Robot and method to recognize and handle abnormal situations

• US 9,132,547 B2
• Filed: 01/10/2013
• Issued: 09/15/2015
• Est. Priority Date: 01/12/2012
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: Jang, Jun Won, Roh, Kyung Shik
• Primary Examiner(s): HOLLOWAY, JASON R
• Application Number:
  US13/738,485
• Publication Number:
  US 20130184867A1
• Time in Patent Office: 978 Days
• Field of Search: 700/246, 700/250, 700252-254, 700/257, 901/1
• US Class Current: 1/1
• CPC Class Codes: B25J 9/163 learning, adaptive, model based, rule based expert control
• CPC Class Codes: B25J 9/1674 characterised by safety, monitoring, diagnostic
• CPC Class Codes: G05B 2219/33002 Artificial intelligence AI, expert, knowledge, rule based system KBS
• CPC Class Codes: G05B 2219/33056 Reinforcement learning, agent acts, receives reward, emotion, action selective
• CPC Class Codes: G05B 2219/45108 Aid, robot for aid to, assist human disabled

Abstract

A robot and method to recognize and handle abnormal situations includes a sensing unit to sense internal and external information of the robot, a storage unit to store the information sensed by the sensing unit, an inference model, a learning model, services providable by the robot, subtasks to provide the services, and handling tasks to handle abnormal situations, and a controller to determine whether an abnormal situation has occurred while the subtasks to provide the selected service are being performed through the learning model in the storage unit and to select a handling task to handle the abnormal situation through the inference model in the storage unit if it is determined that the abnormal situation has occurred. The robot may recognize and handle abnormal situations even though data other than that defined by a designer of the robot is input thereto due to noises or operational environment.

Claims

1. A robot, comprising:

a sensing unit configured to sense information of the robot;

a storage unit configured to store the information sensed by the sensing unit, at least one inference model, at least one learning model, at least one service providable by the robot, at least one subtask to provide the at least one service, and at least one handling task to handle an abnormal situation of the robot; and

a controller configured to select a service to be provided by the robot from among the at least one service providable by the robot by using the at least one inference model, configured to select a subtask to provide the selected service from among the at least one subtask, and configured to determine whether the abnormal situation has occurred while the selected subtask is being performed by using the at least one learning model;

wherein the abnormal situation is a situation in which the robot, which is performing the selected subtask to provide the selected service, is unable to perform a next subtask.

2. The robot according to claim 1, wherein the controller is further configured to determine whether the service to be provided by the robot is selected, and is further configured to determine whether the information satisfies rules included in the at least one inference model in order to select the service to be provided by the robot, if the service to be provided by the robot is not selected.

3. The robot according to claim 1, wherein the controller is further configured to select a learning model corresponding to the selected subtask from among the at least one learning model, is further configured to input the information to the selected learning model, and is further configured to determine whether the abnormal situation has occurred based on an output from the selected learning model.

4. The robot according to claim 3, wherein the controller is further configured to train the selected learning model by mapping the information to occurrence/non-occurrence of abnormal situations if there is no learning model corresponding to the selected subtask in the storage unit, and is further configured to store the trained learning model in the storage unit, in order to thereby create a learning model corresponding to the selected subtask.

5. The robot according to claim 1, wherein the controller is further configured to select a handling task from the at least one handling task to handle the abnormal situation by using the at least one inference model if it is determined that the abnormal situation has occurred.

6. The robot according to claim 5, wherein the controller is further configured to determine whether the information satisfies rules included in the at least one inference model in order to select the handling task from the at least one handling task to handle the abnormal situation.

13. The robot according to claim 1, wherein the information sensed by the sensing unit comprises internal and external information of the robot.

14. The robot according to claim 1, further comprising:

a drive unit configured to drive the robot in order to provide the service selected by the controller.

7. A method, comprising:

sensing, by a sensor, information of a robot;

selecting, by a processor, a service to be provided by the robot by using an inference model stored by the robot;

selecting, by the processor, a subtask to provide the selected service; and

determining, by the processor, by using a learning model stored by the robot, whether an abnormal situation of the robot has occurred while the selected subtask is being performed;

wherein at least one handling task of the robot is configured to handle the abnormal situation, and

wherein the abnormal situation is a situation in which the robot, which is performing the selected subtask to provide the selected service, is unable to perform a next subtask.

8. The method according to claim 7, wherein the selecting of the service comprises:

determining whether the service is selected; and

determining whether the information satisfies rules included in the inference model in order to select the service, if the service to be provided by the robot is not selected.

9. The method according to claim 7, wherein the determining of whether the abnormal situation has occurred comprises:

selecting the learning model corresponding to the selected subtask;

inputting the information to the selected learning model; and

determining whether the abnormal situation has occurred based on an output from the selected learning model.

10. The method according to claim 9, wherein the determining of whether the abnormal situation has occurred further comprises:

training the selected learning model by mapping the information to occurrence/non-occurrence of abnormal situations if no learning model corresponds to the selected subtask; and

storing the trained learning model in order to create a learning model corresponding to the selected subtask.

11. The method according to claim 7, further comprising:

selecting the at least one handling task to handle the abnormal situation by using the inference model if it is determined that the abnormal situation has occurred.

12. The method according to claim 11, wherein the selecting of the at least one handling task comprises:

determining whether the information satisfies rules included in the inference model in order to select the at least one handling task.

15. The method according to claim 7, further comprising:

driving the robot in order to provide the selected service.

16. A non-transitory computer-readable recording medium storing a program to implement a method, the method comprising:

sensing, by a sensor, information of a robot;

selecting, by a processor, a service to be provided by the robot by using an inference model stored by the robot;

selecting, by the processor, a subtask to provide the selected service; and

determining, by the processor, by using a learning model stored by the robot, whether an abnormal situation of the robot has occurred while the selected subtask is being performed;

wherein at least one handling task of the robot is configured to handle the abnormal situation, and

wherein the abnormal situation is a situation in which the robot, which is performing the selected subtask to provide the selected service, is unable to perform a next subtask.