Anomaly detection device, anomaly detection method, and recording medium
View Patent ↗An anomaly detection device includes: an obtainer that obtains vehicle information related to the status of a vehicle and including location data indicating the location of the vehicle; a model storage that stores, for each of a plurality of cells of a grid imposed on a map, an evaluation model for evaluating the vehicle information of the vehicle located at the cell; and a determiner that calculates, based on the vehicle information and evaluation models each being the evaluation model and corresponding to evaluation cells including a first cell including the location of the vehicle indicated in the location data and one or more second cells each having a predetermined positional relationship with the first cell, an anomaly level indicating a degree of anomaly of the vehicle information, determines, based on the anomaly level, whether the vehicle information is anomalous, and outputs a determination result.
1. An anomaly detection device, comprising:
a communication interface that obtains vehicle information related to a status of a vehicle and including location data indicating a location of the vehicle;
a model storage that stores, for each of a plurality of cells of a grid imposed on a map, an evaluation model for evaluating the vehicle information of the vehicle located at a first cell of the plurality of cells; and
a processor that calculates, based on the vehicle information and evaluation models each being the evaluation model, an anomaly level indicating a degree of anomaly of the vehicle information, determines, based on the anomaly level, whether the vehicle information is anomalous, and outputs a determination result, the evaluation models corresponding to evaluation cells including the first cell including the location of the vehicle indicated in the location data and one or more second cells each having a predetermined positional relationship with the first cell.
2. The anomaly detection device according to claim 1 , wherein
when the anomaly level is greater than or equal to a threshold value, the processor determines that the vehicle information is anomalous.
3. The anomaly detection device according to claim 1 , wherein
when the anomaly level is greater than or equal to a threshold value and an evaluation data count is less than a first predetermined number, the processor corrects the anomaly level to reduce the degree of anomaly, and determines, based on the anomaly level corrected, whether the vehicle information is anomalous, the evaluation data count being a sum of data items used to create the evaluation model for each cell included in the evaluation cells.
4. The anomaly detection device according to claim 3 , wherein
when the anomaly level is greater than or equal to the threshold value and the evaluation data count is less than a second predetermined number smaller than the first predetermined number, the processor corrects the anomaly level to reduce the degree of anomaly until the degree of anomaly indicates that the vehicle information is normal.
5. The anomaly detection device according to claim 3 , wherein
when the anomaly level is greater than or equal to the threshold value and the evaluation data count is less than the first predetermined number, the processor corrects the anomaly level by multiplying the anomaly level by a ratio of the evaluation data count to the first predetermined number.
6. The anomaly detection device according to claim 1 , wherein
the vehicle information further includes speed data indicating a travel speed of the vehicle,
when the travel speed indicated in the speed data is less than a first speed, the processor sets a first predetermined positional relationship to the predetermined positional relationship, and when the travel speed is greater than or equal to the first speed, the processor sets a second predetermined positional relationship to the predetermined positional relationship, and
a total number of the second cells each having the second predetermined positional relationship is greater than a total number of the second cells each having the first predetermined positional relationship.
7. The anomaly detection device according to claim 1 , wherein
the vehicle information further includes speed data indicating a travel speed of the vehicle,
when the travel speed indicated in the speed data is less than a first speed, the processor sets a first predetermined positional relationship to the predetermined positional relationship, and when the travel speed is greater than or equal to the first speed, the processor sets a second predetermined positional relationship to the predetermined positional relationship, and
a total number of the second cells arranged in a first direction from the first cell among the second cells each having the second predetermined positional relationship is greater than a total number of the second cells arranged in the first direction from the first cell among the second cells each having the first predetermined positional relationship.
8. The anomaly detection device according to claim 1 , wherein
the communication interface sequentially obtains the vehicle information,
the processor sequentially calculates the anomaly level, sequentially determines whether the vehicle information is anomalous, and sequentially outputs the determination result, and
the anomaly detection device further comprises:
a memory that sequentially stores, in association with each other, the determination result sequentially output from the processor and the location data corresponding to the determination result; and
an anomaly series processor that, in a case where a first determination result output from the processor indicates that the vehicle information is anomalous and a second determination result stored in the memory a last time indicates that the vehicle information is anomalous, when a distance between a first location indicated in first location data corresponding to the first determination result and a second location indicated in second location data associated with the second determination result is less than a predetermined distance, determines the first determination result and the second determination result as being of a same anomaly series.
9. The anomaly detection device according to claim 8 , wherein
the vehicle information further includes speed data indicating a travel speed of the vehicle, and
when the first determination result output from the processor indicates that the vehicle information is anomalous, the anomaly series processor sets the predetermined distance according to the travel speed indicated in the speed data corresponding to the first determination result.
10. The anomaly detection device according to claim 8 , further comprising:
a display controller that causes a display to display at least a portion of the map, wherein
in a case where the anomaly series processor determines the first determination result and the second determination result as being of the same anomaly series, the anomaly series processor determines that each of the first cell corresponding to the first determination result and the first cell corresponding to the second determination result is an anomalous cell, and
the display controller causes the display to display at least a portion of the map in a manner that an anomalous cell is displayed in a display format different from a display format of other cells.
11. The anomaly detection device according to claim 10 , wherein
in the case where the anomaly series processor determines the first determination result and the second determination result as being of the same anomaly series, when a first anomalous cell corresponding to the first determination result and a second anomalous cell corresponding to the second determination result have a predetermined relationship, the anomaly series processor further determines that a cell located between the first anomalous cell and the second anomalous cell is the anomalous cell.
12. The anomaly detection device according to claim 11 , wherein
the vehicle information further includes speed data indicating a travel speed of the vehicle, and
when the first determination result indicates that the vehicle information is anomalous, the anomaly series processor sets the predetermined relationship according to the travel speed indicated in the speed data corresponding to the first determination result.
13. The anomaly detection device according to claim 10 , wherein
in the case where the anomaly series processor determines the first determination result and the second determination result as being of the same anomaly series, when a cell adjacent to a first anomalous cell corresponding to the first determination result and a second anomalous cell corresponding to the second determination result exists and a road area ratio in the cell is less than a predetermined area ratio, the anomaly series processor further determines that the cell is the anomalous cell.
14. An anomaly detection method performed by an anomaly detection device that stores, for each of a plurality of cells of a grid imposed on a map, an evaluation model for evaluating vehicle information related to a status of a vehicle located at a first cell of the plurality of cells and including location data of the vehicle, the anomaly detection method comprising:
obtaining the vehicle information;
calculating, based on evaluation models each being the evaluation model, an anomaly level indicating a degree of anomaly of the vehicle information, the evaluation models corresponding to evaluation cells including the first cell including a location of the vehicle indicated in the location data and one or more second cells each having a predetermined positional relationship with the first cell;
determining, based on the anomaly level, whether the vehicle information is anomalous; and
outputting a determination result.
15. A non-transitory computer-readable recording medium having recorded thereon a program for causing an anomaly detection device to perform an anomaly detection process, the anomaly detection device storing, for each of a plurality of cells of a grid imposed on a map, an evaluation model for evaluating vehicle information related to a status of a vehicle located at a first cell of the plurality of cells and including location data of the vehicle, the anomaly detection process including:
obtaining the vehicle information;
calculating, based on evaluation models each being the evaluation model, an anomaly level indicating a degree of anomaly of the vehicle information, the evaluation models corresponding to evaluation cells including the first cell including a location of the vehicle indicated in the location data and one or more second cells each having a predetermined positional relationship with the first cell;
determining, based on the anomaly level, whether the vehicle information is anomalous; and
outputting a determination result.