Product quality determination methods, electronic devices, storage media and program products

Abstract

This application provides a product quality determination method, electronic device, storage medium, and program product, relating to the field of vehicle technology, and applicable to the quality risk determination of high-level intelligent products. The product quality determination method includes: comparing target characteristics of a target vehicle with those of competing vehicles to obtain a competition coefficient; acquiring test data of the target vehicle; obtaining an anomaly probability coefficient based on the number of anomalies occurring in the test data, wherein the anomaly probability coefficient is positively correlated with the number of anomalies; obtaining a severity coefficient based on the severity of the consequences of anomalies occurring in the test data, wherein the severity of the consequences is positively correlated with the severity coefficient; and calculating a quality quantification value based on the competition coefficient, the anomaly probability coefficient, and the severity coefficient, wherein any one of the competition coefficient, the anomaly probability coefficient, and the severity coefficient is positively correlated with the quality quantification value.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, specifically to a product quality determination method, electronic equipment, storage medium, and program product. Background Technology

[0002] With the rapid development of vehicle intelligence, Level 2 driver assistance technology has entered the commercial stage, and Level 1 to Level 2 smart cockpit technologies have been widely applied in vehicles. The principles and characteristics of higher-level intelligent products are more complex, and their application conditions are also more demanding, rendering the quality standards of lower-level intelligent products inapplicable. For example, current Failure Mode and Effect Analysis (FMEA) methods are qualitative analysis techniques for proactive prevention, primarily applicable to the product design phase, and not suitable for determining the quality risks of higher-level intelligent products. Summary of the Invention

[0003] In view of this, this application provides a product quality determination method, electronic device, storage medium, and program product, which can be applied to the determination of quality risks of high-level intelligent products.

[0004] Firstly, a method for determining product quality is provided, comprising: comparing the target characteristics of a target vehicle with those of competing vehicles to obtain a competition coefficient, wherein the target vehicle is a vehicle applying assisted driving products or intelligent cockpit products; if the competing vehicle does not have the target characteristics, the competition coefficient is the minimum competition coefficient; if the competing vehicle has the target characteristics, and the target characteristics have a target defect in the target vehicle, and the competing vehicle does not have the target defect, the competition coefficient is the maximum competition coefficient; acquiring test data of the target vehicle; obtaining an anomaly probability coefficient based on the number of times the target characteristics anomalies occur in the test data, wherein the anomaly probability coefficient is positively correlated with the number of anomalies; obtaining a severity coefficient based on the severity of the consequences of the anomalies occurring in the test data, wherein the severity of the consequences is positively correlated with the severity coefficient; calculating a quality quantification value based on the competition coefficient, the anomaly probability coefficient, and the severity coefficient, wherein any one of the competition coefficient, the anomaly probability coefficient, and the severity coefficient is positively correlated with the quality quantification value; and determining the corresponding quality risk level based on the range to which the quality quantification value belongs.

[0005] In obtaining the probability coefficient and severity coefficient of anomalies, statistics can be based on standard beta testing by end users. This means quality analysis is based on real anomalies or failures that have already been detected, rather than just preventative analysis and assessment of potential failures or simply evaluating the probability of a failure being detected. A competitiveness coefficient is introduced, using horizontal comparison with competing products as a factor in quality considerations, reflecting users' acceptance of some leading technologies. In addition, the severity coefficient also considers users' acceptance of anomalies or functional failures. Furthermore, a quantifiable quality determination method is provided, which can quickly analyze the acquired data to obtain accurate quality quantification results. Based on these quality quantification results, release strategies can be easily formulated.

[0006] In some possible implementations, the target defect is a target anomaly; if a competitor vehicle has a target characteristic, and the target characteristic has a target anomaly occurring on both the target vehicle and the competitor vehicle, and under the same testing conditions, the probability of the target anomaly occurring on the target vehicle is higher than the probability of it occurring on the competitor vehicle, and the difference between the probability of the target anomaly occurring on the target vehicle and the probability of it occurring on the competitor vehicle is greater than a first threshold, then the competition coefficient is a first competition coefficient; if a competitor vehicle has a target characteristic, and the target characteristic has a target anomaly occurring on both the target vehicle and the competitor vehicle, and under the same testing conditions, the probability of the target anomaly occurring on the competitor vehicle is higher than the probability of it occurring on the competitor vehicle, then the competition coefficient is a first competition coefficient; If the absolute value of the difference in the probability of occurrence of the test on the target vehicle is not greater than the first threshold, then the competition coefficient is the second competition coefficient. If the competitor vehicle has the target characteristic, and the target characteristic has target anomalies on both the target vehicle and the competitor vehicle, and under the same test conditions, the probability of occurrence of the target anomaly on the competitor vehicle is greater than the probability of occurrence on the target vehicle, and the difference between the probability of occurrence of the target anomaly on the competitor vehicle and the probability of occurrence on the target vehicle is greater than the first threshold, then the competition coefficient is the third competition coefficient. The maximum competition coefficient > the first competition coefficient > the second competition coefficient > the third competition coefficient > the minimum competition coefficient.

[0007] In some possible implementations, the target characteristic is a characteristic of assisted driving, and the target anomaly is an abnormal takeover.

[0008] In some possible implementations, obtaining the anomaly probability coefficient based on the number of times the target characteristic anomalies occur in the test data includes: obtaining a first sub-coefficient based on the number of times accidents occur in the test data, where the first sub-coefficient is positively correlated with the number of accidents; if the number of accidents is greater than or equal to a second threshold, then the first sub-coefficient is the maximum probability coefficient; obtaining a second sub-coefficient based on the number of times traffic rule compliance anomalies occur in the test data, where the second sub-coefficient is positively correlated with the number of times traffic rule compliance anomalies occur; if the probability of traffic rule compliance anomalies is 100%, then the second sub-coefficient is the maximum probability coefficient; and obtaining a third sub-coefficient based on the number of times active safety measures anomalies occur in the test data, where the third sub-coefficient is positively correlated with the number of times active safety measures anomalies occur. The frequency of occurrences is positively correlated. If the probability of an abnormality in active safety measures is 100%, then the third sub-coefficient is the maximum probability coefficient. The fourth sub-coefficient is obtained based on the frequency of abnormalities in the functionality of the target characteristic in the test data. The fourth sub-coefficient is positively correlated with the frequency of abnormalities in functionality. If the probability of an abnormality in functionality is 100%, then the fourth sub-coefficient is the maximum probability coefficient. The fifth sub-coefficient is obtained based on the frequency of abnormalities in the perceived experience of the target characteristic in the test data. The fifth sub-coefficient is positively correlated with the frequency of abnormalities in perceived experience. If the probability of an abnormality in perceived experience is 100%, then the fifth sub-coefficient is the maximum probability coefficient. The probability coefficient of anomaly occurrence is the largest among the first, second, third, fourth, and fifth sub-coefficients.

[0009] When setting sub-coefficients, accidents have stricter standards compared to other anomalies. This is determined based on the user's acceptance of different anomalies. Users have a low acceptance of accidents, so a sub-coefficient of 10 is set as long as the number of accidents reaches 5 times or more. For other types of anomalies, a sub-coefficient of 10 is only set when the probability of occurrence reaches 100%, so that the anomaly occurrence probability coefficient can more accurately reflect the user's acceptance of anomalies.

[0010] In some possible implementations, if the number of accidents is less than a second threshold but greater than 1, then the first sub-coefficient is the first coefficient value, which is less than the maximum probability coefficient; if the number of accidents is 1, then the first sub-coefficient is less than the first coefficient value; if the probability of traffic rule compliance exceptions is less than 100%, and the number of traffic rule compliance exceptions is greater than or equal to the second threshold, then the second sub-coefficient is the first coefficient value; if the number of traffic rule compliance exceptions is less than the second threshold but greater than 1, then the second sub-coefficient is the second coefficient value, which is less than the first coefficient value; if the probability of traffic rule compliance exceptions is less than 100%, and the number of traffic rule compliance exceptions is greater than or equal to the second threshold, then the second sub-coefficient is the first coefficient value; if the probability of traffic rule compliance exceptions is less than 100%, and the number of traffic rule compliance exceptions is greater than or equal to the second threshold, then the second sub-coefficient is the second coefficient value, which is less than ... then the probability of traffic rule compliance exceptions is less than 100%. If the probability of rule compliance anomaly occurring is 1, then the second sub-coefficient is less than the second coefficient value; if the probability of proactive safety measures anomaly occurring is less than 100%, and the number of proactive safety measures anomalies is greater than or equal to the third threshold, then the third sub-coefficient is the first coefficient value, and the third threshold is greater than the second threshold; if the number of proactive safety measures anomalies occurring is not greater than the third threshold, but is greater than or equal to the second threshold, then the third sub-coefficient is the second coefficient value; if the number of proactive safety measures anomalies occurring is not greater than the second threshold, but is greater than 1, then the third sub-coefficient is the third coefficient value, and the third coefficient value is less than the second coefficient value; if the probability of proactive safety measures anomalies occurring is less than 100%, and is greater than 1, then the third sub-coefficient is the third coefficient value, and the third coefficient value is less than the second coefficient value; if the probability of proactive safety measures anomalies occurring is less than 100%, and is greater than or equal to the third threshold ... third coefficient value, and the third coefficient value is less than the second coefficient value; if the probability of proactive safety measures anomalies occurring is less than 100%, then the third sub-coefficient is the second coefficient value, and the third sub-coefficient is the third coefficient value, and the third sub-coefficient is the third coefficient value, and the third If the total number of times an anomaly occurs is 1, then the third sub-coefficient becomes the fourth coefficient value, and the fourth coefficient value is less than the third coefficient value; if the probability of an anomaly occurring in the functionality is less than 100%, and the number of times an anomaly occurs in the functionality is greater than or equal to the third threshold, then the fourth sub-coefficient becomes the fifth coefficient value, and the first coefficient value > the fifth coefficient value > the second coefficient value; if the number of times an anomaly occurs in the functionality is not greater than the third threshold, but is greater than or equal to the second threshold, then the fourth sub-coefficient becomes the second coefficient value; if the number of times an anomaly occurs in the functionality is not greater than the second threshold, but is greater than 1 time, then the fourth sub-coefficient becomes the third coefficient value; if the probability of an anomaly occurring in the functionality is less than 100%, and the number of times an anomaly occurs in the functionality is greater than the second threshold ... then the fourth sub-coefficient becomes the second coefficient value; if the probability of an anomaly occurring in the functionality is If the frequency of occurrence is 1, then the fourth sub-coefficient is the fourth coefficient value; if the probability of the perceived abnormality is <100%, and the frequency of the perceived abnormality is greater than or equal to the third threshold, then the fifth sub-coefficient is the fifth coefficient value; if the frequency of the perceived abnormality is not greater than the third threshold, but greater than or equal to the second threshold, then the fifth sub-coefficient is the second coefficient value; if the frequency of the perceived abnormality is not greater than the second threshold, but greater than 1, then the fifth sub-coefficient is the third coefficient value; if the frequency of the perceived abnormality is 1, then the fifth sub-coefficient is the sixth coefficient value, and the sixth coefficient value is < the fourth coefficient value.

[0011] Users have different levels of acceptance of the same number of occurrences of different types of anomalies. Therefore, different sub-coefficient setting standards are set according to different types, so that the anomaly occurrence probability coefficient can more accurately reflect the user's acceptance of the occurrence of anomalies.

[0012] In some possible implementations, if an accident occurs once and is always reproduced in repeated tests at the road segment where the accident occurred, then the first sub-coefficient is the second coefficient value; if an accident occurs once and is only occasionally reproduced in repeated tests at the road segment where the accident occurred, then the first sub-coefficient is the seventh coefficient value, and the second coefficient value > the seventh coefficient value > the third coefficient value; if an accident occurs once and is not reproduced in repeated tests at the road segment where the accident occurred, then the first sub-coefficient is the third coefficient value; if a traffic rule compliance anomaly occurs once and is always reproduced in repeated tests at the road segment where the traffic rule compliance anomaly occurred, then the second sub-coefficient is the seventh coefficient value, and the third coefficient value > the seventh coefficient value > the fourth coefficient value; if a traffic rule compliance anomaly occurs once and is only occasionally reproduced in repeated tests at the road segment where the traffic rule compliance anomaly occurred, then the second sub-coefficient is the fourth coefficient value; if a traffic rule compliance anomaly occurs once and is not reproduced in repeated tests at the road segment where the traffic rule compliance anomaly occurred, then the second sub-coefficient is the sixth coefficient value.

[0013] For accidents or traffic violations that occur only once during the test, the degree of risk can be further determined by whether they recur on the same road segment. This means further determining the user's acceptance of the anomaly and thus determining the sub-coefficient. This allows the anomaly probability coefficient to more accurately reflect the user's acceptance of the anomaly.

[0014] In some possible implementations, the range includes a first range and a second range, where the maximum value of the first range is less than the minimum value of the second range. If the quality quantification value belongs to the first range and none of the competition coefficient, anomaly probability coefficient, and severity coefficient has a maximum value, then the quality risk level is Level 1. If the quality quantification value belongs to the first range and at least one of the competition coefficient, anomaly probability coefficient, and severity coefficient has a maximum value, then the quality risk level is Level 2. If the quality quantification value belongs to the second range and none of the competition coefficient, anomaly probability coefficient, and severity coefficient has a maximum value, then the quality risk level is Level 3. If the quality quantification value belongs to the second range and at least one of the competition coefficient, anomaly probability coefficient, and severity coefficient has a maximum value, then the quality risk level is Level 4.

[0015] The severity of the consequences of an anomaly reflects the user's level of acceptance of the anomaly. Consequences that do not comply with safety and / or traffic regulations are the most unacceptable to users, so they have the highest severity and the corresponding severity coefficient. The severity of the consequences refers to the severity of the impact on the product, or the impact on the user. Based on this principle, a severity coefficient is set according to the severity of the consequences of the anomaly to reflect the user's level of acceptance of the anomaly.

[0016] In some possible implementations, the quality determination method further includes determining that the quality is unqualified if the test data of the target vehicle contains any of the conditions listed in the preset list.

[0017] The preset list is an absolute quality red line list. You can first determine whether there are any situations in the preset list based on the test data. If any abnormality or problem is found in the contents of the preset list, the quality will be rejected directly, and no further quality assessment will be conducted.

[0018] In a second aspect, an electronic device is provided, comprising: a processor and a memory, the memory being used to store at least one instruction, which, when loaded and executed by the processor, causes the electronic device to perform the method described above.

[0019] Thirdly, a computer-readable storage medium is provided, including a program or instructions, wherein the above-described methods are executed when the program or instructions are run on a computer.

[0020] Fourthly, a computer program product is provided, which includes executable instructions that, when executed on a computer, cause the computer to perform the methods described above. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is a flowchart illustrating a quality determination method according to an embodiment of this application;

[0023] Figure 2 This is a flowchart illustrating another quality determination method in an embodiment of this application;

[0024] Figure 3 This is a flowchart illustrating another quality determination method in an embodiment of this application;

[0025] Figure 4 This is a structural block diagram of an electronic device according to an embodiment of this application;

[0026] Figure 5 This is a schematic diagram of an electronic device applied in a system according to an embodiment of this application. Detailed Implementation

[0027] To better understand the technical solution of this application, the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0028] It should be understood that the described embodiments are merely some, not all, of the embodiments in this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0029] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. The singular forms “a,” “the,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise.

[0030] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0031] Before describing the embodiments of this application, the shortcomings of FMEA should first be explained. FMEA is a qualitative analysis technique for prevention, mainly applied in the product design stage. It analyzes and predicts potential design and process failures, but it is not a quality analysis based on actual failures. For example, the frequency statistics in FMEA use probability theory and statistical principles to process and analyze historical data to calculate the probability of occurrence, rather than statistics based on standard beta testing by end users. Furthermore, the detectability risk assessment in FMEA evaluates the probability of a failure being detected after it occurs, rather than a segmented assessment based on already detected failures. Additionally, FMEA does not consider the user's acceptance level of the feature quality. The embodiments of this application can improve upon the shortcomings of FMEA, and the embodiments of this application will be described below.

[0032] Current driver assistance technologies (ADAS) involve equipping vehicles with intelligent systems and various sensor devices to assist driving, such as enabling automatic lane changing and on / off ramp maneuvers in specific scenarios. Intelligent cockpit technology aims to integrate various IT and artificial intelligence technologies to create an integrated digital platform within the vehicle, providing drivers with an intelligent experience and promoting driving safety. Manufacturers continuously develop new features and optimize existing ones for their ADAS or intelligent cockpit products, potentially releasing these intelligent product versions through new car models or new versions. The embodiments in this application can be used to determine the quality of these intelligent products, thereby informing the determination of a product release strategy based on the determined quality.

[0033] This application provides a method for determining product quality, including:

[0034] Step 101: Compare the target characteristics of the target vehicle with those of competing vehicles to obtain the competitiveness coefficient. The target vehicle is a vehicle that uses driver assistance products or smart cockpit products, and the target characteristics are the features of the target vehicle. For example, the target characteristic is the Navigate on Autopilot (NOA) function, which allows the vehicle to drive automatically on highways, including automatic lane changing, overtaking, intersection driving, and automatic entry and exit from highway exits. Each target vehicle using intelligent products will have a corresponding competing vehicle already in the market. The competitiveness coefficient is determined by comparing the target vehicle with the competing vehicle and reflects the perceived superiority of the features.

[0035] If competing vehicles do not possess the target feature, the competition coefficient is set to its minimum value. For example, suppose the target feature is the ability to implement NOA (Noise of Arrival) functionality in all cities nationwide, while competing vehicles only offer NOA functionality in a few cities. In other words, competing vehicles do not possess this target feature, meaning they have a unique feature that others do not. It can also be said that the target vehicle's target feature was commercially implemented first, so the competition coefficient can be set to its minimum value, for example, a minimum competition coefficient of 1.

[0036] If a competing vehicle possesses the target feature, and this target feature has a defect on the target vehicle, while the competing vehicle does not have the target defect, then the competition coefficient is set to its maximum value. For example, suppose both the target and competing vehicles have a city NOA (Noise, Assault, and Alarm) function. However, under the same testing conditions, the target vehicle exhibits an abnormal takeover of the city NOA function, while the competing vehicle does not. This means the target feature has a defect on the target vehicle that causes abnormal takeover, while the competing vehicle does not. In other words, the target feature is perceived as significantly inferior to the competing vehicle on the target vehicle. Therefore, the competition coefficient can be set to its maximum value, for example, 10. Understandably, the competition coefficient can also be further determined between the minimum and maximum values ​​based on a comparison with competing vehicles.

[0037] Step 102: Obtain test data for the target vehicle. The test data comes from actual tests conducted by real users. For example, in each test cycle, 100 real users are selected to test the target vehicle, with a test distance of 10,000 kilometers and more than 100 parking incidents during the test period. It should be noted that if test data is needed in step 101, then step 101 should follow step 102.

[0038] Step 103: Obtain the occurrence probability coefficient based on the number of times the target characteristic anomalies occur in the test data. The occurrence probability coefficient is positively correlated with the number of anomalies. For example, traffic accidents or running red lights occur. The more times anomalies occur, the higher the occurrence probability coefficient.

[0039] Step 104: Obtain the severity coefficient based on the severity of the consequences of the anomalies occurring in the target characteristics of the test data. The severity coefficient is positively correlated with the severity of the consequences. For example, if the anomaly is a traffic accident or a violation of traffic rules, the severity of the consequences is high, so a larger severity coefficient is set. If the anomaly does not affect safety or traffic rules, such as inaccurate parking, the severity of the consequences is low, so a smaller severity coefficient is set. During testing, various types of anomalies may occur; the severity coefficient is set based on the anomaly with the highest severity.

[0040] Step 105: Calculate the quality quantification value based on the competition coefficient, anomaly probability coefficient, and severity coefficient. Any one of these coefficients is positively correlated with the quality quantification value. For example, the quality quantification value can be calculated by multiplying the competition coefficient, anomaly probability coefficient, and severity coefficient together.

[0041] Step 106: Determine the corresponding quality risk level based on the range of the quality quantification value. The quality quantification value is related to market competitiveness, actual test results, and the consequences of anomalies. The better the effect of the target characteristic, the smaller the quality quantification value. Therefore, the range can be pre-defined, and the quality risk level can be determined based on the range of the quality quantification value. For example, the minimum value of any one of the competition coefficient, anomaly probability coefficient, and severity coefficient is 1, and the maximum value is 10. Multiplying the three coefficients yields the quality quantification value. When the quality quantification value is in the range of [1, 150], the quality risk level is low; when it is in the range of [151, 391], the quality risk level is medium; and when it is in the range of [392, 1000], the quality risk level is high. For products with a low quality risk level, they can be fully released or partially released. For products with a medium quality risk level, they can be released after a decision is made. For products with a low quality risk level, they can be either not released or released after a comprehensive fallback plan and decision. Understandably, the division of each coefficient range and interval range, the setting of levels, and the method of formulating release strategies based on quality risk levels can all be set according to actual needs; this is just an example.

[0042] The product quality determination method in this application embodiment can statistically analyze the anomaly occurrence probability coefficient and severity coefficient based on the end-user's standard Beta test. That is, it is based on the actual anomalies or failures that have been detected, rather than just a preventative analysis and assessment of potential failures, or simply an assessment of the probability of a failure being detected. It introduces a competition coefficient, using horizontal comparison with competing products as a factor in quality consideration, reflecting the user's acceptance of some leading technologies. In addition, the severity coefficient also considers the user's acceptance of anomalies or functional failures. Furthermore, it provides a quantifiable quality determination method, which can quickly analyze the acquired data to obtain accurate quality quantification results, and based on these quality quantification results, release strategies can be easily formulated.

[0043] In some embodiments, the target defect is a target anomaly. For example, for the city NOA function, under normal circumstances, the vehicle can automatically change lanes, enter and exit ramps, etc., without driver intervention. However, when the city NOA function malfunctions, driver intervention is required. This abnormal takeover caused by the failure of the normal city NOA function can be considered a target anomaly. For ease of understanding, the following explanation uses the city NOA function as the target characteristic and abnormal takeover as the target anomaly. Table 1 illustrates the competition coefficients corresponding to different scenarios in one embodiment.

[0044] Table 1

[0045]

[0046] As shown in Table 1, the cases of setting the competition coefficient to 1 and 10 have been explained in step 101 above. Other cases will be explained below.

[0047] If a competitor's vehicle has a city NOA (No Assurance of Assistance) function, and abnormal takeovers occur on both the target vehicle and the competitor's vehicle under the same testing conditions, and the probability of abnormal takeovers occurring on the target vehicle is higher than that on the competitor's vehicle, and the difference between the probabilities of abnormal takeovers occurring on the target vehicle and the competitor's vehicle is greater than a first threshold (meaning the target vehicle experiences significantly more abnormal takeovers than the competitor's vehicle), then the competition coefficient is the first competition coefficient. For example, if the target vehicle experiences 10 abnormal takeovers and the competitor's vehicle experiences 2, the former has 8 more than the latter, exceeding the threshold of 5, then the competition coefficient is 8.

[0048] If a competitor's vehicle has a city NOA (No Assistance) function, and abnormal takeovers occur in both the target vehicle and the competitor's vehicle under the same testing conditions, and the absolute value of the difference between the probability of abnormal takeovers occurring in the competitor's vehicle and the probability of them occurring in the target vehicle is not greater than the first threshold, then the competition coefficient is the second competition coefficient. For example, if the target vehicle experiences 5 abnormal takeovers and the competitor's vehicle experiences 4 abnormal takeovers, and the absolute value of the difference between the two is not greater than the threshold of 5, then the competition coefficient is 5.

[0049] If a competitor's vehicle has a city NOA (No Assistance) function, and abnormal takeovers occur on both the target vehicle and the competitor's vehicle under the same testing conditions, and the probability of abnormal takeovers occurring on the competitor's vehicle is higher than that on the target vehicle, and the difference between the probabilities of abnormal takeovers occurring on the competitor's vehicle and the target vehicle is greater than the first threshold, then the competition coefficient is the third competition coefficient. For example, if the competitor's vehicle experiences 10 abnormal takeovers and the target vehicle experiences 2, the former has 8 more than the latter, which is greater than the threshold of 5, then the competition coefficient is 2.

[0050] The maximum value of the competition coefficient is greater than the first competition coefficient, which is greater than the second competition coefficient, which is greater than the third competition coefficient, which is greater than the minimum competition coefficient.

[0051] In some embodiments, the target characteristic is a characteristic of assisted driving, and the target anomaly is an anomaly takeover. For details, please refer to the above embodiments. It can be understood that the target characteristic can be other characteristic functions besides urban NOA functions, such as mapless NOA functions, i.e. NOA functions that do not rely on high-precision maps, such as assisted automatic parking (AVP) functions, etc.

[0052] In some embodiments, the anomaly occurrence probability coefficient in step 103 above can be comprehensively determined based on dimensions such as the probability of accident occurrence, the probability of traffic rule compliance anomaly occurrence, the probability of active safety measure anomaly occurrence, the probability of functional availability anomaly occurrence, and the probability of experience perception anomaly occurrence. Each type of anomaly has a corresponding sub-coefficient, and the anomaly occurrence probability coefficient can be determined based on the sub-coefficient corresponding to each type of anomaly. Table 2 illustrates the correspondence between the number of occurrences of various types of anomalies and the sub-coefficients during the testing process.

[0053] Table 2

[0054]

[0055] As shown in Table 1, different anomaly occurrence categories correspond to different sub-coefficients. The five anomaly occurrence categories of accidents, traffic rule compliance, active safety measures, functions, and experience perception correspond to the first to fifth sub-coefficients, respectively.

[0056] Specifically, step 103 above, obtaining the anomaly occurrence probability coefficient based on the number of times the target characteristic anomalies occur in the test data, includes:

[0057] The first sub-coefficient is derived from the number of accidents occurring for the target characteristic in the test data. This first sub-coefficient is positively correlated with the number of accidents. If the number of accidents is greater than or equal to a second threshold, it indicates a very high risk, and the first sub-coefficient is set to its maximum probability value. For example, in the application of the urban NOA (Noise, Arrival, and Detection) function, if a vehicle experiences a rear-end collision or other accident, the number of accidents during the test is statistically analyzed. The more accidents that occur, the larger the first sub-coefficient becomes. For instance, if the second threshold is set to 5, then 5 or more accidents will be considered. Assuming the coefficient range is 1 to 10, the first sub-coefficient is set to its maximum value of 10.

[0058] The second sub-coefficient is derived from the frequency of traffic rule compliance anomalies in the test data. This second sub-coefficient is positively correlated with the frequency of these anomalies. If the probability of a traffic rule compliance anomaly occurring is 100%, indicating a very high risk, then the second sub-coefficient is set to its maximum probability value. For example, in the application of the city's NOA (No-Occupational-Avoidance) function, a vehicle's unexplained traffic violations, such as running a red light, constitute a traffic rule compliance anomaly. Therefore, the frequency of these anomalies during testing is statistically analyzed. The more frequent the traffic violations, the larger the second sub-coefficient. If the probability of a traffic rule compliance anomaly occurring is guaranteed, meaning an anomaly will occur in every test (i.e., a 100% probability), then the second sub-coefficient is set to its maximum value of 10.

[0059] The third sub-coefficient is derived from the number of times the active safety measures of the target characteristics occur abnormally in the test data. This third sub-coefficient is positively correlated with the number of abnormal occurrences. If the probability of an abnormal occurrence is 100%, indicating a very high risk, then the third sub-coefficient is set to its maximum probability value. For example, in the application of the city's NOA (Noise, Arrival, and Response) function, in the absence of safety risks, a misidentification of an obstacle could lead to erroneous Automatic Emergency Braking (AEB). This abnormal AEB triggering constitutes an active safety measure anomaly. Therefore, the number of abnormal occurrences of active safety measures during testing is statistically analyzed. The more frequent the abnormal occurrences, the larger the third sub-coefficient. If an active safety measure anomaly occurs in every test (i.e., the probability is 100%), then the third sub-coefficient is set to its maximum value of 10.

[0060] The fourth sub-coefficient is derived from the number of times the functional availability anomalies of the target feature occur in the test data. This fourth sub-coefficient is positively correlated with the number of occurrences. If the probability of a functional availability anomaly occurring is 100%, indicating a very high risk, then the fourth sub-coefficient is set to its maximum probability value. For example, during testing, situations such as the city NOA function failing to activate or the vehicle failing to park correctly during automatic parking constitute functional availability anomalies. The more frequent these anomalies, the larger the fourth sub-coefficient. If a functional availability anomaly occurs in every test, then the fourth sub-coefficient is set to its maximum value of 10.

[0061] The fifth sub-coefficient is derived from the frequency of perceived abnormalities in the test data. This fifth sub-coefficient is positively correlated with the frequency of perceived abnormalities. If the probability of a perceived abnormality is 100%, indicating a very high risk, then the fifth sub-coefficient is set to its maximum value. For example, in the application of the city NOA (Noise, Arbitration, and Autonomy) function, an unexplained lane change during normal driving constitutes a perceived abnormality. The more frequent the perceived abnormalities, the larger the fifth sub-coefficient. If a perceived abnormality occurs in every test, then the fifth sub-coefficient is set to its maximum value of 10.

[0062] The anomaly occurrence probability coefficient is the largest among the first, second, third, fourth, and fifth sub-coefficients. That is, in obtaining the anomaly occurrence probability coefficient, the corresponding sub-coefficient is first determined based on the frequency of occurrence of each anomaly type, and then the maximum value among them is used as the final anomaly occurrence probability coefficient. As can be seen from the data in Table 2, when setting the sub-coefficients, accidents have a stricter standard compared to other anomalies. This is determined based on users' acceptance of different anomaly occurrences. Users have a low acceptance of accidents, so a sub-coefficient of 10 is set as long as the number of accidents reaches 5 or more. For other types of anomalies, a sub-coefficient of 10 is only set when the occurrence probability reaches 100%, so that the anomaly occurrence probability coefficient can more accurately reflect users' acceptance of anomaly occurrences.

[0063] In some embodiments, as shown in Table 2, if the number of accidents is less than the second threshold 5 and greater than 1, it indicates a high risk. In this case, the first sub-coefficient is the first coefficient value, for example, 9. The first coefficient value 9 is less than the maximum probability coefficient value 10.

[0064] If the accident occurs once, it indicates a medium risk. In this case, the first sub-coefficient should be less than the first coefficient value. For example, the first sub-coefficient can be set to a value between 4 and 6.

[0065] If the probability of traffic rule compliance violations is less than 100%, and the number of violations is greater than or equal to the second threshold of 5, it indicates a high risk, and the second sub-coefficient is the first coefficient value of 9.

[0066] If the number of traffic rule compliance violations is less than the second threshold of 5 but greater than 1, it indicates a medium risk. In this case, the second sub-coefficient is the second coefficient value of 6, which is less than the first coefficient value.

[0067] If the probability of a traffic rule violation occurring abnormally is 1, it indicates a low risk. In this case, the second sub-coefficient should be less than the value of the second coefficient. For example, the second sub-coefficient can be set to a value between 1 and 3.

[0068] If the probability of an abnormality in proactive safety measures is less than 100%, and the number of abnormalities in proactive safety measures is greater than or equal to the third threshold of 10, it indicates a high risk. In this case, the third sub-coefficient is the first coefficient value of 9, and the third threshold of 10 is greater than the second threshold of 5.

[0069] If the number of abnormal occurrences of proactive safety measures is no greater than the third threshold of 10, and is greater than or equal to the second threshold of 5, it indicates that the risk is moderate to high. In this case, the third sub-coefficient is the second coefficient value of 6.

[0070] If the number of abnormal occurrences of proactive safety measures is no greater than the second threshold of 5, and is greater than 1, it indicates that the risk is moderate to low. In this case, the third sub-coefficient is the third coefficient value of 4, which is less than the second coefficient value of 6.

[0071] If the number of abnormal occurrences of proactive safety measures is 1, it indicates low risk. In this case, the third sub-coefficient is the fourth coefficient value 2, which is less than the third coefficient value 4.

[0072] If the probability of a functional availability anomaly occurring is less than 100%, and the number of functional availability anomalies occurring is greater than or equal to the third threshold of 10, it indicates a high risk. In this case, the fourth sub-coefficient is the fifth coefficient value of 8, and the first coefficient value of 9 > the fifth coefficient value of 8 > the second coefficient value of 6.

[0073] If the number of times the functionality is available is no greater than the third threshold of 10, and is greater than or equal to the second threshold of 5, it indicates that the risk is moderate to high. In this case, the fourth sub-coefficient is the second coefficient value of 6.

[0074] If the number of times the functional availability anomaly occurs is no greater than the second threshold of 5, but greater than 1 time, it indicates that the risk is moderate to low. In this case, the fourth sub-coefficient is the third coefficient value of 4.

[0075] If the number of times the functional availability exception occurs is 1, it indicates low risk, and the fourth sub-coefficient is the fourth coefficient value of 2.

[0076] If the probability of an abnormal experience is less than 100%, and the number of abnormal experiences is greater than or equal to the third threshold of 10, it indicates a high risk, and the fifth sub-coefficient is the fifth coefficient value of 8.

[0077] If the number of abnormal experiences is no greater than the third threshold of 10 and is greater than or equal to the second threshold of 5, it indicates that the risk is moderate to high. In this case, the fifth sub-coefficient is the second coefficient value of 6.

[0078] If the number of abnormal experiences is no greater than the second threshold of 5, but greater than 1 time, it indicates that the risk is moderate to low, and the fifth sub-coefficient is the third coefficient value of 4.

[0079] If the number of abnormal experiences is 1, it indicates low risk. In this case, the fifth sub-coefficient is the sixth coefficient value 1, and the sixth coefficient value 1 < the fourth coefficient value 2.

[0080] Users have different levels of acceptance of the same number of occurrences of different types of anomalies. Therefore, different sub-coefficient setting standards are set according to different types, so that the anomaly occurrence probability coefficient can more accurately reflect the user's acceptance of the occurrence of anomalies.

[0081] In some embodiments, as shown in Table 2, if the accident occurs once and repeated tests on the same road segment will always reproduce the accident, then the first sub-coefficient is the second coefficient value of 6. For example, if the vehicle only has one collision with the guardrail on a certain road segment during the test, and subsequent additional tests show that the collision with the guardrail occurs on the same road segment each time, then the risk is considered to be moderate to high. Therefore, a relatively large value of 6 can be selected from 4 to 6 as the corresponding sub-coefficient.

[0082] If an accident occurs only once, and repeated tests on the same road segment where the accident occurred occasionally recur, then the first sub-coefficient is the seventh coefficient value of 5, and the second coefficient value is 6 > the seventh coefficient value of 5 > the third coefficient value of 4. For example, if a vehicle only has one collision with a guardrail on a certain road segment during the test, and subsequent additional tests occasionally result in collisions with guardrails on the same road segment, then the risk is considered moderate. Therefore, a sub-coefficient of 5, which is relatively middle between 4 and 6, can be selected as the corresponding sub-coefficient.

[0083] If the accident occurs once, and repeated tests on the same road segment do not reproduce the accident, then the first sub-coefficient is the third coefficient value of 4. For example, if the vehicle only crashes into the guardrail once on a certain road segment during the test, and subsequent additional tests do not result in a crash on the same road segment, then the risk is considered to be moderate to low. Therefore, a relatively smaller sub-coefficient of 4 can be selected from 4 to 6 as the corresponding sub-coefficient.

[0084] If the number of traffic rule compliance violations occurs once, and repeated testing on the same road segment where the violation occurred will always reproduce the violation, then the second sub-coefficient is the seventh coefficient value of 3, and the third coefficient value of 4 > the seventh coefficient value of 3 > the fourth coefficient value of 2. For example, if a vehicle commits a traffic violation by crossing a double yellow line only once on a certain road segment during the test, and subsequent additional tests show that the violation occurs on the same road segment every time, then the risk is considered low to high. Therefore, a relatively larger sub-coefficient of 3 can be selected from 2 to 3 as the corresponding sub-coefficient.

[0085] If the number of traffic rule compliance violations occurs once, and repeated tests on the same road segment where the violation occurred only occasionally show this pattern, then the second sub-coefficient is the fourth coefficient value of 2. For example, if a vehicle commits a traffic violation by crossing a double yellow line only once on a certain road segment during the test, and subsequent additional tests occasionally show the same violation on the same road segment, then the risk is considered low to very low. Therefore, a smaller value of 2 can be chosen from 2 to 3 as the corresponding sub-coefficient.

[0086] If the number of traffic rule compliance violations occurs once, and the violation does not recur in repeated tests on the same road segment where the violation occurred, then the second sub-coefficient is the sixth coefficient value of 1. For example, if a vehicle commits a traffic violation by crossing a double yellow line only once on a certain road segment during the test, and subsequent additional tests do not show any violations on the same road segment, then the risk is very low, and therefore the smallest value of 1 can be selected as the corresponding sub-coefficient.

[0087] For accidents or traffic violations that occur only once during the test, the degree of risk can be further determined by whether they recur on the same road segment. This means further determining the user's acceptance of the anomaly and thus determining the sub-coefficient. This allows the anomaly probability coefficient to more accurately reflect the user's acceptance of the anomaly.

[0088] In some embodiments, since various types of anomalies may occur during testing, some leading to serious consequences while others have minor ones, and users have varying degrees of acceptance of these consequences, a severity coefficient can be determined based on the severity of the consequences of the anomaly. This allows for the deriving of the final quality quantification value based on the severity coefficient. Table 3 illustrates the correspondence between consequence severity and severity coefficient.

[0089] Table 3

[0090]

[0091]

[0092] As shown in Table 3, the severity coefficient is determined by the highest severity of the consequences of the anomaly in the test data. The consequences corresponding to the severity coefficients from high to low include: non-compliance with safety and / or traffic regulations, loss or degradation of the target feature, unsatisfactory experience of the target feature, and interference and / or differences in experience caused by the target feature. The severity of the consequences of an anomaly reflects the user's acceptance of the anomaly. The consequence of non-compliance with safety and / or traffic regulations is the most unacceptable to users, therefore it has the highest severity and the highest corresponding severity coefficient. The severity of the consequence refers to the severity of its impact on the product, or in other words, its impact on the user. Based on this principle, the severity coefficient is set according to the severity of the consequences of the anomaly to reflect the user's acceptance of the anomaly.

[0093] In some embodiments, as shown in Table 3, the target characteristic is a characteristic of assisted driving. The consequences corresponding to the severity coefficients from high to low include:

[0094] Anomalies that affect the safe operation of the vehicle and / or include situations that do not comply with traffic rules, occur without warning, and are the most serious consequences. The severity coefficient can be set to the maximum value, such as 10. Specific scenarios include consequences caused by traffic rule compliance anomalies such as running a red light or driving into the oncoming lane; active safety measures anomalies that seriously affect safety, such as AEB anomalies on highways; and accident anomalies such as hitting a stone block while parking.

[0095] An anomaly that affects the safe operation of a vehicle and / or includes situations that do not comply with traffic rules will be warned when the anomaly occurs and the consequences will be relatively serious. The severity coefficient can be set to 9. Specific scenarios include anomalies that affect safety, such as densely packed vehicles in adjacent lanes or changing lanes across a solid yellow line after signaling. The consequences corresponding to severity coefficients of 10 and 9 are non-compliance with safety and / or traffic rule requirements.

[0096] The loss of basic functions of the target characteristics does not affect the safe operation of the vehicle. The severity coefficient can be set to 8. Specific scenarios include the inability to activate the parking function in a normal parking space or the abnormal exit of the assisted driving mode under normal road conditions. Although the basic functions of assisted driving are lost, the safe operation of the vehicle is not affected.

[0097] The performance of the main function of the target feature is reduced. The severity coefficient can be set to 7. Specific scenarios include the inability to recognize narrow parking spaces and the inability of the automatic parking function to start when encountering narrow parking spaces. This means that the basic function of the target feature is available, but it is running in a degraded manner and the performance level is reduced. The consequences corresponding to the severity coefficients of 8 and 7 are that the target feature is lost or degraded.

[0098] If the driving or parking path is unstable or out of control, the severity level can be set to 6. Specific scenarios include situations such as sudden lane changes, abnormal lane changes, and abnormal AEB during normal cruising in urban areas.

[0099] If the driving or parking path does not meet the requirements, such as being unnatural or not intelligent enough, the severity coefficient can be set to 5. Specific scenarios include aggressive lane cutting, constantly following slow vehicles, sluggish turning at intersections / merging into ramps, not slowing down when exiting ramps, and parking over the lines. The consequences of severity coefficients of 6 and 5 are that the experience of the target characteristics does not meet the requirements.

[0100] If the navigation information is incorrect, such as an error in the navigation map or navigation prompts, you can set the severity level to 4. For example, if you are driving south but the navigation prompts you to go north.

[0101] If the navigation information is unclear, such as the navigation prompts, alarm icons or alarm sounds not being clear enough, the severity coefficient can be set to 3. In specific scenarios, such as when the icons for manual takeover are dark or small, the consequences of severity coefficients of 4 and 3 are that the target characteristics are interfered with and / or there are differences in the user experience.

[0102] No consequences, meaning there are no identifiable consequences. The severity level can be set to 2. Specific scenarios include situations where the layout of a navigation map interface is unscientific and inconvenient to look at.

[0103] No impact, meaning there is no identifiable impact. The severity coefficient can be set to the minimum value of 1. Specific scenarios include situations where the assisted driving report lacks abnormal takeover times.

[0104] Setting severity coefficients based on a more detailed classification of the severity of the consequences of anomalies can more accurately reflect the user's level of acceptance of anomalies.

[0105] In some embodiments, based on the range defined by the quality quantification value, the quality risk level can be further determined by whether the competition coefficient, the anomaly probability coefficient, and the severity coefficient contain a maximum value as a constraint. As shown in Table 4, Table 4 illustrates the specific methods for classifying different quality risk levels.

[0106] Table 4

[0107]

[0108] As shown in Table 4, the interval range includes the first interval range [1, 150] and the second interval range [151, 391]. The maximum value of the first interval range, 150, is less than the minimum value of the second interval range, 151. First, the interval range is determined by multiplying the severity coefficient, the anomaly probability coefficient, and the competition coefficient to obtain the quality quantification value. Then, the quality risk level is determined based on whether each coefficient has a maximum value of 10.

[0109] If the quality quantification value falls within the first interval range [1, 150], and none of the competition coefficient, anomaly probability coefficient, and severity coefficient has a maximum value of 10, for example, the combination of the severity coefficient, anomaly probability coefficient, and competition coefficient is 5×6×5, and the severity, anomaly probability, and competition are all moderate, then the quality risk level is Level 1, which has the lowest risk and can be fully released / published.

[0110] If the quality quantification value belongs to the first interval range [1,150], at least one of the competition coefficient, the probability coefficient of anomaly occurrence, and the severity coefficient is the maximum value of 10. For example, the combination of the severity coefficient, the probability coefficient of anomaly occurrence, and the competition coefficient is 10×3×5, which means the severity is the highest, the probability of anomaly occurrence is low, and the competition is moderate. Then the quality risk level is the second level, which can be a small amount or a part.

[0111] If the quality quantification value falls within the second interval range [151,391], and none of the competition coefficient, anomaly probability coefficient, and severity coefficient has a maximum value of 10, then the quality risk level is the third level, and it can be released / published after decision-making at the L1 level. L1 is the designated decision-making level.

[0112] If the quality quantification value falls within the second interval range [151, 391], and at least one of the competition coefficient, anomaly probability coefficient, and severity coefficient is at its maximum value of 10, then the quality risk level is level four. It can be released / published after decision-making at level L2. L2 is a higher decision-making level than L1. Since there is a maximum value of 10 in the coefficients, level four is riskier than level three. Therefore, a higher decision-making level can decide whether to release / publish it.

[0113] If the quality quantification value falls within the third interval range [392, 1000], then the quality risk level is level five. Level five has the highest risk and may not be issued / released, or it may be released / released after a comprehensive fallback plan and a decision by the L2 layer.

[0114] In terms of the technical range of the quality quantification value, combining whether each coefficient contains the maximum value as a constraint condition to refine the quality risk level can more accurately reflect the user's acceptance of the occurrence of anomalies, thereby improving the reliability of product release strategies based on risk levels.

[0115] In some embodiments, the product quality determination method further includes: if the test data of the target vehicle contains a condition from a preset list, then the quality is determined to be unqualified.

[0116] The preset list is an absolute quality red line list. You can first determine whether there are any situations in the preset list based on the test data. If any abnormality or problem is found in the contents of the preset list, the quality will be rejected directly, and no further quality assessment will be conducted.

[0117] As shown in Table 5, Table 5 illustrates some scenario examples of the preset list.

[0118] Table 5

[0119]

[0120] For example, version V2.X of the target vehicle's assisted driving features mainly introduces the new target feature AVP. Quality is assessed using the product quality determination method described in this application to determine whether commercial release is feasible. In the test data, one serious issue occurred: a car collided with a standard square post next to a parking space in a regular accessible parking space. Analysis of the improvement plan concluded that this anomaly was a single incident, caused by…, and is planned to be resolved in next month's version. The quality assessment is: there is an absolute quality red line issue; commercial release is not permitted until the issue is resolved. An emergency undecided release is recommended, and a quality assessment will be conducted after the red line issue is resolved.

[0121] The quality determination method of this application embodiment will be described below based on a real-world scenario.

[0122] like Figure 2 As shown, a version of an intelligent product may include multiple features, such as new features, which are features added to the current version compared to the previous version. Multiple features may also include updated features, which are features that the current version has compared to the previous version. These features may have been optimized and updated in the current version compared to the previous version. The quality of new features and updated features can be determined separately.

[0123] After the quality assessment of the new features of intelligent products is initiated, the first step is to test or verify the new features, i.e., to execute step 102 above. Then, based on the test data and the preset list, an absolute red line assessment is performed. If there is an absolute red line problem, i.e., an anomaly or problem in the content of the preset list appears, the quality is directly rejected. If there is no absolute red line problem, i.e., no anomaly or problem in the content of the preset list appears, a COS quality assessment is performed. The COS quality assessment includes steps 101, 103, 104, 105, and 106 above. If the quality quantification value obtained by multiplying the competition coefficient C, the anomaly occurrence probability coefficient O, and the severity coefficient S is higher than the first threshold, it belongs to the third interval range, i.e., high quality risk. If the quality quantification value is between the first and second thresholds, it belongs to the second interval range, i.e., medium quality risk. If the quality quantification value is lower than the second threshold, it belongs to the first interval range, i.e., low quality risk.

[0124] The quality assessment of updated features in intelligent products is similar to that of new features. After the quality assessment of updated features is initiated, testing or verification is first performed based on the updated features, i.e., step 102 above is executed. Then, based on the test data and the preset list, an absolute red line assessment is conducted. If there is an absolute red line problem, i.e., an anomaly or problem in the content of the preset list appears, the quality is directly rejected. If there is no absolute red line problem, i.e., no anomaly or problem in the content of the preset list appears, a COS quality assessment is conducted to obtain a quality quantification value. If the quality quantification value is significantly worse than the previous version, the quality risk is high. If the quality quantification value is basically the same as the previous version, the quality risk is medium. If the quality quantification value is significantly better than the previous version, the quality risk is low.

[0125] In addition, such as Figure 3 As shown, a version of an intelligent product may have multiple features. For each feature, a quality assessment can be performed using the quality determination method described in this application. The overall quality risk assessment, considering all newly added and updated features in the version, can then be used as the basis for determining the commercial release strategy for that version. During the quality assessment for the commercial release of an intelligent product version, an absolute red-line assessment is performed first. If one or more features exhibit absolute red-line issues, the quality is directly rejected, meaning the entire product version will not be released. Only when all features are free of absolute red-line issues does the COS (Coefficient of Performance) quality assessment begin. The quality of all features is reviewed collectively to comprehensively assess the quality risk of the version. For example, in controlling the quality of version releases, the feature with the highest quality quantification value is used as the standard to determine the version release strategy.

[0126] like Figure 4 As shown, this application embodiment also provides an electronic device 100, including: a processor 110 and a memory 121. The memory 121 is used to store at least one instruction. When the instruction is loaded and executed by the processor 110, it causes the electronic device 100 to perform the above-described product quality determination method.

[0127] It is understood that the structures illustrated in the embodiments of the present invention do not constitute a specific limitation on the electronic device 100. In other embodiments of this application, the electronic device 100 may include more or fewer components than illustrated, or combine some components, or split some components, or have different component arrangements. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.

[0128] Processor 110 may include one or more processing units, such as application processor (AP), modem processor, graphics processing unit (GPU), image signal processor (ISP), controller, video codec, digital signal processor (DSP), baseband processor, and / or neural network processing unit (NPU). Different processing units may be independent devices or integrated into one or more processors.

[0129] The controller can generate operation control signals based on the instruction opcode and timing signals to complete the control of instruction fetching and execution.

[0130] The processor 110 may also include a memory for storing instructions and data. In some embodiments, the memory in the processor 110 is a cache memory. This memory can store instructions or data that the processor 110 has just used or that are used repeatedly. If the processor 110 needs to use the instruction or data again, it can retrieve it directly from the memory. This avoids repeated accesses, reduces the waiting time of the processor 110, and thus improves the efficiency of the system.

[0131] Internal memory 121 can be used to store computer executable program code, which includes instructions. Internal memory 121 may include a program storage area and a data storage area. The program storage area may store the operating system, applications required for at least one function, etc. The data storage area may store data created during the use of electronic device 100, etc. In addition, internal memory 121 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, universal flash storage (UFS), etc. Processor 110 executes various functional applications and data processing of electronic device 100 by running instructions stored in internal memory 121 and / or instructions stored in memory disposed in the processor.

[0132] like Figure 5As shown, the electronic device can be the COS quality assessment device in the system. The system also includes a problem ticket device, which pre-sets the setting rules for the competition coefficient, anomaly probability coefficient, and severity coefficient into the COS quality assessment device. Select a feature type that needs to be commercially released, and enter the test problem of that feature type into the problem ticket device. The COS quality assessment device obtains the test problem, calculates the quality quantification value, generates a quality risk level, and provides suggestions for commercial release / release strategies. When all features meet the release conditions, the version can be released. When a feature is risky, the version is evaluated based on the feature with the highest risk.

[0133] This application also provides a computer-readable storage medium, including a program or instructions, wherein the methods of any of the above embodiments are executed when the program or instructions are run on a computer.

[0134] This application also provides a computer program product containing executable instructions that, when executed on a computer, cause the computer to perform the methods of any of the above embodiments.

[0135] The control methods in the above embodiments can be implemented entirely or partially through software, hardware, firmware, or any combination thereof. When implemented in software, they can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state disk).

[0136] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, the simultaneous existence of A and B, or the existence of B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, and c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0137] The above are merely preferred embodiments of this application and are not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

Claims

1. A method for determining product quality, characterized in that, include: The target vehicle's characteristics are compared with those of competing vehicles to obtain a competition coefficient. The target vehicle is a vehicle that uses driver assistance products or smart cockpit products. If the competing vehicle does not possess the target characteristic, then the competition coefficient is the minimum value of the competition coefficient; If the competing vehicle has the target characteristic, and the target characteristic has a target defect in the target vehicle, and the competing vehicle does not have the target defect, then the competition coefficient is the maximum value of the competition coefficient. Obtain test data for the target vehicle; An anomaly probability coefficient is obtained based on the number of times the target characteristic anomalies occur in the test data, and the anomaly probability coefficient is positively correlated with the number of times the anomalies occur. A severity coefficient is obtained based on the severity of the consequences of the abnormal occurrence of the target characteristic in the test data, and the severity of the consequences is positively correlated with the severity coefficient; A quality quantification value is calculated based on the competition coefficient, the anomaly probability coefficient, and the severity coefficient, wherein any one of the competition coefficient, the anomaly probability coefficient, and the severity coefficient is positively correlated with the quality quantification value. The corresponding quality risk level is determined based on the range to which the quality quantification value belongs.

2. The method according to claim 1, characterized in that, The target defect is a target anomaly; If the competing vehicle has the target characteristic, and the target characteristic has the target anomaly occurring in both the target vehicle and the competing vehicle, and under the same testing conditions, the probability of the target anomaly occurring in the target vehicle is greater than the probability of it occurring in the competing vehicle, and the difference between the probability of the target anomaly occurring in the target vehicle and the probability of it occurring in the competing vehicle is greater than a first threshold, then the competition coefficient is the first competition coefficient. If the competing vehicle has the target characteristic, and the target characteristic has the target anomaly occurring in both the target vehicle and the competing vehicle, and the absolute value of the difference between the probability of the target anomaly occurring in the competing vehicle and the probability of it occurring in the target vehicle under the same testing conditions is not greater than the first threshold, then the competition coefficient is the second competition coefficient. If the competing vehicle has the target characteristic, and the target characteristic has the target anomaly occurring in both the target vehicle and the competing vehicle, and under the same testing conditions, the probability of the target anomaly occurring in the competing vehicle is greater than the probability of it occurring in the target vehicle, and the difference between the probability of the target anomaly occurring in the competing vehicle and the probability of it occurring in the target vehicle is greater than the first threshold, then the competition coefficient is the third competition coefficient. The maximum value of the competition coefficient is greater than the first competition coefficient, the second competition coefficient, the third competition coefficient, and the minimum competition coefficient.

3. The method according to claim 2, characterized in that, The target characteristic is a characteristic of assisted driving, and the target anomaly is an abnormal takeover.

4. The method according to any one of claims 1 to 3, characterized in that, The process of obtaining the anomaly probability coefficient based on the number of times the target characteristic anomalies occur in the test data includes: A first sub-coefficient is obtained based on the number of accidents occurring for the target characteristic in the test data. The first sub-coefficient is positively correlated with the number of accidents occurring. If the number of accidents occurring is greater than or equal to a second threshold, then the first sub-coefficient is the maximum probability coefficient. The second sub-coefficient is obtained based on the number of traffic rule compliance anomalies of the target characteristic in the test data. The second sub-coefficient is positively correlated with the number of traffic rule compliance anomalies. If the probability of the traffic rule compliance anomaly is 100%, then the second sub-coefficient is the maximum value of the probability coefficient. A third sub-coefficient is obtained based on the number of times the active safety measures of the target characteristics in the test data are abnormal. The third sub-coefficient is positively correlated with the number of times the active safety measures are abnormal. If the probability of the active safety measures being abnormal is 100%, then the third sub-coefficient is the maximum value of the probability coefficient. The fourth sub-coefficient is obtained based on the number of times the functional availability anomaly of the target characteristic occurs in the test data. The fourth sub-coefficient is positively correlated with the number of times the functional availability anomaly occurs. If the probability of the functional availability anomaly occurring is 100%, then the fourth sub-coefficient is the maximum value of the probability coefficient. The fifth sub-coefficient is obtained based on the number of times the perceived abnormality of the target characteristic occurs in the test data. The fifth sub-coefficient is positively correlated with the number of times the perceived abnormality occurs. If the probability of the perceived abnormality occurring is 100%, then the fifth sub-coefficient is the maximum value of the probability coefficient. The probability coefficient of an anomaly occurrence is the largest of the first sub-coefficient, the second sub-coefficient, the third sub-coefficient, the fourth sub-coefficient, and the fifth sub-coefficient.

5. The method according to claim 4, characterized in that, If the number of times the accident occurs is less than the second threshold and greater than 1, then the first sub-coefficient is the first coefficient value, and the first coefficient value is less than the maximum value of the probability coefficient; If the accident occurs once, then the first sub-coefficient is less than the first coefficient value; If the probability of the traffic rule compliance exception occurring is less than 100%, and the number of traffic rule compliance exceptions occurring is greater than or equal to the second threshold, then the second sub-coefficient is the first coefficient value; If the number of traffic rule compliance violations is less than the second threshold but greater than 1, then the second sub-coefficient is the second coefficient value, which is less than the first coefficient value. If the probability of an abnormal traffic rule compliance occurs is 1, then the second sub-coefficient is less than the value of the second coefficient. If the probability of the active safety measure failing is less than 100%, and the number of times the active safety measure fails is greater than or equal to the third threshold, then the third sub-coefficient is the first coefficient value, and the third threshold is greater than the second threshold. If the number of abnormal occurrences of the active safety measures is not greater than the third threshold, but is greater than or equal to the second threshold, then the third sub-coefficient is the value of the second coefficient. If the number of abnormal occurrences of the active safety measures is not greater than the second threshold and is greater than 1, then the third sub-coefficient is the third coefficient value, and the third coefficient value is less than the second coefficient value; If the number of times the active safety measures fail is 1, then the third sub-coefficient is the fourth coefficient value, and the fourth coefficient value is less than the third coefficient value; If the probability of the function being available is less than 100%, and the number of times the function is available is greater than or equal to the third threshold, then the fourth sub-coefficient is the fifth coefficient value, and the first coefficient value > the fifth coefficient value > the second coefficient value; If the number of times the function is available is not greater than the third threshold and is greater than or equal to the second threshold, then the fourth sub-coefficient is the value of the second coefficient. If the number of times the function is available is not greater than the second threshold and is greater than 1, then the fourth sub-coefficient is the value of the third coefficient. If the number of times the function can be used to cause an exception is 1, then the fourth sub-coefficient is the value of the fourth coefficient; If the probability of the perceived abnormality occurring is less than 100%, and the number of occurrences of the perceived abnormality is greater than or equal to the third threshold, then the fifth sub-coefficient is the value of the fifth coefficient. If the number of times the perceived abnormality occurs is not greater than the third threshold, but is greater than or equal to the second threshold, then the fifth sub-coefficient is the value of the second coefficient. If the number of times the perceived abnormality occurs is not greater than the second threshold and is greater than 1 time, then the fifth sub-coefficient is the value of the third coefficient. If the number of times the perceived abnormality occurs is 1, then the fifth sub-coefficient is the sixth coefficient value, and the sixth coefficient value is less than the fourth coefficient value.

6. The method according to claim 5, characterized in that, If the accident occurs once, and repeated testing on the road segment where the accident occurred will always reproduce the accident, then the first sub-coefficient is the value of the second coefficient. If the accident occurs once, and repeated testing on the road segment where the accident occurred shows occasional occurrences, then the first sub-coefficient is the seventh coefficient value, and the second coefficient value > the seventh coefficient value > the third coefficient value; If the accident occurs once, and repeated testing at the road segment where the accident occurred does not reproduce the accident, then the first sub-coefficient is the value of the third coefficient. If the number of times the traffic rule compliance anomaly occurs is 1, and repeated testing under the road segment where the traffic rule compliance anomaly occurs will always reproduce the anomaly, then the second sub-coefficient is the seventh coefficient value, and the third coefficient value > the seventh coefficient value > the fourth coefficient value; If the number of times the traffic rule compliance anomaly occurs is 1, and the anomaly is repeatedly tested and occurs sporadically in the road segment where the traffic rule compliance anomaly occurs, then the second sub-coefficient is the fourth coefficient value; If the number of times the traffic rule compliance anomaly occurs is 1, and repeated testing on the road segment where the traffic rule compliance anomaly occurred does not reproduce the anomaly, then the second sub-coefficient is the sixth coefficient value.

7. The method according to any one of claims 1 to 6, characterized in that, The interval range includes a first interval range and a second interval range, wherein the maximum value of the first interval range is less than the minimum value of the second interval range; If the quality quantification value falls within the first interval range, and none of the competition coefficient, the anomaly occurrence probability coefficient, and the severity coefficient have a maximum value, then the quality risk level is Level 1. If the quality quantification value falls within the first interval range, and at least one of the competition coefficient, the anomaly occurrence probability coefficient, and the severity coefficient is at its maximum value, then the quality risk level is Level 2. If the quality quantification value falls within the second interval range, and none of the competition coefficient, the anomaly probability coefficient, and the severity coefficient have a maximum value, then the quality risk level is level three. If the quality quantification value falls within the second interval range, and at least one of the competition coefficient, the anomaly occurrence probability coefficient, and the severity coefficient is at its maximum value, then the quality risk level is level four.

8. The method according to any one of claims 1 to 7, characterized in that, Also includes: If the test data of the target vehicle contains any of the conditions listed in the preset list, then the quality is determined to be substandard.

9. An electronic device, characterized in that, include: A processor and a memory, the memory being used to store at least one instruction that, when loaded and executed by the processor, causes the electronic device to perform the method as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that, Includes a program or instructions that, when run on a computer, execute the method as described in any one of claims 1 to 8.

11. A computer program product, characterized in that, The computer program product includes executable instructions that, when executed on a computer, cause the computer to perform the method described in any one of claims 1 to 8.

Images