초록 ▼

Data during normal driving is generated using travel data during the occurrence of a failure or the like, the travel data being accumulated into ordinary vehicles traveling in a city or other places on a daily basis. A failure diagnosis of a vehicle is performed by comparing a reference value with t

Data during normal driving is generated using travel data during the occurrence of a failure or the like, the travel data being accumulated into ordinary vehicles traveling in a city or other places on a daily basis. A failure diagnosis of a vehicle is performed by comparing a reference value with the time sequence electronic control unit (ECU) data of a plurality of driving parameters during the occurrence of a failure, the data being stored into a storage device in the ECU of the vehicle when a failure of the vehicle has occurred. The time sequence ECU data obtained from a large number of vehicles are sequentially accumulated and stored and the numerical vectors of the accumulated time sequence ECU data are generated. The numerical vectors are clustered and classified into a plurality of clusters according to the characteristics. In each of the plurality of clusters, a range of high occurrence rate values is obtained for the values of the respective driving parameters. The obtained range of high occurrence rate values is then stored as a range of the normal values of the driving parameters, and the range of the normal values is set as the reference value for a failure diagnosis.

대표청구항 ▼

1. A system for generating reference values to be used for diagnosis of trouble in vehicles, the system comprising: an accumulating storage for storing onboard snap shot (OBS) data of a plurality of driving parameters that are stored in storage devices of electronic control units (ECUs) of vehicles

1. A system for generating reference values to be used for diagnosis of trouble in vehicles, the system comprising: an accumulating storage for storing onboard snap shot (OBS) data of a plurality of driving parameters that are stored in storage devices of electronic control units (ECUs) of vehicles when trouble takes place, the OBS data read out from a large number of vehicles in trouble that are brought for diagnosis; anda reference value generating unit that is configured toa) generate numerical vectors of the driving parameters for each one of predetermined time periods for each one of the large number of vehicles in trouble that are brought for diagnosis using the OBS data accumulated in the accumulating storage;b) group the numerical vectors into a plurality of clusters based on characteristics of the parameters included in the numerical vectors;c) identify, for each driving parameter in each one of the plurality of clusters, one or more ranges of values of high occurrence rate; andd) store said one or more ranges of values of high occurrence rate as a range of normal values of the driving parameters in each one of the clusters, said range of normal values representing certain driving conditions,wherein, because said OBS data stored in the accumulating storage contain a larger number of normal values than the abnormal values, the abnormal values included in the OBS data that are read out from the large number of vehicles are diluted or excluded by performance of said reference value generating unit of performing steps a), b), c) and d); andwherein the range of normal values of the driving parameters thus stored is used as a range of said reference values for diagnosing troubles of vehicles. 2. The system according to claim 1, wherein the driving parameters include at least one of engine revolution speed, vehicle speed and cooling water temperature for a predetermined time period before a failure of a vehicle occurs. 3. The system according to claim 2, wherein the reference value generating unit is further configured to divide the onboard snap shot (OBS) data into a plurality of time zones for each record, to calculate characteristics of the driving parameters for each time zone, and to generate numerical vectors of the characteristics. 4. The system according to claim 3, wherein the reference value generating unit is further configured to calculate distance between the numerical vectors of the characteristics for each time zone and a plurality of first clusters each having core vectors of initial values, and to group each numerical vector of the characteristics into a first cluster that is nearest to the numerical vector of the characteristics, thereby producing the plurality of first clusters. 5. The system according to claim 4, wherein the reference value generating unit is further configured to: e) calculate for each one of the first clusters an average of the numerical vectors of the characteristics belonging to the each one cluster, said average replacing the initial value of the core vectors of a corresponding first cluster,f) calculate distance between the numerical vectors of the driving parameters in each record for each time zone and the core vectors of each one of the first clusters, and group each numerical vector into a first cluster that is nearest to the numerical vector, thereby updating the plurality of first clusters. 6. The system according to claim 5, wherein g) the processes of e) and f) are repeated until the centers of the core vectors converge or until a preset number of times of trial is done. 7. The system according to claim 6, wherein grouping of numerical vectors into the plurality of first clusters is performed for one or more of predetermined driving parameters, and for the plurality of first cluster a secondary clustering is performed for a larger number of driving parameters for each one of the first clusters to group into a plurality of second clusters. 8. The system according to claim 7, wherein for the numerical vector of each second cluster, rate of occurrence is calculated for each driving parameter, and the range of numerical values having high occurrence rate are allocated as reference values for a corresponding driving parameter. 9. A method for generating reference values that are to be used in diagnosis of a trouble of a vehicle by comparing the reference values with onboard snap shot (OBS) data, the method comprising: a) reading out, by a data accumulation system, for storing in an accumulating storage, onboard snap shot (OBS) data of a plurality of driving parameters that are stored in a storage device of an electronic control unit (ECU) of the vehicle when the trouble takes place, wherein the reading out comprises reading out the OBS data from a large number of vehicles in trouble that are brought for diagnosis;b) generating, by a reference value generating system, numerical vectors of the driving parameters for each one of predetermined time periods for each one of the vehicles that are brought for diagnosis using the OBS data accumulated in the accumulating storage;c) grouping, by the reference value generating system, the numerical vectors into a plurality of clusters based on characteristics of the parameters included in the numerical vectors;d) identifying, by the reference value generating system, for each driving parameter in each cluster one or more ranges of values of a corresponding driving parameter that have large counts of numerical vectors; ande) storing, by the reference value generating system, said one or more ranges of values as a range of normal values of the driving parameters in a corresponding cluster, which represents certain driving condition,wherein, because said OBS data stored in the accumulating storage contain a larger number of normal values than the abnormal values, the abnormal values included in the OBS data that are read out from the large number of vehicles are diluted or excluded in the process of b), c), d), and e); andwherein the range of normal values is used as said reference values in diagnosis of troubles. 10. The method according to claim 9, wherein the driving parameters include at least one of engine revolution speed, vehicle speed and cooling water for a predetermined time period before a failure of a vehicle occurs. 11. The method according to claim 10, wherein the method further comprises: dividing, by the reference value generating system, the onboard snap shot (OBS) data into a plurality of time zones for each record;calculating, by the reference value generating system, characteristics of the driving parameters for each time zone; andgenerating, by the reference value generating system, numerical vectors of the characteristics. 12. The method according to claim 11, wherein the method further comprises: calculating, by the reference value generating system, distance between the numerical vectors of the characteristics for each time zone and a plurality of first clusters each having core vectors of initial values; andgrouping, by the reference value generating system, each numerical vector of the characteristics into a first cluster that is nearest to the numerical vector of the characteristics, thereby producing the plurality of first clusters. 13. The method according to claim 12, wherein the method further comprises: a1) calculating, by the reference value generating system, for each one of the first clusters an average of the numerical vectors of the characteristics belonging to the cluster, said average replacing the initial values of the core vectors of a corresponding first cluster,b1) calculating, by the reference value generating system, distance between the numerical vectors of the driving parameters in each record for each time zone and the core vectors of each one of the first clusters, and group each numerical vector into a first cluster that is nearest to the numerical vector, thereby updating the plurality of first clusters. 14. The method according to claim 13, wherein c1) the processes of a1) and b1) are repeated until the centers of the core vectors converge or until a preset number of times of trial is done. 15. The method according to claim 14, wherein grouping of numerical vectors into the plurality of first clusters is performed for one or more of predetermined driving parameters, and for the plurality of first cluster secondary clustering is performed for a larger number of driving parameters for each one of the first clusters to group into a plurality of second clusters. 16. The method according to claim 15, wherein for the numerical vector of each second cluster, rate of occurrence is calculated for each driving parameter, and the range of numerical values having high occurrence rate are allocated as reference values for a corresponding driving parameter.