Granted Patent B2

US 12,447,604 · App. 17/977,747 · Granted Oct 21, 2025

Inventor: Grant Kenji Larsen (Austin, TX)

B25J9/0006B25J9/163B25J9/1697B25J13/086G05B19/4155B25J15/00G05B2219/40269

Abstract

Claims (58)

1. A robotic assistance device, comprising:

a robotic arm configured assist a human being, the robotic arm including at least a first operative piece;

one or more sensors configured to receive input data, the input data comprising information related to one or more activities being performed by the human being, a workspace in which the one or more activities are performed, a workpiece being affected by the one or more activities, and an object being held by the human being; and

a processor operatively connected to the one or more sensors and the first robotic arm, the processor configured to:

process data received from the one or more sensors,

apply a machine learning model to the processed data to determine that a task should be performed by the robotic assistance device,

determine, based on the sensor data, a classification associated with an object held by the human being and configure the operative piece of the robot based on the classification; and

cause the operative piece to perform the task to reduce cognitive load on the human being.

2. The robotic assistance device of claim 1 , wherein the operative piece comprises a gripper configured to grip the object, chosen from one of the following:

an impactive gripper,

an ingressive gripper,

an astrictive gripper, or

a contigutive gripper.

3. The robotic assistance device of claim 2 , wherein the one or more sensors comprises an inertial monitoring unit; and

wherein determining that the task is complete comprises sensing a pattern of one or more tugs on the object using the inertial monitoring unit.

4. The robotic assistance device of claim 1 , wherein the operative piece comprises a gripper, and wherein the task comprises delivering a tool gripped by the gripper to the human being.

5. The robotic assistance device of claim 1 , wherein the operative piece comprises a gripper, and wherein the task comprises retrieving a tool from the human being.

6. The robotic assistance device of claim 1 , wherein causing the operative piece to perform the task comprises the operative piece moving from a first position to a second position, and

wherein the processor is further configured to:

determine that the task is complete; and

responsive to determining that the task is complete, cause the operative piece to move from the second position to the first position.

7. The robotic assistance device of claim 1 , wherein the one or more sensors comprises one or more of:

a person-zone proximity sensor; and

a gripper-zone object proximity sensor.

8. The robotic assistance device of claim 1 , wherein the one or more sensors comprises one or more cameras.

9. The robotic assistance device of claim 1 , wherein the robotic arm is mounted to a fixed location within a workspace.

10. The robotic assistance device of claim 1 , wherein the arm is configured to be removably attached to the human being.

11. A method of operating a robotic assistance device comprising:

receiving input data, including:

information related to one or more activities being performed by a human being,

information related to a workspace in which the one or more activities are performed,

information related to a workpiece being affected by the one or more activities, and

information related to a classification associated with an object held by the human being;

processing the received input;

applying a machine learning model to the processed input to determine one or more assistance tasks based at least in part on the one or more activities being performed by the human being and the classification;

determining a response based, at least in part, on the processed input;

configuring an end effector of the robot based on the classification; and

causing the robotic assistance device to perform the one or more assistance tasks in response to the received input data to reduce cognitive load on the human being.

12. The method of claim 11 , wherein the input data comprises one or more of the following:

sensor data from one or more sensors connected to the robotic assistance device,

communicated data from another device in signal communication with the robotic assistance device,

external data from one or more external sources, or

command data for augmenting one or more of the sensor data, the communicated data, or the external data.

13. The method of claim 12 ,

wherein the one or more sensors comprises an inertial monitoring unit; and

wherein determining that the task is complete comprises sensing a pattern of one or more tugs on the robotic assistance device using the inertial monitoring unit.

14. The method of claim 12 , wherein the one or more sensors comprises one or more cameras.

15. The method of claim 11 , wherein processing the received input comprises one or more of the following:

processing the input data using a local machine learning model,

processing the input data using a remote machine learning model, or

processing the input data using an algorithmic processing model.

16. The method of claim 11 , further comprising:

determining that the task is complete; and

responsive to determining that the task is complete, causing the robotic assistance device to move from an operative position to a retracted position.

17. The method of claim 11 , wherein the robotic assistance device comprises a gripper, and wherein the task comprises delivering a tool gripped by the gripper to the human being.

18. The method of claim 11 , wherein the robotic assistance device comprises a gripper, and wherein the task comprises retrieving a tool from the human being.

19. The method of claim 11 , wherein the robotic assistance device is mounted to a fixed location within a workspace.

20. The method of claim 11 , wherein the robotic assistance device is configured to be removably attached to the human being.

Assignments (3)

Continuity (2)

References Cited (42)