Proactive acquisition of data for maintenance of appearance model by mobile robot

Patent metadata

Field	Value
Publication number	US-10816974-B2
Application number	US-201715623088-A
Country	US
Kind code	B2
Filing date	Jun 14, 2017
Priority date	Feb 28, 2017
Publication date	Oct 27, 2020
Grant date	Oct 27, 2020

Abstract

The novel technology described in this disclosure includes an example method comprising selecting a target of interest having an obsolete appearance model, the obsolete appearance model describing a prior appearance of the target of interest, navigating a first mobile robot to a location the first mobile robot including a mechanical component providing motive force to the first mobile robot and an image sensor, and searching for the target of interest at the location. The method may include collecting, in the location by the image sensor of the first mobile robot, appearance data of the target of interest, and updating the obsolete appearance model using the appearance data of the target of interest. In some implementations, the method may, in a subsequent meeting between the target of interest and a second mobile robot at a later point in time, recognizing the target of interest using the updated appearance model.

First claim

What is claimed is:

1. A computer-implemented method comprising:
selecting, by one or more processors, a target of interest having an obsolete appearance model, the obsolete appearance model describing a prior appearance of the target of interest;
navigating a first mobile robot to a location, the first mobile robot including a mechanical component providing motive force to the first mobile robot and an image sensor;
searching for the target of interest at the location including navigating at the location while collecting identification data and identifying the target of interest with a threshold level of confidence in the identification of the target of interest based on the identification data;
collecting, in the location by the image sensor of the first mobile robot, appearance data of the target of interest;
updating, by the one or more processors, the obsolete appearance model using the appearance data of the target of interest; and
in a subsequent meeting between the target of interest and a second mobile robot at a later point in time, recognizing, by the one or more processors, the target of interest using the updated appearance model.

2. The computer-implemented method of claim 1 , wherein selecting the target of interest is based on a predicted probability that the obsolete appearance model of the target of interest is obsolete.

3. The computer-implemented method of claim 2 , wherein the predicted probability that the obsolete appearance model of the target of interest is obsolete is based on an event detectable by the first mobile robot.

4. The computer-implemented method of claim 2 , wherein the predicted probability that the obsolete appearance model of the target of interest is obsolete is based on a predicted time of an appearance of the target of interest changing.

5. The computer-implemented method of claim 1 , wherein selecting the target of interest includes
predicting a likelihood that the obsolete appearance model of the target of interest is obsolete, and
determining whether the likelihood that the obsolete appearance model of the target of interest is obsolete satisfies a defined threshold.

6. The computer-implemented method of claim 1 , further comprising, after updating the obsolete appearance model using the appearance data of the target of interest, selecting a second target of interest on a target of interest list.

7. The computer-implemented method of claim 1 , further comprising updating, by the one or more processors, a target of interest list to indicate that the target of interest has the updated appearance model.

8. The computer-implemented method of claim 1 , further comprising:
physically moving the first mobile robot to the location by the mechanical component; and
searching, by the first mobile robot, for the target of interest by physically moving the first mobile robot around the location.

9. The computer-implemented method of claim 1 , wherein the location is predicted based on a history of locations where the first mobile robot has recognized the target of interest previously.

10. The computer-implemented method of claim 1 , wherein collecting the appearance data of the target of interest includes physically moving the first mobile robot around the location while collecting the appearance data until sufficient appearance data has been collected to identify the target of interest as a particular subject and update the obsolete appearance model.

11. The computer-implemented method of claim 1 , wherein the first mobile robot and the second mobile robot are the same mobile robot.

12. A system comprising:
one or more non-transitory memories;
one or more processors;
one or more mobile robots including a mechanical component providing motive force to the one or more mobile robots and an image sensor;
a target identifier executable to select a target of interest having an obsolete appearance model, the obsolete appearance model describing a prior appearance of the target of interest;
a navigator executable to navigate a first mobile robot to a location and search for the target of interest at the location, searching for the target of interest including navigating at the location while collecting identification data and identifying the target of interest with a threshold level of confidence in the identification of the target of interest based on the identification data;
a model updater executable to collect, in the location, appearance data of the target of interest and update, by the one or more processors, the obsolete appearance model using the appearance data of the target of interest; and
a subject detector executable to, in a subsequent meeting between the target of interest and a second mobile robot at a later point in time, recognize, by the one or more processors, the target of interest using the updated appearance model.

13. The system of claim 12 , wherein selecting the target of interest is based on a predicted probability that the obsolete appearance model of the target of interest is obsolete.

14. The system of claim 13 , wherein the predicted probability that the obsolete appearance model of the target of interest is obsolete is based on an event detectable by the first mobile robot.

15. The system of claim 13 , wherein the predicted probability that the obsolete appearance model of the target of interest is obsolete is based on a predicted time of an appearance of the target of interest changing.

16. The system of claim 12 , wherein selecting the target of interest includes
predicting a likelihood that the obsolete appearance model of the target of interest is obsolete, and
determining whether the likelihood that the obsolete appearance model of the target of interest is obsolete satisfies a defined threshold.

17. The system of claim 12 , wherein the target identifier is further executable to, after updating the obsolete appearance model using the appearance data of the target of interest, select a second target of interest on a target of interest list.

18. The system of claim 12 , wherein the target identifier is further executable to update, by the one or more processors, a target of interest list to indicate that the target of interest has the updated appearance model.

19. The system of claim 12 , wherein the navigator is further executable to
physically move the first mobile robot to the location by the mechanical component, and
search, by the first mobile robot, for the target of interest by physically moving the first mobile robot around the location.

20. The system of claim 12 , wherein the location is predicted based on a history of locations where the first mobile robot has recognized the target of interest previously.

21. The system of claim 12 , wherein collecting the appearance data of the target of interest includes physically moving the first mobile robot around the location while collecting the appearance data until sufficient appearance data has been collected to identify the target of interest as a particular subject and update the obsolete appearance model.

22. The system of claim 12 , wherein the first mobile robot and the second mobile robot are the same mobile robot.

23. A computer-implemented method comprising:
selecting, by one or more processors, a target of interest having an obsolete appearance model, the obsolete appearance model describing a prior appearance of the target of interest;
searching an environment using a mobile computing device for the target o

Assignees

• Toyota Motor Co Ltd

Inventors

• Martinson Eric
• Guo Rui
• Nakano Yusuke

Classifications

• G06V30/1914

  Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries, e.g. user dictionaries · CPC title

• G06F18/28

  Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries · CPC title

• G06V30/194

  References adjustable by an adaptive method, e.g. learning · CPC title

• G06V40/103

  Static body considered as a whole, e.g. static pedestrian or occupant recognition · CPC title

• G06V20/10

  Terrestrial scenes (scenes under surveillance with static cameras G06V20/52; scenes perceived from the exterior of a vehicle G06V20/56; scenes perceived from the interior of a vehicle G06V20/59) · CPC title