System and method for establishing a gripping behavior model

TWI946047BActive Publication Date: 2026-11-11DELTA ELECTRONICS INC(CN)
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
TW114134905
Authority / Receiving Office
TW · TW
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-09-11
Publication Date
2026-11-11
Estimated Expiration
2045-09-10

Smart Images

  • Figure TWG2TB001920799_001
    Figure TWG2TB001920799_001
  • Figure TWG2TB001920799_002
    Figure TWG2TB001920799_002
  • Figure TWG2TB001920799_003
    Figure TWG2TB001920799_003
Patent Text Reader

Abstract

A method for establishing a clamping behavior model is provided, comprising: performing a clamping operation on a target object using an end effector of a robotic arm; and acquiring dynamic motion data of the target object in a predetermined direction during the clamping operation by the end effector using sensors. The method further comprises performing the following steps using a computer device: acquiring dynamic motion data from sensors; calculating the critical acceleration of the target object in the predetermined direction at the moment of release based on the dynamic motion data; calculating the critical clamping force based on the mass of the target object and the critical acceleration; writing the dynamic motion data and the critical clamping force into a clamping behavior database; and establishing a clamping behavior model using the clamping behavior database.
Need to check novelty before this filing date? Find Prior Art

Claims

1. A system for establishing a clamping behavior model, the system comprising: A robotic arm includes: an end effector configured to perform a gripping operation on a target object; a sensor configured to acquire dynamic motion data of the target object in a predetermined direction during the gripping operation performed by the end effector; and a controller; a computer device includes: a processing unit; and a storage unit storing a computer program containing multiple instructions that, when loaded into the processing unit, cause the processing unit to perform the following steps: acquiring the dynamic motion data from the sensor; calculating a critical acceleration of the target object along the predetermined direction at the moment of release based on the dynamic motion data; calculating a critical gripping force based on the mass of the target object and the critical acceleration; writing the dynamic motion data and the critical gripping force into a gripping behavior database; and establishing a gripping behavior model using the gripping behavior database; and converting the established gripping behavior model into a robotic arm language and inputting the robotic arm language into the controller, causing the controller to control the end effector to reproduce the gripping behavior associated with the gripping behavior model.

2. The system as claimed in claim 1, wherein the critical acceleration is the maximum acceleration reached during a period when a clamping state signal of the end effector indicates a clamping state.

3. The system as described in claim 2, wherein the instructions further cause the processing unit to perform the following steps: compare the clamping state signal of the end effector with the dynamic motion data captured by the sensor in time to estimate an I / O delay parameter; and write the I / O delay parameter, the dynamic motion data and the critical clamping force together into the clamping behavior database.

4. The system as described in any one of claims 1 to 3, wherein the sensor is an encoder built into the robotic arm.

5. The system as claimed in claim 1, wherein the clamping operation performed by the end effector on the target object is a simple harmonic motion or an approximate simple harmonic motion.

6. A method for establishing a clamping behavior model, comprising: A gripping operation is performed on a target object using an end effector of a robotic arm; A sensor is used to acquire dynamic motion data of the target object in a predetermined direction during the clamping operation performed by the end effector. A computer device is used to perform the following steps: acquiring the dynamic motion data from the sensor; calculating a critical acceleration of the target object along the predetermined direction at the moment of release based on the dynamic motion data; calculating a critical clamping force based on the mass of the target object and the critical acceleration; writing the dynamic motion data and the critical clamping force into a clamping behavior database; and establishing a clamping behavior model using the clamping behavior database; and converting the established clamping behavior model into a robotic arm language and inputting the robotic arm language into a controller of the robotic arm, causing the controller to control the end effector to reproduce the clamping behavior associated with the clamping behavior model.

7. The method as described in claim 6, wherein the critical acceleration is the maximum acceleration reached during a period when a clamping state signal of the end effector indicates a clamping state.

8. The method as described in claim 7, wherein the steps performed by the computer device further include: The clamping status signal of the end effector is compared with the dynamic motion data captured by the sensor to estimate an I / O delay parameter; and the I / O delay parameter, the dynamic motion data and the critical clamping force are written into the clamping behavior database.

Citation Information

Patent Citations

  • Nasopharynx swab sampling robot man-machine cooperative interaction control method and system

    CN113133787A

  • Manipulator as well as clamping suite and control system matched with multiple manipulators

    CN113319850A

  • Robot motion control system, and operational method thereof

    TW201718206A

  • Method and system for dexterous manipulation by a robot

    US20250065506A1