Control method for a robotic system

Patent metadata

Field	Value
Publication number	US-11865723-B2
Application number	US-201917040075-A
Country	US
Kind code	B2
Filing date	Apr 30, 2019
Priority date	Apr 30, 2018
Publication date	Jan 9, 2024
Grant date	Jan 9, 2024

Abstract

An exemplary robotic system includes a plurality of controllable joints and a controller. An exemplary control method provides for controlling the controllable joints by the controller. The control method provides for determining a configuration space for the robotic system and determining a reference movement path within the configuration space. The control method then provides for assigning a plurality of streamlines in the configuration space to yield a flow field based on the reference movement path. The control method then provides for measuring actual velocity vectors of the robotic system in the configuration space. The control method then provides for determining an error velocity vector based on a difference between the actual velocity vector and the desired velocity vector given by the flow field corresponding to the current robot configuration. The control method then provides for applying a total control vector at the plurality of controllable joints, by the controller, based on the error velocity vector.

First claim

What is claimed is:

1. A control method for generating a desired movement in a robotic system including a plurality of controllable joints and a controller, the control method comprising:
determining a configuration space for the robotic system, the configuration space including dimensionality equal to a number of the plurality of controllable joints;
determining, at the controller, a reference movement path within the configuration space;
assigning a plurality of streamlines in the configuration space to yield a flow field, each of the plurality of streamlines converging to the reference movement path and comprising a velocity magnitude;
measuring an actual velocity vector in the configuration space based on a plurality of velocity components corresponding to the plurality of controllable joints;
determining an error velocity vector based on a difference between the actual velocity vector and selected streamlines in the plurality of streamlines, the selected streamlines having a corresponding location in the configuration space to the actual velocity vector; and
applying a total control vector at the plurality of controllable joints, by the controller, based on the error velocity vector,
wherein applying the total control vector further comprises:
receiving, at the controller, at least one drag coefficient;
determining the total control vector based on a product of the at least one drag coefficient and the error velocity vector; and
applying torque or force at each joint of the plurality of controllable joints based on the total control vector.

2. The method of claim 1 , wherein the applied total control vector for each joint comprises a dot product of the total control vector and an axis associated with each joint.

3. The method of claim 1 , wherein the applied total control vector is strictly dissipative relative to the flow field.

4. The method of claim 1 , wherein streamlines in the plurality of streamlines include identical velocity magnitudes.

5. The method of claim 1 , wherein the at least one drag coefficient further comprises:
an axial drag coefficient in a first direction, the first direction being tangential to the plurality of streamlines; and
a lateral drag coefficient in a second direction, the second direction being normal to the plurality of streamlines.

6. The method of claim 5 , wherein determining the total control vector further comprises:
identifying a first component of the error velocity vector and a second component of the error velocity vector, wherein the first component of the error velocity vector is in the first direction and the second component of the error velocity vector is in the second direction;
determining a first component of the total control vector and a second component of the total control vector, wherein the first component of the total control vector is equal to a product of the axial drag coefficient and the first component of the error velocity vector, and
wherein the second component of the total control vector is equal to a product of the lateral drag efficient and the second component of the error velocity vector; and
determining a combined control vector comprising a vector sum of the first component of the total control vector and the second component of the total control vector.

7. The method of claim 6 , wherein applying the total control vector for each joint comprises a dot product of the combined control vector and a controllable degree of freedom associated with each joint.

8. A control method for generating a desired movement in a robotic system including a plurality of joints and a controller, the control method comprising:
determining a configuration space for the robotic system, the configuration space including dimensionality equal to a number of the plurality of joints;
determining, at the controller, a reference movement path within the configuration space;
assigning a plurality of streamlines in the configuration space to yield a flow field, each of the plurality of streamlines converging to the reference movement path and comprising a velocity magnitude;
measuring an actual velocity vector in the configuration space based on a plurality of velocity components corresponding to controllable joints of the plurality of joints;
determining an error velocity vector based on a difference between the actual velocity vector and selected streamlines in the plurality of streamlines, the selected streamlines having a corresponding location in the configuration space to the actual velocity vector;
applying a total control vector at the controllable joints in the plurality of joints, by the controller, based on the error velocity vector,
wherein applying the total control vector further comprises:
receiving, at the controller, at least one drag coefficient;
determining the total control vector based on a product of the at least one drag coefficient and the error velocity vector; and
applying torque or force at each controllable joint of the controllable joints based on the total control vector.

9. The method of claim 8 , wherein the applied total control vector for each controllable joint comprises a dot product of the total control vector and an axis associated with each controllable joint.

10. The method of claim 8 , wherein the applied total control vector is strictly dissipative relative to the flow field.

11. The method of claim 8 , wherein streamlines in the plurality of streamlines include identical velocity magnitudes.

12. The method of claim 8 , wherein the at least one drag coefficient further comprises:
an axial drag coefficient in a first direction, the first direction being tangential to the plurality of streamlines; and
a lateral drag coefficient in a second direction, the second direction being normal to the plurality of streamlines.

13. The method of claim 12 , wherein determining the total control vector further comprises:
identifying a first component of the error velocity vector and a second component of the error velocity vector, wherein the first component of the error velocity vector is in the first direction and the second component of the error velocity vector is in the second direction;
determining a first component of the total control vector and a second component of the total control vector, wherein the first component of the total control vector is equal to a product of the axial drag coefficient and the first component of the error velocity vector, and
wherein the second component of the total control vector is equal to a product of the lateral drag efficient and the second component of the error velocity vector; and
determining a combined control vector comprising a vector sum of the first component of the total control vector and the second component of the total control vector.

14. The method of claim 13 , wherein applying the total control vector for each controllable joint comprises a dot product of the combined control vector and a controllable degree of freedom associated with each controllable joint.

Assignees

• Univ Vanderbilt

Inventors

• Goldfarb Michael
• Martinez-Guerra Andres
• Lawson Brian

Classifications

• B25J9/1664 Primary

  characterised by motion, path, trajectory planning · CPC title

• B25J9/1633

  compliant, force, torque control, e.g. combined with position control · CPC title

• B25J9/1651

  acceleration, rate control · CPC title

• B25J9/1653

  parameters identification, estimation, stiffness, accuracy, error analysis · CPC title

• B25J9/0006 Primary

  Exoskeletons, i.e. resembling a human figure · CPC title