Vision-guided alignment system and method

Patent metadata

Field	Value
Publication number	US-9508148-B2
Application number	US-201113817833-A
Country	US
Kind code	B2
Filing date	Aug 25, 2011
Priority date	Aug 27, 2010
Publication date	Nov 29, 2016
Grant date	Nov 29, 2016

Abstract

A vision-guided alignment system to align a plurality of components includes a robotic gripper configured to move one component relative to another component and a camera coupled to a processor that generates an image of the components. A simulated robotic work cell generated by the processor calculates initial calibration positions that define the movement of the robotic gripper such that position errors between the actual position of the robotic gripper and the calibration positions are compensated by a camera space manipulation based control algorithm executed by the processor to control the robotic gripper to move one component into alignment with another component based on the image of the components.

First claim

What is claimed is:

1. A vision-guided alignment system to align a plurality of components comprising:
a robotic gripper configured to move one component relative to another component;
a camera coupled to a processor, said camera generating an image of the components; and
a computer aided design (CAD) model of the components stored on a non-transitory computer readable medium readable by the processor;
wherein a computer-simulated robotic work cell representing a real or potential robotic work cell is generated by said processor using said CAD model, said computer-simulated robotic work cell calculates initial calibration positions that define the movement of said robotic gripper, such that position errors between the actual position of the robotic gripper and said calibration positions are compensated by a camera space manipulation based control algorithm executed by said processor, to control said robotic gripper to move one component into alignment with another component based on said image of the components.

2. The vision-guided alignment system of claim 1 , wherein said initial calibration positions are verified and tested in said computer-simulated robotic work cell.

3. The vision-guided alignment system of claim 1 , wherein said robot gripper assembles the aligned components together at a predetermined insertion position identified by said processor using said CAD model associated with at least one of the components.

4. The vision-guided alignment system of claim 3 , wherein said CAD model identifies the relationship between at least one feature viewable by said camera and at least one feature not viewable by said camera.

5. The vision-guided alignment system of claim 1 , wherein said processor is configured to identify visual features from said image using said CAD model associated with at least one of the components.

6. The vision-guided alignment system of claim 1 , wherein said camera has a field of view that encompasses the component grasped by said robotic gripper.

7. The vision-guided alignment system of claim 1 , wherein said initial calibration positions are generated by said processor in accordance with a data fitting method selected from the group consisting of a least-squares method, a Levenberg-Marquardt method, a linear regression method, or a simplex method.

8. A method of aligning a plurality of components comprising:
providing camera and a robotic gripper each controlled by a processor, said robotic gripper configured to move a grasped component relative to a mounted component, said grasped component comprising first and second opposing sides;
transferring a computer aided design (CAD) model of the grasped component to the processor;
identifying the relative position of at least one feature on said first side of said grasped component viewable by said camera with at least one feature on said second side of said grasped component not viewable by said camera from said CAD model of said grasped component processed by said processor;
generating calibration positions from a camera space manipulation control algorithm for said gripper by a work cell simulated at said processor that define the movement of said grasped component by said robotic gripper to said mounted component based on said CAD model of said grasped component;
moving said grasped component by said robotic gripper in accordance with said calibration positions;
monitoring the movement of said grasped component relative to said mounted component by said camera; and
joining said at least one said feature on said second side of said grasped component to at least one feature of said mounted component.

9. The method of claim 8 , wherein said calibration positions generated at said generating step are constrained by one or more of the position of said camera, the position of one or more of said components, or the position of said robotic gripper.

10. The method of claim 8 , further comprising verifying said calibration positions in said work cell after said generating step.

11. The method of claim 8 , wherein said calibration positions are generated at said generating step in accordance with a data fitting method selected from the group consisting of a least-squares method, a Levenberg-Marquardt method, a linear regression method, or a simplex method.

12. The method of claim 8 , further comprising identifying the position of said at least one feature of said mounted component by said CAD model of said mounted component processed by said processor.

13. A vision-guided alignment system to align a plurality of components comprising:
a robotic gripper configured to move one component relative to another component;
a camera generating an image of the components;
a computer aided design (CAD) model of the components;
a processor generating a computer-simulated robotic work cell from said CAD model representing a real or potential robotic work cell, wherein said processor calculates initial calibration positions that define the movement of said robotic gripper, and wherein said processor executes a camera space manipulation based algorithm to compensate position errors between actual positions of the robotic gripper and said initial calibration positions to control said robotic gripper to move one component into alignment with another component based on said image of the components; and
said robot gripper assembles the components together at a predetermined insertion position identified by said processor using said CAD model, the insertion position comprising a viewable feature on one of the components and a not viewable feature on the other component.

14. The system of claim 13 , wherein said computer-simulated robotic work cell verifies said initial calibration positions.

Assignees

• Zhang Biao
• Wang Jianjun
• Rossano Gregory F
• Fuhlbrigge Thomas A
• Abb Research Ltd

Inventors

• Zhang Biao
• Wang Jianjun
• Rossano Gregory F
• Fuhlbrigge Thomas A

Classifications

• G05B2219/40616

  Sensor planning, sensor configuration, parameters as function of task · CPC title

• Y10S901/47

  Optical · CPC title

• G05B2219/35059

  Convert pcb design data to control data for surface mounting machine · CPC title

• B25J9/1687

  Assembly, peg and hole, palletising, straight line, weaving pattern movement · CPC title

• B25J9/1661 Primary

  characterised by task planning, object-oriented languages · CPC title