JP2010014498A - Failure analysis server for vehicle, failure analysis system for vehicle, and rule information storage method - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a failure analysis server for a vehicle for correlating efficiently vehicle detection data with a cause of malfunction, and to provide a failure analysis system for the vehicle, and a rule information storage means. <P>SOLUTION: This failure analysis server 50 for the vehicle for generating rule information for estimating the cause of malfunction from vehicle detection data and repair information, based on the vehicle detection data detected by a vehicle component 21, corresponding to generation of malfunction and on the repair information for solving the malfunction, has a reception means 31 for receiving from the vehicle, position information of the vehicle and the vehicle detection data; an area discriminating means 32 for discriminating an area where the vehicle is traveling, based on the position information; a storage means 60 for storing rule information classified by area generated, by using the vehicle detection data and the repair information of the vehicle traveling in the same area; and an estimation means 24 for reading out the rule information generated for a discriminated area from the storage means 60, and estimating the cause of the malfunction, by applying the rule information to the vehicle detection data received. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a vehicle failure analysis server that generates rule information for estimating the cause of a failure from vehicle detection data and repair information, and more particularly to a vehicle failure analysis server that generates rule information for each region characterized by a failure. The present invention relates to a vehicle failure analysis system and a rule information storage method.

  When a failure is detected in the vehicle, vehicle detection data indicating the state of the vehicle component when the failure is detected, a failure code, etc. are stored in an ECU (electronic control unit) or the like in order to analyze the cause of the failure It has become. However, with the complexity and sophistication of vehicle parts, it has become difficult to identify the cause of a malfunction only by referring to vehicle detection data and failure codes. In order to analyze a defect more efficiently, a technique for creating a database of environmental information such as weather and vehicle detection data when a defect is detected has been proposed (for example, see Patent Document 1).

  In addition, a failure analysis system has been proposed in which a phenomenon when a failure occurs is input and stored in association with vehicle detection data, and the cause of the failure detected in another vehicle is estimated (for example, patent document). 2). By using the phenomenon of failure and various accumulated information, the cause of the failure is estimated in a shorter time.

  In addition, from the viewpoint of quantifying the association between the vehicle detection data and the cause of the failure, the vehicle detection data in a specific operation state is compared with the membership function data stored in advance, and converted into a fitness level. There has been proposed a vehicle failure diagnosis apparatus that calculates failure cause data by comparing with rule data determined in a specific operating state, and estimates the cause of the failure as a larger cause of failure (for example, patents) Reference 3).

In addition, when registering diagnostic information in the center (server), a technology has been proposed for analyzing the failure in consideration of the area where the diagnostic information is generated and the surrounding environment by adding the current position by the navigation system. (For example, refer to Patent Document 4).
JP 2006-27391 A JP 2006-251918 A JP-A-3-267734 Japanese Patent No. 3151831

  However, just making environmental information and vehicle detection data or the like into a database as described in Patent Document 1 does not necessarily relate to a defect and its environment, and is not necessarily effective in analyzing the defect. Further, as in the failure analysis system described in Patent Document 2, it is inefficient to search for the cause of the failure from all the accumulated vehicle detection data, and there is a problem that the processing load is large. For example, like a vehicle failure diagnosis apparatus described in Patent Document 3, a large amount of data and a high level of processing capability are required to learn the relationship between a failure and its cause by data mining.

  Specifically, since vehicle detection data detected in accordance with national laws cannot be used for failure analysis even if it is accumulated outside that country, it may be preferable not to accumulate it for efficient failure analysis.

  In addition, features that occur due to driving conditions such as geographical conditions (for example, high altitudes), traffic rules (for example, vehicle speed restrictions), and traffic environments (for example, heavy traffic congestion) in the country in which the vehicle travels may occur. is there. In this case, by accumulating vehicle detection data and the like for each country and learning for each country, it is possible to expect a reduction in the data amount of learning results and a reduction in processing load when estimating the cause of failure.

  In addition, Patent Document 4 describes that the diagnosis information stored in the server is analyzed in consideration of the area and surrounding environment. However, it is not possible to estimate the cause of the trouble of complicated and advanced vehicle parts. Have difficulty.

  In view of the above problems, an object of the present invention is to provide a vehicle failure analysis server, a vehicle failure analysis system, and a rule information storage method that more efficiently associate vehicle detection data with a cause of a failure.

  In view of the above problems, the present invention is based on vehicle detection data detected by a vehicle component in response to occurrence of a failure and repair information for solving the failure, and rule information for estimating the cause of the failure from the vehicle detection data and the repair information. A vehicle failure analysis server that generates (for example, teacher data), receiving means for receiving the vehicle position information and the vehicle detection data from the vehicle, and the vehicle is running based on the position information Discriminating from the area discriminating means for discriminating the area, the storage means for storing the rule information for each area generated using the vehicle detection data and the repair information of vehicles traveling in the same area, and the storage means An estimation means (for example, a data mining unit 24) that reads out the rule information generated for the selected area and applies it to the received vehicle detection data to estimate the cause of the malfunction; Characterized in that it has a.

  According to the present invention, the teacher data that reflects the characteristics of each region that is likely to cause defects can be stored for each region, thereby reducing the data amount of each teacher data and the processing load of failure analysis using the teacher data. .

  Further, according to one aspect of the present invention, there is provided an area discriminating unit that discriminates an area where the vehicle is traveling based on the vehicle position information transmitted from the vehicle, and is generated for the area discriminated from the storage unit. The rule information is read out and transmitted to the vehicle.

  According to the present invention, since the vehicle can acquire the teacher data generated for the traveling region, it is not necessary to hold the teacher data having a large amount of data generated assuming the traveling in another region. The processing load of failure analysis can be reduced.

  It is possible to provide a vehicle failure analysis server, a vehicle failure analysis system, and a rule information storage method that more efficiently associate vehicle detection data with a cause of a failure.

  The best mode for carrying out the present invention will be described below with reference to the accompanying drawings.

  FIG. 1 is a diagram for explaining an outline of a vehicle failure analysis system 200. First, vehicle detection data detected by the vehicle failure analysis apparatus 100 mounted on the vehicle 40 differs depending on national traffic laws, regulations, climate, and the like. For example, since the vehicle 40 traveling in the United States is subjected to strict exhaust gas regulations, the vehicle 40 shipped to the United States includes a sensor for detecting whether the exhaust gas satisfies the regulations. In addition, since Mexico adjacent to the United States has many mountainous areas, the vehicle 40 often travels in high altitudes, and the vehicle 40 shipped (possibly traveled) to Mexico realizes ignition control according to atmospheric pressure. Therefore, an atmospheric pressure sensor is provided. In cold and tropical countries, parts and controls appropriate to the climate may be used.

  On the other hand, since Europe has an autobahn with no vehicle speed limitation, vehicles 40 traveling in Europe may continue to travel at high speeds for a long time, and traffic jams will occur in emerging countries with undeveloped roads. Because there are many cases, start and stop are often repeated.

  In this way, the vehicle detection data detected by the vehicle 40 varies from country to country, and even if the same vehicle detection data is detected, the driving pattern can vary from country to country. A spot color will occur. For this reason, in the learning using the vehicle detection data and the repair information for solving the problem that has occurred, it is preferable to accumulate and process the vehicle detection data for each country.

  Therefore, the vehicle failure analysis system 200 of the present embodiment groups the vehicles (vehicle detection data and repair information) 40 for each country in which the vehicle travels, generates teacher data for each country, and travels in that country. 40, the teacher data dedicated to the country is distributed.

  For example, the vehicle failure analysis apparatus 100 for the vehicle 40 traveling in Mexico stores Mexican teacher data, and using the Mexican teacher data, whether or not a failure has occurred, a vehicle component in which the failure has occurred, and The state of the vehicle parts (hereinafter simply referred to as the cause of the malfunction) is estimated. When moving to the United States, the vehicle failure analysis apparatus 100 receives the American teacher data from the server 50, replaces the Mexican teacher data, and thereafter uses the American teacher data to estimate the cause of the failure.

  Therefore, the cause of the failure can be estimated using the teacher data optimized for estimating the cause of the failure of the vehicle traveling in the country. Further, since the teacher data is replaced in the vehicle 40, the storage capacity is not compressed, and the amount of teacher data is not increased, so that the processing load does not increase.

  The vehicle detection data is a detection value detected by an on-board sensor, a control signal of an actuator, a signal indicating a state of a sensor / actuator / ECU (electronic control unit) or the like (hereinafter referred to as a vehicle part), etc. Any data that can be acquired by the vehicle failure analysis apparatus 100 on the vehicle may be used. For example, freeze frame data (hereinafter referred to as FFD) is known as vehicle detection data at the time of detecting a malfunction. In the present embodiment, the generation of teacher data using data mining is sometimes referred to as learning, and the cause of failure using the teacher data is referred to as data mining.

  As a general definition, “country” refers to a political society with a certain territory and its exclusive power organization and governing power. The vehicle detection data handled in this embodiment has been described as being affected by the topography and climate caused by the territory, and affected by laws and regulations resulting from the governing authority, but what causes the vehicle detection data to be unique Not limited to these. For example, there may be a culture that protects the vehicle speed regardless of the vehicle speed restriction by law, and depending on the country's religion (eg Islam), driving by women may not be permitted. May be obtained. Therefore, the “country” in this embodiment is wider than the general definition.

  On the other hand, topography, climate, laws, culture, and religion can vary from region to region within one “country” in the general definition. For example, when the country is divided by the sea, desert, mountainous area, etc., the topography and the climate may change greatly depending on the divided area. In addition, for a region (for example, a state in the United States) overseen by a local government, the local government may also define local laws and regulations. Also, it is not uncommon in the world that cultures and religions are different within one “country” in the general definition. In addition, in an area where the country wishes to be independent (for example, the relationship between China and Taiwan), it may be difficult to objectively distinguish whether the area is a country or an area.

  Therefore, the “country” in the present embodiment is a concept that is wider than the country in the general definition and includes one region in the country. It also includes the concept of regions larger than the country (eg, Eurozone, Oceania).

[Teacher data]
The generation of teacher data will be briefly described. FIG. 2 is an example of a diagram for schematically explaining a procedure for generating teacher data by the server 50. The vehicle failure analysis apparatus 100 that has detected some malfunction stores vehicle detection data at the time of malfunction detection. You may memorize | store together the diagnosis code which classifies the content of a malfunction. The vehicle detection data (FFD) stored at the time of failure detection is, for example, engine speed, cooling water temperature, throttle valve opening, travel speed, and the like.

  Further, the vehicle failure analysis apparatus 100 stores vehicle detection data such as the engine speed over a predetermined cycle time while overwriting and saving the vehicle detection data such as the engine rotational speed from the oldest, regardless of the detection of the failure separately from the FFD. When the defect is detected and before and after the vehicle detection data, overwriting is prohibited, and is used together with the FFD to estimate the cause of the defect. The FFD and the vehicle detection data stored periodically are often partially duplicated, and in the present embodiment, it is not necessary to distinguish them, so both will be referred to as vehicle detection data below. .

  When a malfunction that requires inspection of the vehicle 40 is detected, a warning lamp that lights the occurrence of the malfunction on the meter panel of the vehicle 40 is lit, so that the driver can bring the vehicle 40 into the service factory. In the service factory, vehicle detection data is read using a diagnostic tool or the like, and the cause of the malfunction is specified as a result of inspection by a service person or replacement of vehicle parts.

  The cause of the failure is disconnection of connection, contact failure, deterioration or breakage of vehicle parts. The service person resolves (repairs) the problem that occurred according to the cause of the problem. Repair information is obtained by quantifying the contents of the repair by a method suitable for learning and data mining. Accordingly, the repair information and the cause of the defect include equivalent information, but are referred to as repair information when the teacher data is generated, and are referred to as the cause of the problem when data mining.

  The vehicle (service factory) 40 transmits the read vehicle detection data, repair information, and position information at the time of occurrence of a failure to the server 50. Note that the content of the repair may be transmitted to the server 50 and the server 50 may generate repair information.

  The position information transmitted to the server 50 is preferably stored together with the FFD when a malfunction is detected. Thereby, when transmitting to the server 50, the vehicle failure analysis apparatus 100 does not need to detect the position information again. If the vehicle 40 is brought into the service factory without crossing the border due to the occurrence of a problem, the position information of the service factory may be transmitted.

  The server 50 determines the country in which the vehicle 40 was traveling based on the position information. Depending on the determined country and the cause of the problem, vehicle detection data that is unnecessary in that country can be excluded (for example, the detected value of the gas in a country without specific exhaust gas regulations) or clearly unnecessary for the cause of the problem. Vehicle detection data is eliminated. That is, by selecting the vehicle detection data according to the accumulated vehicle detection data and the cause of the failure, it is possible to shorten the learning time and to generate optimal teacher data for the vehicle detection data in that country.

  Then, the server 50 groups the vehicle detection data and repair information of the vehicle 40 traveling in the same country, and generates teacher data for estimating the cause of the malfunction from the vehicle detection data.

  Learning by data mining is known for generating teacher data. For example, the server 50 learns the estimation logic that the vehicle detection data and the change amount of the vehicle detection data reach the repair information. A number of methods such as association rules, clustering, neural networks, and genetic algorithms have been proposed as statistical / analysis algorithms for data mining, but any of these may be used in the present embodiment.

  Therefore, the teacher data of the present embodiment is an estimation logic for estimating the cause of the defect from the vehicle detection data detected by the vehicle 40, for example, and a parameter associated with the estimation logic. The server 50 updates the teacher data every predetermined period (for example, every day, every week, every month). This teacher data is generated for each country. In addition, even if it is the teacher data of the same country, several teacher data may be produced | generated according to a vehicle model. In this case, when the vehicle-type teacher data is generated, the vehicle-type teacher data of the country is distributed to the vehicle 40.

[Vehicle failure analysis device 100]
The vehicle failure analysis apparatus 100 will be described. FIG. 3 is an example of a functional block diagram of the vehicle failure analysis apparatus 100. The vehicle failure analysis apparatus 100 is controlled by the control unit 20. Sensors 21, a position detector 22, and a communication device 23 are connected to the control unit 20 via an in-vehicle LAN such as a CAN (Controller Area Network) and a dedicated line. A part of the sensors 21 is connected to the control unit 20 via another ECU (electronic control unit). The controller 20 communicates with the sensors 21 and other ECUs and other ECUs by time division multiplex communication using, for example, CSMA / CD (Carrier Sense Multiple Access / Collision Detection) as a communication procedure, and receives vehicle detection data.

  The sensors 21 include, for example, a rotation speed sensor that detects the engine speed, a throttle valve opening sensor that detects the throttle valve opening, an exhaust temperature sensor that detects the exhaust temperature, a water temperature sensor that detects the cooling water temperature, and a vehicle speed. A vehicle speed sensor for detecting, a remaining charge amount sensor for detecting a remaining battery level, an acceleration sensor for detecting acceleration or deceleration, a yaw rate sensor for detecting a yaw rate, a master cylinder pressure sensor for detecting a master cylinder pressure, and the like.

  The position detector 22 includes a GPS receiver and an autonomous sensor, detects the position of the vehicle 40 based on the arrival time of radio waves received from GPS satellites, and accumulates the moving direction and distance detected by the autonomous sensor. Is used to estimate the position of the vehicle 40 with high accuracy. Since the position information detected by the position detector 22 is latitude, longitude, and altitude coordinate information, the control unit 20 determines the country based on the position information, or the server 50 that has received the position information travels the vehicle 40. Determine the country in which it is located. Note that the identification information of the country in which the vehicle 40 is traveling may be transmitted to the server 50.

  The communication device 23 communicates with the server 50 via a base station of a communication network such as a mobile phone or a wireless LAN access point. The communication device 23 specifies the identification information of the server 50 (for example, IP address), connects to the server of the communication carrier, and transmits the identification information (telephone number and the like) of the vehicle 40 together with the position information. And the server 50 receives the teacher data selected and transmitted based on the position information of the vehicle 40.

  Further, the control unit 20 is a computer having a CPU, an ASIC (Application Specific Integrated Circuit), a RAM, a ROM, a nonvolatile memory such as a hard disk drive, an input / output interface, and the like, and the CPU executes a program, or It has a data mining unit 24, a teacher data request unit 25, and a teacher data update unit 26, which are realized by hardware such as ASIC. In addition, a teacher data storage unit 27 is mounted in the nonvolatile memory.

  The data mining unit 24 accumulates the vehicle detection data detected by the sensors 21 and applies the teacher data to the vehicle detection data every predetermined cycle time or when a defect is detected, thereby detecting the presence or absence of the defect and the defect. If so, the cause of the problem is estimated. Note that “failure is detected” means that multiple vehicle detections are detected when it is not necessarily definite that a failure has been detected, for example, when one vehicle detection data deviates from a predetermined normal value range. Any situation is included when it is definite that a defect has been detected, such as when the data clearly shows an abnormal value. Even in an uncertain situation, the cause of the failure can be estimated by data mining.

  The teacher data request unit 25 transmits the position information of the vehicle 40 to the server 50 and requests the server 50 for teacher data. The teacher data request unit 25 can request teacher data at two timings. One timing is when it is detected that the vehicle 40 has entered a different country based on the position information detected by the position detector 22. The vehicle failure analysis apparatus 100 stores border position information in advance or can acquire border position information by inquiring a predetermined server. The other timing is a regular timing. For example, the regular interval is preferably about the same as the interval at which the server 50 updates the teacher data. By periodically requesting teacher data from the server 50, it is possible to always perform data mining using the latest teacher data. In addition, when regularly requesting teacher data, teacher data of the country can be acquired before a long time has passed since the crossing of the border, so it is necessary for the teacher data requesting unit 25 to detect that the border has been crossed. There is no.

  The teacher data update unit 26 replaces the teacher data stored in the teacher data storage unit 27 with the teacher data requested by the teacher data request unit 25 and received by the communication device 23. In other words, it may be overwritten. Since the overwriting is performed, the storage capacity of the teacher data storage unit 27 only needs to be secured for the teacher data in the country, and can be suppressed to the minimum.

  The server 50 will be described. The server 50 has a computer equipped with a CPU or the like, and has an information communication unit 31 and a country determination unit 32 that are executed by the CPU or implemented by hardware such as an ASIC, and store teacher data of each country. A master data storage unit 60 is included.

  The information communication unit 31 is a NIC (Network Interface Card) connected to a network such as the Internet, and forms a peer-to-peer network with the communication device 23 of the vehicle 40. The information communication unit 31 receives position information included in the IP packet and the TCP packet, and similarly transmits teacher data to the communication device 23 of the vehicle 40.

  The country determination unit 32 determines the country in which the vehicle 40 is traveling based on the position information received from the vehicle 40. For example, the country discriminating unit 32 detects the closest coordinates to the position information from the coordinate information of the borders of each country stored in advance, and traces the coordinates of one of the two countries whose coordinates are part of the border. Go. If the finally received position information is surrounded by the border of the country, it is determined that the vehicle is traveling in the country. If it is not enclosed, it follows the border coordinates of the other country and repeats the same decision. And the country discrimination | determination part 32 reads the teacher data of the discriminated country from the master data memory | storage part 60, and transmits to the failure analysis apparatus 100 for vehicles via the information communication part 31. FIG.

[Teacher data update procedure]
FIG. 4 shows an example of a sequence diagram illustrating a procedure in which the vehicle failure analysis apparatus 100 updates teacher data. The sequence diagram of FIG. 4 is repeatedly executed at predetermined cycle times in a state where the ignition of the vehicle 40 is on, for example.

  First, the teacher data request unit 25 determines whether the vehicle 40 has crossed the border based on the position information, or whether it is time to acquire teacher data (S10). When crossing the border, or when it is time to acquire teacher data (Yes in S10), the teacher data request unit 25 requests teacher data from the server 50 (S20). At the time of the request, the teacher data request unit 25 transmits the position information of the vehicle 40 to the server 50.

  Moving to the processing of the server 50, the server 50 determines the country in which the vehicle 40 travels based on the received position information (S210), and for the country generated from the vehicle detection data and repair information of the vehicle 40 traveling in that country. Are read from the master data storage unit 60 and transmitted to the vehicle 40 (S220).

  Returning to the processing of the vehicle failure analysis apparatus 100, the teacher data update unit 26 receives the teacher data via the communication device 23 (S30), and updates the teacher data in the teacher data storage unit 27 with the received teacher data (S40). ).

  In step S10, when the border is not crossed and it is not the timing to acquire the teacher data (No in S10), and after updating the teacher data in step S40, the control unit 20 uses the sensors 21 and the like. The vehicle travels while acquiring and accumulating vehicle detection data (S50). When the vehicle detection data is accumulated, the data mining unit 24 determines whether or not to perform data mining (S60). Data mining is executed, for example, every predetermined cycle time or when a failure is detected.

  When a failure is detected at every predetermined cycle time (Yes in S60), the data mining unit 24 applies teacher data to the accumulated vehicle detection data and executes data mining to estimate the cause of the failure (S70). . If no defect is detected as a result of data mining (No in S80), the process is terminated. As a result of data mining, when a defect is detected (Yes in S80), the control unit 20 executes a series of defect detection control (S90). For example, the control unit 20 notifies the server 50 of malfunction information (such as malfunctioned vehicle parts, vehicle detection data, etc.), and if there is a malfunction that requires immediate repair, notifies the occupant accordingly. If there is a problem that prompts inspection, turn on the warning lamp on the meter panel. The estimation of the cause of a defect by data mining is highly accurate, so it is possible to instruct the occupant to take an appropriate action.

  As described above, the vehicle failure analysis system 200 according to the present embodiment generates teacher data for each country that is characterized by a failure that has occurred. Therefore, the efficiency of teacher data generation (the accuracy of failure cause estimation is high, the accuracy of High teacher data can be generated in a short time, and teacher data can be generated with less vehicle detection data). Further, since the vehicle 40 performs data mining using the teacher data optimized for the vehicle detection data of the country, it is possible to reduce the processing load and estimate the cause of the problem.

  In the first embodiment, the vehicle failure analysis apparatus 100 performs data mining. In the present embodiment, a vehicle failure analysis system 200 in which the server 50 performs data mining will be described.

  FIG. 5 shows an example of a functional block diagram of the vehicle failure analysis system 200. 5 that are the same as those in FIG. 3 are assigned the same reference numerals, and descriptions thereof are omitted. The vehicle failure analysis apparatus 100 in FIG. 5 includes a data mining request unit 28. Since the vehicle failure analysis apparatus 100 does not perform data mining, it is not necessary for the vehicle failure analysis apparatus 100 to store the teacher data, and the data mining unit 24, the teacher data request unit 25, the teacher data update unit 26, and the teacher data storage unit. 27 is not provided.

  The data mining request unit 28 transmits position information and vehicle detection data to the server 50 via the communication device 23 and receives data mining result information from the server 50 when a predetermined cycle time or a malfunction is detected. .

  The server 50 receives the position information and the vehicle detection data, and the country determination unit 32 determines the country as in the first embodiment. The country discrimination result detected by the country discrimination unit 32 is sent to the data mining unit 24, and the data mining unit 24 reads the teacher data of the country from the master data storage unit 60. Then, the data mining unit 24 applies teacher data to the vehicle detection data received from the vehicle failure analysis apparatus 100, and executes data mining. Data mining result information is transmitted to the vehicle failure analysis apparatus 100 via the information communication unit 31. If the vehicle has a problem, the data mining result information includes the cause of the problem.

  Therefore, the vehicle failure analysis apparatus 100 according to the present embodiment does not need to have the data mining unit 24, so that the processing load can be reduced, and it is not necessary to store teacher data, so that an increase in cost can be suppressed.

  FIG. 6 shows an example of a sequence diagram showing a procedure in which the vehicle failure analysis system 200 performs data mining. The sequence diagram of FIG. 6 is repeatedly executed at predetermined cycle times with the ignition of the vehicle 40 turned on, for example.

  The controller 20 travels while acquiring vehicle detection data from the sensors 21 and the like and accumulating it (S310). When the vehicle detection data is accumulated, the data mining request unit 28 determines whether to perform data mining (S320). As in the first embodiment, data mining may be executed, for example, every predetermined cycle time or when a failure is detected, or when the server 50 permits the processing to be concentrated on the server 50. It may be determined that the data is mined. When data mining is not performed (No in S320), accumulation of vehicle detection data is repeated.

  When every predetermined cycle time or a failure is detected (Yes in S320), the data mining request unit 28 transmits the accumulated vehicle detection data and position information to the server 50 (S330).

  Shifting to the process of the server 50, the information communication part 31 of the server 50 receives position information and vehicle detection data (S410). And the country discrimination | determination part 32 discriminate | determines the country of the vehicle 40 (S420).

  The data mining unit 24 reads out the teacher data of the determined country from the master data storage unit 60 and applies it to the received device data to execute data mining (S430). The information communication unit 31 transmits the data mining result information to the vehicle failure analysis apparatus 100 (S440).

  Returning to the process of the vehicle failure analysis apparatus 100, if it is determined from the data mining result information that no defect has occurred (No in S340), the process ends. When it is determined from the data mining result information that a defect has occurred (Yes in S340), the control unit 20 executes a series of defect detection control (S350) as in the first embodiment.

  In addition to the effects of the first embodiment, the vehicle failure analysis system 200 of the present embodiment can simplify the configuration of the vehicle failure analysis apparatus 100 and suppress an increase in cost.

It is an example of the figure for demonstrating the outline of the failure analysis system for vehicles. It is an example of the figure which illustrates typically the production | generation procedure of the teacher data by a server. It is an example of the functional block diagram of the failure analysis apparatus for vehicles. It is an example of the sequence diagram which shows the procedure in which the failure analysis apparatus for vehicles updates teacher data. It is an example of a functional block diagram of a failure analysis system for vehicles. It is an example of the sequence diagram which shows the procedure in which the failure analysis system for vehicles performs data mining.

Explanation of symbols

DESCRIPTION OF SYMBOLS 20 Control part 21 Sensors 22 Position detector 24 Data mining part 27 Teacher data storage part 40 Vehicle 50 Server 60 Master data storage part 100 Failure analysis apparatus for vehicles 200 Failure analysis system for vehicles

Claims (7)

Vehicle failure analysis for generating rule information for estimating the cause of a failure from the vehicle detection data and the repair information based on vehicle detection data detected by a vehicle component in response to the occurrence of the failure and repair information for solving the failure A server,
Receiving means for receiving position information of the vehicle and the vehicle detection data from a vehicle;
Area discriminating means for discriminating the area where the vehicle is traveling based on the position information;
Storage means for storing the rule information for each region generated using the vehicle detection data and the repair information of a vehicle traveling in the same region;
An estimation unit that reads out the rule information generated for the determined region from the storage unit and applies the received rule detection data to the received vehicle detection data; and
The vehicle failure analysis server according to claim 1, further comprising: Based on vehicle position information transmitted from the vehicle, the vehicle has a region discriminating unit that discriminates the region in which the vehicle is traveling,
Read the rule information generated for the region determined from the storage means, and transmit to the vehicle,
The vehicle failure analysis server according to claim 1. The region is one country or a region where a local government governs in one country,
The vehicle failure analysis server according to claim 1, wherein the vehicle failure analysis server is a vehicle failure analysis server. A vehicle for receiving vehicle detection data detected by a vehicle component in response to occurrence of a failure and repair information for solving the failure, and generating rule information for estimating the cause of the failure from the vehicle detection data and the repair information A vehicle failure analysis system having a failure analysis server and a vehicle failure analysis device,
The vehicle failure analysis server includes:
Receiving means for receiving position information of the vehicle and the vehicle detection data from a vehicle;
Area discriminating means for discriminating the area where the vehicle is traveling based on the position information;
Storage means for storing the rule information for each region generated using the vehicle detection data and the repair information of a vehicle traveling in the same region;
An estimation unit that reads out the rule information generated for the determined region from the storage unit and applies the received rule detection data to the received vehicle detection data; and
A vehicle failure analysis system comprising: The vehicle failure analysis server includes:
Transmitting the estimation result of the cause of the defect estimated by the estimation means to the vehicle failure analysis device;
The vehicle failure analysis system according to claim 4. The vehicle failure analysis server includes:
Based on vehicle position information transmitted from the vehicle, the vehicle has a region discriminating unit that discriminates the region in which the vehicle is traveling,
From the storage means, the rule information generated for the determined area is read and transmitted to the vehicle,
The vehicle failure analysis device includes:
Replacing pre-stored rule information with the transmitted rule information,
The vehicle failure analysis system according to claim 4. Vehicle failure analysis for generating rule information for estimating the cause of a failure from the vehicle detection data and the repair information based on vehicle detection data detected by a vehicle component in response to the occurrence of the failure and repair information for solving the failure A server rule information storage method comprising:
Receiving means for receiving position information of the vehicle and the vehicle detection data from a vehicle;
A step of determining a region where the vehicle is traveling based on the position information;
From the storage means for storing the rule information for each region generated using the vehicle detection data and the repair information of vehicles traveling in the same region, an estimation unit is generated for the determined region. Reading the rule information and applying it to the received vehicle detection data to estimate the cause of the failure;
A rule information storage method characterized by comprising:

Images