Machine object determination based on human interaction

Patent metadata

Field	Value
Publication number	US-9975241-B2
Application number	US-201514958440-A
Country	US
Kind code	B2
Filing date	Dec 3, 2015
Priority date	Dec 3, 2015
Publication date	May 22, 2018
Grant date	May 22, 2018

Abstract

This disclosure pertains to machine object determination based on human interaction. In general, a device such as a robot may be capable of interacting with a person (e.g., user) to select an object. The user may identify the target object for the device, which may determine whether the target object is known. If the device determines that target object is known, the device may confirm the target object to the user. If the device determines that the target object is not known, the device may then determine a group of characteristics for use in determining the object from potential target objects, and may select a characteristic that most substantially reduces a number of potential target objects. After the characteristic is determined, the device may formulate an inquiry to the user utilizing the characteristic. Characteristics may be selected until the device determines the target object and confirms it to the user.

First claim

What is claimed:

1. A device to be controlled by human interaction, comprising:
user interface circuitry configured to receive input from a human user, the input identifying a target object and generate an output to the human user;
memory circuitry configured to store data at least regarding objects known to the device; and
processing circuitry configured to at least:
receive the input from the user interface circuitry;
access the memory circuitry to determine if the target object identified in the input from the human user is known to the device; and
when it is determined that the target object is unknown to the device, the processing circuitry is further configured to:
formulate a first inquiry to said human user, wherein the first inquiry is to reduce a number of potential objects that may correspond to the target object and comprises at least a first object characteristic;
cause the user interface circuitry to present the first inquiry to said human user;
determine, in response to receipt of a reply to said first inquiry from said human user via said user interface circuitry, whether said target object is known to the device;
when it is determined that said target object is still unknown to the device following receipt of said reply to said first inquiry;
formulate a second inquiry to said human user, wherein the second inquiry is to further reduce the number of potential objects that may correspond to the target object and comprises at least a second object characteristic that is different from the first object characteristic;
cause the user interface circuitry to present the second inquiry to said human user; and
determine, in response to receipt of a reply to said second inquiry from said human user via said user interface circuitry, whether said target object is known to the device.

2. The device of claim 1 , wherein the device is a robot.

3. The device of claim 1 , wherein the first and second object characteristics are based on a characteristic of the target object.

4. The device of claim 3 , wherein the first object characteristic, the second object characteristic, or both the first and second object characteristics is/are selected from a group of characteristics based on a characteristic most substantially reducing a number of potential target objects.

5. The device of claim 1 , wherein the processing circuitry is further to:
when it is determined that more than one potential target object corresponds to the target object following receipt of the reply to the second inquiry, formulate a third inquiry to the human user, wherein the third inquiry is to further reduce the number of potential objects that correspond to the target object, and cause the user interface circuitry to present the third inquiry to the human user.

6. The device of claim 4 , wherein the processing circuitry is to determine the group of characteristics based on potential target object characteristics.

7. The device of claim 6 , further comprising sensing circuitry and wherein the processing circuitry is to determine the group of characteristics based on capabilities of the sensing circuitry.

8. The device of claim 6 , wherein the processing circuitry is to determine the potential target objects by causing the sensing circuitry to scan an area in which the device is operating.

9. The device of claim 4 , wherein the processing circuitry is to determine the potential target objects based on a context corresponding to an area in which the device is operating.

10. A method for object determination, comprising:
initiating object determination in a device;
receiving an input from a human user via user interface circuitry in the device, the input identifying a target object;
accessing memory circuitry in the device to determine if the target object identified in the input from the human user is known to the device;
when it is determined that the target object is unknown to the device, the method further comprises:
formulating, with the device, a first inquiry to the human user, wherein the first inquiry is to reduce a number of potential objects that may correspond to the target object and comprises at least a first object characteristic;
causing the user interface circuitry to present the first inquiry to the human user;
in response to receipt of a reply to the first inquiry from the human user via the user interface circuitry, determining with the device whether said target object is known to the device;
when it is determined that said target object is still unknown to the device following receipt of said reply to said first inquiry, the method further comprises:
formulating, with the device, a second inquiry to the human user, wherein the second inquiry is to further reduce the number of potential objects that may correspond to the target object and comprises at least a second object characteristic that is different from the first object characteristic;

causing the user interface circuitry to present the second inquiry to the human user; and
determining, in response to receipt of a reply to said second inquiry from said human user via the user interface circuitry, whether said target object is known to the device.

11. The method of claim 10 , wherein the first and second object characteristics are based on a characteristic of the target object.

12. The method of claim 11 , further comprising:
selecting, with the device the first characteristic, the second characteristic, or both the first and second characteristics from a group of characteristics based on the characteristic most substantially reducing a number of potential target objects.

13. The method of claim 9 , further comprising:
when it is determined that more than one potential target object corresponds to the target object following receipt of the reply to the second inquiry, the method further comprises formulating a third inquiry to the user, wherein the third inquiry is to further reduce the number of potential objects that correspond to the target object, and causing the user interface circuitry of the device to present the third inquiry to the human user.

14. The method of claim 12 , further comprising:
determining the group of characteristics based on at least one of potential target object characteristics or capabilities of sensing circuitry in the device.

15. The method of claim 14 , further comprising:
determining the potential target objects based on at least one of scanning an area in which the device is operating with the sensing circuitry or a context corresponding to the area in which the device is operating.

16. At least one non-transitory machine-readable storage medium having stored thereon, individually or in combination, instructions for object determination that, when executed by one or more processors, cause the one or more processors to perform the following operations comprising:
initiating object determination in a device;
receiving input from a human user via user interface circuitry in the device, the input identifying a target object;
accessing memory circuitry in the device to determine if the target object identified in the input from the human user is known to the device;
when it is determined that the target object is unknown to the device, the instructions when executed further cause the one or more processors to perform the following operations comprising:
formulating, with the device, a first inquiry to the human user, wherein the first inquiry is to reduce a number of potential objects that may correspond to the target object and comprises at least a first object characteristic;

Assignees

• Intel Corp

Inventors

• Kamhi Gila
• Moran Amit
• Nistel Kobi
• Chettrit David

Classifications

• G06N3/008 Primary

  based on physical entities controlled by simulated intelligence so as to replicate intelligent life forms, e.g. based on robots replicating pets or humans in their appearance or behaviour · CPC title

• B25J9/1661

  characterised by task planning, object-oriented languages · CPC title

• B25J9/161 Primary

  Hardware, e.g. neural networks, fuzzy logic, interfaces, processor · CPC title

• G05B2219/40408

  Intention learning · CPC title

• G05B2219/40411

  Robot assists human in non-industrial environment like home or office · CPC title