ROBOTIC SYSTEM AND METHOD FOR OBSERVING, LEARNING, AND SUPPORTING HUMAN ACTIVITIES

• US 20090210090A1
• Filed: 02/18/2008
• Published: 08/20/2009
• Est. Priority Date: 02/18/2008
• Status: Active Grant

• Current Assignee: Toyota Motor Corporation
• Original Assignee: Toyota Motor Corporation, Toyota Motor Engineering & Manufacturing North America Incorporated (Toyota Motor Corporation)
• Inventors: Takemitsu, Mori, Kalik, Steven F.
• Granted Patent: US 8,140,188 B2
• Time in Patent Office: Days
• US Class Current: 700/245
• CPC Class Codes: B25J 9/1661 characterised by task planning, object-oriented languages
• CPC Class Codes: G05B 2219/40298 Manipulator on vehicle, wheels, mobile
• CPC Class Codes: G05B 2219/40543 Identification and location, position of components, objects
• CPC Class Codes: G05B 2219/45108 Aid, robot for aid to, assist human disabled

Abstract

An example method for allowing a robot to assist with a task, the task being carried out in an environment including one or more non-human objects each having associated object locations, comprises detecting one or more changes in object locations within the environment, predicting a task requirement (such as a future object location change, or task goal) by comparing the change in the object location with stored data, the stored data including object location changes associated with previously observed tasks; and providing robotic assistance to achieve the task requirement. Example apparatus are also disclosed.

Claims

1. A method for allowing a robot to assist with a task, the task being carried out in an environment including one or more non-human objects each having an associated object location, the method comprising:

detecting a change in an object location;

predicting a task requirement by comparing the change in the object location with stored data, the stored data including patterns of object location changes associated with previously performed tasks; and

providing robotic assistance to achieve the task requirement.

2. The method of claim 1, wherein detecting a change in an object location comprises:

sensing the environment using a sensor, the sensor providing sensor data; and

determining object locations within the environment using the sensor data.

3. The method of claim 2, the sensor being located on the robot.

4. The method of claim 2, the sensor being an electromagnetic radiation sensor.

5. The method of claim 4, the sensor being an optical sensor.

6. The method of claim 2, the sensor being a capacitive sensor, an inductive sensor, an acoustic sensor, or an ultrasound sensor.

7. The method of claim 2, wherein the sensor is a lidar sensor or a radar sensor.

8. The method of claim 1, wherein the method includes detecting a plurality of object location changes.

9. The method of claim 8, wherein the method includes using a pattern-matching algorithm to match the plurality of object location changes to the patterns of object location changes in the stored data.

10. The method of claim 8, further comprising determining a statistical likelihood of the task requirement from a distribution of the plurality of object location changes within the stored data.

11. The method of claim 1, wherein the stored data is collected by the robot while sensing the environment during previous human activity within the environment.

12. The method of claim 1, wherein the robotic assistance is provided to a human who has started to perform the task, the robotic assistance being provided without receiving a direct request from the human.

13. The method of claim 1, wherein the task requirement is a change in location of a second object.

14. The method of claim 13, wherein the robotic assistance provides the change in location of the second object.

15. A method for allowing a robot to predict how to assist with a task, the task being carried out by a human in an environment including non-human objects each having an associated object location, the method comprising:

sensing the environment using a sensor, the sensor being a component of the robot;

determining object locations within the environment;

detecting changes in object locations within the environment; and

predicting how to assist with the task by comparing the changes in object locations with stored data, the stored data including object location changes associated with previously performed tasks.

16. The method of claim 15, wherein the task is identified by comparing detected changes in object locations with patterns of object location changes in the stored data.

17. The method of claim 16, wherein patterns of object location changes in the stored data are used to predict future object location changes.

18. A robot for assisting with a task within an environment including objects, the objects being non-human objects, the robot comprising:

a sensor, operable to provide sensor data representing the objects;

an electronic circuit, operable to provide;

a location change detector, operable to provide object location change data for the objects within the environment using the sensor data,

a comparator, operable to compare object location change data with stored data, and

a predictor, operable to predict a future object location change for the task from a comparison of the object location change data with the stored data; and

an actuator, in electronic communication with the electronic circuit, operable to effect the future object location change on receipt of a signal from the electronic circuit.

19. The apparatus of claim 18, the location change detector, comparator, and predictor being provided by one or more algorithms executable by a processor within the electronic circuit.

20. The apparatus of claim 18, the stored data being stored within a memory of the electronic circuit.

21. The apparatus of claim 18, wherein the sensor is an electromagnetic radiation sensor, a capacitive sensor, an inductive sensor, an acoustic sensor, an ultrasound sensor, a lidar sensor, or a radar sensor.