Humanoid robot push recovery on level and non-level ground

• US 8,849,454 B2
• Filed: 03/20/2012
• Issued: 09/30/2014
• Est. Priority Date: 03/21/2011
• Status: Active Grant

• Current Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Original Assignee: Honda Motor Co., Ltd. (Honda Motor Company)
• Inventors: Yun, Seungkook, Goswami, Ambarish, Lee, Sung-Hee
• Primary Examiner(s): HOLLOWAY, JASON R
• Application Number:
  US13/425,383
• Publication Number:
  US 20120245734A1
• Time in Patent Office: 924 Days
• Field of Search: 700/250, 700/253, 700/254, 700/258, 318/568.12, 901/1, 901/50
• US Class Current: 700/253
• 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: Y10S 901/01 Mobile robot

Abstract

A robot controller controls a robot to maintain balance in response to an external disturbance (e.g., a push) on level or non-level ground. The robot controller determines a predicted stepping location for the robot such that the robot will be able to maintain a balanced upright position if it steps to that location. As long as the stepping location predicted stepping location remains within a predefined region (e.g., within the area under the robot'"'"'s feet), the robot will maintain balance in response to the push via postural changes without taking a step. If the predicted stepping location moves outside the predefined region, the robot will take a step to the predicted location in order to maintain its balance.

Claims

1. A computer-implemented method for controlling a humanoid robot in response to an external disturbance, the robot standing on a first section of non-level ground having a first slope and within stepping range of a second section of the non-level ground having a second slope, the method comprising:

receiving, by a robot controller, state information of a humanoid robot;

determining, by the robot controller, whether to step on the first section or the second section of the non-level ground and determining a predicted stepping location of the robot on the first or second section of the non-level ground, the predicted stepping location based on the state information of the humanoid robot and the first and second slopes, the predicted stepping location comprising a location such that the humanoid robot will maintain balance if it steps to the predicted stepping location;

responsive to the predicted stepping location falling within a predefined region, controlling the robot to make a postural change without stepping to maintain balance in response to the external disturbance; and

responsive to the predicted stepping location falling outside the predefined region, controlling the robot to take a step to the predicted stepping location to maintain balance in response to the external disturbance.

2. The computer-implemented method of claim 1, wherein determining the predicted stepping location comprises:

modeling the humanoid robot as a rimless wheel model representing a first leg and a second leg of a same length, the rimless wheel model having a point mass corresponding to the center of mass of the humanoid robot, a first spoke corresponding to the first leg of the humanoid robot, and a second spoke corresponding to the second leg of the humanoid robot.

3. The computer-implemented method of claim 2, wherein determining the predicted stepping location comprises:

determining a plane on which the rimless wheel model resides;

determining an anchor point for the rimless wheel model;

applying a dynamic model to model motion of the rimless wheel through the plane about the anchor point; and

determining, based on the modeled motion, the stepping location as a point on the ground surface where the rimless wheel model will come to a stop with the point mass above the stepping location.

4. The computer-implemented method of claim 1, wherein controlling the humanoid robot to make a postural change comprises:

determining desired rates of change of linear and angular momentum of the robot based on the desired motion;

determining admissible rates of change of linear and angular momentum based on the desired rates of change and physical limitations of the humanoid robot for balance maintenance, wherein the desired rate of change of linear momentum is respected with preference over the desired rate of change of angular momentum;

generating, by the balance controller, a control output indicating joint torques for achieving the admissible rates of change of linear and angular momentum; and

outputting, by the balance controller, the control output to joint actuators, the joint actuators causing the robot to move according to joint torques.

5. The computer-implemented method of claim 1, wherein controlling the humanoid robot to take the step to the predicted stepping location comprises:

determining desired rates of change of linear and angular momentum of the robot based on the desired motion and to achieve a step to the predicted stepping location;

determining admissible rates of change of linear and angular momentum based on the desired rates of change and physical limitations of the humanoid robot for balance maintenance, wherein the desired rate of change of angular momentum is respected with preference over the desired rate of change of linear momentum;

generating, by the balance controller, a control output indicating joint torques for achieving the admissible rates of change of linear and angular momentum; and

outputting, by the balance controller, the control output to joint actuators, the joint actuators causing the robot to move according to joint torques.

6. The computer-implemented method of claim 5, further comprising:

determining a trajectory of a center of mass of the humanoid robot; and

determining a trajectory for a swing foot of the robot executing the step to the predicted stepping location.

7. A non-transitory computer-readable storage medium for storing computer program instructions for controlling a humanoid robot in response to an external disturbance, the robot standing on a first section of non-level ground having a first slope and within stepping range of a second section of the non-level ground having a second slope, the program instructions when executed by a processor cause the processor to perform steps including:

receiving, by a robot controller, state information of the humanoid robot;

determining, by the robot controller, whether to step on the first section or the second section of the non-level ground and determining a predicted stepping location of the robot on the first or second section of the non-level ground, the predicted stepping location based on the state information of the humanoid robot and the first and second slopes, the predicted stepping location comprising a location such that the humanoid robot will maintain balance if it steps to the predicted stepping location;

responsive to the predicted stepping location falling within a predefined region, controlling the robot to make a postural change without stepping to maintain balance in response to the external disturbance; and

responsive to the predicted stepping location falling outside the predefined region, controlling the robot to take a step to the predicted stepping location to maintain balance in response to the external disturbance.

8. The non-transitory computer-readable storage medium of claim 7, wherein determining the predicted stepping location comprises:

modeling the humanoid robot as a rimless wheel model representing a first leg and a second leg of a same length, the rimless wheel model having a point mass corresponding to the center of mass of the humanoid robot, a first spoke corresponding to the first leg of the humanoid robot, and a second spoke corresponding to the second leg of the humanoid robot.

9. The non-transitory computer-readable storage medium of claim 8, wherein determining the predicted stepping location comprises:

determining a plane on which the rimless wheel model resides;

determining an anchor point for the rimless wheel model;

applying a dynamic model to model motion of the rimless wheel through the plane about the anchor point; and

determining, based on the modeled motion, the stepping location as a point on the ground surface where the rimless wheel model will come to a stop with the point mass above the stepping location.

10. The non-transitory computer-readable storage medium of claim 7, wherein controlling the humanoid robot to make a postural change comprises:

determining desired rates of change of linear and angular momentum of the robot based on the desired motion;

determining admissible rates of change of linear and angular momentum based on the desired rates of change and physical limitations of the humanoid robot for balance maintenance, wherein the desired rate of change of linear momentum is respected with preference over the desired rate of change of angular momentum;

generating, by the balance controller, a control output indicating joint torques for achieving the admissible rates of change of linear and angular momentum; and

outputting, by the balance controller, the control output to joint actuators, the joint actuators causing the robot to move according to joint torques.

11. The non-transitory computer-readable storage medium of claim 7, wherein controlling the humanoid robot to take the step to the predicted stepping location comprises:

determining desired rates of change of linear and angular momentum of the robot based on the desired motion and to achieve a step to the predicted stepping location;

determining admissible rates of change of linear and angular momentum based on the desired rates of change and physical limitations of the humanoid robot for balance maintenance, wherein the desired rate of change of angular momentum is respected with preference over the desired rate of change of linear momentum;

generating, by the balance controller, a control output indicating joint torques for achieving the admissible rates of change of linear and angular momentum; and

outputting, by the balance controller, the control output to joint actuators, the joint actuators causing the robot to move according to joint torques.

12. The non-transitory computer-readable storage medium of claim 11, further comprising:

determining a trajectory of a center of mass of the humanoid robot; and

determining a trajectory for a swing foot of the robot executing the step to the predicted stepping location.

13. A computer-implemented method for determining a stepping location on non-level ground for a humanoid robot in response to an external disturbance, the method comprising:

modeling the humanoid robot as a rimless wheel model, the rimless wheel model having a point mass corresponding to the center of mass of the humanoid robot, a first spoke corresponding to a first leg of the humanoid robot, and a second spoke corresponding to a second leg of the humanoid robot;

determining a plane on which the rimless wheel model resides based on a direction of the humanoid robot'"'"'s center of mass projected onto the non-level ground;

determining an anchor point for the rimless wheel model based on the center of pressure of the humanoid robot;

applying a dynamic model to model motion of the rimless wheel through the plane about the anchor point;

selecting, based on the modeled motion, which of a first section of the non-level ground and a second section of non-level ground to step to, the first section of the non-level ground having a first slope and the second section of the non-level ground having a second slope; and

determining the stepping location on the selected section of the non-level ground above which the point mass of the rimless wheel model will come to rest after executing the modeled motion.

14. The method of claim 13, wherein determining the stepping location comprises:

determining a possible range of stepping locations based on physical limitations of the robot; and

determining the stepping location within the possible range.

15. The method of claim 13, further comprising:

determining a leg angle of the humanoid robot to achieve a step to the stepping location; and

determining a duration of time to complete the step to the stepping location.

16. The method of claim 13, wherein applying the dynamic model comprises:

applying a first model to model a first phase of a stepping motion where the humanoid robot balances on a support leg while stepping with a swing leg, the first model modeling the humanoid robot as an inverted pendulum; and

applying a second model to model a second phase of the stepping motion where the swing leg impacts the non-level ground, the second model modeling an instantaneous collision between the swing leg and the non-level ground.