Method, apparatus, computer program and computer readable storage medium for determining a defective vehicle

Abstract

The invention presents a method for determining defective vehicles, wherein the defective vehicles are a subset of a plurality of vehicles (2) and the vehicles (2) are divided into a plurality of vehicle types, the method comprising: providing (S1) for each vehicle type an expected value of a number of predefined messages to be sent; determining (S2) for each vehicle (2) an actual value of a number of predefined messages that have been sent; determining (S3) for each vehicle (2) a deviation value, wherein each deviation value represents a difference between the actual value and the expected value; determining (S4) from the deviation values a defective subgroup of defective vehicles, the actual values of which differ from the expected values. Furthermore a device, a computer program and a computer-readable storage medium are presented.

Description

Technical Field

[0001] This invention provides a method for identifying defective vehicles. Furthermore, an apparatus, a computer program, and a computer-readable storage medium are also provided. Background Technology

[0002] The objective is to provide a method in which defective vehicles can be identified particularly simply and quickly. Furthermore, an apparatus and a computer program capable of implementing such a method should be provided. Additionally, a computer-readable storage medium having such a computer program should also be provided. Summary of the Invention

[0003] The objective is achieved by a method for determining a defective vehicle according to the present application, an apparatus for determining a defective vehicle according to the present application, a computer program product according to the present application, and a computer-readable storage medium according to the present application.

[0004] First, a method for identifying defective vehicles is described. A defective vehicle is a subset of multiple vehicles. These multiple vehicles, for example, form a set, wherein the vehicles in the set are interconnected, for example, by means of a device. For example, each vehicle is connected to the device, particularly an external device, by means of a communication connection. The device is, for example, a server, particularly a backend server. Vehicles are categorized into various vehicle types. The vehicle type is, for example, a vehicle model. For example, each vehicle type includes multiple vehicles. Vehicles included in one vehicle type are, for example, not included in another vehicle type. For example, vehicle types group vehicles.

[0005] According to at least one embodiment of the method, an expected value for the number of pre-given messages to be sent is provided for each vehicle type. For example, the expected value is an average of the number of pre-given messages to be sent for each vehicle type. For example, the expected value is determined temporally before the provision. That is, the expected value is determined separately before implementing the method. The expected value is determined, for example, based on a shorter past time period, such as the past 4 weeks or the past 12 weeks, before implementing the method. Advantageously, the expected value is therefore not fixed but can change dynamically.

[0006] According to at least one embodiment of the method, an actual value for the number of pre-given messages sent is determined for each vehicle. For example, the actual value is determined based on a time interval. This time interval is, for example, at least one hour and at most 48 hours, particularly 24 hours. Specifically, the time interval represents a weekday.

[0007] For example, each vehicle sends pre-defined messages to an external device during vehicle operation, and this information is stored and processed in the external device. For example, to determine the actual value, the pre-defined messages sent by each vehicle within a time interval are processed separately. The actual value is, for example, the number of pre-defined messages sent within the time interval.

[0008] The pre-given message to be sent and / or the pre-given message already sent is, for example, the corresponding vehicle status data. If the vehicle is running, the vehicle sends, for example, a pre-given message containing status data to an external device. The status data is, for example, data of the vehicle displayed to a user in the vehicle and / or on a mobile device.

[0009] According to at least one embodiment of the method, a deviation value is determined for each vehicle, wherein each deviation value represents the difference between an actual value and an expected value. For example, each vehicle has one deviation value.

[0010] In addition to deviation values, each vehicle has at least one attribute. These attributes may include, for example, one of the following: vehicle type, product update, software version, controller combination, domestic market, backend-hub, and production date. Each vehicle may, for example, include at least some of the same attributes.

[0011] For example, information including product updates indicates what kind of updates have been made to the vehicle's software, particularly the control software. Similarly, information including software versions indicates the current status of the vehicle's software, particularly the control software. Information including controller combinations indicates, for example, which controllers are installed in the vehicle and / or which controllers communicate with each other. Information including the domestic market indicates the geographical region in which the vehicle primarily operates. Information including back-end hubs indicates which external back-end hubs the vehicle connects to.

[0012] According to at least one embodiment of the method, a defective subgroup of defective vehicles is determined based on deviation values. The defective subgroup, for example, includes defective vehicles with relatively large deviation values.

[0013] According to at least one embodiment of the method, the actual value of the defective subgroup differs from the expected value.

[0014] For example, the defective subgroup may include defective vehicles that send significantly fewer pre-given messages compared to the pre-given message count. That is, the defective vehicles in the defective subgroup send relatively few messages. In this case, the actual value of the defective subgroup is less than the expected value.

[0015] Alternatively, the actual value can be greater than the expected value. For example, a defective vehicle might send more of the pre-given messages that have already been sent in this situation.

[0016] For example, the actual value of the defective subgroup is at least two standard deviations larger or smaller than the expected value.

[0017] This method allows for the particularly simple and rapid identification of defective subgroups, including defective vehicles. These defective vehicles, for example, share common attributes of responsibility for the fault. If the fault involves only a subset of vehicles, particularly a subset of interconnected vehicles in a fleet, the given method can identify the defective subgroup. The defective subgroup includes, for example, defective vehicles that are specific vehicle types with only a specific software version and / or a specific production date for the back-end hub.

[0018] Therefore, the properties of defective vehicles are known, and targeted remedies can be introduced.

[0019] According to at least one embodiment of the method, when providing the expected value, a standard deviation of the pre-given number of messages to be sent is additionally provided for each vehicle type. Specifically, when providing the value, both the expected value and the standard deviation of the pre-given number of messages to be sent are provided separately for each vehicle type.

[0020] Alternatively, when providing the expected value, a width for the pre-given number of messages to be sent is additionally provided for each vehicle type. Here, the width corresponds, for example, to the distance between the 20th percentile and the 80th percentile of the expected value.

[0021] According to at least one embodiment of the method, each deviation value represents a certain number of expected value standard deviations.

[0022] According to at least one embodiment of the method, each expected value is determined based on a plurality of time intervals, each time interval representing a weekday. For example, a separate expected value, specifically a separate expected value and a separate standard deviation, is assigned to each weekday for each vehicle type.

[0023] According to at least one embodiment of the method, when identifying a defective subgroup, a global median representing all deviation values ​​for each vehicle is determined. Specifically, the global median is the median of all deviation values.

[0024] According to at least one embodiment of the method, when determining a defective subgroup, at least one first subgroup is determined based on a first attribute.

[0025] According to at least one embodiment of the method, the at least one first subgroup has a first median that differs most from the global median.

[0026] According to at least one embodiment of the method, when determining a defective subgroup, at least one second subgroup and at least one third subgroup are generated from the at least one first subgroup.

[0027] According to at least one embodiment of the method, the at least one second subgroup has a second median based on a second attribute.

[0028] According to at least one embodiment of the method, the at least one third subgroup has a third median based on a third attribute.

[0029] According to at least one embodiment of the method, the difference between the second median and the third median is maximized.

[0030] For example, further subgroups can be generated from the at least one second subgroup and / or the at least one third subgroup based on other properties.

[0031] According to at least one embodiment of the method, the defective subgroup is determined based on a pre-given population size of the second and third subgroups.

[0032] For example, the pre-defined group size is the termination criterion of the method. In this case, the defective subgroup is formed by a subgroup smaller than the pre-defined group size. The pre-defined group size of the defective subgroup is pre-defined based on statistical correlation. For example, the pre-defined group size is such that overfitting does not occur. For example, the pre-defined group size of the defective subgroup includes at least 500 vehicles and at most 5000 vehicles.

[0033] If the population size of the first subgroup and / or the second group is greater than a predetermined population size, further subgroups are generated from the second subgroup and / or the third subgroup. For example, the method terminates when the population size of at least one of the further subgroups is less than the predetermined population size.

[0034] According to at least one embodiment of the method, the first, second, and third attributes each include at least one of the following: vehicle type, product update, software version, controller combination, domestic market, backend hub, and production date. If further subgroups arise, the additional attributes may also include one of the above information.

[0035] According to at least one embodiment of the method, the first, second, and third properties are different from each other. In particular, the other properties are also different from each other when generating the further subgroups.

[0036] Furthermore, an apparatus for identifying defective vehicles is provided. This apparatus is constructed to implement the method described herein. Therefore, features of the embodiments disclosed in connection with the method are also disclosed in connection with the apparatus, and vice versa.

[0037] Furthermore, a computer program is provided, which includes instructions that, when executed by a computer, cause the computer to perform the methods described herein.

[0038] Furthermore, a computer-readable storage medium is provided on which the computer program described herein is stored. Attached Figure Description

[0039] Embodiments of the present invention will now be described in more detail with reference to the schematic diagrams. In the diagrams:

[0040] Figure 1 A flowchart illustrating a method according to one embodiment is shown;

[0041] Figure 2 A schematic diagram of a system having a device according to one embodiment is shown; and

[0042] Figure 3 A schematic diagram is shown in a method according to one embodiment for determining a defective subgroup. Detailed Implementation

[0043] According to Figure 1 In the flowchart of the method of the embodiment, method step S1 is first implemented, wherein an expected value for the number of pre-given messages to be sent is provided for each vehicle type.

[0044] For example, each vehicle 2 in an interconnected fleet is configured to send information during operation. This information consists of pre-defined messages and, for example, includes the status data of the corresponding vehicle 2. An expected value can be formed for each vehicle type, comprising the number of pre-defined messages to be sent, where each pre-defined message represents a different vehicle type. Furthermore, the expected value also includes the number of pre-defined messages to be sent within a week.

[0045] The expected value can be generated in time before the defective subgroup of vehicle 2 is identified.

[0046] In the subsequent method step S2, the actual value of the number of pre-given messages sent is determined for each vehicle 2.

[0047] For example, each vehicle 2 sends pre-defined messages to device 1, particularly an external device, during operation; these pre-defined messages are stored in device 1. For instance, the number of pre-defined messages sent within a week is determined, corresponding to an actual value. In a further method step S3, a deviation value is then determined for each vehicle 2, where each deviation value represents the difference between the actual value and the expected value. That is, the number of pre-defined messages to be sent within a week is determined by the number of pre-defined messages already sent within a week, particularly within the same week.

[0048] Subsequently, in method step S4, a defective subgroup of defective vehicles is determined based on the deviation value, wherein the actual value of the defective subgroup is less than the expected value. Each defective vehicle includes at least one identical attribute. If the number of pre-given messages sent within a week is significantly less than the number of pre-given messages to be sent within that week, particularly within the same week, the probability of errors induced by this attribute increases.

[0049] according to Figure 2 The system of one embodiment includes a device 1, particularly an external device, which is connected to a vehicle 2 via a communication connection 3. The communication connection 3 is configured to transmit a pre-given message to the device 1.

[0050] Device 1 is constructed for implementing according to Figure 1 The method. The device is, for example, constructed in a backend server.

[0051] For this purpose, device 1 specifically includes a computing unit, a program and data memory, and, for example, one or more communication interfaces. The program and data memory and / or the computing unit and / or the communication interfaces can be constructed in a single structural unit and / or distributed across multiple structural units.

[0052] In order to implement the method, the program for identifying defective vehicles is stored in the program and data memory of the device 1, and the program performs the method described above.

[0053] according to Figure 3 First, the first subgroup SG1 is determined based on the first attribute. For example, the first attribute is information about the first product update. All vehicles with this attribute are included in the first subgroup SG1.

[0054] In particular, the first subgroup SG1 has a first median that differs most from the global median, where the global median represents all deviation values ​​of vehicle 2. For example, the first subgroup SG1 includes 600,000 vehicles.

[0055] Subsequently, a second subgroup SG2 and a third subgroup SG3 are generated from the first subgroup SG1. For example, the second subgroup SG2 includes vehicles in the first subgroup SG1 whose actual values ​​correspond to expected values. The third subgroup SG3 includes, for example, vehicles in the first subgroup SG1 whose actual values ​​differ from expected values. The second subgroup SG2 includes vehicles related to a second attribute. For example, the second attribute is information about updates to a second product. Therefore, all vehicles in the second subgroup SG2 include vehicles with both the first and second attributes. Furthermore, the second subgroup SG2 has a second median.

[0056] The third subgroup SG3 includes vehicles related to a third attribute. For example, the third attribute might be information about the domestic market. Therefore, all vehicles in the third subgroup SG3 include vehicles with both the first and third attributes. Furthermore, the third subgroup SG3 has a third median.

[0057] Here, the second subgroup SG2 and the third subgroup SG3 are chosen to maximize the difference between the second median and the third median.

[0058] Subsequently, the third subgroup SG3 generates further subgroups, namely the fourth subgroup SG4 and the fifth subgroup SG5, based on additional attributes. The fourth subgroup SG4, for example, includes vehicles in the third subgroup SG3 whose actual values ​​correspond to their expected values. The fifth subgroup SG5, for example, includes vehicles in the first subgroup SG1 whose actual values ​​differ from their expected values. For example, the fifth subgroup SG5 may include only 15,000 vehicles.

[0059] According to the combination Figure 1 The method of the embodiment, wherein the fifth subgroup SG5 corresponds to the defective subgroup of defective vehicles.

[0060] List of reference numerals

[0061] 1 device

[0062] 2 vehicles

[0063] 3 communication device

[0064] SG1 first subgroup

[0065] SG2 second subgroup

[0066] SG3 third subgroup

[0067] SG4 fourth subgroup

[0068] SG5 fifth subgroup

[0069] S1...S4 Method Steps

Claims

1. A method for identifying defective vehicles, wherein, The defective vehicles are a subset of multiple vehicles (2), and the multiple vehicles (2) are divided into multiple vehicle types, the method comprising: - Provide the expected number of pre-given messages to be sent for each vehicle type (S1); - For each vehicle (2), determine the actual value of the number of pre-given messages that have been sent (S2); - For each vehicle (2), determine (S3) a deviation value, where each deviation value represents the difference between the actual value and the expected value; - Determine (S4) a defective subgroup of defective vehicles based on the deviation value, wherein the actual value of the defective subgroup is different from the expected value; Specifically, when identifying defective subgroups: - Determine the global median of all deviation values ​​representing each vehicle (2); - Determine at least one first subgroup (SG1) based on the first attribute, where: --The at least one first subgroup (SG1) has a first median that differs most from the global median; - At least one second subgroup (SG2) and at least one third subgroup (SG3) are generated from the at least one first subgroup (SG1), wherein: --The at least one second subgroup (SG2) has a second median according to the second attribute; --The at least one third subgroup (SG3) has a third median based on the third attribute, and --The difference between the second median and the third median is maximized.

2. The method according to claim 1, wherein, In addition to providing the expected value, the standard deviation of the amount of pre-given information to be sent is provided for each vehicle type.

3. The method according to claim 2, wherein, Each deviation value represents the number of standard deviations from the expected value.

4. The method according to any one of claims 1 to 3, wherein, Each expected value is determined based on multiple time intervals, each time interval representing a weekday.

5. The method according to any one of claims 1 to 3, wherein, The defective subgroup is determined based on the pre-given population size of the second subgroup (SG2) and the third subgroup (SG3).

6. The method according to any one of claims 1 to 3, wherein, - The first, second, and third attributes each include at least one of the following information: vehicle type, product update, software version, controller combination, domestic market, backend hub, and production date. - The first, second, and third attributes are different from each other.

7. An apparatus (1) for identifying defective vehicles, the apparatus being configured to implement the method according to any one of claims 1 to 6.

8. A computer program product comprising a computer program, the computer program including instructions that, when executed by a computer, cause the computer program to perform the method according to any one of claims 1 to 6.

9. A computer-readable storage medium on which a computer program product according to claim 8 is stored.

Images