Anomaly detection data identification device, anomaly identification system, and storage medium that stores anomaly identification program

Patent metadata

Field	Value
Publication number	US-10979442-B2
Application number	US-201916242263-A
Country	US
Kind code	B2
Filing date	Jan 8, 2019
Priority date	Feb 23, 2018
Publication date	Apr 13, 2021
Grant date	Apr 13, 2021

Abstract

A piece of anomaly detection data that is the subject of determination is referred to as target data. The position of a vehicle when an anomaly occurred is grasped based on the target data. The number of pieces of the anomaly detection data is defined as the number of pieces of the anomaly detection data output within a predetermined area from the position of the vehicle when the anomaly occurred that is grasped based on the target data among the anomaly detection data stored in the memory. It is determined that the possibility that the target data is generated due to repair of the vehicle is higher when the number of pieces of the anomaly detection data is greater than or equal to a predetermined number than when the number of pieces of the anomaly detection data is less than the predetermined number.

First claim

The invention claimed is:

1. An anomaly detection data identification device configured to receive anomaly detection data from a vehicle-mounted anomaly notification device, the anomaly detection data identification device comprising:
a processing circuit and a memory, the processing circuit being configured to:
receive the anomaly detection data from the vehicle-mounted anomaly notification device, the anomaly detection data including a number of pieces of information that include target data,
store the received anomaly detection data in the memory,
determine whether a degree of probability possibility that the target data in the received anomaly detection data is generated due to repair of the vehicle is greater or less than a predetermined occurrence threshold,
acquire a position of the vehicle when the anomaly occurred based on the target data, and
when an anomaly is detected, output anomaly detection data including the position of a vehicle and information indicating a presence of the anomaly, wherein:

the number of pieces of the anomaly detection data is defined as the number of pieces of the anomaly detection data output within a predetermined area from the position of the vehicle when the anomaly occurred is acquired based on the target data among the anomaly detection data stored in the memory, and
the degree of probability that the target data is generated due to repair of the vehicle is: (i) greater than the predetermined occurrence threshold when the number of pieces of the anomaly detection data is greater than or equal to a predetermined number in the predetermined area from the position of the vehicle, and (ii) less than the predetermined occurrence threshold when the number of pieces of the anomaly detection data is less than the predetermined number in the predetermined area from the position of the vehicle.

2. The anomaly detection data identification device according to claim 1 , wherein
the anomaly detection data includes information about time at which an anomaly occurred, and
the processing circuit is configured to:
specify a repair site candidate based on the target data,
aggregate a number of occurrences of the anomaly by time periods based on the anomaly detection data output within a specified area from the repair site candidate among the anomaly detection data stored in the memory, the number of occurrences of the anomaly being defined as the number of occurrences of the anomaly that corresponds to a time indicated by the target data among the categorized time periods, and
determine that the possibility that the target data is generated due to repair of the vehicle is higher when the number of occurrences of the anomaly is greater than a predetermined value.

3. The anomaly detection data identification device according to claim 2 , wherein
the processing circuit is configured to:
calculate a centroid based on positions included within a predetermined distance from the position of the vehicle identified by the target data among positions of the vehicle identified by past anomaly detection data stored in the memory, the centroid being a mean value of the positions, and
specify the repair site candidate based on the calculated centroid.

4. The anomaly detection data identification device according to claim 3 , wherein
the calculation of the centroid includes:
setting the calculated centroid as an initial value of a first centroid,
calculating a second centroid based on the positions included within the predetermined distance from the first centroid among the positions of the vehicle identified by the anomaly detection data of the past stored in the memory, the second centroid being a mean value of the positions of the vehicle,
selecting the first centroid as the repair site candidate if-when a distance between the position of the vehicle identified by the target data and the second centroid is greater than a threshold value, and
selecting the second centroid as the repair site candidate when a distance between the position of the vehicle identified by the target data and the second centroid is less than or equal to the threshold value, and a distance between the first centroid and the second centroid is less than or equal to a specified value, and

when a distance between the first centroid and the second centroid is greater than the specified value, the processing circuit substitutes the second centroid for the first centroid and executes the calculates another second centroid.

5. The anomaly detection data identification device according to claim 1 , wherein
the anomaly detection data includes information about a vehicle speed when the anomaly occurred, and
the processing circuit is configured to:
determine that the possibility that the target data is generated due to repair of the vehicle is higher when the vehicle speed is small than when the vehicle speed is great.

6. The anomaly detection data identification device according to claim 1 , wherein
the anomaly detection data includes information about the a day of week on which the anomaly occurred, and
the processing circuit is configured to:
specify a repair site candidate based on the target data,
aggregate a number of occurrences of the anomaly by day of week based on the anomaly detection data output within a specified area from the repair site candidate among the anomaly detection data stored in the memory, and
determine that the possibility that the target data is generated due to repair of the vehicle is higher when the number of occurrences of the anomaly is greater than a predetermined value on the day corresponding to the day indicated by the target data among the categories of the day.

7. The anomaly detection data identification device according to claim 1 , wherein
the anomaly detection data includes information that identifies the vehicle, and
the processing circuit is configured to:
aggregate a number of occurrences of anomaly of the same vehicle based on the anomaly detection data stored in the memory, and
determine that the possibility that the anomaly detection data is generated due to repair of the vehicle is higher when the number of occurrences of the anomaly of the same vehicle as the vehicle identified by the target data is greater than a predetermined value.

8. The anomaly detection data identification device according to claim 1 , wherein
the processing circuit is configured to
output, in on a display, data that links information indicating the occurrence of the anomaly based on the anomaly detection data and a determination result of the determination process corresponding to the information.

9. An anomaly identification system comprising:
a vehicle-mounted anomaly notification device configured to anomaly is detected, output anomaly detection data including a position of a vehicle and information indicating an occurrence of the anomaly when an anomaly is detected, and
an anomaly detection data identification device is-configured to receive the anomaly detection data from the vehicle-mounted anomaly notification device, the anomaly detection data identification device including a processing circuit and a memory, the processing circuit being configured to:
receive the anomaly detection data from the vehicle-mounted anomaly notification device, the anomaly detection data including a number of pieces of information that include target data,
store the received anomaly detection data in the memory,
determine whether a degree of probability that the target data in the received anomaly detection data is generated due to repair of the vehicle is greater or less than a predetermined occurrence-threshold,
acquire a position of the vehicle when the anomaly occurred based on the target

Assignees

• Toyota Motor Co Ltd

Inventors

• Yamanashi Takahiro
• Tahara Jun

Classifications

• G07C5/0808

  Diagnosing performance data (testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005) · CPC title

• G07C5/008 Primary

  communicating information to a remotely located station (transmission systems for measured values G08C) · CPC title

• H04L67/12

  specially adapted for proprietary or special-purpose networking environments, e.g. medical networks, sensor networks, networks in vehicles or remote metering networks · CPC title

• G06Q10/0635

  Risk analysis of enterprise or organisation activities · CPC title

• H04L63/1425 Primary

  Traffic logging, e.g. anomaly detection · CPC title