmitted back, and determines the feedback data for adjusting the importance coefficient corresponding to the M initial data types of the vehicle terminal. It can be understood that after receiving the target data to be transmitted back, the cloud analyzes the cause of the fault of the current scenario according to the measured data of the N target data types in the target data to be transmitted back. When analyzing the cause of the fault of the current scenario, it may be due to the low importance of the received target data to be transmitted back for the cause of the fault analysis, resulting in inaccurate cause of the fault analysis or the inability to analyze the cause of the fault. It is necessary for the cloud to return the feedback data to the vehicle terminal in real time, so that the vehicle terminal can adjust the measured data of the target data type transmitted back to the cloud in real time, and then the cloud can accurately analyze the cause of the fault of the current scenario according to the measured data of the target data type transmitted back by the vehicle terminal.

[0029] As an example, in step S104, after the vehicle end sends the target feedback data to the cloud, it monitors in real time whether the cloud returns feedback data. After receiving the feedback data sent by the cloud, according to the feedback data, the importance degree coefficients corresponding to the M initial data types in the data to be feedback are dynamically adjusted in real time to determine the adjusted importance degree coefficients corresponding to each initial data type. So that in the next data feedback, according to the adjusted importance degree coefficients, the target feedback data to be sent to the cloud can be accurately and efficiently determined, so that the cloud can accurately and efficiently locate and analyze the cause of the current scene failure based on the target feedback data. It can be understood that after the cloud analyzes the target feedback data, the feedback data is determined. The vehicle end dynamically adjusts the importance degree coefficients corresponding to the M initial data types in the data to be feedback according to the feedback data, and a dynamic feedback adjustment mechanism is established between the cloud and the vehicle end, so that the vehicle end can, according to the feedback data of the cloud, correct the importance degree coefficients of the initial data types in real time, ensure the priority feedback of the measured data corresponding to the initial data types with higher importance degree coefficients, and there is no need to feedback all the data to be feedback by the vehicle end, which can effectively avoid the system crashes and data losses of the systems to which the vehicle end and the cloud belong caused by more feedback data, and achieve the purpose of efficient and accurate feedback of the target feedback data.

[0030] In this embodiment, the measured data corresponding to the N target data types selected from the M initial data types is used as the target feedback data, so that the total data volume of the measured data in the target feedback data is less than the target data volume, and the sum of the importance degree coefficients corresponding to the N target data types is the largest, which can ensure that the target feedback data with higher importance degree coefficients is feedback to the cloud to the greatest extent under the constraints of the target data volume and the importance degree coefficients, so as to achieve the efficient and accurate feedback of the target feedback data. The importance degree coefficients of the M initial data types in the data to be feedback are dynamically adjusted according to the feedback data, and a dynamic feedback adjustment mechanism is established between the cloud and the vehicle end, so that the vehicle end can, according to the feedback data of the cloud, correct the importance degree coefficients corresponding to the M initial data types in the data to be feedback in real time, ensure the priority feedback of the target feedback data with higher importance degree coefficients, and there is no need to feedback all the data to be feedback by the vehicle end, which can effectively avoid the system crashes and data losses of the systems to which the vehicle end and the cloud belong caused by more feedback data, and achieve the purpose of efficient and accurate feedback of the target feedback data, so that the cloud can efficiently and accurately locate and analyze the cause of the current scene failure based on the efficiently and accurately feedback target feedback data.

[0031] In one embodiment, as Figure 2 shown, step S101, that is, based on the current scene, determining the importance degree coefficients and measured data corresponding to the M initial data types in the data to be feedback, includes: S201: Based on the current scene, determine the data priorities corresponding to the M initial data types in the data to be feedback; S202: Determine the importance degree coefficients and measured data of the M initial data types in the data to be transmitted back based on the data priorities corresponding to the M initial data types in the data to be transmitted back.

[0032] Among them, the data priority is used to characterize the correlation degree between different initial data types and the current scenario.

[0033] As an example, in step S201, the vehicle terminal determines the data priorities corresponding to the M initial data types in the data to be transmitted back according to the correlation between the M initial data types and the current scenario.

[0034] For example, the data to be transmitted back includes major data categories such as critical safety data, regular driving data, and optimization-related data. Each major data category contains multiple data types. For example, critical safety data includes safety-related data types such as front-view camera data and rear-view camera data. Regular driving data includes vehicle driving-related data types such as environmental change data during vehicle driving. Optimization-related data includes vehicle driving performance data under different road conditions and other performance optimization-related data types. The correlation between critical safety data and the current scenario is relatively high, and the data priority is relatively large. The correlation between regular driving data and the current scenario is medium, and the data priority is medium. The correlation between optimization-related data and the current scenario is relatively low, and the data priority is relatively low. That is, the data priorities corresponding to the safety data type, driving data type, and optimization data type decrease in sequence. The vehicle terminal pre-determines the priority corresponding to each major data category according to the correlation between each major data category and the current scenario. After obtaining the M initial data types, the vehicle terminal matches each initial data type with the data types included in each major data category, determines the major data category to which each initial data type belongs, and determines the priority corresponding to the major data category to which each initial data type belongs as the data priority corresponding to each initial data type.

[0035] Another example is that the vehicle terminal respectively performs a correlation analysis on the M initial data types and the current scenario, determines the correlation between each initial data type and the current scenario, determines a higher data priority for the initial data type with a higher correlation with the current scenario, and determines a lower data priority for the initial data type with a lower correlation with the current scenario.

[0036] As an example, in step S202, the vehicle end determines the importance degree coefficient of the M initial data types in the data to be transmitted back and the measured data corresponding to each initial data type according to the high or low data priorities corresponding to the M initial data types in the data to be transmitted back. It can be understood that the higher the data priority of the initial data type, the higher the importance of the data to be transmitted back corresponding to the initial data type for positioning and analyzing the cause of the fault in the current scenario, and the larger the importance degree coefficient of the initial data type. For example, the vehicle end determines the importance degree coefficient corresponding to the initial data type with a relatively high data priority within the interval of (0.7, 1], determines the importance degree coefficient corresponding to the initial data type with a medium data priority within the interval of (0.5, 0.7], and determines the importance degree coefficient corresponding to the initial data type with a relatively low data priority within the interval of (0, 0.5).

[0037] As shown in Table 1 below, it includes the importance degree coefficients corresponding to some initial data types in the data to be transmitted back in some current scenarios. It can be seen from Table 1 that the initial data types include but are not limited to front view camera data, rear view camera data, planned path data, decision-making data, and control data. represents the importance degree coefficient corresponding to the i-th initial data type in the j-th current scenario. For example, when the vehicle end determines that the current scenario is a forward collision, it determines that among the initial data types, the front view camera data is vehicle system fault data and belongs to critical safety data, corresponding to a relatively high data priority, and determines that the front view camera data corresponds to a relatively large importance degree coefficient . When the vehicle end determines that the current scenario is an abnormal exit of the APA function, it determines that among the data to be transmitted back, the decision-making data is vehicle system fault data and belongs to critical safety data, corresponding to a relatively high data priority, and determines that the decision-making data corresponds to a relatively large importance degree coefficient . In the above manner, the importance degree coefficient A corresponding to each data to be transmitted back in the current scenario is determined ij It can be understood that since the cloud needs to receive the data transmitted back by the vehicle end to analyze the fault in the current scenario, the larger the importance degree coefficient of the data transmitted back to the cloud, the more efficient and accurate the analysis and positioning of the cause of the fault in the current scenario. Therefore, the importance degree coefficient corresponding to the data to be transmitted back is determined, so as to screen the target data to be transmitted back for transmission to the cloud from the data to be transmitted back according to the importance degree coefficient corresponding to each initial data type, and reduce the transmission of the data with a relatively low relevance to the current scenario in the data to be transmitted back, so as to improve the accuracy and efficiency of the vehicle end data transmission, so that the cloud can accurately and efficiently perform the positioning and analysis of the cause of the fault according to the target data to be transmitted back accurately and efficiently transmitted by the vehicle end.

[0038] Table 1 In one embodiment, the target data volume is the product of the system operation index and a preset calibration coefficient; the system operation index is determined based on system operation data.

[0039] Among them, the preset calibration coefficient is a coefficient that is pre-calibrated to correct the system operation index. The system operation index is an index value used to characterize the operation status of the system to which the vehicle end and the cloud end belong. The system operation data is data related to the operation of the system to which the vehicle end and the cloud end belong.

[0040] As an example, the vehicle end obtains the system operation data between the vehicle end and the cloud end, processes the system operation data to determine the system operation index C, and uses a preset calibration coefficient k pre-calibrated according to the transmission effect between the vehicle end and the cloud end to perform a correction process on the system operation index C to determine the target data volume D, that is, D = k C. It can be understood that the target data volume is used to characterize the total data volume transmitted from the vehicle end to the cloud end and is related to the operation status between the vehicle end and the cloud end. Therefore, according to the system operation index used to characterize the operation status of the system to which the vehicle end and the cloud end belong, the target data volume can be determined more accurately.

[0041] In one embodiment, the system operation data includes network connection data and system load data; The system operation index is the difference between the network connection index and the system load index; The network connection index is determined based on the network connection data, and the system load index is determined based on the system load data; Among them, the network connection data is data related to the connection status of the communication network between the vehicle end and the cloud end; the system load data is data related to the load between the vehicle end and the cloud end.

[0042] Among them, the network connection index is index data used to characterize the network connection status between the vehicle end and the cloud end. The system load index is index data used to characterize the load status of the vehicle end.

[0043] As an example, the vehicle end obtains the network connection data, analyzes and processes the network connection data to determine the network connection index used to characterize the network connection status between the vehicle end and the cloud end. It can be understood that since the network connection index is used to characterize the network connection status between the vehicle end and the cloud end, the network connection status is used to reflect the system operation status, the better the network connection status, the better the system operation status, and the system operation index is used to characterize the system operation status. Therefore, the network connection index can effectively affect the system operation index between the vehicle end and the cloud end. Therefore, determining the network connection index between the vehicle end and the cloud end helps to accurately determine the system operation index.

[0044] As an example, the vehicle end obtains system load data, analyzes the system load data, and determines a system load indicator for the load condition of the vehicle end. Understandably, since the system load indicator is used to characterize the system load size between the vehicle end and the cloud end, and the system load size is used to reflect the system operation condition, the larger the system load, the worse the system operation condition, and the system operation indicator is used to characterize the system operation condition. Therefore, the system load indicator can effectively affect the system operation indicator between the vehicle end and the cloud end. Therefore, determining the system load indicator between the vehicle end and the cloud end helps to accurately determine the system operation indicator.

[0045] As an example, the vehicle end takes the network connection indicator and the system load indicator The difference between , and determines it as the system operation indicator C between the vehicle end and the cloud end, that is, C = . Understandably, the network connection indicator is used to characterize the network connection condition between the vehicle end and the cloud end. The larger the network connection indicator , the better the network connection condition, and further indicates that the system operation indicator is better; the smaller the network connection indicator , the worse the network connection condition, and further indicates that the system operation indicator is worse. The system load indicator is used to characterize the load occupancy condition of the system to which the vehicle end belongs. The larger the system load indicator , the more the load occupancy condition, and further indicates that the system operation indicator is worse. The smaller the system load indicator , the less the network connection condition, and further indicates that the system operation indicator is better. Therefore, taking the network connection indicator and the system load indicator The difference between , and determining it as the system operation indicator C between the vehicle end and the cloud end can more reasonably determine the system operation condition between the vehicle end and the cloud end.

[0046] In this embodiment, according to the network connection indicator for characterizing the network connection condition between the vehicle end and the cloud end and the system load indicator for characterizing the load condition of the vehicle end, the system operation indicator for characterizing the operation condition of the system to which the vehicle end and the cloud end belong is accurately determined, so as to subsequently reasonably determine the target data to be transmitted back to the cloud end among the data to be transmitted back according to the accurate system operation indicator.

[0047] In one embodiment, the network connection data includes available bandwidth, network latency, and network latency variance; The network connection indicator is the sum value of the weighted processing of the available bandwidth, network latency, and network latency variance.

[0048] As an example, the vehicle terminal obtains the amount of successfully transmitted data Y within the target time period T, and the time when the vehicle terminal sends data to the cloud , and the time when the cloud receives the data ; the ratio of the amount of successfully transmitted data Y to the target time period T is determined as the available bandwidth B; the time when the cloud receives the data and the time when sending data to the cloud The difference is determined as the network latency L, that is, L = - ; it is also necessary to obtain the network latency m times, calculate the variance corresponding to the network latency m times, and obtain the network latency variance S, where m > 1. In this example, the available bandwidth, network latency, and network latency variance are determined to determine the network connection metrics.

[0049] As an example, the vehicle terminal uses a preset first weight coefficient to correct the available bandwidth B to obtain the first value B, uses a second weight coefficient to correct the network latency L to obtain the second value L, uses a third weight coefficient to correct the network latency variance S to obtain the third value S, according to = B - L - S, determine the network connection metric between the vehicle terminal and the cloud, where 0 < < 1, 0 < < 1, 0 < < 1, + + = 1. Understandably, the network connection metric is used to characterize the network connection status between the vehicle terminal and the cloud. The larger the network connection metric , the better the network connection status between the vehicle terminal and the cloud. In this example, based on the available bandwidth, network latency, and network latency variance, combined with different preset weight coefficients, the network connection metric between the vehicle terminal and the cloud is dynamically determined, so that the calculated network connection metric is related to metrics such as available bandwidth, network latency, and network latency variance, and can comprehensively and effectively reflect the network connection state between the vehicle terminal and the cloud.

[0050] In one embodiment, the system load data includes the CPU occupancy rate and the encoding resource occupancy rate; the system load metric is the sum value of the weighted processing of the CPU occupancy rate and the encoding resource occupancy rate.

[0051] Among them, the CPU occupancy rate refers to the data used to characterize the usage of the CPU on the vehicle side. The encoding resource occupancy rate refers to the data used to characterize the usage of encoding resources when data is transmitted back on the vehicle side.

[0052] As an example, the vehicle side processes a preset weight coefficient to obtain a processed coefficient (1 - ), and uses the preset weight coefficient to correct the CPU occupancy rate R to obtain a first corrected data R, and uses (1 - ) to correct the encoding resource occupancy rate E to obtain a second corrected data (1 - ). E is used to reasonably determine the system load index according to the first corrected data and the second corrected data. The vehicle side determines the sum of the first corrected data R and the second corrected data (1 - ) E as the system load index between the vehicle side and the cloud side . That is = R + (1 - ) E. Understandably, the system load index is used to characterize the load occupancy size of the vehicle side. The larger the system load index , the greater the load occupancy of the vehicle side. In this example, based on the CPU occupancy rate and the encoding resource occupancy rate, combined with the preset weight coefficient , the system load index between the vehicle side and the cloud side is dynamically determined, so that the calculated system load index is related to the CPU occupancy rate and the encoding resource occupancy rate, and can comprehensively reflect the system load index between the vehicle side and the cloud side.

[0053] In this embodiment, by correcting the CPU occupancy rate and the encoding resource occupancy rate based on the preset weight coefficient, the system load index between the vehicle side and the cloud side can be determined more accurately.

[0054] In one embodiment, as shown in Figure 3 , step S104, that is, based on the feedback data sent by the cloud, updating the importance degree coefficients corresponding to the M initial data types in the data to be transmitted back, includes: S301: If the feedback data includes critical data types and non-critical data types, increase the importance coefficient corresponding to the critical data types, and / or decrease the importance coefficient corresponding to the non-critical data types, where the critical data types are target data types whose fault cause correlation determined based on the target feedback data is greater than a preset threshold, and the non-critical data types are target data types whose fault cause correlation determined based on the target feedback data is not greater than the preset threshold; S302: If the feedback data does not include critical data types and non-critical data types, decrease the importance coefficient corresponding to the target data types, and / or increase the importance coefficient corresponding to the remaining data types; where the remaining data types are the data types in the initial data types other than the target data types.

[0055] Among them, the preset threshold is a threshold preset for judging the correlation between the target feedback data of the target data type and the fault cause. The target feedback data types include critical data types and non-critical data types; As an example, in step S301, when the vehicle end determines that the received feedback data is critical data types and non-critical data types in the target feedback data, it judges whether the feedback data includes a first correction coefficient for correcting the importance coefficient corresponding to the critical data types, and / or a first correction coefficient for correcting the importance coefficient corresponding to the non-critical data types. When it is determined that the feedback data includes the first correction coefficient and / or the second correction coefficient, the first correction coefficient is adopted to adjust the importance coefficient corresponding to the critical data types to obtain the adjusted importance coefficient corresponding to the critical data types , and / or adopt the second correction coefficient to adjust the importance coefficient corresponding to the non-critical data types to obtain the adjusted importance coefficient corresponding to the non-critical data types . Among them, the first correction coefficient is greater than 1, and the second correction coefficient is less than 1. In this example, the cloud determines the target data types with a fault cause correlation greater than the preset threshold in the target feedback data as critical data types, and the target data types with a fault cause correlation not greater than the preset threshold as non-critical data types. For example, the cloud determines that among the target data types, the critical data type is the front-view camera data, the non-critical data type is the rear-view camera data, takes the front-view camera data as the critical data type, takes the rear-view camera data as the non-critical data type, and determines the first correction coefficient for correcting the critical data types, and / or , send the front view camera data as the critical data type, the rear view camera data as the non-critical data type, the first correction coefficient and / or the second correction coefficient to the vehicle terminal, and the vehicle terminal adopts the first correction coefficient to adjust the importance coefficient corresponding to the initial data type of the front view camera data at the vehicle terminal, and / or adopt the second correction coefficient to adjust the importance coefficient corresponding to the initial data type of the rear view camera data at the vehicle terminal.

[0056] For another example, when the vehicle terminal determines that the feedback data does not include the first correction coefficient and the second correction coefficient, it adopts a preset rule to determine the first correction coefficient and / or the second correction coefficient. For example, the preset rule is to select the first correction coefficient corresponding to the critical data type within a preset interval greater than 1 , and / or select the second correction coefficient corresponding to the non-critical data type within a preset interval less than 1 , adopt the first correction coefficient to adjust the importance coefficient corresponding to the critical data type to obtain the adjusted importance coefficient corresponding to the critical data type , and / or adopt the second correction coefficient to adjust the importance coefficient corresponding to the non-critical data type to obtain the adjusted importance coefficient corresponding to the non-critical data type . For example, in the target data types determined by the cloud, the front view camera data is the critical data type and the rear view camera data is the non-critical data type. The front view camera data is sent to the vehicle terminal as the critical data type, and the rear view camera data is sent to the vehicle terminal as the non-critical data type. The vehicle terminal adopts a preset rule to determine the first correction coefficient and / or the second correction coefficient, and uses the first correction coefficient to adjust the importance coefficient corresponding to the initial data type of the front view camera data at the vehicle terminal, and / or uses the second correction coefficient to adjust the importance coefficient corresponding to the initial data type of the rear view camera data at the vehicle terminal.

[0057] It can be understood that in the target backhaul data, the measured data corresponding to the critical data type has a greater correlation with the cause of the fault in the current scenario than the preset threshold, indicating that the critical data type has a higher critical degree for fault cause analysis in the current scenario. Therefore, it is necessary to increase the importance coefficient corresponding to the critical data type in the vehicle terminal, and / or decrease the importance coefficient corresponding to the non-critical data type, so as to increase the priority of the critical data type in the vehicle terminal, and / or decrease the priority of the non-critical data type in the vehicle terminal, so that when the subsequent data of the current scenario is backhauled, as much target backhaul data corresponding to the critical data type can be backhauled as possible, so as to improve the efficiency and accuracy of data backhaul, so that the cloud can accurately and efficiently locate and analyze the cause of the fault.

[0058] As an example, in step S302, when the vehicle end determines that the feedback data does not include the key data type and the non-key data type, it determines whether the feedback data includes a third correction coefficient for correcting the importance coefficient corresponding to the target data type, and / or a fourth correction coefficient for correcting the importance coefficient corresponding to the remaining data types. When it is determined that the feedback data includes the third correction coefficient and / or the fourth correction coefficient it uses the third correction coefficient to correct the importance coefficient corresponding to the target data type to obtain the adjusted importance coefficient corresponding to the target data type so as to reduce the importance coefficient corresponding to the target data type, and / or uses the fourth correction coefficient to correct the importance coefficient corresponding to the remaining data types to obtain the adjusted importance coefficient corresponding to the remaining data types so as to increase the importance coefficient corresponding to the remaining data types other than the target data type in the initial data type, where the third correction coefficient is less than 1 and the fourth correction coefficient is greater than 1. When the vehicle end determines that the feedback data does not include the third correction coefficient and the fourth correction coefficient it uses a preset rule to determine the third correction coefficient and / or the fourth correction coefficient For example, the preset rule is to select the third correction coefficient corresponding to the target data type within a preset interval less than 1 and / or select the fourth correction coefficient corresponding to the remaining data types within a preset interval greater than 1 It uses the third correction coefficient to correct the importance coefficient corresponding to the target data type to obtain the adjusted importance coefficient corresponding to the target data type so as to reduce the importance coefficient corresponding to the target data type, and / or uses the fourth correction coefficient to correct the importance coefficient corresponding to the remaining data types to obtain the adjusted importance coefficient corresponding to the remaining data types so as to increase the importance coefficient corresponding to the remaining data types other than the target data type in the initial data type.

[0059] Understandably, if there is no critical data type and non-critical data type in the feedback data, it indicates that the cloud cannot locate and analyze the cause of the fault based on the measured data corresponding to the target data type. It is necessary to increase the importance coefficient corresponding to the remaining data types that have not been transmitted back to the cloud in the data to be transmitted back, and / or reduce the importance coefficient corresponding to the target data type in the data to be transmitted back, so as to transmit as many of the remaining data types in the data to be transmitted back as possible, reduce the transmission of data unrelated to the cause of the fault, and enable the cloud to locate and analyze the cause of the fault corresponding to the current scenario more accurately and efficiently.

[0060] In this embodiment, according to the received feedback data from the cloud, the importance coefficient corresponding to the initial data type in the data to be transmitted back is adjusted in real time, so as to ensure that the target transmitted-back data of the target data type transmitted back to the cloud plays a key role in the fault analysis of the current scenario, and realize the efficient and accurate transmission of the target transmitted-back data by the vehicle side, so that the cloud can locate and analyze the cause of the fault corresponding to the current scenario more accurately and efficiently.

[0061] In another embodiment, a data transmission method is provided. Taking the cloud in Figure 5 as an example, the following steps executed by the cloud are included: S1031: Receive the target transmitted-back data sent by the vehicle side. The target transmitted-back data includes N target data types determined from M initial data types and the measured data corresponding to the N target data types. Among them, the total data volume of the target transmitted-back data is less than the target data volume, and the sum of the importance coefficients corresponding to the N target data types is the largest, 1≤N≤M; S1032: Analyze the measured data of the N target data types to determine the feedback data, and send the feedback data to the vehicle side, so that the vehicle side updates the importance coefficient corresponding to the M initial data types in the data to be transmitted back based on the feedback data.

[0062] As an example, in step S1031, the cloud receives in real time the measured data of the N target data types in the target transmitted-back data sent by the vehicle side, so as to locate and analyze the cause of the fault for the current scenario according to the measured data of the N target data types. In this example, the current scenario includes, but is not limited to, the abnormal exit of the ACC function, the abnormal exit of the ICA function, the abnormal exit of the APA function, and the occurrence of a forward collision shown in Table 1 above. The target data types include, but are not limited to, the front view camera data, the rear view camera data, the planned path data, the decision-making data, and the control data shown in Table 1 above.

[0063] As an example, in step S1032, the cloud performs fault analysis and processing based on the measured data of N target data types. According to whether the measured data of the N target data types can locate the fault cause corresponding to the current scenario of the vehicle terminal, the feedback data to be sent to the vehicle terminal is determined, and the feedback data is sent to the vehicle terminal, so that the vehicle terminal adaptively adjusts the importance coefficient corresponding to M initial data types in the data to be transmitted back according to the feedback data of the cloud, and transmits the target data to be transmitted back that can efficiently and accurately locate the fault cause corresponding to the current scenario of the vehicle terminal to the cloud. Understandably, the cloud analyzes the measured data of the N target data types in the target data to be transmitted back, determines the feedback data for enabling the vehicle terminal to adjust the importance coefficient, and realizes the real-time adjustment of the importance coefficient corresponding to the initial data types in the data to be transmitted back by the vehicle terminal, so that after multiple rounds of adjustment, the vehicle terminal can adaptively transmit the target data to be transmitted back to the cloud more efficiently and accurately, so that the cloud can efficiently and accurately locate the fault cause corresponding to the current scenario of the vehicle terminal according to the target data to be transmitted back.

[0064] In this embodiment, the cloud receives the target data to be transmitted back sent by the vehicle terminal, analyzes the target data to be transmitted back, determines the feedback data, and transmits the feedback data to the vehicle terminal, so that the vehicle terminal can adaptively transmit the target data to be transmitted back to the cloud more efficiently and accurately, facilitating the cloud to efficiently and accurately locate the fault cause corresponding to the current scenario of the vehicle terminal.

[0065] In one embodiment, step S1032, that is, analyzing the measured data of N target data types, determining the feedback data, and sending the feedback data to the vehicle terminal, so that the vehicle terminal updates the importance coefficient corresponding to M initial data types in the data to be transmitted back based on the feedback data, includes: S1032A: Perform fault analysis on the measured data of N target data types to determine the data analysis result; S1032B: If the data analysis result is that the fault cause can be determined, then determine that the feedback data includes critical data types and non-critical data types, and send the feedback data to the vehicle terminal, so that the vehicle terminal increases the importance coefficient corresponding to the critical data types and / or decreases the importance coefficient corresponding to the non-critical data types; where the critical data type is a target data type whose fault cause correlation determined based on the measured data in the target data to be transmitted back is greater than the preset threshold, and the non-critical data type is a target data type whose fault cause correlation determined based on the measured data in the target data to be transmitted back is not greater than the preset threshold; S1032C: If the data analysis result is that the cause of the fault cannot be determined, it is determined that the feedback data does not include the critical data type and the non-critical data type, and the fact that the cause of the fault cannot be determined is sent to the vehicle terminal as the feedback data, so that the vehicle terminal reduces the importance coefficient corresponding to the target data type and / or increases the importance coefficient of the remaining data types; wherein, the remaining data types are the data types in the initial data types except the target data type.

[0066] Wherein, the data analysis result is used to represent whether the cloud can locate and analyze the cause of the fault corresponding to the current scenario based on the measured data in the target feedback data.

[0067] As an example, in step S1032A, the cloud locates and analyzes the cause of the fault corresponding to the current scenario based on the measured data of N target data types in the target feedback data transmitted back by the vehicle terminal, and determines whether the cause of the fault corresponding to the current scenario can be located to obtain the data analysis result. The data analysis result includes being able to locate the cause of the fault and not being able to locate the cause of the fault.

[0068] As an example, in step S1032B, when the cloud determines that the data analysis result is that the cause of the fault can be located, it further analyzes the importance of the measured data of N target data types, determines the critical data types that are more important for fault cause location and the non-critical data types that are less important for fault cause location among the N target data types, and determines the critical data types and the non-critical data types as the feedback data and sends them to the vehicle terminal, so that the vehicle terminal increases the importance coefficient corresponding to the critical data type and / or reduces the importance coefficient corresponding to the non-critical data type. In this example, the cloud analyzes the correlation between the measured data of each target data type and the cause of the fault, obtains the fault cause correlation corresponding to each target data type, determines the target data type with the fault cause correlation greater than the preset threshold as the critical data type, and determines the target data type with the fault cause correlation not greater than the preset threshold as the non-critical data type.

[0069] In this example, after the cloud determines the critical data type and the non-critical data type, within a preset interval greater than 1, it selects the first correction coefficient for adjusting the importance coefficient corresponding to the critical data type, and within a preset interval less than 1, it selects the second correction coefficient for adjusting the importance coefficient corresponding to the non-critical data type, and determines the critical data type, the first correction coefficient corresponding to the critical data type, the non-critical data type, and / or the second correction coefficient corresponding to the non-critical data type as the feedback data and sends it to the vehicle terminal.

[0070] Understandably, the cloud hopes to receive as much measured data corresponding to the key data types transmitted back from the vehicle side as possible. Therefore, it is necessary to increase the importance coefficient corresponding to the key data types on the vehicle side, and the first correction coefficient for correcting the importance coefficient corresponding to the key data types is greater than 1. The cloud hopes to receive as little measured data corresponding to the non-key data types transmitted back from the vehicle side as possible. Therefore, it is necessary to decrease the importance coefficient corresponding to the non-key data types on the vehicle side, and the second correction coefficient for correcting the importance coefficient corresponding to the non-key data types is less than 1. For example, if the current scenario is a forward collision, after the cloud receives the target transmitted-back data and determines that the data analysis result is that the cause of the forward collision can be located, then based on the measured data of each target data type in the target transmitted-back data, the importance analysis of the target data type is further carried out to determine that among the target data types, the front-view camera data is a key data type, the rear-view camera data is a non-key data type, and the first correction coefficient greater than 1 corresponding to the key data type is determined, the second correction coefficient less than 1 corresponding to the non-key data type is determined, and the front-view camera data, the rear-view camera data, the first correction coefficient, and / or the second correction coefficient are determined as feedback data, so that the vehicle side can adjust the importance coefficient corresponding to the front-view camera data according to the first correction coefficient, and / or adjust the importance coefficient corresponding to the rear-view camera data according to the second correction coefficient.

[0071] As an example, in step S1032C, when the cloud determines that the data analysis result is that the cause of the failure cannot be located, it sends the inability to determine the cause of the failure as feedback data to the vehicle side, so that the vehicle side reduces the importance coefficient corresponding to the target data type and / or increases the importance coefficient of the remaining data types. In this example, when the cloud determines that the data analysis result is that the cause of the failure cannot be determined, it can also determine the third correction coefficient less than 1 and the fourth correction coefficient greater than 1 as feedback data, where the third correction coefficient is used to correct the importance coefficient corresponding to the target data type in the data to be transmitted back from the vehicle side, and the fourth correction coefficient is used to correct the importance coefficient corresponding to the remaining data types other than the target data type among the M initial data types in the data to be transmitted back from the vehicle side. Understandably, since the cause of the failure cannot be located based on the measured data corresponding to the currently transmitted target data type, it is necessary to transmit as much of the remaining data types in the initial data types of the data to be transmitted back that have not been transmitted to the cloud as possible, and transmit as little of the data that has already been transmitted, in order to locate and analyze the cause of the failure for the current scenario more efficiently and accurately. Therefore, the third correction coefficient for adjusting the importance coefficient corresponding to the target data type is less than 1, and the fourth correction coefficient for adjusting the importance coefficient corresponding to the remaining data types is greater than 1.

[0072] In this embodiment, different feedback data is determined according to whether the cause of the fault can be determined, so that the vehicle end adjusts the importance coefficient corresponding to M initial data types in the data to be transmitted back according to the feedback data corresponding to different analysis results. According to the adjusted importance coefficient, the target data to be transmitted back is transmitted to the cloud more efficiently and accurately, so that the cloud can accurately and efficiently locate and analyze the cause of the fault corresponding to the current scenario based on the target data to be transmitted back. After the vehicle end adjusts the importance coefficient corresponding to the initial data type in the data to be transmitted back multiple times according to the feedback data, the cloud can receive the target data to be transmitted back with a high correlation with the current scenario, and accurately analyze and locate the cause of the fault in the current scenario, which is relatively efficient and convenient.

[0073] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution is prior or posterior. The execution order of each process should be determined according to its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0074] In one embodiment, a data transmission-back device is provided, and the data transmission-back device corresponds one-to-one with the data transmission-back method in the above embodiment. As Figure 4 shown, the data transmission-back device includes an importance coefficient determination module 401, a target data transmission-back determination module 402, a target data transmission-back sending module 403, and an importance coefficient update module 404. The detailed description of each functional module is as follows: The importance coefficient determination module 401 determines the importance coefficient and measured data corresponding to M initial data types in the data to be transmitted back based on the current scenario; The target data transmission-back determination module 402 is used to determine N target data types from M initial data types, form target data to be transmitted back based on the measured data corresponding to the N target data types, the total data volume of the target data to be transmitted back is less than the target data volume, and the sum of the importance coefficients corresponding to the N target data types is the largest, 1≤N≤M; The target data transmission-back sending module 403 is used to send the target data to be transmitted back to the cloud, so that the cloud analyzes the measured data corresponding to the N target data types to determine feedback data; The importance coefficient update module 404 updates the importance coefficient corresponding to M initial data types in the data to be transmitted back based on the feedback data sent by the cloud.

[0075] In one embodiment, the importance coefficient determination module 401 includes: The data priority determination sub-module determines the data priority and measured data corresponding to M initial data types in the data to be transmitted back based on the current scenario; The importance coefficient determination sub-module determines the importance coefficients of the M initial data types in the data to be transmitted back based on the data priorities corresponding to the M initial data types in the data to be transmitted back.

[0076] In one embodiment, the importance coefficient update module 404 includes: The first update sub-module is used to increase the importance coefficient corresponding to the critical data type and / or decrease the importance coefficient corresponding to the non-critical data type if the feedback data includes the critical data type and the non-critical data type, where the critical data type is the target data type with a failure cause correlation greater than the preset threshold determined based on the target data to be transmitted back, and the non-critical data type is the target data type with a failure cause correlation not greater than the preset threshold determined based on the target data to be transmitted back; The second update sub-module is used to decrease the importance coefficient corresponding to the target data type and / or increase the importance coefficient corresponding to the remaining data types if the feedback data does not include the critical data type and the non-critical data type; where the remaining data types are the data types other than the target data type among the initial data types.

[0077] For the specific limitations of the data transmission back device, reference can be made to the limitations on the data transmission back method in the above text, which will not be elaborated here. Each module in the above data transmission back device can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in the processor in the computer device in hardware form or independent of it, or stored in the memory in the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0078] In one embodiment, a vehicle is provided, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the data transmission back method in the above embodiment is implemented, such as Figure 1 S101 - S104 shown, or Figures 2 to 3 as shown in, to avoid repetition, it will not be elaborated here. Or, when the processor executes the computer program, the functions of each module / unit in this embodiment of the above data transmission back device are implemented, such as Figure 4 the functions of the importance coefficient determination module 401, the target data to be transmitted back determination module 402, the target data to be transmitted back sending module 403, and the importance coefficient update module 404 shown, to avoid repetition, it will not be elaborated here.

[0079] In one embodiment, a computer-readable storage medium is provided. A computer program is stored on the computer-readable storage medium. When the computer program is executed by the processor, the data transmission back method in the above embodiment is implemented, such as Figure 1 S101 - S104 shown, or Figures 2 to 3As shown, to avoid repetition, it will not be elaborated here. Alternatively, when the computer program is executed by a processor, it implements the functions of the various modules / units in the above-described embodiment of the data feedback device. For example Figure 4 the functions of the importance degree coefficient determination module 401, the target feedback data determination module 402, the target feedback data sending module 403, and the importance degree coefficient update module 404 shown. To avoid repetition, it will not be elaborated here.

[0080] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium used in the various embodiments provided in the present application can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0081] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the above division of the functional units and modules is used as an example. In actual applications, the above functions can be allocated to different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above.

[0082] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A data feedback method, characterized in that, Including the following steps performed at the vehicle end: Based on the current scenario, determine the importance degree coefficients and measured data corresponding to M initial data types in the data to be transmitted back; Determine N target data types from the M initial data types, and form target data to be transmitted back based on the measured data corresponding to the N target data types. The total data volume of the target data to be transmitted back is less than the target data volume, and the sum of the importance degree coefficients corresponding to the N target data types is the largest, where 1 ≤ N ≤ M; Send the target data to be transmitted back to the cloud so that the cloud can analyze the measured data corresponding to the N target data types to determine the feedback data; Based on the feedback data sent by the cloud, update the importance degree coefficients corresponding to the M initial data types in the data to be transmitted back.

2. The data feedback method according to claim 1, wherein The step of determining the importance degree coefficients and measured data corresponding to the M initial data types in the data to be transmitted back based on the current scenario includes: Based on the current scenario, determine the data priorities corresponding to the M initial data types in the data to be transmitted back; Based on the data priorities corresponding to the M initial data types in the data to be transmitted back, determine the importance degree coefficients and measured data corresponding to the M initial data types in the data to be transmitted back.

3. The data feedback method according to claim 1, wherein The target data volume is the product of the system operation index and the preset calibration coefficient; The system operation index is determined based on the system operation data; 4. The data feedback method according to claim 3, wherein The system operation data includes network connection data and system load data; The system operation index is the difference between the network connection index and the system load index; The network connection index is determined based on the network connection data, and the system load index is determined based on the system load data; Wherein, the network connection data is data related to the communication network between the vehicle end and the cloud; the system load data is data related to the load between the vehicle end and the cloud.

5. The data feedback method according to claim 4, wherein The network connection data includes available bandwidth, network latency, and network latency variance; The network connection index is the sum value obtained by weighted processing of the available bandwidth, the network latency, and the network latency variance.

6. The data feedback method according to claim 4, wherein The system load data includes CPU occupancy rate and encoding resource occupancy rate; The system load index is the sum value obtained by weighted processing of the CPU occupancy rate and the encoding resource occupancy rate.

7. The data feedback method according to claim 1, wherein The step of updating the importance degree coefficients corresponding to the M initial data types in the data to be transmitted back based on the feedback data sent by the cloud includes: If the feedback data includes key data types and non-key data types, increase the importance degree coefficients corresponding to the key data types and / or decrease the importance degree coefficients corresponding to the non-key data types, where the key data types are target data types whose fault cause correlation based on the target data to be transmitted back is greater than the preset threshold, and the non-key data types are target data types whose fault cause correlation based on the target data to be transmitted back is not greater than the preset threshold; If the feedback data does not include key data types and non-key data types, decrease the importance degree coefficients corresponding to the target data types and / or increase the importance degree coefficients corresponding to the remaining data types; where the remaining data types are the data types in the initial data types other than the target data types.

8. A data feedback device, characterized in that, Including: An importance coefficient determination module determines the importance coefficients and measured data corresponding to M initial data types in the data to be transmitted back based on the current scenario. A target transmitted-back data determination module is used to determine N target data types from the M initial data types, and form target transmitted-back data based on the measured data corresponding to the N target data types. The total data volume of the target transmitted-back data is less than the target data volume, and the sum of the importance coefficients corresponding to the N target data types is the largest, where 1 ≤ N ≤ M. A target transmitted-back data sending module is used to send the target transmitted-back data to the cloud so that the cloud analyzes the measured data corresponding to the N target data types to determine feedback data. An importance coefficient update module updates the importance coefficients corresponding to the M initial data types in the data to be transmitted back based on the feedback data sent by the cloud.

9. A vehicle, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the data transmission-back method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the data transmission-back method according to any one of claims 1 to 7.

Images