Failure detection device for an external sensor and a failure detection method for an external sensor

Patent metadata

Field	Value
Publication number	US-11130499-B2
Application number	US-201916688655-A
Country	US
Kind code	B2
Filing date	Nov 19, 2019
Priority date	Dec 26, 2018
Publication date	Sep 28, 2021
Grant date	Sep 28, 2021

Abstract

A failure detection device that detects a failure in a plurality of external sensors onboard a vehicle, the failure detection device comprises: an overlapping region storage unit that stores an overlapping region of detection areas of the external sensors; an environment performance storage unit that stores an environment-dependent performance of sensing of the external sensors; an environment information acquisition unit that acquires environment information about the vehicle; a recognition result comparison unit that compares recognition results of an object in the overlapping region from the external sensors; a failure likelihood computation unit that computes a failure likelihood of each external sensor based on a comparison result of the recognition results, the environment information, and the environment-dependent performance; and a failure determination unit that determines a failure in each external sensor based on the failure likelihood of each of the external sensors.

First claim

What is claimed is:

1. A failure detection device that detects a failure in a plurality of external sensors onboard a vehicle, the failure detection device comprising:
an overlapping region storage unit that stores an overlapping region of detection areas of the plurality of external sensors;
an environment performance storage unit that stores an environment-dependent performance of sensing of the plurality of external sensors;
an environment information acquisition unit that acquires environment information about the vehicle;
a recognition result comparison unit that compares recognition results of an object in the overlapping region from the plurality of external sensors;
a failure likelihood computation unit that computes a failure likelihood of each of the external sensors based on comparison results of the recognition results, the environment information, and the environment-dependent performance; and
a failure determination unit that determines a failure in each of the external sensors based on the failure likelihood of each of the plurality of external sensors.

2. The failure detection device according to claim 1 , wherein
the failure determination unit determines that a failure occurs in one of the external sensors when a cumulative value of the failure likelihood for the external sensor exceeds a predetermined threshold value.

3. A failure detection device that detects a failure in a plurality of external sensors onboard a vehicle, the failure detection device comprising:
an overlapping region storage unit that stores an overlapping region of detection areas of the plurality of external sensors;
an environment performance storage unit that stores an environment-dependent performance of sensing of the plurality of external sensors;
an environment information acquisition unit that acquires environment information about the vehicle;
a recognition result comparison unit that compares recognition results of an object in the overlapping region from the plurality of external sensors;
a failure likelihood computation unit that computes a failure likelihood of each of the external sensors based on comparison results of the recognition results, the environment information, and the environment-dependent performance; and
a driving mode determination unit that determines one or more driving modes of automatic driving adoptable by the vehicle based on the failure likelihood of each of the plurality of external sensors.

4. The failure detection device according to claim 3 , wherein
the driving mode determination unit determines the driving mode according to the failure likelihood of each of the external sensors and the detection area of each of the external sensors.

5. The failure detection device according to claim 1 , wherein
when each of the external sensors in the overlapping region has a different object recognition result, the failure likelihood computation unit determines that the external sensor having subordinate environment-dependent performance with respect to the environment information is an abnormal sensor, and
increases the failure likelihood for the abnormal sensor.

6. The failure detection device according to claim 5 , wherein
the failure likelihood computation unit
increases the failure likelihood by a small value in a case of unsuccessful recognition in an environment for which the abnormal sensor exhibits low environment-dependent performance, and
increases the failure likelihood by a large value in a case of unsuccessful recognition in an environment for which the abnormal sensor exhibits high environment-dependent performance.

7. The failure detection device according to claim 1 , wherein
a wireless communication unit provided in the vehicle acquires, over a wireless network, data corresponding to a current environment of the vehicle from among data storing past unsuccessful recognitions in at least one different vehicles in association with a contemporaneous environment accumulated in a data center, and increases the failure likelihood when a plurality of abnormalities have previously occurred in the current environment of the vehicle.

8. The failure detection device according to claim 7 , wherein
an increase value in the failure likelihood computed by the failure likelihood computation unit is updated in the data center.

9. A failure detection method that detects a failure in a plurality of external sensors onboard a vehicle, the failure detection method comprising:
a first step of acquiring environment information about the vehicle;
a second step of comparing detection results of an object in an overlapping region of detection areas of the plurality of external sensors, based on the overlapping region stored in advance, an environment-dependent performance dependent on an external environment of the plurality of external sensors stored in advance, and the environment information acquired in the first step;
a third step of specifying an external sensor having an incorrect detection result as an abnormal sensor when, in the second step, detection results of the object are different among the external sensors;
a fourth step computing a failure likelihood with respect to the abnormal sensor specified in the third step based on the environment-dependent performance, the failure likelihood acting as an indicator that indicates a possibility of a state in which repair is required;
a fifth step of determining a failure in the abnormal sensor based on a comparison between a cumulative value of the failure likelihood computed in the fourth step and a predetermined threshold value; and
a sixth step of determining one or more driving modes of automatic driving adoptable by the vehicle from the cumulative value of the failure likelihood computed in the fourth step and the detection areas of the external sensors.

10. The failure detection device according to claim 3 , wherein
when each of the external sensors in the overlapping region has a different object recognition result, the failure likelihood computation unit determines that the external sensor having subordinate environment-dependent performance with respect to the environment information is an abnormal sensor, and increases the failure likelihood for the abnormal sensor.

11. The failure detection device according to claim 3 , wherein
a wireless communication unit provided in the vehicle acquires, over a wireless network, data corresponding to a current environment of the vehicle from among data storing past unsuccessful recognitions in at least one different vehicle in association with a contemporaneous environment accumulated in a data center, and increases the failure likelihood when a plurality of abnormalities have previously occurred in the current environment of the vehicle.

Assignees

• Hitachi Ltd

Inventors

• Nishida Takehisa
• Okude Mariko
• Muto Kazuo

Classifications

• B60W2556/45

  External transmission of data to or from the vehicle · CPC title

• B60W2556/10

  Historical data · CPC title

• B60W2050/0292

  Fail-safe or redundant systems, e.g. limp-home or backup systems · CPC title

• B60W50/023 Primary

  Avoiding failures by using redundant parts · CPC title

• B60W2556/50

  of positioning data, e.g. GPS [Global Positioning System] data · CPC title