Assistance for robot manipulation

Patent metadata

Field	Value
Publication number	US-12109682-B2
Application number	US-202217816727-A
Country	US
Kind code	B2
Filing date	Aug 2, 2022
Priority date	Feb 3, 2020
Publication date	Oct 8, 2024
Grant date	Oct 8, 2024

Abstract

A robot control system includes circuitry configured to: acquire an input command value indicating a manipulation of a robot by a subject user; acquire a current state of the robot and a target state associated with the manipulation of the robot; determine a state difference between the current state and the target state; acquire from a learned model, a degree of distribution associated with a motion of the robot, based on the state difference, wherein the learned model is generated based on a past robot manipulation; set a level of assistance to be given during the manipulation of the robot by the subject user, based on the degree of distribution acquired; and generate an output command value for operating the robot, based on the input command value and the level of assistance.

First claim

What is claimed is:

1. A robot control system comprising circuitry configured to:
acquire an input command value indicating a manipulation of a robot by a subject user;
acquire a current state of the robot and a target state associated with the manipulation of the robot;
determine a state difference between the current state and the target state;
acquire from a learned model, a degree of distribution associated with a motion of the robot, based on the state difference, wherein the learned model includes a first model generated based on a past manipulation by another user different from the subject user, and a second model generated based on a past manipulation by the subject user, and wherein the degree of distribution includes a first degree of distribution acquired from the first model and a second degree of distribution acquired from the second model;
set a level of assistance to be given during the manipulation of the robot by the subject user, based on the degree of distribution acquired, wherein the level of assistance includes a first level of assistance that is set based on the first degree of distribution, and a second level of assistance that is set based on the second degree of distribution; and
generate an output command value for operating the robot, based on the input command value and the level of assistance.

2. The robot control system according to claim 1 ,
wherein the circuitry is further configured to acquire from the learned model, a predicted command value which is predicted based on the past manipulation by the other user, and
wherein the output command value is generated based further on the predicted command value.

3. The robot control system according to claim 2 , wherein the circuitry is further configured to:
calculate weighted values of the input command value and of the predicted command value according to respective weights associated with the level of assistance;
add the weighted values to obtain a combined command value; and
generate the output command value based on the combined command value.

4. The robot control system according to claim 1 , wherein the circuitry is further configured to:
identify a task to be executed by the robot; and
select the learned model from a plurality of learned models corresponding to a plurality of tasks executable by the robot, based on the task identified.

5. The robot control system according to claim 4 , wherein the task to be executed by the robot is identified based on sensor data indicating a response of the robot to the input command value.

6. The robot control system according to claim 1 , wherein the degree of distribution is a variance calculated based on a Gaussian process.

7. The robot control system according to claim 1 , wherein the degree of distribution indicates a degree of certainty of manipulation by the other user that is associated with the state difference.

8. The robot control system according to claim 1 , wherein the level of assistance is set to increase in response to a decrease in the degree of distribution.

9. The robot control system according to claim 1 , wherein, the other user has a greater skill level in robot manipulation than the subject user.

10. The robot control system according to claim 9 ,
wherein the output command value is generated based on the input command value, the first level of assistance, and the second level of assistance.

11. The robot control system according to claim 10 , wherein the second degree of distribution indicates a degree of similarity between the manipulation indicated by the input command value and the past manipulation by the subject user that is associated with the state difference.

12. The robot control system according to claim 10 , wherein the second level of assistance is set to increase in response to a decrease in the second degree of distribution.

13. The robot control system according to claim 10 , wherein the output command value is generated based on a comparison of the second level of assistance with a threshold.

14. The robot control system according to claim 13 , wherein the circuitry is further configured to:
acquire from the first model, a predicted command value which is predicted based on the past manipulation by the other user;
calculate weighted values of the input command value and of the predicted command value according to respective weights associated with the first level of assistance; and
add the weighted values to obtain a combined command value,
wherein the output command value is generated based on the combined command value, in response to the second level of assistance being equal to or greater than the threshold.

15. The robot control system according to claim 13 , wherein the output command value is generated by setting the input command value as the output command value, in response to the second level of assistance being less than the threshold.

16. The robot control system according to claim 1 , further comprising the robot.

17. The robot control system according to claim 1 , further comprising a manipulation controller configured to receive the manipulation by the subject user.

18. A robot control method executable by a robot control system including at least one processor, the method comprising:
acquiring an input command value indicating a manipulation of a robot by a subject user;
acquiring a current state of the robot and a target state associated with the manipulation of the robot;
determining a state difference between the current state and the target state;
acquiring from a learned model, a degree of distribution associated with a motion of the robot, based on the state difference, wherein the learned model includes a first model generated based on a past manipulation by another user different from the subject user, and a second model generated based on a past manipulation by the subject user, and wherein the degree of distribution includes a first degree of distribution acquired from the first model and a second degree of distribution acquired from the second model;
setting a level of assistance to be given during the manipulation of the robot by the subject user, based on the degree of distribution acquired, wherein the level of assistance includes a first level of assistance that is set based on the first degree of distribution, and a second level of assistance that is set based on the second degree of distribution; and
generating an output command value for operating the robot, based on the input command value, the first level of assistance, and the second level of assistance.

19. A non-transitory computer-readable storage medium storing processor-executable instructions to:
acquire an input command value indicating a manipulation of a robot by a subject user;
acquire a current state of the robot and a target state associated with the manipulation of the robot;
determine a state difference between the current state and the target state;
acquire from a learned model, a degree of distribution associated with a motion of the robot, based on the state difference, wherein the learned model includes a first model generated based on a past manipulation by another user different from the subject user, the other user having a greater skill level in robot manipulation than the subject user, and a second model generated based on a past manipulation by the subject user, and wherein the degree of distribution includes a first degree of distribution acquired from the first model and a second degree of distribution acquired from the second mod

Assignees

• Tokyo Inst Tech
• Yaskawa Electric Corp

Inventors

• Fujita Masayuki
• Hatanaka Takeshi
• Yamauchi Junya
• Noda Kosei
• SHIMAMOTO Keita
• SOKABE Koji
• Hirata Ryokichi
• Adachi Masaru

Classifications

• B25J9/163 Primary

  learning, adaptive, model based, rule based expert control · CPC title

• G05B2219/40527

  Modeling, identification of link parameters · CPC title

• B25J9/1605

  Simulation of manipulator lay-out, design, modelling of manipulator · CPC title

• B25J13/06 Primary

  Control stands, e.g. consoles, switchboards · CPC title