KR102738516B1 - System and method for teaching sealing robots - Google Patents

Abstract

A sealing robot teaching system and method thereof are disclosed.
A sealing robot teaching system for generating robot passes of a plurality of sealing robots arranged in a painting process line according to an embodiment of the present invention includes: a sealing robot having a sealer and a 3D scanner mounted on an end effector via a bracket; a robot controller controlling a 3D scan operation and a sealing operation of a loaded body through kinematic posture control of the sealing robot; and a server correcting the robot pass to match a surface shape caused by at least one of an alignment error and an assembly error of the body identified by 3D matching a master sample in which a robot pass for the sealing operation is defined to 3D scan data of the body measured by the 3D scanner, and then transmitting the corrected robot pass to the robot controller.

Description

{SYSTEM AND METHOD FOR TEACHING SEALING ROBOTS}

The present invention relates to a sealing robot teaching system and method thereof, and more specifically, to a sealing robot teaching system and method thereof that automates the teaching of a sealing robot for applying sealant placed in a painting process of a car body by utilizing a 3D scanner.

In general, in factories that produce automobiles, a sealing process is essential in which sealing material is applied using a sealing robot to the gaps that occur in the overlapping and connecting parts of the steel plates of the assembled automobile body.

As a representative example, in the body painting process, a number of sealing robots are deployed and the lower part of the body is sealed with sealant to enhance durability, prevent rust, and block noise.

Since the lower sealing work requires identifying the exact location of the gap in the body and performing the sealing work using a robot, a vision system for aligning the body's exact location and a teaching work for setting the sealer application path for each robot are applied.

For example, in the case of a fixed camera vision system used in a conventional painting process, the alignment status of the entire body is corrected by measuring the positions/sizes of several body holes. However, errors occur due to accumulated manufacturing variations as the body goes through several preceding processes, but the conventional fixed camera vision system has a problem in that it is impossible to correct the sealer application path according to the assembly errors of the body.

In addition, in the painting process, the application of OLP (Off-Line Programming) technology used in welding parts at conventional body shops is impossible due to errors in body position and accumulated assembly errors due to manufacturing dispersion, so teaching work for the sealer robot is done manually.

That is, in the case of sealer robots in paint shops, manual teaching is done one by one by specialized personnel without going through a separate OLP simulation, which causes excessive costs for teaching robots for new car models and maintenance costs.

In addition, in the painting process, there are frequent cases where robot teaching needs to be modified due to various reasons such as body assembly errors and continuous quality assurance. However, in the current factory, workers check the parts to be modified with their own eyes and manually modify the robot path using a teaching pendant, which has the disadvantage of causing complex problems such as increased workload/time, occurrence of human errors, and decreased yield due to delays in line waiting work.

The information contained in this background section has been prepared to promote understanding of the background of the invention and may include matters that are not prior art and are already known to those skilled in the art.

An embodiment of the present invention provides a sealing robot teaching system and method that automatically corrects the pass of a sealing robot to fit the position and shape of a body by matching design data of a body in which a pass of a sealing robot is defined to 3D scan data of a surface of a body loaded into a painting process line.

According to one aspect of the present invention, a sealing robot teaching system for generating robot passes of a plurality of sealing robots arranged in a painting process line includes: a sealing robot having a sealer and a 3D scanner mounted on an end effector through a bracket; a robot controller for controlling a three-dimensional scan operation and a sealing operation of a loaded body through kinematic posture control of the sealing robot; and a server for correcting the robot pass according to a surface shape caused by at least one of an alignment error and an assembly error of the body identified by three-dimensionally matching a master sample in which a robot pass for the sealing operation is defined to 3D scan data of the body measured by the 3D scanner, and then transmitting the result to the robot controller.

The above master sample may be composed of CAD data including at least one robot pass defined for the sealing operation and feature points of the body for the three-dimensional matching on the 3D shape of the sample body before mass production.

Additionally, the 3D scanner can be calibrated to generate 3D scan data in the same coordinate system as the coordinate system of the sealing robot.

Additionally, the robot controller can detect the relative coordinates of the sealer and the 3D scanner based on the coordinates of the end effector according to the posture information of the sealing robot.

In addition, the server may include an external interface unit including one or more communication modules for communicating with the 3D scanner and robot controller for each of the sealing robots; a scan processing unit for filtering out noise and background parts through preprocessing of the 3D scan data and generating a 3D body shape for generating the robot path; a robot teaching unit for displaying the 3D body shape on a simulation tool and generating a robot path according to line information input through a computing input device; a database (DB) for storing various programs and data for setting, modifying, and managing the sealing work of the robot path for each of the sealing robots; a display unit for displaying a screen generated through the simulation tool to an operator; and a control unit for correcting the coordinates of the robot path by attaching the 3D shape of the robot path defined in the master sample to the surface coordinates of the current 3D body shape after the 3D matching.

In addition, the scan processing unit can divide the entire image of the vehicle body into parts or robot pass units during the preprocessing process of the 3D scan data to create multiple independent individual 3D shapes.

In addition, the robot teaching unit can display a 2D plan view of the 3D body shape through the simulation tool, and define the robot path by applying a pass line input to the overlapping portion of the body to the surface of the 3D body shape.

In addition, the robot teaching unit can input the pass line using any one of an input device, a mouse, a pen, and a digitizer, at the overlapping portion of the first body and the second body.

Additionally, the pass line can be input in a point-to-point manner, connecting points one after another from a start point to an end point, or in a drawing manner, in which the pass line is input continuously.

In addition, the robot teaching unit can align a pre-fabricated body sample to a correct position and then input pass lines to various parts of the body shape of a 3D point cloud type acquired by scanning with the 3D scanner to create a master sample in which the robot path is defined.

In addition, the control unit can three-dimensionally match feature points defined in the 3D shape of the master sample to feature points corresponding to the current 3D body shape by using at least one of the PPF (Point Pair Features) and ICP (Iterative Closest Point) algorithm techniques.

In addition, the control unit can generate 6-axis coordinates converted into a robot coordinate system of the sealing robot using the robot path corrected to fit the 3D body shape and transmit the same to the robot controller through the external interface unit.

In addition, the control unit can detect an assembly error of a second body overlapping the first body when the 3D body shape and the 3D shape of the master sample do not match, and can delete a pass line where the assembly error occurred using the simulation tool and modify the pass line based on the feature points of the overlapping second body.

In addition, if the control unit is repeated a predetermined number of times or more than a standard number of times as to a non-matching judgment of the same pattern during a predetermined cycle or the three-dimensional matching, it can compare the 3D coordinates of the 3D scanner using a fixed calibration target with the robot coordinates of the sealing robot to check whether an error has occurred, and correct the 3D coordinates of the 3D scanner by the amount of error that has occurred.

Meanwhile, according to an aspect of the present invention, a sealing robot teaching method in which a server controlling a painting process line generates robot passes of a plurality of sealing robots includes the steps of: a) generating a master sample in which at least one robot pass defined for a sealing operation on a 3D shape of a sample body and feature points of each body surface component are defined; b) collecting, from a robot controller, 3D scan data of the body scanned by a 3D scanner mounted on a sealing robot when the body is loaded during mass production; c) filtering out noise and background portions as a preprocessing step of the 3D scan data and generating a 3D body shape for generating the robot pass; and d) correcting the robot pass to match a surface shape caused by at least one of an alignment error and an assembly error of the body detected by three-dimensionally matching the 3D body shape and the 3D shape of the master sample, and then transmitting the robot pass to the robot controller.

In addition, the step a) may include a step of displaying a 3D body shape of the sample body scanned through the 3D scanner in 2D on a simulation tool; a step of inputting a pass line using any one of an input device of a mouse, a pen, and a digitizer on an overlapping portion of the bodies; and a step of applying the input pass line to a surface of the 3D body shape of the sample body to define a 3D robot path.

Additionally, the step c) may include a step of dividing the entire image of the body into parts or robot pass units to generate a plurality of independent individual 3D shapes.

In addition, the step d) may include the step of attaching the three-dimensional shape of the robot pass defined in the master sample to the surface coordinates of the 3D body shape, and correcting the coordinates of the robot pass to match the alignment position of the current body.

In addition, the step d) may include a step d2) of detecting an assembly error of the body due to a mismatch between the 3D body shape and the 3D shape of the master sample during the 3D matching; and a step d3) of correcting a robot pass of a part where an assembly error has occurred through a simulation tool of a robot teaching unit.

In addition, the step d3) may include a step of deleting a previously defined 3D robot path in the assembly error occurrence area through the simulation tool and generating a modified 3D robot path based on feature points of the overlapping body area.

According to an embodiment of the present invention, the precision of a sealing operation can be improved by automatically correcting the coordinates of a robot pass defined in CAD data of a master sample to match the current position of the body, regardless of an alignment error in which the current position of the body acquired by a 3D scan does not match the reference alignment position.

In addition, by detecting assembly errors of the body through three-dimensional matching of the 3D shape of the body and the 3D shape of the master sample, and by supporting the operator to easily modify, delete, and add robot passes through a simulation tool, problems dependent on conventional teaching pendant manual work can be solved, and production yield and product quality can be improved.

In addition, by distributing various part-specific robot pass operations generated through body samples to a robot controller while considering the spacing between sealing robots and interference in posture control, it is possible to generate an optimized robot pass to prevent collisions between sealing robots.

Figure 1 schematically shows the configuration of the sealing robot teaching system of the present invention.
FIG. 2 is a block diagram schematically showing the configuration of a server according to an embodiment of the present invention.
FIG. 3 is a drawing for explaining an example of 3D scan data processing according to an embodiment of the present invention.
FIG. 4 is a drawing for explaining a process of generating a 3D robot path according to an embodiment of the present invention.
Figure 5 shows the results of a 3D robot path tracking test according to an embodiment of the present invention.
FIG. 6 is a drawing for explaining a process of modifying a 3D robot path according to an embodiment of the present invention.
Figure 7 illustrates a calibration target according to an embodiment of the present invention.
Figure 8 is a flow chart schematically illustrating a sealing robot teaching method according to an embodiment of the present invention.

Hereinafter, with reference to the attached drawings, embodiments of the present invention will be described in detail so that those skilled in the art can easily practice the present invention. However, the present invention may be implemented in various different forms and is not limited to the embodiments described herein. In addition, in order to clearly describe the present invention in the drawings, parts that are not related to the description are omitted, and similar parts are assigned similar drawing reference numerals throughout the specification.

Throughout the specification, when a part is said to "include" a certain component, this does not mean that it excludes other components, but rather that it may include other components, unless otherwise specifically stated. In addition, terms such as "part," "unit," and "module" described in the specification mean a unit that processes at least one function or operation, which may be implemented by hardware, software, or a combination of hardware and software.

Throughout the specification, terms such as first, second, A, B, etc. may be used to describe various components, but the components should not be limited by these terms. These terms are only used to distinguish one component from another, for example, without departing from the scope of the rights according to the concept of the present invention, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component.

Now, a sealing robot teaching system and method according to an embodiment of the present invention will be described in detail with reference to the drawings.

Figure 1 schematically shows the configuration of the sealing robot teaching system of the present invention.

Referring to FIG. 1, the sealing robot teaching system (100) includes a sealing robot (110), a 3D scanner (120), a robot controller (130), and a server (140).

The sealing robot (110) can be composed of a six-degree-of-freedom multi-joint manipulator, and a sealer (112) is mounted on the end effector (111).

For example, sealing robots (110) are arranged in multiples on both sides of the painting process line and perform a sealing operation of applying a sealant to the lower part of a car body transported by a conveyor jig. Here, the car body refers to the object when the sealing operation in the painting process line is assumed as an example, and has a general meaning of an object such as a unit decoration or part attached to the car body in a sealing operation in a process other than the painting process line.

The 3D scanner (120) is fixedly mounted to the end effector (111) of the sealing robot (110) via a bracket, similar to the sealer (112).

A 3D scanner (120) scans the surface of a vehicle body including a sealant application surface using at least one 3D structured light laser sensor to obtain 3D scan data.

The 3D scanner (120) has the same point of view as the sealer (112) mounted on the end effector (111) and moves according to the attitude control of the sealing robot (110), and is calibrated to generate 3D scan data with the same coordinates as the coordinate system of the sealing robot (110).

The robot controller (130) stores kinematic posture control information for the sealing work and 3D scanning work of the sealing robot (110), and controls the movement path of the robot through posture control according to each work mode. That is, the movement path control of the robot includes the movement path control of the sealer (112) for the sealing work and the movement path of the 3D scanner (120) for the 3D scanning work.

Here, the movement path for the sealing operation includes a loading/unloading path for returning the sealer (112) to the starting point (SP) of the taught robot pass or to the end point (EP) and a robot pass path for applying the sealant through the actual sealer (112). Accordingly, the robot controller (130) can control the sealer (112) to operate only in the section from the starting point (SP) to the end point (EP) defined by the robot pass in the coordinate system.

The robot controller (130) can detect the relative coordinates of the sealer (112) and the 3D scanner (120) based on the coordinates (x, y, z) of the end effector (111) according to the posture information currently being taken by the sealing robot (110).

The server (140) is a computing system that controls the sealing work of at least one sealing robot (110) placed in the painting process line, and automatically controls robot teaching for the sealing work area of each sealing robot (110).

The server (140) merges a master sample in which a robot path for the sealing operation of the sealing robot (110) is defined into the 3D scan data of the vehicle body collected through the 3D scanner (120) by three-dimensional matching, automatically corrects the robot path to match the surface shape due to the alignment error and assembly error of the vehicle body identified from the 3D scan data, and transmits the result to the robot controller (130).

The above master sample means CAD data including at least one 3D robot pass defined for sealing work on the 3D shape of a sample body before mass production and feature points of the body surface for the 3D matching.

Through this, the server (140) can control the sealing work of each sealing robot (110) adapted to various errors due to the alignment error of the body position occurring in the painting process line and the accumulation of dispersion in the production of the body.

For example, FIG. 2 is a block diagram schematically showing the configuration of a server according to an embodiment of the present invention.

Referring to FIG. 2, the server (140) includes an external interface unit (141), a scan processing unit (142), a robot teaching unit (143), a database (DB) (144), a display unit (145), and a control unit (146).

The external interface unit (141) includes one or more wired/wireless communication modules for communication with the 3D scanner (120) and robot controller (130) of each sealing robot (110).

The external interface unit (141) can receive 3D scan data of a body scanned by a 3D scanner (120) and transmit a robot pass processed by the control unit (146) to the robot controller (130).

Hereinafter, a method of collecting and processing 3D scan data and using the same to generate a robot path according to an embodiment of the present invention will be described with reference to FIGS. 3 and 4.

FIG. 3 is a drawing for explaining an example of 3D scan data processing according to an embodiment of the present invention.

First, referring to FIG. 3, 3D scan data acquired using a 3D structured light laser sensor of a 3D scanner (120) includes a surface image of a body sample including a sealant application surface and its 3D coordinates.

The scan processing unit (142) filters out noise and unnecessary background parts from the 3D scan data received from the external interface unit (141) through a preprocessing operation, and leaves only the 3D body shape for generating a robot pass on the sealant application surface. Here, the body is manufactured in a state where a second body (b2) is overlapped on a part of a first body (b1) having a folded surface, and a robot pass for the sealing operation must be set on the overlapped part (c).

The robot teaching unit (143) displays the 3D body shape preprocessed in the scan processing unit (142) on the operator's simulation tool, and receives line information displayed through a computing input device on the displayed body shape to generate a 3D robot path.

For example, FIG. 4 is a drawing for explaining a process of generating a 3D robot path according to an embodiment of the present invention.

Referring to Fig. 4, the robot teaching unit (143) displays the 3D body shape plan view in 2D through a simulation tool (S1). However, the 3D body shape is not limited to the plan view, and the operator may display a desired part, i.e., a part where a 3D robot path needs to be generated, through rotation of 2D images or 3D body shapes at various angles such as the front, left/right sides, bottom, and back.

The robot teaching unit (143) inputs a pass line (S2) using one of an input device among a mouse, a pen, and a digitizer at the part (c) where the first body (b1) and the second body (b2) overlap on the 2D plane diagram. The pass line can be input by a point-to-point method connecting points one after another from a start point (SP) to an end point (EP), or by a drawing method in which the pass line is input continuously.

The robot teaching unit (143) applies the input pass line to the surface of the 3D body shape to define a 3D robot path (S3).

The robot teaching unit (143) creates a master sample in which the 3D robot path defined on the surface of the 3D body shape and the corresponding feature points of each component on the body surface are defined and stores the master sample in the DB (144).

In the same manner as above, the robot teaching unit (143) aligns a pre-fabricated body sample in the correct position when applying a new car process in the painting process line, and then inputs pass lines to various parts of the body shape of a 3D point cloud type acquired by scanning with a 3D scanner (120) to create a master sample with a defined high-precision robot path.

DB (144) stores various programs and data for setting/modifying robot passes and managing sealing work according to automatic teaching of each sealing robot (110) of the server (140) according to an embodiment of the present invention, and stores data generated according to the operation of the server (140).

DB (144) can store a program for executing the functions of the scan processing unit (142) and robot teaching unit (143), and store data generated according to the execution.

DB (144) can store a master sample in which at least one robot pass required for the sealing work of each sealing robot (110) generated by the robot teaching unit (143) is defined and can be provided for the sealing work of each robot.

The display unit (145) displays data stored in the database (144) and data generated according to the operation of the server (140).

The display unit (145) can provide the operator with a visual pass line input environment by displaying a screen generated through the simulation tool of the robot teaching unit (143) to the operator.

The control unit (146) is a central processing unit that controls the overall operation of each part of the server (140) according to an embodiment of the present invention, and can implement the functions of each part by executing various programs stored in the DB (144).

The control unit (146) performs 3D matching of the 3D body shape obtained by analyzing 3D scan data received through the 3D scanner (120) during operation of the painting process line for actual mass production and the master sample stored in the DB (144). Here, the 3D matching can be processed using at least one of the PPF (Point Pair Features) and ICP (Iterative Closest Point) algorithm techniques, and as a result, the feature points corresponding to the current 3D body shape can be precisely matched based on the feature points defined in the 3D shape of the master sample.

At this time, the control unit (146) can precisely match the above two shapes and then attach and merge the three-dimensional shape of the robot path defined in the master sample to the surface coordinates of the current 3D body shape. This has the effect of allowing the robot path defined in the master sample to be applied by correcting the coordinates to match the current alignment position of the body, even if the current body position is in any state where it does not match the reference alignment position.

In addition, the control unit (146) can control a more precise sealing operation of the sealing robot (110) by an additional operation of attaching the position of the 3D robot pass to the surface of the current body in order to compensate for the manufacturing error of the current body shape based on the shape of the body sample used in the definition of the robot pass. In other words, this can be expressed as attaching the 3D robot pass defined in the master sample to the surface of the current 3D body shape.

The control unit (146) generates 6-axis coordinates (x, y, z, Rx, Ry, Rz) converted into the robot coordinate system of the sealing robot (110) using the 3D robot path corrected to the current 3D body shape and transmits them to the robot controller (130) through the external interface unit (141).

At this time, the robot controller (130) receives the coordinates of the 3D robot path corrected from the server (140), positions the sealer (112) of the sealing robot (110) at the starting point (SP) of the robot path through posture control, and then tracks the 3D robot path to control the sealing operation up to the end point (EP).

Meanwhile, FIG. 5 shows the results of a 3D robot path tracking test according to an embodiment of the present invention.

Referring to FIG. 5, the results of testing a sealing robot (110) and a robot controller (130) to verify the accuracy of a 3D robot path generated by a server (140) according to an embodiment of the present invention are shown.

The server (140) sequentially tested the tracking performance by transmitting the 3D robot path generated at each position to the robot controller (130) for a tracking test (A) when the body is aligned to the normal position, a tracking test (B) after the body is changed to the first position, and a tracking test (C) after the body is changed to the third position. At this time, the test conditions were that the gap between the focus of the sealer (112) and the sealing part of the body was maintained within 1 mm for the entire section, the normal vector between the sealer (112) and the body was maintained for the entire section, and the robot speed was 300 mm/s or more.

As a result of the test, it was confirmed that a 3D robot path precisely defined on the surface of a body according to an embodiment of the present invention satisfies the above conditions and performs robot tracking without collision as long as it is defined only within the FOV (Field of View) area of the 3D scanner (120) even when the location of the part is changed as in the above tests (B) and (C).

Meanwhile, as explained in the background technology above, in the conventional painting process line, if the robot pass is pushed or the actual sealant application area and the robot pass do not match due to the accumulation of errors in the alignment position or assembly of parts in the preceding process of assembling the body, various teaching tasks such as adding/deleting robot passes on site are required. On the other hand, most of the teaching tasks depend on manual work by specialists using a teaching pendant, resulting in problems such as cost and work delays.

Accordingly, the server (140) according to an embodiment of the present invention provides a function that can simply teach all tasks that can occur in a mass production plant within a margin of error of ±1 mm on the simulation tool program of the robot teaching unit (143) based on accurate 3D scanning and robot path generation.

For example, FIG. 6 is a drawing for explaining a process of modifying a 3D robot path according to an embodiment of the present invention.

Referring to Fig. 6, it is explained assuming that the 3D robot path defined on the body as described above in Fig. 4 is applied to an actual mass-produced painting process line.

The control unit (146) performs a three-dimensional matching between the current 3D body shape acquired through 3D scanning and the 3D shape of the master sample stored in the DB (144), and if the matching is normal, transmits the defined 3D robot path to the robot controller (130) (S4).

On the other hand, if the current 3D body shape and the 3D shape of the master sample stored in the DB (144) do not match, the control unit (146) detects an assembly error of the second body (b2) overlapping the first body (b1) and corrects the pass line based on the feature point (Sp) of the overlapping second body (b2) (S5).

At this time, modification of the pass line may include a process of deleting an existing 3D robot path and generating a modified 3D robot path by executing a simulation tool of the robot teaching unit (143).

The control unit (146) transmits the modified 3D robot path to the robot controller (130), so that the robot controller (130) controls the sealer (112) to perform the sealing operation with the modified 3D robot path (S6).

Through this, it is possible to detect not only alignment errors of the body but also assembly errors of the body, and to support the operator to easily modify, delete, and add robot passes through simulation tools, thereby resolving the problem of relying on conventional manual teaching pendant work and improving production yield and product quality.

In the above description, for easy understanding of the embodiments of the present invention, it is assumed that one body sample and one 3D robot pass are generated and explained concisely. However, in an actual painting process line, a robot pass for applying sealant to a body may be tens of meters long and the number of passes may reach hundreds, so sealing robots (110) arranged on both sides cooperate to perform the sealing work.

Therefore, since the entire image of the 3D scan data received from the 3D scanner (120) may include a number of parts and corresponding robot passes, the preprocessing work of the 3D scan data processed by the scan processing unit (142) may include a step of dividing the entire image of the vehicle body into parts/robot passes to generate a plurality of independent individual 3D shapes.

That is, various parts (decoration items) included in the entire car body image can be divided into a single car body part image as in Fig. 3, and then each image can be filtered to leave only the 3D car body shape.

This is because it is possible to segment the 3D body shape of various parts (upholstery) from the entire image of the current body acquired through 3D scanning and perform 3D matching with the 3D shape of the corresponding master sample, so that the robot pass for each part can be corrected according to the assembly error of each part.

Therefore, the embodiment of the present invention has a significant meaning in that it can not only correct the misaligned position of the entire vehicle body, but also detect assembly errors of each part of the entire vehicle body and correct the corresponding robot pass of each part.

In addition, the control unit (146) can generate an optimal robot path for preventing collisions between sealing robots (110) by distributing various part-specific robot passes generated through the initial body sample to the controller of each sealing robot (110) while considering the spacing between sealing robots (110) and interference in posture control.

Meanwhile, in the task of mounting a 3D scanner (120) on a sealing robot (110) and performing a scan, maintaining a constant 3D coordinate relationship between the sealing robot (110) and the mounted 3D scanner (120) is the key to determining the precision of the sealing task.

Even if a 3D scanner (120) is mounted on the end effector (111) of a sealing robot (110) with a rigid bracket, the mounting position may become misaligned due to the robot's repeated sealing operations, and if the errors resulting from this accumulate, it will have a negative impact on the reliability of the scan accuracy.

Accordingly, the control unit (146) checks whether the robot coordinates of the sealing robot (110) and the 3D coordinates of the 3D scanner (120) are aligned through the calibration target at regular intervals.

Figure 7 illustrates a calibration target according to an embodiment of the present invention.

Referring to FIG. 7, the control unit (146) compares the 3D coordinates of the 3D scanner (120) using the fixed calibration target (15) at regular intervals with the robot coordinates of the sealing robot (110), checks whether an error has occurred, and corrects the 3D coordinates of the 3D scanner (120) by the amount of error that has occurred.

At this time, a checkerboard-shaped calibration target (15) is placed in an area that can be captured by a 3D scanner (120) mounted on each sealing robot (110), and then the sealing robot (110) moves the 3D scanner (120) about 10 times to acquire multiple images from various angles.

Then, after recognizing the relative position of the calibration target (15) at each sealing robot (110) location, a transformation matrix for the image of the 3D scanner (120) according to the movement of the robot can be obtained to define the correlation between the 3D coordinates of the 3D scanner (120) and the robot coordinates.

In addition, the control unit (146) can check whether the 3D coordinates of the 3D scanner (120) are aligned through the calibration if the same pattern of non-matching is repeated more than a standard number of times during 3D matching of the 3D body shape and the master sample.

Meanwhile, based on the configuration of the aforementioned sealing robot teaching system (100), a sealing robot teaching method according to an embodiment of the present invention will be described through Fig. 8 below, with the server (140) as the subject.

Figure 8 is a flow chart schematically illustrating a sealing robot teaching method according to an embodiment of the present invention.

Referring to FIG. 8, a server (140) according to an embodiment of the present invention sets a master sample in which at least one 3D robot pass is defined for sealing work on a 3D shape of a sample body before mass production and corresponding body surface component-specific feature points are defined (S101). Here, the process of setting the master sample can be referred to the explanation through FIG. 3 and FIG. 4.

When a car body is loaded through a conveyor when the painting process line for actual mass production is operated, the server (140) collects 3D scan data of the car body scanned by a 3D scanner (120) mounted on a sealing robot (110) (S102).

The server (140) performs a preprocessing operation on the 3D scan data to filter out noise and unnecessary background parts, and obtains a 3D body shape for generating a robot pass on the sealant application surface (S103). At this time, the server (140) can divide the entire image of the 3D scan data into parts that make up the body or robot pass units to obtain 3D shapes for multiple parts.

The server (140) three-dimensionally matches the measured 3D body shape with the 3D shape of the master sample stored in the DB (144) (S104). At this time, the server (140) can precisely match the feature points defined in the 3D shape of the master sample to the corresponding feature points of the measured 3D body shape.

After the server (140) precisely aligns the two 3D shapes, it attaches the 3D shape of the robot pass defined in the master sample to the surface coordinates of the current 3D body shape, and automatically corrects the coordinates of the robot pass to match the alignment position of the current body (S105).

In addition, if the server (140) detects an assembly error of the body due to a mismatch between the current 3D body shape and the 3D shape of the master sample during the 3D matching (S106; example), the server (140) modifies the 3D robot path of the part where the assembly error occurred through the simulation tool of the robot teaching unit (143) (S109). At this time, the server (140) can delete the 3D robot path previously defined for the part where the assembly error occurred through the simulation tool, and generate a modified 3D robot path based on the feature points of the overlapping body (b2) part.

Thereafter, the server (140) can transmit the modified 3D path to the robot controller (130) to control the sealing operation according to the posture control of the sealing robot (110).

Meanwhile, in the step S106, if an assembly error of the body is not detected during the three-dimensional matching (S106; No), the server (140) can control the sealing operation by transmitting the coordinates of the robot pass corrected to match the current alignment position of the body to the robot controller (130).

After the sealing work of the above body is completed, the server (140) returns to the S102 step when the next body is loaded, and this is repeated until the painting process line stops operating.

In this way, according to the embodiment of the present invention, the coordinates of the robot pass defined in the CAD data of the master sample are automatically corrected to match the current body position regardless of the alignment error in which the current body position acquired by 3D scanning does not match the reference alignment position, thereby improving the precision of the sealing operation.

In addition, by detecting assembly errors of the body through three-dimensional matching of the 3D shape of the body and the 3D shape of the master sample, and by supporting the operator to easily modify, delete, and add robot passes through a simulation tool, it is possible to solve the problem of relying on the conventional manual teaching pendant work and improve the production yield and product quality.

In addition, by distributing various part-specific robot pass operations generated through the first body sample to the robot controller while considering the spacing between sealing robots and interference in posture control, it is possible to generate an optimized robot pass to prevent collisions between sealing robots.

Embodiments of the present invention are not implemented only through the devices and/or methods described above, and may also be implemented through a program for realizing a function corresponding to the configuration of an embodiment of the present invention, a recording medium on which the program is recorded, etc., and such implementation can be easily implemented by an expert in the technical field to which the present invention belongs, based on the description of the embodiments described above.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements made by those skilled in the art using the basic concept of the present invention defined in the following claims also fall within the scope of the present invention.

100: Ceiling Robot Teaching System 110: Ceiling Robot
111: End effector 112: Sealer
120: 3D scanner 130: Robot controller
140: Server 141: External interface section
142: Scan processing unit 143: Robot teaching unit
144: Database 145: Display
146: Control unit 150: Calibration target
b1: first body b2: second body
c: overlapping part

Claims (20)

In a sealing robot teaching system that generates robot passes of multiple sealing robots placed in a painting process line,
Sealing robot with sealer and 3D scanner mounted via bracket on end effector;
A robot controller that controls the three-dimensional scanning and sealing operations of a loaded body through the kinematic posture control of the above sealing robot; and
A server that corrects the robot pass to match the surface shape caused by at least one of the alignment error and assembly error of the body identified by three-dimensionally matching a master sample in which a robot pass for the sealing operation is defined to the 3D scan data of the body measured by the 3D scanner, and then transmits the result to the robot controller;
A sealing robot teaching system including: In the first paragraph,
A sealing robot teaching system in which the above master sample is composed of CAD data including at least one robot pass defined for the sealing operation and feature points of the body for the three-dimensional matching on the 3D shape of the sample body before mass production. In the first paragraph,
The above 3D scanner
A sealing robot teaching system that is calibrated to generate 3D scan data in the same coordinate system as the coordinate system of the above sealing robot. In the first paragraph,
The above robot controller
A sealing robot teaching system that detects the relative coordinates of the sealer and the 3D scanner based on the coordinates of the end effector according to the detailed information of the sealing robot. In any one of claims 1 to 4,
The above server is
An external interface unit including one or more communication modules for communicating with the 3D scanner and robot controller for each of the above sealing robots;
A scan processing unit that filters out noise and background parts through preprocessing of the above 3D scan data and creates a 3D body shape for creating the robot path;
A robot teaching unit that displays the above 3D body shape on a simulation tool and generates a robot path according to line information input through a computing input device;
A database (DB) that stores various programs and data for setting, modifying, and managing the sealing work of each robot pass for each sealing robot;
A display unit that displays the screen generated through the above simulation tool to the operator; and
A control unit that corrects the coordinates of the robot path by attaching the three-dimensional shape of the robot path defined in the master sample to the surface coordinates of the current 3D body shape after the three-dimensional matching;
A sealing robot teaching system including: In paragraph 5,
The above scan processing unit
A sealing robot teaching system that divides the entire image of the vehicle body into parts or robot pass units during the preprocessing process of the above 3D scan data to create multiple independent 3D shapes. In paragraph 5,
The above robot teaching unit
A sealing robot teaching system that displays a 2D plan view of the 3D body shape through the above simulation tool and defines the robot path by applying a pass line input to the overlapping portion of the body to the surface of the 3D body shape. In Article 7,
The above robot teaching unit
A sealing robot teaching system in which the pass line is input using one of an input device among a mouse, a pen, and a digitizer at the overlapping portion of the first body and the second body. In Article 7,
The above pass line is
A ceiling robot teaching system that inputs a point-to-point method that connects points one by one from a start point to an end point, or a drawing method that inputs a pass line continuously. In paragraph 5,
The above robot teaching unit
A sealing robot teaching system that aligns a pre-fabricated body sample in a correct position and then inputs pass lines to various parts of a body shape of a 3D point cloud type acquired by scanning with the 3D scanner to create a master sample in which the robot path is defined. In paragraph 5,
The above control unit
A sealing robot teaching system that three-dimensionally matches feature points defined in the 3D shape of the master sample to feature points corresponding to the current 3D body shape using at least one of the PPF (Point Pair Features) and ICP (Iterative Closest Point) algorithm techniques. In paragraph 5,
The above control unit
A sealing robot teaching system that generates 6-axis coordinates converted into a robot coordinate system of the sealing robot using the robot path corrected to the 3D body shape and transmits the same to the robot controller through the external interface unit. In paragraph 5,
The above control unit
A sealing robot teaching system that detects an assembly error of a second body overlapping a first body when the 3D body shape and the 3D shape of the master sample do not match, deletes a pass line where the assembly error occurred using the simulation tool, and modifies the pass line based on the feature points of the overlapping second body. In paragraph 5,
The above control unit
A sealing robot teaching system that compares the 3D coordinates of the 3D scanner using a fixed calibration target with the robot coordinates of the sealing robot to check whether an error has occurred, and corrects the 3D coordinates of the 3D scanner by the amount of error that has occurred, when a non-matching judgment of the same pattern is repeated a standard number of times or more at a predetermined interval or during the above-mentioned 3D matching. In a sealing robot teaching method in which a server controlling a painting process line generates robot passes for multiple sealing robots,
a) a step of creating a master sample in which at least one robot pass is defined for a sealing operation on a 3D shape of a sample body and feature points for each component on the body surface are defined;
b) a step of collecting 3D scan data of the body scanned by a 3D scanner mounted on a sealing robot from a robot controller when the body is loaded during mass production;
c) a step of filtering out noise and background parts as a preprocessing task of the above 3D scan data and creating a 3D body shape for generating the robot path; and
d) a step of correcting the robot path to match the surface shape caused by at least one of the alignment error and assembly error of the body detected by three-dimensionally matching the 3D body shape and the 3D shape of the master sample, and then transmitting the result to the robot controller;
A teaching method for a sealing robot including: In Article 15,
Step a) above,
A step of displaying the 3D body shape of the sample body scanned through the 3D scanner in 2D on a simulation tool;
A step of inputting a pass line using one of an input device among a mouse, a pen, and a digitizer on the overlapping part of the body; and
A step of defining a 3D robot path by applying the input pass line to the surface of the 3D body shape of the sample body;
A teaching method for a sealing robot including: In Article 15,
Step c) above,
A sealing robot teaching method comprising a step of dividing the entire image of the above body into parts or robot pass units to generate a plurality of independent individual 3D shapes. In Article 15,
Step d) above,
d1) A sealing robot teaching method including a step of attaching a three-dimensional shape of a robot path defined in the master sample to the surface coordinates of the 3D body shape and correcting the coordinates of the robot path to match the current alignment position of the body. In any one of Articles 15 to 18,
Step d) above,
d2) a step of detecting an assembly error of the body due to a mismatch between the 3D body shape and the 3D shape of the master sample during the 3D matching; and
d3) A step of correcting the robot path in the area where an assembly error occurred through a simulation tool of the robot teaching section;
A teaching method for a sealing robot including: In Article 19,
Step d3) above,
A sealing robot teaching method comprising the step of deleting a previously defined 3D robot path in the assembly error occurrence area through the simulation tool and generating a modified 3D robot path based on feature points of the overlapping body area.

Images