Humanoid Robot that can Dynamically Walk with Limited Available Footholds in the Presence of Disturbances

• US 20130184861A1
• Filed: 07/06/2012
• Published: 07/18/2013
• Est. Priority Date: 07/06/2011
• Status: Active Grant

• Current Assignee: Florida Institute Of Human & Machine Cognition
• Original Assignee: Florida Institute Of Human & Machine Cognition
• Inventors: Pratt, Jerry E., Koolen, Twan, Rebula, John
• Granted Patent: US 8,942,848 B2
• Time in Patent Office: Days
• US Class Current: 700/245
• CPC Class Codes: B62D 57/032 with alternately or sequentially lifted supporting base and legs; with alternately or sequentially lifted feet or skid B62D57/024 takes precedence
• CPC Class Codes: G05D 1/021 specially adapted to land vehicles
• CPC Class Codes: G05D 1/0891 specially adapted for land vehicles

Abstract

A control system for a bipedal humanoid robot that utilizes certain fundamental characteristics of bipedal motion to provide a robust and relatively simple balancing and walking mechanism. The system primarily utilizes the concept of “capturability,” which is defined as the ability of the robot to come to a stop without falling by taking N or fewer steps. This ability is considered crucial to legged locomotion and is a useful, yet not overly restrictive criterion for stability. In the preferred embodiment, the bipedal robot is maintained in a 1-step capturable state. This means that future step-locating and driving decisions are made so that the robot may always be brought to a balanced halt with the taking of one step. Other embodiments maintain the bipedal robot in an N-step capturable state, in which the robot may always brought to a balanced halt by taking N or fewer steps.

Claims

1. A method for controlling the motion of a bipedal humanoid robot having a center of mass, a first leg and a second leg, said first leg ending in a first support foot bounded by a first support polygon, and said second leg ending in a second support foot bounded by a second support polygon, comprising:

a. determining an instantaneous capture point of said humanoid robot based on a position of said center of mass, and a velocity of said center of mass;

b. determining a set of vertices for said first polygon that are visible from said instantaneous capture point;

c. defining a set of reflected visible vertices by reflecting each of said visible vertices through said instantaneous capture point;

d. using said instantaneous capture point and said set of reflected visible vertices to define first and second capture region boundary rays;

e. determining a maximum step boundary;

f. defining a capture region bounded by said set of reflected visible vertices, said first capture region boundary ray, said second capture region boundary ray, and said maximum step boundary; and

g. placing at least a portion of said second support polygon into said capture region.

2. A method for controlling the motion of a bipedal humanoid robot as recited in claim 1, further comprising:

a. determining a set of polygons which provide appropriate footing for said humanoid robot;

b. intersecting said set of polygons which provide appropriate footing with said capture region to form a second set of polygons that lie within said capture region and that provide appropriate footing; and

c. placing at least a portion of said second support foot region into said second set of polygons.

3. A method for controlling the motion of a bipedal humanoid robot as recited in claim 1, wherein:

a. said first support foot is connected to said first leg by a first pivoting ankle joint; and

b. said second support foot is connected to said second leg by a second pivoting ankle joint.

4. A method for controlling the motion of a bipedal humanoid robot as recited in claim 3, wherein said capture region is expanded to account for the fact said second support foot can pivot with respect to said second leg.

5. A method for controlling the motion of a bipedal humanoid robot as recited in claim 2, wherein:

a. said first support foot is connected to said first leg by a first pivoting ankle joint; and

b. said second support foot is connected to said second leg by a second pivoting ankle joint.

6. A method for controlling the motion of a bipedal humanoid robot as recited in claim 5, wherein said capture region is expanded to account for the fact said second support foot can pivot with respect to said second leg.

7. A method for controlling the motion of a bipedal humanoid robot as recited in claim 1 wherein:

a. said bipedal robot receives a disturbing force while swinging said second support polygon toward said capture region;

b. after receiving said disturbing force, recalculating said instantaneous capture point and said capture region to determine a revised instantaneous capture point and a revised capture region; and

c. placing at least a portion of said second support polygon into said revised capture region.

8. A method for controlling the motion of a bipedal humanoid robot as recited in claim 2 wherein:

a. said bipedal robot receives a disturbing force while swinging said second support polygon toward said second set of polygons;

b. after receiving said disturbing force, recalculating said instantaneous capture point and said capture region to determine a revised instantaneous capture point and a revised capture region;

c. intersecting said set of polygons which provide appropriate footing with said revised capture region to form a revised second set of polygons that lie within said revised capture region and that provide appropriate footing; and

d. placing at least a portion of said second support foot region into said revised second set of polygons.

9. A method for controlling the motion of a bipedal humanoid robot as recited in claim 8, wherein:

a. said first support foot is connected to said first leg by a first pivoting ankle joint; and

b. said second support foot is connected to said second leg by a second pivoting ankle joint.

10. A method for controlling the motion of a bipedal humanoid robot as recited in claim 3 wherein:

a. said bipedal robot receives a disturbing force while swinging said second support polygon toward said capture region;

b. after receiving said disturbing force, recalculating said instantaneous capture point and said capture region to determine a revised instantaneous capture point and a revised capture region; and

c. placing at least a portion of said second support polygon into said revised capture region.

11. A method for controlling the motion of a bipedal humanoid robot having a center of mass, a first leg and a second leg, said first leg ending in a first support foot bounded by a first support polygon, and said second leg ending in a second support foot bounded by a second support polygon, comprising:

a. determining an instantaneous capture point of said humanoid robot based on a position of said center of mass, and a velocity of said center of mass;

b. determining a maximum step boundary;

c. defining a capture region based on said instantaneous capture point, said maximum step boundary, and a requirement that said humanoid robot be at all times N-step capturable; and

d. placing at least a portion of said second support polygon into said capture region.

12. A method for controlling the motion of a bipedal humanoid robot as recited in claim 11, further comprising:

a. determining a set of polygons which provide appropriate footing for said humanoid robot;

b. intersecting said set of polygons which provide appropriate footing with said capture region to form a second set of polygons that lie within said capture region and that provide appropriate footing; and

c. placing at least a portion of said second support foot region into said second set of polygons.

13. A method for controlling the motion of a bipedal humanoid robot as recited in claim 11, wherein:

a. said first support foot is connected to said first leg by a first pivoting ankle joint; and

b. said second support foot is connected to said second leg by a second pivoting ankle joint.

14. A method for controlling the motion of a bipedal humanoid robot as recited in claim 13, wherein said capture region is expanded to account for the fact said second support foot can pivot with respect to said second leg.

15. A method for controlling the motion of a bipedal humanoid robot as recited in claim 12, wherein:

a. said first support foot is connected to said first leg by a first pivoting ankle joint; and

b. said second support foot is connected to said second leg by a second pivoting ankle joint.

16. A method for controlling the motion of a bipedal humanoid robot as recited in claim 15, wherein said capture region is expanded to account for the fact said second support foot can pivot with respect to said second leg.

17. A method for controlling the motion of a bipedal humanoid robot as recited in claim 11 wherein:

a. said bipedal robot receives a disturbing force while swinging said second support polygon toward said capture region;

b. after receiving said disturbing force, recalculating said instantaneous capture point and said capture region to determine a revised instantaneous capture point and a revised capture region; and

c. placing at least a portion of said second support polygon into said revises capture region.

18. A method for controlling the motion of a bipedal humanoid robot as recited in claim 12 wherein:

a. said bipedal robot receives a disturbing force while swinging said second support polygon toward said second set of polygons;

b. after receiving said disturbing force, recalculating said instantaneous capture point and said capture region to determine a revised instantaneous capture point and a revised capture region;

c. intersecting said set of polygons which provide appropriate footing with said revised capture region to form a revised second set of polygons that lie within said revised capture region and that provide appropriate footing; and

d. placing at least a portion of said second support foot region into said revised second set of polygons.

19. A method for controlling the motion of a bipedal humanoid robot as recited in claim 18, wherein:

a. said first support foot is connected to said first leg by a first pivoting ankle joint; and

b. said second support foot is connected to said second leg by a second pivoting ankle joint.

20. A method for controlling the motion of a bipedal humanoid robot as recited in claim 13 wherein:

a. said bipedal robot receives a disturbing force while swinging said second support polygon toward said second set of polygons;

b. after receiving said disturbing force, recalculating said instantaneous capture point and said capture region to determine a revised instantaneous capture point and a revised capture region;

c. intersecting said set of polygons which provide appropriate footing with said revised capture region to form a revised second set of polygons that lie within said revised capture region and that provide appropriate footing; and

d. placing at least a portion of said second support foot region into said revised second set of polygons.