阅读说明：本技术 一种汽车挡风玻璃自动装配系统及自动装配方法 (Automatic assembling system and automatic assembling method for automobile windshield ) 是由 徐劲力 刘畋渔 卢杰 丁刚强 袁智军 黄丰云 潘昊 詹强民 黄辑 刘万志 于 2021-09-13 设计创作，主要内容包括：本发明提供一种汽车挡风玻璃自动装配系统及自动装配方法,自动装配系统包括3D视觉模块、机器人、机器人控制模块、挡风玻璃抓取模块和图像处理模块。本发明利用结构光和双目视觉系统组成的3D视觉模块,在挡风玻璃抓取模块抓取汽车挡风玻璃与车体车窗框装配过程中,拍摄装配图像,获取多组特征点对,通过特征点检测车窗与车窗框之间的装配间隙,机器人在机器人控制模块控制下,根据视觉检测系统获得的装配间隙对机器人装配路线进行调整,最终达到装配要求。应用本发明可以实现汽车挡风玻璃的自动装配,无需人为干预,并可提高汽车挡风玻璃的自动化装配精度。(The invention provides an automatic assembly system and an automatic assembly method for an automobile windshield. According to the invention, a 3D vision module consisting of a structured light and a binocular vision system is utilized, in the process of assembling the automobile windshield and the automobile window frame grabbed by the windshield grabbing module, an assembly image is shot, a plurality of groups of characteristic point pairs are obtained, the assembly gap between the automobile window and the automobile window frame is detected through the characteristic points, and the robot adjusts the assembly route of the robot according to the assembly gap obtained by the vision detection system under the control of the robot control module, so that the assembly requirement is finally met. The invention can realize the automatic assembly of the automobile windshield without human intervention and can improve the automatic assembly precision of the automobile windshield.)

1. An automatic assembly system for an automobile windshield is characterized by comprising a 3D vision module, a robot control module, a windshield grabbing module and an image processing module;

the 3D vision module is used for grabbing an assembly image in the process that the windshield grabbing module grabs the automobile windshield and the automobile window frame of the automobile body;

the image processing module is used for extracting a plurality of corresponding characteristic point pairs according to the assembly image and calculating an assembly gap between the automobile windshield and the automobile window frame of the automobile body based on the plurality of characteristic point pairs;

and the robot control module is used for planning an assembly route of the robot according to the assembly clearance between the automobile windshield and the automobile window frame of the automobile body and controlling the robot to assemble the automobile windshield according to the planned assembly route.

2. The automatic assembly system of claim 1, wherein the 3D vision modules comprise 4, each of the 3D vision modules comprises a structured light and a binocular vision system, and the 4 3D vision modules are mounted on a robotic arm of a robot;

and 4, the 3D vision modules are used for acquiring a plurality of groups of characteristic point pairs in the installation process of the automobile windshield.

3. The vehicle windshield assembly system of claim 2, wherein the structured light is a one-dimensional laser position finder, and wherein 4 of the 3D vision modules are configured to acquire a plurality of sets of pairs of feature points during installation of the vehicle windshield, comprising:

when the windshield grabbing module grabs the automobile windshield, acquiring a first image containing a plurality of boundary points on the automobile windshield, wherein the boundary points on the automobile windshield are a plurality of characteristic points of the windshield;

when the windshield grabbing module grabs the automobile windshield to move above the automobile body window frame according to the planned assembly line, acquiring a second image containing a plurality of boundary points of the automobile body window frame, wherein the plurality of boundary points of the automobile body window frame are a plurality of characteristic points of the automobile body window frame;

correspondingly, the image processing module is configured to:

extracting a plurality of characteristic points of an automobile windshield from the first image and extracting a plurality of characteristic points of a vehicle body window frame from the second image respectively, wherein the plurality of characteristic points of the automobile windshield and the plurality of characteristic points of the vehicle body window frame form a plurality of groups of characteristic point pairs;

and calculating the assembling clearance between the automobile windshield and the automobile body window frame based on the plurality of characteristic points of the automobile windshield and the corresponding plurality of characteristic points of the automobile body window frame.

4. The automated automobile windshield assembly system according to claim 1, wherein the windshield gripper module comprises four glass suction cups, the four glass suction cups being disposed at four corners of the windshield gripper module for gripping the four corners of the automobile windshield.

5. The automatic assembling system for automobile windshield glass according to claim 1, wherein the robot is a six-degree-of-freedom robot and comprises two parts, the first part is an arm with 3 control axes, and the spatial position of the reference point at the tail end of the robot is determined by the movement of the 3 control axes; the second part is a wrist with 3 control axes, and the posture of the robot end tool is determined by the movement of the 3 control axes.

6. The automated automotive windshield assembly system of claim 1, further comprising a windshield centering and positioning module comprising a lateral centering and clamping mechanism and a longitudinal centering and clamping mechanism;

the transverse centering and clamping mechanism clamps the automobile windshield through the clamping rollers on the two sides to finish transverse centering and clamping, and the longitudinal centering and clamping mechanism clamps the automobile windshield through the clamping rollers to finish longitudinal centering and clamping so as to finish centering and positioning of the automobile windshield.

7. An automatic assembling method for an automobile windshield, which is characterized by comprising the following steps:

the vehicle body lifting appliance suspends a vehicle body into an automatic assembly area, the lifting appliance is statically and mechanically positioned after the vehicle body lifting appliance enters a preset position, and a vehicle body window frame reaches the preset position to finish positioning;

the automobile windshield is arranged on the centering workbench, and the centering and positioning of the automobile windshield are completed through the windshield centering and positioning module;

the windshield grabbing module grabs the automobile windshield after centering and positioning and conveys the grabbed automobile windshield into the shooting range of the 3D vision module;

shooting a preset area in the assembly process by the plurality of 3D vision modules, and capturing an assembly image;

the image processing module extracts a plurality of corresponding characteristic point pairs according to the assembly image and calculates an assembly gap between the automobile windshield and the automobile window frame of the automobile body based on the plurality of characteristic point pairs;

the robot control module plans an assembly route of the robot according to the assembly clearance between the automobile windshield and the automobile window frame of the automobile body, and controls the robot to grab the automobile windshield for assembly according to the planned assembly route.

8. The automatic assembling method for automobile windshields according to claim 7, wherein the 3D vision module comprises a structured light and a binocular vision system, and the structured light is a one-dimensional laser position finder; the plurality of 3D vision modules shoot a preset area in the assembly process, capture an assembly image, and comprise:

when the windshield grabbing module grabs the automobile windshield, the one-dimensional laser position finder projects a linear light plane on the automobile windshield to form a characteristic laser straight line, images the characteristic laser straight line on a camera imaging plane and acquires a first image containing a plurality of boundary points on the automobile windshield, wherein the boundary points on the automobile windshield are a plurality of characteristic points of the automobile windshield;

when the windshield grabbing module grabs the automobile windshield and moves above the automobile body window frame according to the planned assembly line, the one-dimensional laser position finder projects a linear light plane on the automobile windshield and the automobile body window frame respectively, a characteristic laser straight line is formed on the automobile windshield and the automobile body window frame respectively and is imaged on a camera imaging plane, a second image containing a plurality of boundary points of the automobile body window frame except the boundary points on the automobile windshield is obtained, the plurality of boundary points of the automobile body window frame are a plurality of characteristic points of the automobile body window frame, and the plurality of characteristic points of the automobile body window frame correspond to the plurality of characteristic points of the automobile windshield one by one.

9. The automatic assembling method for automobile windshields according to claim 8, wherein the image processing module extracts corresponding plural sets of feature point pairs from the assembling image and calculates the assembling gap between the automobile windshield and the automobile body window frame based on the plural sets of feature point pairs, including:

according to the optical trigonometry principle, obtaining the spatial position coordinates of each point on the characteristic laser straight line from the first image and the second image based on the camera imaging principle, obtaining the spatial position coordinates of a plurality of characteristic points of the automobile windshield and the spatial position coordinates of a plurality of characteristic points of the automobile body window frame, and forming a plurality of groups of characteristic point pairs;

and dividing the distance between two characteristic points of each group of characteristic point pairs into an assembly horizontal gap l and an assembly vertical height difference h to obtain a plurality of groups of assembly horizontal gaps l and assembly vertical height differences h.

10. The automatic assembling method for automobile windshields according to claim 9, wherein the robot control module plans a robot assembling route according to the assembling gap between the automobile windshields and the automobile body window frame, and controls the robot to grab the automobile windshields for assembling according to the planned assembling route, further comprising:

according to the obtained multiple groups of assembly vertical height differences h, adjusting the track of the robot to enable the four groups of assembly vertical height differences h to be within the same numerical error range;

acquiring the adjusted assembly horizontal gaps l, adjusting the track of the robot to enable the multiple groups of assembly horizontal gaps l to meet the error range allowed by the installation standard, and if 1 or 2 of the multiple groups of assembly horizontal gaps l cannot enable the multiple groups of assembly horizontal gaps l to meet the error range allowed by the installation standard through adjusting the track of the robot, reporting errors by the system and stopping assembly;

and if the plurality of groups of assembly horizontal gaps enable the plurality of groups of assembly horizontal gaps l to be in the error range allowed by the installation standard by adjusting the track of the robot, descending for a certain distance along the direction of the assembly vertical height difference h under the control of the robot, and repeating the assembly gap detection, the error judgment and the descending along the assembly vertical height until the assembly vertical height difference h is zero to complete the assembly.

Technical Field

The invention relates to the field of automatic assembly of automobiles, in particular to an automatic assembly system and an automatic assembly method for an automobile windshield.

Background

For the field of automatic assembly of automobile windshields, most of the existing automobile windshields are basically subjected to automatic gluing in a gluing stage, and in the traditional windshield installation, most of glass feeding and installation depend on manual semi-automatic installation. Due to the participation of manpower, the installation consistency is poor, and the efficiency is low. Only a few newly built finished automobile production lines realize full process automation including automobile windshield installation.

For the existing automatic assembly of the windshield by means of machine vision, the automobile body window frame is mainly positioned by means of the machine vision and compared with a set window frame template, and the installation track of a robot is planned according to the method.

The problem that the existing technology relying on machine vision to assemble the automobile windshield at home is that:

(1) in the manufacturing process of the automobile body, due to manufacturing errors, the consistency of the window frames of the automobile body is poor, the route planned by the robot controller cannot be universal, and the adjustment needs to be carried out according to the manufacturing errors.

(2) In the manufacturing process of the automobile windshield, the consistency of the automobile windshield is poor due to manufacturing errors, and the assembly precision is low due to the fact that only the manufacturing errors of a vehicle body are generally considered and the manufacturing errors of the windshield are ignored in the traditional technology of assembling the automobile windshield by relying on machine vision.

(3) The existing assembly technology plans the assembly track of a robot by calculating the space position coordinate system of a vehicle body and a vehicle window frame, lacks an assembly detection device and cannot carry out closed-loop adjustment.

Disclosure of Invention

The invention provides an automatic assembling system and an automatic assembling method for an automobile windshield, aiming at the technical problems in the prior art.

According to a first aspect of the invention, an automatic assembly system for an automobile windshield is provided, comprising a 3D vision module, a robot control module, a windshield grabbing module and an image processing module;

the 3D vision module is used for grabbing an assembly image in the process that the windshield grabbing module grabs the automobile windshield and the automobile window frame of the automobile body;

the image processing module is used for extracting a plurality of corresponding characteristic point pairs according to the assembly image and calculating an assembly gap between the automobile windshield and the automobile window frame of the automobile body based on the plurality of characteristic point pairs;

and the robot control module is used for planning an assembly route of the robot according to the assembly clearance between the automobile windshield and the automobile window frame of the automobile body and controlling the robot to assemble the automobile windshield according to the planned assembly route.

On the basis of the technical scheme, the invention can be improved as follows.

Optionally, the number of the 3D vision modules is 4, each of the 3D vision modules includes a structured light and a binocular vision system, and the 4 3D vision modules are all mounted on a mechanical arm of the robot; and 4, the 3D vision modules are used for acquiring a plurality of groups of characteristic point pairs in the installation process of the automobile windshield.

Optionally, the structured light is a one-dimensional laser position finder, 4 the 3D vision module is used for obtaining a plurality of groups of feature point pairs in the installation process of the automobile windshield, including:

when the windshield grabbing module grabs the automobile windshield, acquiring a first image containing a plurality of boundary points on the automobile windshield, wherein the boundary points on the automobile windshield are a plurality of characteristic points of the windshield;

when the windshield grabbing module grabs the automobile windshield to move above the automobile body window frame according to the planned assembly line, acquiring a second image containing a plurality of boundary points of the automobile body window frame, wherein the plurality of boundary points of the automobile body window frame are a plurality of characteristic points of the automobile body window frame;

correspondingly, the image processing module is configured to:

extracting a plurality of characteristic points of an automobile windshield from the first image and extracting a plurality of characteristic points of a vehicle body window frame from the second image respectively, wherein the plurality of characteristic points of the automobile windshield and the plurality of characteristic points of the vehicle body window frame form a plurality of groups of characteristic point pairs;

and calculating the assembling clearance between the automobile windshield and the automobile body window frame based on the plurality of characteristic points of the automobile windshield and the corresponding plurality of characteristic points of the automobile body window frame.

Optionally, the windshield grabbing module comprises four glass suckers, and the four glass suckers are arranged at four corners of the windshield grabbing module and used for grabbing four corners of the automobile windshield.

Optionally, the robot is a six-degree-of-freedom robot, and includes two parts, the first part is an arm with 3 control axes, and the spatial position of the reference point at the tail end of the robot is determined by the motion of the 3 control axes; the second part is a wrist with 3 control axes, and the posture of the robot end tool is determined by the movement of the 3 control axes.

Optionally, the device further comprises a windshield centering and positioning module, which comprises a transverse centering and clamping mechanism and a longitudinal centering and clamping mechanism; the transverse centering and clamping mechanism clamps the automobile windshield through the clamping rollers on the two sides to finish transverse centering and clamping, and the longitudinal centering and clamping mechanism clamps the automobile windshield through the clamping rollers to finish longitudinal centering and clamping so as to finish centering and positioning of the automobile windshield.

According to a second aspect of the present invention, there is provided an automatic assembling method for an automobile windshield, comprising:

the vehicle body lifting appliance suspends a vehicle body into an automatic assembly area, the lifting appliance is statically and mechanically positioned after the vehicle body lifting appliance enters a preset position, and a vehicle body window frame reaches the preset position to finish positioning;

the automobile windshield is arranged on the centering workbench, and the centering and positioning of the automobile windshield are completed through the windshield centering and positioning module;

the windshield grabbing module grabs the automobile windshield after centering and positioning and conveys the grabbed automobile windshield into the shooting range of the 3D vision module;

shooting a preset area in the assembly process by the plurality of 3D vision modules, and capturing an assembly image;

the image processing module extracts a plurality of corresponding characteristic point pairs according to the assembly image and calculates an assembly gap between the automobile windshield and the automobile window frame of the automobile body based on the plurality of characteristic point pairs;

the robot control module plans an assembly route of the robot according to the assembly clearance between the automobile windshield and the automobile window frame of the automobile body, and controls the robot to grab the automobile windshield for assembly according to the planned assembly route.

Optionally, the 3D vision module includes a structured light and a binocular vision system, and the structured light is a one-dimensional laser position finder; the plurality of 3D vision modules shoot a preset area in the assembly process, capture an assembly image, and comprise:

when the windshield grabbing module grabs the automobile windshield, the one-dimensional laser position finder projects a linear light plane on the automobile windshield to form a characteristic laser straight line, the characteristic laser straight line is imaged on a camera image plane, a first image containing a plurality of boundary points on the automobile windshield is obtained, and the boundary points on the automobile windshield are a plurality of characteristic points of the automobile windshield;

when the windshield grabbing module grabs the automobile windshield and moves above the automobile body window frame according to the planned assembly line, the one-dimensional laser position finder projects a linear light plane on the automobile windshield and the automobile body window frame respectively, a characteristic laser straight line is formed on the automobile windshield and the automobile body window frame respectively and is imaged on a camera imaging plane, a second image containing a plurality of boundary points of the automobile body window frame except the boundary points on the automobile windshield is obtained, the plurality of boundary points of the automobile body window frame are a plurality of characteristic points of the automobile body window frame, and the plurality of characteristic points of the automobile body window frame correspond to the plurality of characteristic points of the automobile windshield one by one.

Optionally, the image processing module extracts corresponding multiple groups of feature point pairs according to the assembly image, and calculates an assembly gap between the windshield and the window frame of the vehicle body based on the multiple groups of feature point pairs, including:

according to the optical trigonometry principle, obtaining the spatial position coordinates of each point on the characteristic laser straight line from the first image and the second image based on the camera imaging principle, obtaining the spatial position coordinates of a plurality of characteristic points of the automobile windshield and the spatial position coordinates of a plurality of characteristic points of the automobile body window frame, and forming a plurality of groups of characteristic point pairs;

and dividing the distance between two characteristic points of each group of characteristic point pairs into an assembly horizontal gap l and an assembly vertical height difference h to obtain a plurality of groups of assembly horizontal gaps l and assembly vertical height differences h.

Optionally, the robot control module is according to the fitting gap between car windshield and the automobile body car window frame plans the robot assembly route to control the robot and snatch car windshield according to the assembly route of planning and assemble, still include:

according to the obtained multiple groups of assembly vertical height differences h, adjusting the track of the robot to enable the four groups of assembly vertical height differences h to be within the same numerical error range;

acquiring the adjusted assembly horizontal gaps l, adjusting the track of the robot to enable the multiple groups of assembly horizontal gaps l to meet the error range allowed by the installation standard, and if 1 or 2 of the multiple groups of assembly horizontal gaps l cannot enable the multiple groups of assembly horizontal gaps l to meet the error range allowed by the installation standard through adjusting the track of the robot, reporting errors by the system and stopping assembly;

and if the plurality of groups of assembly horizontal gaps enable the plurality of groups of assembly horizontal gaps l to be in the error range allowed by the installation standard by adjusting the track of the robot, descending for a certain distance along the direction of the assembly vertical height difference h under the control of the robot, and repeating the assembly gap detection, the error judgment and the descending along the assembly vertical height until the assembly vertical height difference h is zero to complete the assembly.

According to the automatic assembling system and the automatic assembling method for the automobile windshield, provided by the invention, the assembling clearance between the windshield and the automobile body window frame in the installation process of the automobile windshield can be detected, and the movement track of the robot is adjusted according to the measured assembling clearance, so that the assembling efficiency is improved, the assembling precision is improved, and the manual labor intensity is reduced.

Drawings

FIG. 1 is a schematic structural view of an automatic assembling system for a windshield of an automobile according to the present invention;

FIG. 2 is a schematic view of a characteristic point of an automobile windshield acquired;

FIG. 3 is a schematic view of a plurality of sets of pairs of feature points acquired for an automotive windshield;

FIG. 4 is a schematic view of a characteristic point of a vehicle body window frame obtained;

FIG. 5 is a schematic diagram of a plurality of sets of feature point pairs of a vehicle window frame of a vehicle body;

FIG. 6 is a schematic view of the calculated assembly horizontal clearance l and assembly vertical height difference h from a set of pairs of characteristic points;

fig. 7 is a flowchart of the process of assembling the windshield of the automobile.

In the drawings, the names of the components represented by the respective reference numerals are as follows:

1. one-dimensional laser position finder, 2, binocular vision system.

Detailed Description

The following detailed description of embodiments of the present invention is provided in connection with the accompanying drawings and examples. The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.

Fig. 1 is an automatic assembly system for an automobile windshield provided by the invention, and the automatic assembly system mainly comprises a 3D vision module, a robot control module, a windshield grabbing module and an image processing module.

The 3D vision module is used for grabbing an assembly image in the process that the windshield grabbing module grabs the automobile windshield and the automobile window frame of the automobile body; the image processing module is used for extracting a plurality of corresponding characteristic point pairs according to the assembly image and calculating an assembly gap between the automobile windshield and the automobile window frame of the automobile body based on the plurality of characteristic point pairs; and the robot control module is used for planning an assembly route of the robot according to the assembly clearance between the automobile windshield and the automobile window frame of the automobile body and controlling the robot to assemble the automobile windshield according to the planned assembly route.

It can be understood that, based on the defects in the background art, the embodiment of the invention provides an automatic assembling system for an automobile windshield, which can realize the full process automation including the installation of the automobile windshield.

The automatic assembling system for the automobile windshield mainly comprises a 3D vision module, a robot control module, a windshield grabbing module and an image processing module. The 3D vision module is installed on the robot, collects an assembly image between a windshield and a vehicle body and vehicle window frame grabbed by the robot, and transmits the collected image to the image processing module. The image processing module receives the assembly image transmitted by the 3D vision module, calculates the spatial position relation between the windshield and the vehicle window frame of the vehicle body, calculates a spatial position coordinate system of the windshield, converts the information of the sensor coordinate system into a robot coordinate system by means of hand-eye calibration, and calculates the assembly clearance of the robot. The robot control module controls the robot, the windshield grabbing module is used for grabbing the windshield to be installed, the robot assembly route is planned according to the assembly clearance obtained by the image processing module, and the automobile windshield is installed.

According to the invention, a 3D vision module consisting of a structured light and a binocular vision system 2 is utilized, in the process of assembling an automobile windshield and an automobile body window frame grabbed by a windshield grabbing module, an assembly image is shot, a plurality of groups of characteristic point pairs are obtained, the assembly gap between the automobile windshield and the automobile body window frame is detected through the characteristic points, and the assembly route of the robot is adjusted by the robot under a robot control module according to the assembly gap obtained by a vision detection system, so that the assembly requirement is finally met. The invention can realize the automatic assembly of the automobile windshield without human intervention and can improve the automatic assembly precision of the automobile windshield.

In a possible embodiment mode, the number of the 3D vision modules is 4, each 3D vision module comprises a structured light and a binocular vision system 2, the 4 3D vision modules are all installed on a mechanical arm of the robot, and the 4 3D vision modules are used for acquiring multiple groups of feature point pairs in the installation process of an automobile windshield.

It can be understood that the 3D vision module is composed of structured light and a binocular vision system 2, the whole vision system is composed of 4 sets of 3D vision modules, and the vision system is installed on the mechanical arm to photograph 4 regions and 8 key points in the installation process of the windshield. Wherein the 3D vision module needs to be calibrated and merged into a common coordinate system.

The structured light in the 3D vision module is the one-dimensional laser position finder 1, specifically, when 4 3D vision modules are used to obtain an image (called as a first image) containing a plurality of characteristic points of the windshield of the automobile and an image (called as a second image) containing a plurality of characteristic points of the window frame of the automobile body, when the windshield grabbing module grabs the automobile windshield, the one-dimensional laser position finder 1 projects a straight light plane on the windshield to form a characteristic laser straight line, and imaging on the camera image plane, and obtaining a plurality of boundary points on the windshield as a plurality of feature points of the automobile windshield, as shown in fig. 2, four feature points obtained by the four sets of 3D vision modules are as shown in fig. 3, the 3D vision module is relatively stationary on the windshield, and a plurality of feature points on the windshield are always located at the same position in the obtained imaging image.

When the windshield grabbing module grabs the automobile windshield to move above the automobile body window frame according to a preset track, the one-dimensional laser position finder 1 projects a linear light plane on the windshield and the automobile body window frame, at the moment, a characteristic laser straight line is formed on the windshield and the automobile body window frame respectively and is imaged on a camera imaging plane, automobile body window frame boundary points except the boundary points on the windshield are obtained, the automobile body window frame boundary points are used as characteristic points, and as shown in fig. 4, a plurality of characteristic points of the automobile body window frame obtained by the four groups of 3D vision modules are shown in fig. 5.

The method comprises the steps that a first image and a second image are obtained through 4 3D vision modules, a plurality of feature points of an automobile windshield are extracted from the first image and a plurality of feature points of an automobile body window frame are extracted from the second image respectively through an image processing module, and the plurality of feature points of the automobile windshield and the plurality of feature points of the automobile body window frame form a plurality of groups of feature point pairs. The fitting gap between the automobile windshield and the vehicle body window frame is calculated based on a plurality of characteristic points of the automobile windshield and a plurality of corresponding characteristic points of the vehicle body window frame, as shown in fig. 6.

The windshield grabbing module comprises four glass suckers, and the four glass suckers are arranged at 4 corners of the windshield grabbing module and used for grabbing four corners of the automobile windshield.

The robot is a six-degree-of-freedom robot and comprises two parts, wherein the first part is an arm with 3 control axes, and the spatial position of a reference point at the tail end of the robot is determined by the motion of the 3 control axes; the second part is a wrist with 3 control axes, and the posture of the robot end tool is determined by the movement of the 3 control axes.

In one possible embodiment, the automatic assembly system for the automobile windshield further comprises a windshield centering and positioning module, which comprises a transverse centering and clamping mechanism and a longitudinal centering and clamping mechanism. The transverse centering and clamping mechanism clamps the windshield through the clamping rollers on the two sides to finish transverse centering and clamping, and the longitudinal centering and clamping mechanism clamps the windshield through the clamping rollers to finish longitudinal centering and clamping so as to finish centering and positioning of the automobile windshield and guarantee position consistency when the six-freedom-degree robot grabs the automobile windshield by means of the grabbing device.

The image processing module receives the assembly image transmitted from the 3D vision module, detects the gap and the height difference of the assembly horizontal plane of the windshield, and adjusts the assembly route of the preset robot so as to improve the assembly precision of the automobile windshield.

Referring to fig. 7, the assembling process of the automatic automobile windshield assembling system specifically includes the following steps:

step A, conveying a car body window frame to a preset automation area by a car body lifting appliance, and mechanically positioning the lifting appliance in a static state;

step B, the windshield is positioned in the windshield centering and positioning module;

c, the robot grabs the windshield in the glass centering and positioning module by means of the windshield grabbing module;

step D, the robot grabs the windshield according to a preset track and reaches the shooting range of the 3D vision module above the car window frame of the car body;

step E, the 3D vision module shoots 4 areas at the assembly edge to obtain 8 characteristic points, as shown in figure 3;

step F, the image processing module receives the assembly image transmitted from the 3D vision module, and detects the assembly gap and the height difference of the automobile windshield according to the obtained 4 area images and 8 characteristic points;

step G, the assembly path planning of the robot is modified according to the detected assembly gap and height difference of the automobile windshield, and the step E, F is repeatedly executed after the automobile windshield descends a certain height under the movement of the robot;

step H, when the vehicle body window frame and the windshield are assembled at a height difference of 20mm according to the height difference and the modified path, the step E, F is executed again;

step I, recalculating the assembling clearance, if the assembling clearance is larger than the assembling technical requirement and cannot be adjusted to meet the installation requirement, executing a step J, and if the assembling clearance of the windshield is detected to meet the assembling technical requirement, executing a step K;

step J, prompting by system alarm information that the assembly process cannot be completed due to overlarge manufacturing error of a car window frame or windshield of the car body;

and step K, assembling according to the modified assembly route to finish assembly.

Wherein, when the working procedure E is executed, the method comprises the following steps:

completing camera calibration and structured light plane calibration;

and secondly, calculating the relative positions of the 4 3D vision modules, and converting the three-dimensional measurement value of each sensing system into coordinate values in a common coordinate system.

And step three, completing hand-eye calibration, realizing conversion between a robot coordinate system and a camera coordinate system, and operating the robot by using the camera coordinate system.

And step four, 4 images are obtained by 4 cameras at preset positions, each image has two separated laser lines projected on a car body window frame and a car windshield as shown in fig. 4, boundary points on the car body window frame and the car windshield are selected as characteristic points, 8 characteristic points are obtained as shown in fig. 5, and the 8 characteristic points are used for determining the assembling clearance and the height difference between the car windshield and the car body window frame.

The invention also provides an automatic assembling method of the automobile windshield, which mainly comprises the following steps:

firstly, a car body is suspended by a car body hanger to enter an automatic assembly area, the hanger is statically positioned mechanically after entering a preset position, and a car body window frame reaches the preset position to finish positioning.

The automobile windshield is arranged on the centering workbench to complete centering and positioning of the automobile windshield. The transverse centering and clamping mechanism clamps the windshield through clamping rollers on two sides to finish transverse centering and clamping; the longitudinal centering and clamping mechanism clamps the windshield through the clamping roller to finish longitudinal centering and clamping.

The windshield grabbing module grabs the automobile windshield which is centered and positioned through the glass sucker. The windshield grabbing module comprises 4 glass suckers, and the 4 glass suckers are arranged at 4 corners of the windshield grabbing module and correspond to 4 corners of the automobile windshield. Each sucker is controlled by a separate controller to prevent accidental falling due to faults.

The robot is controlled by the robot control module, the windshield grabbing module is used for grabbing the centered and positioned windshield, and the windshield is conveyed to the shooting range of the 3D vision module according to a preset track.

4 3D vision modules are installed at the tail end of the robot and used for shooting a preset area. The 3D vision module needs to complete the calculation of partial parameters of the camera module in advance, including parameters of camera calibration, hand-eye calibration and binocular vision. It is first necessary to know the relative positional relationship between the 4 3D vision modules for converting the three-dimensional measurements obtained by the 3D vision system into a common coordinate system. The camera in one of the 3D vision modules needs to be selected as a camera reference coordinate system, and the camera in each 3D vision module is calibrated by using the same world coordinate system. Then, the relative position coordinates of each camera in the same world coordinate system are calculated, and then the coordinates of each camera in the same world coordinate system are converted into the coordinates in the reference coordinate system of the selected camera. And finally, converting the parameters obtained by hand-eye calibration into coordinates in a robot coordinate system.

After the windshield is conveyed to the shooting range of the 3D vision module according to a preset track, the 3D vision module shoots a preset area to obtain 4 images. The specific shooting process is that, the 3D vision module includes structured light and binocular vision system, the structured light is one-dimensional laser position finder,

when the windshield grabbing module grabs the automobile windshield, the one-dimensional laser position finder projects a linear light plane on the automobile windshield to form a characteristic laser straight line, the characteristic laser straight line is imaged on a camera image plane, a first image containing a plurality of boundary points on the automobile windshield is obtained, and the boundary points on the automobile windshield are a plurality of characteristic points of the automobile windshield;

when the windshield grabbing module grabs the automobile windshield and moves above the automobile body window frame according to the planned assembly line, the one-dimensional laser position finder projects a linear light plane on the automobile windshield and the automobile body window frame respectively, a characteristic laser straight line is formed on the automobile windshield and the automobile body window frame respectively and is imaged on a camera imaging plane, a second image containing a plurality of boundary points of the automobile body window frame except the boundary points on the automobile windshield is obtained, the plurality of boundary points of the automobile body window frame are a plurality of characteristic points of the automobile body window frame, and the plurality of characteristic points of the automobile body window frame correspond to the plurality of characteristic points of the automobile windshield one by one.

The image processing module obtains the spatial position coordinates of each point on the characteristic laser straight line from the first image and the second image based on the camera imaging principle according to the optical trigonometry principle, obtains the spatial position coordinates of a plurality of characteristic points of the automobile windshield and the spatial position coordinates of a plurality of characteristic points of the automobile body window frame, and forms a plurality of groups of characteristic point pairs; and dividing the distance between two characteristic points in each group of characteristic point pairs into an assembly horizontal gap l and an assembly vertical height difference h to obtain a plurality of groups of assembly horizontal gaps l and assembly vertical height differences h.

The image processing module calculates and obtains an assembly gap and a height difference through 4 images and 8 characteristic points, a new assembly path is planned, and the robot control module controls the robot to move according to the planned path. This step is to reduce the height, improve the detection accuracy, and further improve the assembly accuracy.

After a plurality of groups of assembling horizontal gaps l and vertical height differences h are calculated, the process of assembling the automobile windshield is that the tracks of the robot are adjusted according to the obtained plurality of groups of assembling vertical height differences h so that the four groups of assembling vertical height differences h are within the same numerical error range; acquiring the adjusted assembly horizontal gaps l, adjusting the track of the robot to enable the multiple groups of assembly horizontal gaps l to meet the error range allowed by the installation standard, and if 1 or 2 of the multiple groups of assembly horizontal gaps l cannot enable the multiple groups of assembly horizontal gaps l to meet the error range allowed by the installation standard through adjusting the track of the robot, reporting errors by the system and stopping assembly; and if the plurality of groups of assembly horizontal gaps enable the plurality of groups of assembly horizontal gaps l to be in the error range allowed by the installation standard by adjusting the track of the robot, descending for a certain distance along the direction of the assembly vertical height difference h under the control of the robot, and repeating the assembly gap detection, the error judgment and the descending along the assembly vertical height until the assembly vertical height difference h is zero to complete the assembly.

In summary, the automatic assembling method for the automobile windshield provided by the embodiment of the invention mainly comprises the following steps:

step one, when the windshield grabbing module grabs the automobile windshield, the one-dimensional laser position finder projects a straight light plane on the windshield to form a characteristic laser straight line, images the characteristic laser straight line on a camera image plane, obtains a plurality of boundary points on the windshield, and uses the boundary points as a plurality of characteristic points of the automobile windshield, as shown in fig. 2, the characteristic points obtained by four groups of 3D vision modules are as shown in fig. 3, the 3D vision modules are relatively static on the windshield, and the plurality of characteristic points on the windshield are always located at the same position in the obtained imaging image.

And step two, when the windshield grabbing module grabs the automobile windshield to move above the automobile body window frame according to a preset track, the one-dimensional laser position finder projects a linear light plane on the windshield and the automobile body window frame, at the moment, a characteristic laser straight line is formed on the windshield and the automobile body window frame respectively and is imaged on a camera imaging plane, a plurality of boundary points of the automobile body window frame except the boundary points on the windshield are obtained, the plurality of boundary points of the automobile body window frame are used as a plurality of characteristic points of the automobile body window frame, and as shown in fig. 4, a plurality of characteristic points of the automobile body window frame obtained by the four groups of 3D vision modules are shown in fig. 5.

And step three, obtaining the spatial position coordinates of each point on the characteristic laser straight line according to the optical trigonometry principle and the camera imaging, and obtaining the spatial position coordinates of all the characteristic points. The distance between two feature points in each set of feature point pairs is divided into an assembly horizontal gap l and an assembly vertical height difference h as shown in fig. 6, and four sets of assembly horizontal gaps l and assembly vertical height differences h are obtained.

And step four, according to the four groups of assembly vertical height differences h, adjusting the track of the robot to enable the four groups of assembly vertical height differences h to be within the same numerical value error range, and ensuring the vertical installation precision of the windshield.

And step five, repeating the step three to obtain the assembly horizontal clearance l after being adjusted in the step four, and adjusting the track of the robot to enable the four groups of assembly horizontal clearances l to be in the error range allowed by the installation standard. If 1 or 2 of the four groups of assembly horizontal gaps l cannot enable the four groups of assembly horizontal gaps l to be in the error range allowed by the installation standard by adjusting the track of the robot, executing a sixth procedure; and if the four groups of assembly horizontal gaps can be adjusted through adjusting the track of the robot so that the four groups of assembly horizontal gaps l meet the error range allowed by the installation standard, executing a seventh step.

And step six, the manufacturing error of the car body window frame or the car windshield is too large, so that normal assembly cannot be carried out, a system reports errors, and assembly is stopped.

And step seven, vertically descending the windshield for a certain distance along the direction of the vertical height difference h of the assembly under the control of the robot, approaching the car window frame of the car body to be installed, and repeating the step three, the step four and the step five until the vertical height difference h of the assembly is zero to finish the assembly.

According to the automatic assembling system and the automatic assembling method for the automobile windshield, provided by the embodiment of the invention, in the assembling process of the automobile windshield and the automobile body window frame, the 3D vision system is used for shooting the assembling image to obtain a plurality of groups of characteristic point pairs, the assembling gap between the window and the window frame is detected through the characteristic points, and the robot adjusts the assembling route according to the assembling gap obtained by the vision detection system under the robot control module to finally meet the assembling requirement. The invention can realize the automatic assembly of the automobile windshield without human intervention and can improve the automatic assembly precision of the automobile windshield.

It should be noted that, in the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and reference may be made to relevant descriptions of other embodiments for parts that are not described in detail in a certain embodiment.

As will be appreciated by one skilled in the art, embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.

The present invention is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.

It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.