Method for processing plans having multiple end points and electronic device applying the same method

Patent metadata

Field	Value
Publication number	US-11264031-B2
Application number	US-202016783364-A
Country	US
Kind code	B2
Filing date	Feb 6, 2020
Priority date	Mar 6, 2019
Publication date	Mar 1, 2022
Grant date	Mar 1, 2022

Abstract

A method for processing plan is provided. The method includes multiple end points including obtaining an intent corresponding to a user utterance, generating the plan for setting a connection order of actions for executing the intent, determining whether there are multiple end points of the actions, determining whether the actions having different end points are independent, and combining actions having the same end points to generate a plurality of sub plans again and then processing the plurality of sub plans in parallel when the actions having the different end points are independent.

First claim

What is claimed is:

1. A method for processing a plan including multiple end points, the method comprising:
obtaining an intent corresponding to a user utterance;
generating the plan for setting a connection order of actions for executing the intent;
determining whether there are multiple end points of the actions;
determining whether the actions having different end points are independent; and
when the actions having the different end points are independent, combining actions having the same end points to generate a plurality of sub plans again and then processing the plurality of sub plans in parallel.

2. The method of claim 1 , wherein a storage location of the actions is set depending on the end points having the actions.

3. The method of claim 1 , further comprising:
when the actions having the different end points are independent, collecting actions having the same end points to generate a single sub plan.

4. The method of claim 1 , further comprising:
when the actions having the different end points are not independent, processing the actions having the different end points in an order set depending on dependency.

5. The method of claim 1 , further comprising:
dividing a plurality of actions included in the plan depending on the end points; and
when it is determined that there is no dependency between the actions having the different end points, separating the actions having different end points.

6. The method of claim 1 , further comprising:
processing the actions included in a first sub plan among the plurality of sub plans to generate a first result; and
processing the actions included in a second sub plan among the plurality of sub plans to generate a goal action based on the first result.

7. The method of claim 6 , further comprising:
processing the actions included in the second sub plan among the plurality of sub plans to generate a second result; and
merging the first result and the second result to generate the goal action.

8. The method of claim 1 , wherein the multiple end points include a user terminal and/or an intelligence server.

9. The method of claim 1 , further comprising:
receiving the user utterance from an input device connected to a plurality of home appliances;
establishing the plan such that the plurality of home appliances perform a corresponding operation; and
dividing actions included in the plan depending on the end points to generate the plurality of sub plans again.

10. An electronic device applying a method for processing plans including multiple end points, the electronic device comprising:
a client module operationally connected to an intelligence server; and
an execution engine operatively connected to the client module,
wherein the client module is configured to:
receive a user utterance to transmit the user utterance to the intelligence server,
set a connection order of concepts and/or actions, using the plans,
receive a first sub plan generated using first actions having the electronic device as an end point among the actions from the intelligence server,
generate a second sub plan, using second actions having the intelligence server as the end point among the actions, and
when the actions having the different end points are independent, process the first sub plan and the second sub plan in parallel.

11. The electronic device of claim 10 , wherein the client module is further configured to receive a first result generated by processing the first sub plan from the intelligence server.

12. The electronic device of claim 11 ,
wherein the client module is further configured to deliver the first sub plan and/or the first result to the execution engine, and
wherein the execution engine is further configured to process the first sub plan to generate a goal action by reflecting the first result.

13. The electronic device of claim 12 , wherein the execution engine includes:
a parallel processing module configured to generate a second result by processing the second sub plan; and
a result merging module configured to generate the goal action by merging the first result and the second result.

14. The electronic device of claim 10 , wherein the client module is further configured to receive a sub plan that adjusts an order such that actions having the same end points are first processed when the actions having the different end points are dependent.

15. A method of processing plans including multiple end points, the method comprising:
determining whether there are a plurality of end points of actions, of which a connection order is set by the plans;
when there is dependency between the actions having different end points, identifying an end point of a goal action;
performing actions belonging to an end point other than the end point of the goal action to shrink and update a plan establishing and/or setting an execution order of the actions for performing an intent;
delivering the updated plan to an end point performing the goal action; and
performing actions belonging to the end point performing the goal action and performing the goal action.

16. The method of claim 15 , further comprising:
first performing actions having an end point different from the end point of the goal action.

17. The method of claim 15 , further comprising:
performing actions belonging to an end point other than the end point of the goal action to generate a first result; and
performing actions belonging to an end point performing the goal action based on the first result.

18. The method of claim 17 , further comprising:
performing actions belonging to an end point performing the goal action to generate a second result; and
combining the first result with the second result to generate the goal action.

19. The method of claim 15 , wherein the plan sets a processing order and/or processing method of a plurality of concepts or actions.

20. The method of claim 15 , wherein the performing of the goal action includes transmitting a result including whether the goal action succeeds or fails.

Assignees

• Samsung Electronics Co Ltd

Inventors

• Jaygarl Hojun
• Kim Jinwoong
• Lee Kwangbin
• Suh Donghee
• Choe Sungyul

Classifications

• G06F9/451

  Execution arrangements for user interfaces · CPC title

• G10L15/1815

  Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning · CPC title

• G10L15/22 Primary

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G06N5/00

  Computing arrangements using knowledge-based models · CPC title

• G10L2015/223

  Execution procedure of a spoken command · CPC title