Detecting robot grasp of very thin object or feature

Patent metadata

Field	Value
Publication number	US-2021122056-A1
Application number	US-202017029415-A
Country	US
Kind code	A1
Filing date	Sep 23, 2020
Priority date	Oct 25, 2019
Publication date	Apr 29, 2021
Grant date	—

Abstract

A plurality of sensors are configured to provide respective outputs that reflect a sensed value associated with engagement of a robotic arm end effector with an item. The respective outputs of the plurality of sensors are used to make a determination associated with engagement of a robotic arm end effector with an item. A first value measured by a first sensor is used to determine a first input associated with a first factor. A second value measured by a second sensor is used to determine a second input associated with a second factor. The first input and the second input are provided to a multi-factor model configured to provide, based at least in part on the first input and the second input, an output associated with engagement of the robotic arm end effector with the item. The output of the multi-factor model is used to make the determination associated with engagement of the robotic arm end effector with the item.

First claim

What is claimed is:

1 . A robotic system, comprising:
a plurality of sensors, each of the plurality of sensors configured to provide an output that reflects a sensed value associated with engagement of a robotic arm end effector with an item, the plurality of sensors including a first sensor configured to measure a first value associated with a first factor indicative of engagement and a second sensor configured to measure a second value associated with a second factor indicative of engagement, wherein the first and second factors are different; and a processor coupled to the plurality of sensors and configured to use the respective outputs of the plurality of sensors to make a determination associated with engagement of the robotic arm end effector with the item, at least in part by:
using the first value to determine a first input associated with the first factor and using the second value to determine a second input associated with the second factor;
providing the first input and the second input to a multi-factor model configured to provide, based at least in part on the first input and the second input, an output associated with engagement of the robotic arm end effector with the item; and
using the output of the multi-factor model to make the determination associated with engagement of the robotic arm end effector with the item.

2 . The robotic system of claim 1 , wherein a tactile sensing unit includes the plurality of sensors, wherein the tactile sensing unit is comprised of a plurality of layers, wherein each of the plurality of layers includes one or more sensors of the plurality of sensors.

3 . The robotic system of claim 1 , wherein the first factor and the second factor include at least two of weight, deformation, continuity, conductivity, resistance, inductance, and capacitance.

4 . The robotic system of claim 1 , wherein the first sensor includes a plurality of magnets and one or more magnetic sensors.

5 . The robotic system of claim 4 , wherein the plurality of magnets are arranged in a grid.

6 . The robotic system of claim 4 , wherein at least one of the plurality of magnets is located on a boundary associated with a tactile sensing unit.

7 . The robotic system of claim 1 , wherein the first sensor includes an emitter and a receiver.

8 . The robotic system of claim 1 , wherein the first sensor includes a plurality of markers and an image sensor.

9 . The robotic system of claim 1 , wherein the first sensor includes one or more electro- mechanical sensors.

10 . The robotic system of claim 1 , wherein the first sensor includes a pressure sensor coupled to a flexible membrane that includes a gas or a liquid.

11 . The robotic system of claim 1 , wherein the second sensor includes a conductive material.

12 . The robotic system of claim 1 , wherein the plurality of sensors includes a force sensor attached to a wrist portion of the robotic arm end effector.

13 . The robotic system of claim 1 , wherein the first factor is associated with a first coefficient and the second factor is associated with a second coefficient.

14 . The robotic system of claim 13 , wherein the first coefficient and the second coefficient are equal.

15 . The robotic system of claim 13 , wherein the first coefficient and the second coefficient are different.

16 . The robotic system of claim 1 , wherein the respective outputs of the plurality of sensors are provided to the processor when a static measurement is performed and when a dynamic measurement is performed.

17 . The robotic system of claim 1 , wherein the output of the multi-factor model indicates the robotic arm end effector successfully grasped the item.

18 . The robotic system of claim 1 , wherein the output of the multi-factor model indicates the robotic arm end effector did not successful grasp the item.

19 . The robotic system of claim 1 , wherein the output of the multi-factor model indicates that human intervention is needed.

20 . The robotic system of claim 1 , wherein the output of the multi-factor model indicates the robotic arm end effector is to engage a different item.

21 . A method, comprising:
using a first value sensed by a first sensor of a plurality of sensors to determine a first input associated with a first factor and using a second value sensed by a second sensor of the plurality of sensors to determine a second input associated with a second factor, wherein each of the plurality of sensors are configured to provide an output that reflects a sensed value associated with engagement of a robotic arm end effector with an item, wherein the first and second factors are different; providing the first input and the second input to a multi-factor model configured to provide, based at least in part on the first input and the second input, an output associated with engagement of the robotic arm end effector with the item; and using the output of the multi-factor model to make a determination associated with engagement of the robotic arm end effector with the item.

22 . A computer program product embodied in a non-transitory computer readable medium and comprising computer instructions for:
using a first value sensed by a first sensor of a plurality of sensors to determine a first input associated with a first factor and using a second value sensed by a second sensor of the plurality of sensors to determine a second input associated with a second factor, wherein each of the plurality of sensors are configured to provide an output that reflects a sensed value associated with engagement of a robotic arm end effector with an item, wherein the first and second factors are different; providing the first input and the second input to a multi-factor model configured to provide, based at least in part on the first input and the second input, an output associated with engagement of the robotic arm end effector with the item; and using the output of the multi-factor model to make a determination associated with engagement of the robotic arm end effector with the item.

Assignees

• Dexterity Inc

Inventors

• Menon Samir
• Sun Zhouwen
• Su Harry Zhe

Classifications

• B25J13/084 Primary

  Tactile sensors (in general G01L5/16, G01L5/22) · CPC title

• B25J13/082

  Grasping-force detectors (in general G01L5/16, G01L5/22) · CPC title

• B25J19/027

  Electromagnetic sensing devices · CPC title

• B25J9/1612

  characterised by the hand, wrist, grip control · CPC title

• B25J19/023

  including video camera means · CPC title