KR20230103788A - System for mounting room mirror on vehicle glass - Google Patents

Abstract

The present invention relates to a vehicle glass room mirror mounting system. According to the present invention, a first 3D scan module and a gripper are mounted on the arm of an articulated robot, and the changed robot posture is changed by changing a preset robot posture for product picking based on a first position deviation obtained by scanning the rear surface of a room mirror. The end of the rod located at the rear of the room mirror when the arm of the first cooperating robot that controls the arm part with the furnace and grips the room mirror and the arm part of the second cobot that grasps the room mirror received from the first cobot approaches the front The second 3D scan module that precisely scans the side surface and the gripper are attached to the arm of the articulated robot, and the arm that holds the room mirror approaches the second 3D scan module to be precisely scanned, and then controls the arm to move the room mirror. It is mounted on the vehicle glass, but based on the second position deviation obtained by scanning the end surface of the rod, the preset product mounting robot posture is changed and the arm part is controlled with the changed robot posture, so that the end surface of the room mirror is the fixing piece of the vehicle glass Controls the second cobot mounted on the robot, and the first and second cobots and the first and second 3D scan modules, and analyzes each positional deviation from the scanning results of the first and second 3D scan modules to determine the first and second cobots. , and a control module that is transmitted to the second cooperative robot.
According to the present invention, it is possible to accurately mount the end of the rod of the room mirror to the fastening part of the inside of the vehicle glass by using a plurality of cooperative robots interlocking with the 3D scan module.

Description

Vehicle glass room mirror mounting system {System for mounting room mirror on vehicle glass}

The present invention relates to a vehicle glass room mirror mounting system, and more particularly, to a vehicle glass room mirror capable of accurately coupling and mounting the rod end of the room mirror to the room mirror fastening part of the vehicle glass using a collaborative robot and a 3D scan module. It's about the mounting system.

The room mirror of the vehicle is inserted and mounted into a mounting portion attached to the front glass of the vehicle. For example, the end of the rod of the room mirror is inserted into and fastened to the fastening socket attached to the inside of the front glass of the vehicle.

Until now, most of the room mirrors have been manually installed by workers, but some of them are also installed by robots.

In general, in the case of a part picking task using a robot, part picking is performed by calculating a change in the position of an object to be picked and transmitting the position deviation to the robot.

However, in the case of a room mirror, since the angular positions of the rods pivotally coupled to the rear are different, it is necessary to obtain a deviation for this part and then correct and mount it.

Accordingly, there is a need for an automated robot system capable of accurately aiming and mounting a fastening piece formed at an end of a rod to a mounting portion of a vehicle glass in consideration of a change in angle or position of a rod coupled to a rear surface of the room mirror after picking the room mirror.

The background technology of the present invention is disclosed in Korean Patent Publication No. 2011-0035461 (published on April 6, 2011).

An object of the present invention is to provide a vehicle glass room mirror mounting system capable of accurately coupling and mounting a fastening piece of a rod end of a room mirror to a room mirror fastening portion of a vehicle glass by using a collaborative robot and a 3D scan module.

In the present invention, a first 3D scan module and a gripper are mounted on the arm of an articulated robot, the arm is moved to scan the back of a room mirror waiting on a tray, and a first position deviation of a current scanning model compared to a reference model is measured. A first collaborative robot that grips the room mirror by changing the posture of the robot for product picking that has been previously set based on the changed robot posture and controlling the arm with the changed robot posture, and a second collaborative robot that grips the room mirror received from the first collaborative robot A second 3D scan module that scans the rear surface of the room mirror when the arm unit approaches the front and precisely scans the end surface of the rod of the room mirror that rotates while mounted on the vehicle glass, and the arm unit of the articulated robot The gripper is mounted on the rearview mirror, and the arm part gripping the room mirror is approached by the second 3D scan module to be precisely scanned, and then the arm part is controlled according to the scanning result to mount the room mirror on the vehicle glass, and the rod end surface reference model Based on the second distorted position deviation of the current rod end surface scanning model, the preset product mounting robot posture is changed to control the arm with the changed robot posture, so that the end surface of the room mirror is mounted on the fixing piece of the vehicle glass. Controls a second cooperative robot, the first and second cooperative robots, and the first and second 3D scan modules, and analyzes positional deviations from scanning results of the first and second 3D scan modules to determine the first and second cooperative robots. , It provides a vehicle glass room mirror mounting system including a control module transmitted to the second cooperative robot.

In addition, the first position deviation or the second position deviation may include the 3D position and rotational movement amount of the current scanning model compared to the reference position of the preset reference model.

In addition, the control module performs 3D matching between a scanning model obtained by scanning a rear edge shape of the room mirror through the first 3D scan module and a pre-stored reference model using an Iterative Closest Point (ICP) algorithm to obtain the first position. The deviation is calculated, and the second position deviation is obtained by 3D matching the scanning model obtained by scanning the shape of the end surface of the rod on the rear side of the room mirror through the second 3D scan module and the reference model of the end surface of the rod previously stored using the ICP algorithm. can be computed.

In addition, the first collaborative robot pre-stores the posture of the robot for product picking generated based on the reference model as a default posture, and the second collaborative robot stores the posture of the robot for product picking generated based on the reference model at the end of the rod. The posture of the robot for product installation can be stored in advance as a default posture.

In addition, the first cooperative robot controls the arm part according to the scanning result to grip the room mirror, and then transfers the grip to the second cooperative robot disposed adjacent to the side, and the second cooperative robot controls the first cooperative robot. After receiving the room mirror in a state in which the dark parts face each other with the cooperating robot, the dark part may be moved toward the front of the second 3D scan module at the set position through a posture change.

In addition, the first cooperative robot and the second cooperative robot may each gripper grip the edge of the room mirror, but grip different edge portions of the room mirror so that there is no interference between the grippers in the transfer process of the room mirror. there is.

In addition, the vehicle glasses may be prepared in advance by being placed on a cradle provided between the first and second cooperative robots with a concave surface facing upward.

In addition, coordinate calibration between the first cooperative robot or the second cooperative robot and the first 3D scan module or the second 3D scan module is performed in advance based on the coordinate system of the first cooperative robot or the second cooperative robot, and the cooperative robot and the second cooperative robot 3D scan modules may have a common coordinate system.

In addition, the control module may scan a plurality of objects obtained by scanning the object through the first or second 3D scan module while changing the posture of the robot in a state in which a three-dimensional calibration object is mounted on the first or second cooperative robot. The calibration may be performed based on a result of scanning a robot pose and a plurality of objects.

According to the present invention, the rod end surface of the room mirror can be accurately coupled and mounted to the fastening portion of the room mirror of the vehicle glass using the collaborative robot and the 3D scan module.

In addition, based on the 3D scanning of the shape of the end of the rod of the room mirror, the current scanned model analyzes the positional deviation from the reference model to change the robot posture, so that the end of the rod can be precisely aimed at the fastening part of the glass and fastened. .

According to this, even if there is a positional deviation of the end of the rod with respect to the room mirror for each room mirror product, the posture of the robot can be adaptively changed during the installation process of the room mirror according to the distorted state of the product, thereby increasing the accuracy of the mounting operation.

1 is a diagram showing the configuration of a vehicle glass room mirror mounting system according to an embodiment of the present invention.
FIG. 2 is a view showing the room mirror mounting system shown in FIG. 1 in detail.
3 is a diagram showing an actual implementation of FIG. 2 by way of example.
4 is a diagram illustrating the structure of a general room mirror.
5 is a diagram explaining a product picking operation of the first collaborative robot according to an embodiment of the present invention.
6 is a diagram explaining a product mounting operation of the second collaborative robot according to an embodiment of the present invention.
7 is a diagram showing a calibration object in an embodiment of the present invention.
8 is a conceptual diagram of calibration according to an embodiment of the present invention.
9 is a conceptual diagram of a picking operation algorithm in an embodiment of the present invention.
10 is a conceptual diagram of a mounting operation algorithm in an embodiment of the present invention.
11 is a view illustrating a process of installing a glass room mirror of a vehicle according to an embodiment of the present invention by way of example.

Then, with reference to the accompanying drawings, an embodiment of the present invention will be described in detail so that those skilled in the art can easily practice it. However, the present invention may be implemented in many different forms and is not limited to the embodiments described herein. And in order to clearly explain the present invention in the drawings, parts irrelevant to the description are omitted, and similar reference numerals are attached to similar parts throughout the specification.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected" but also the case where it is "electrically connected" with another element interposed therebetween. . In addition, when a certain component is said to "include", this means that it may further include other components without excluding other components unless otherwise stated.

1 is a diagram showing the configuration of a vehicle glass room mirror mounting system according to an embodiment of the present invention, FIG. 2 is a diagram showing the room mirror mounting system shown in FIG. 1 in detail, and FIG. 3 is an actual implementation of FIG. It is a drawing showing the

1 to 3, the vehicle glass room mirror mounting system 100 according to an embodiment of the present invention includes a first cooperative robot 110 equipped with a first 3D scan module 115, and a second 3D scan module. 120, the second collaborative robot 130, and a control module 140.

The control module 140 can control the operation of each cooperative robot 110 and 130 and each 3D scan module 115 and 120, and is connected to them through a wired, wireless or wired/wireless mixed network to communicate with each other and to give various information in real time. can receive

The first and second cooperative robots 110 and 130 are implemented as 6-axis multi-joint robots, and grippers 111 and 131 for gripping the room mirror 10 are mounted on the arms, respectively.

The first collaborative robot 110 is a picking robot that picks the room mirror 10 prepared on the front tray and delivers it to the second collaborative robot 130, and the second collaborative robot 110 picks up the room mirror 10 that has been delivered. (10) corresponds to a mounting robot for mounting the fixing piece 21 of the vehicle front glass 20 (hereinafter, vehicle glass) prepared on one side.

The first 3D scan module 115 and the second 3D scan module 120 are used to increase the accuracy of the picking operation through the first cooperative robot 110 and the mounting operation through the second cooperative robot 130, respectively. These 3D scan modules correspond to general 3D scan sensors and 3D scanner devices, and structured light scanning may be used.

The vehicle glass 20 is prepared in advance before work. For example, as shown in FIGS. 2 and 3, the concave surface is placed on the cradle 22 provided between the first and second cooperative robots 110 and 130 so that the concave surface faces upward. and can be prepared in advance. On one side of the concave surface, a fixing piece 21 is provided to which the end surface corresponding to the end of the rod 11 of the room mirror 10 is coupled.

In the embodiment of the present invention, the end surface 12 of the end of the rod 11 of the room mirror 10 is slidably inserted into the fixing piece 21 provided on the concave inner surface of the vehicle glass 20. as a representative example. Here, the end surface 12 of the end of the rod means a fastening piece fastened to the fixing piece 21 of the glass.

4 is a diagram illustrating the structure of a general room mirror. As shown in FIG. 4 , in the room mirror 10, a rod 11 capable of adjusting an angle is installed on the rear side opposite to the front side where the mirror is installed. At this time, the room mirror 10 is mounted while the end surface formed at the end of the rod, that is, the fastening piece 12 is coupled to the fixing piece 21 of the vehicle glass 20. At this time, male and female fastening, sliding coupling method, etc. are mainly applied.

However, the angle of the rod 11 located on the rear surface of the room mirror 10 or the end surface 12 of the end of the rod 11 may be deformed by an external force or the like. For this reason, the finished room mirrors 10 manufactured are shipped in a state in which the rod 11 or the bent direction or angle of the end 12 thereof is different.

In this way, since there is a positional deviation of the rod 11 with respect to the room mirror 10 for each completed product, and the positional deviation is different, the end surface of the rod 11 is the fixing piece 21 of the glass 20 In order to be accurately aimed and mounted on the robot, it is essential to correct the posture of the robot for each product.

According to the present invention, a plurality of collaborative robots accurately approach and mount the end 12 of the rod 11 of the room mirror 10 to the fixing piece 21 inside the vehicle glass in conjunction with the scan result of the 3D scan module. can do. The operation of the system according to the present invention will be described in more detail below.

As shown in FIGS. 2 and 3 , the first collaborative robot 110 mounts the first 3D scan module 115 and the gripper 111 on the arm of the articulated robot, moves the arm and waits on the tray 30. Scanning the back side of the room mirror 10, and changing the preset posture of the robot for product picking based on the deviation of the first position of the current scanning model compared to the reference model, and controlling the arm part with the changed robot posture to move the room mirror 10 to grip

The robot posture for product picking represents a robot posture pre-registered based on the reference model, and may be registered in the first cooperative robot 110 in advance. In an embodiment of the present invention, the robot posture includes time-series posture information necessary for a picking operation, and the posture information includes information such as angles and positions of robot joints or arms controlled in real time.

The rear surface of the room mirror 10 is prepared on the top of the tray 30 in an upside down state, and the first collaborative robot 110 moves the arm part toward the room mirror 10 so that the room mirror 10 scan the back Accordingly, when the rear shape of the room mirror 10 is scanned, the shape of the rear border is scanned and the scanned model is compared with a pre-stored reference model to check the degree of distortion compared to the reference model.

The first cooperative robot 110 controls the dark part according to the scanning result to hold the room mirror in place, and then transfers the rearview mirror to the second cooperative robot 130 disposed adjacent to the side. The first cooperative robot 110 grips the edge portion of the room mirror 10 through the gripper 111 and transfers it to the second cooperative robot 130 .

Here, the operation timings of the first cooperative robot 110 and the second cooperative robot 130 may be controlled by the control module 140, which controls the robots 110 and 130 and the 3D scan modules 115 and 120. ) and real-time information exchange to determine the operation status and perform various data processing and calculation operations.

When the second collaborative robot 130 receives the room mirror 10 from the first collaborative robot 110, the room mirror 10 is delivered by manipulating the gripper 131 while facing the first collaborative robot 110. receive That is, the first cooperative robot 110 and the second cooperative robot 130 exchange the room mirror 10 in a state where the dark parts face each other.

At this time, the second cooperative robot 130 also allows the gripper 131 to grip the edge of the room mirror 10 in the process of handing over the room mirror 10 . However, each of the grippers 111 and 131 grips the other edge portion of the room mirror 10 so that there is no interference between the grippers 111 and 131 in the process of receiving the transfer. This is possible through adjusting the distance between the grippers. For example, the second cooperative robot 130 may grip the room mirror by controlling the grippers at a wider distance than the first cooperative robot 110 .

When the arm part of the second collaborative robot 130 gripping the room mirror 10 received from the first cooperative robot 110 approaches the front, the second 3D scan module 120 moves the rear surface of the room mirror 10 While scanning, the end face 12 of the rod 11 of the room mirror 10 rotating while mounted on the vehicle glass 20 is precisely scanned.

The second cooperative robot 130 mounts the gripper 131 on the arm of the articulated robot, and receives the room mirror 10 in a state where the first cooperative robot 110 and the arm face each other. Accordingly, the rear surface of the room mirror 10 faces the outside. Then, the second collaborative robot 130 approaches the front of the second 3D scan module 120 at a previously known position of the arm holding the room mirror 10 so that the back side of the room mirror 10 is precisely scanned. .

In this way, the second cooperative robot 130 receives the room mirror 10 in a state where the first cooperative robot 110 and the arm part face each other, and then changes the robot posture to move the arm part to the set position of the second 3D scan module ( 120) to achieve precise scanning.

Accordingly, the second 3D scan module 120 scans the shape of the rear surface of the room mirror 10. In particular, the shape of the rod end surface 12 on the rear surface is scanned, and the scanned model is compared with the pre-stored rod end surface reference model. By comparing, it is possible to check the degree of deviation compared to the reference model.

The second collaborative robot 130 removes the dark part according to the scanning result through the second 3D scan module 120 and mounts the room mirror 10 on the vehicle glass 20 provided at the lower side, but compares the end surface of the rod to the reference model Based on the second distorted position deviation of the current rod end surface scanning model, the preset product mounting robot posture is changed to control the arm part with the changed robot posture, so that the rod end surface 12 of the room mirror 10 is the vehicle glass ( 20) to be mounted on the fixing piece 21.

That is, the rod end 12 of the room mirror 10 is accurately aimed at the fixing piece 21 attached to the inside of the front glass 20 of the vehicle, and then inserted and fastened. Here, the posture of the robot for product mounting indicates a posture of the robot pre-registered based on the reference model of the end surface of the rod, and may be calculated in advance and registered to the second collaborative robot 130 .

The control module 140 controls the first cooperative robot 110, the second cooperative robot 130, the first 3D scan module 115, and the second 3D scan module 120, and controls the first and second 3D scan modules. Each position deviation is analyzed from the scanning result of the scan module 115 and transmitted to the first and second cooperative robots 115 and 120 respectively.

5 is a diagram explaining a product picking operation of the first collaborative robot according to an embodiment of the present invention.

In FIG. 5 , the room mirror part master corresponds to the room mirror reference model M1_ref. Here, the reference model M1_ref is, for example, a model obtained by scanning a room mirror placed at a predetermined position (reference position) through the first 3D scan module 115 or a model obtained by converting a design drawing into point cloud data. , which is a reference model used to obtain a position deviation compared to a predetermined reference position for a room mirror placed at an arbitrary position on the tray 30 by an operator. Here, the positional deviation is a concept including positional movement and rotational movement of the corresponding part.

The control module 140 obtains a scanning model M1 obtained by scanning the rear edge shape of the room mirror through the first 3D scanning module 115 of the first collaborative robot 110, and obtains a scanning model M1 and a previously known The pre-stored reference model (M1_ref) is 3D matched through an Iterative Closest Point (ICP) algorithm and a 3D pattern matching algorithm to calculate the amount of movement relative to the reference position of the reference model, and the first position deviation according to the amount of movement is calculated by the first collaborative robot 110. forward to At this time, the movement amount includes positional and rotational movement amounts.

That is, the position deviation includes the 3-dimensional position and rotational movement amount of the current scanning model M1 compared to the reference position of the reference model M1_ref, and includes a total of 6 parameters including 3-axis position deviation and 3-axis tilt deviation. (Δx, Δy, Δz, ΔRx, ΔRy, ΔRz).

Here, the first collaborative robot 110 pre-stores the pre-created posture of the robot for product picking as a default posture based on the reference model M1_ref. Such a default posture may refer to a robot posture required for picking an edge portion of a room mirror corresponding to a reference model to a proper position.

Therefore, the first collaborative robot 110 corrects the pre-stored product picking robot posture corresponding to the reference model M1_ref using the first position deviation, thereby changing the robot posture according to the degree of distortion of the currently placed product for picking. will perform In this way, the first collaborative robot 110 changes the robot posture by reflecting the first position deviation in the robot posture for product picking, and controls the arm based on this to perform product picking.

6 is a diagram explaining a product mounting operation of the second collaborative robot according to an embodiment of the present invention. The master shown in FIG. 6 corresponds to the rod end surface reference model M2_ref of the room mirror. The rod end surface reference model (M2_ref) is scanned through the second 3D scan module 120 by approaching it to the second 3D scan module 120 while holding the reference model in the correct position in the second collaborative robot 130. is obtained Here, of course, the rod end surface reference model (M2_ref) corresponds to a rod end surface 12 obtained in an in-position state with no angle distortion with respect to the room mirror body, and the end surface 12 is enlarged to form a scanned image. may apply.

The rod end surface reference model M2_ref is a reference model used to obtain a positional deviation relative to a predetermined reference position of the end surface 12 on the rear surface of the room mirror. The positional deviation includes positional movement and rotational movement of the corresponding part.

The control module 140 obtains a scanning model M2 obtained by enlarging and scanning the shape of the rod end surface 12 on the rear surface of the room mirror through the second 3D scanning module 120, and obtains the scanning model M2 and the pre-stored reference model. (M2_ref) is 3D matched through the ICP algorithm and the 3D pattern matching algorithm to calculate the amount of movement relative to the reference position of the reference model, and transmits the first position deviation according to the amount of movement to the first collaborative robot 110. At this time, the movement amount includes positional and rotational movement amounts.

That is, the position deviation includes the 3-dimensional position and rotational movement amount of the current scanning model M2 compared to the reference position of the reference model M2_ref, which also includes a total of six parameters (Δx, Δy, Δz, ΔRx, ΔRy, ΔRz).

Here, the second collaborative robot 130 pre-stores the pre-created product mounting robot posture as a default posture based on the rod end face reference model M2_ref. Such a default posture may represent a robot posture required for the rod end face 12 of the room mirror corresponding to the reference model to be mounted on the fixing piece 21 of the vehicle glass 20 while aiming at the correct position.

Therefore, the second collaborative robot 130 corrects the pre-stored product mounting robot posture corresponding to the reference model M2_ref using the second position deviation, thereby adjusting the robot posture according to the degree of distortion of the current end surface 12. It can be modified and accurately mounted on the fixing piece 21 of the vehicle glass 20. In this way, the second collaborative robot 130 changes the robot posture by reflecting the second position deviation to the pre-registered product mounting robot posture, and controls the arm based on this to perform product picking.

According to an embodiment of the present invention, the entire configuration may have a common coordinate system based on the coordinate system of the collaborative robot. To this end, coordinate calibration between the first cooperative robot 110 or the second cooperative robot 130 and the first 3D scan module 115 or the second 3D scan module 120 is performed by the first cooperative robot 110 or the second cooperative robot 110. It is made in advance based on the coordinate system of the cooperative robot 130, and may have a common coordinate system between the cooperative robot and the 3D scan module. Accordingly, the coordinates of the scanning image obtained through the 3D scan module may be changed based on the coordinate system of the robot base of the cooperative robot.

The calibration process may be performed through a process of scanning the calibration object through a 3D scan module while changing the posture of the robot while the calibration object is mounted on the cobot. At this time, in the embodiment of the present invention, a three-dimensional structured calibration object is utilized.

Specifically, the control module 140 is a plurality of robots obtained by scanning the object through the first or second 3D scan module while changing the robot pose while the three-dimensional calibration object is mounted on the first or second cooperative robot. Calibration can be performed based on a set of poses and multiple object scan results.

The first 3D scan module 115 is mounted on the robot tool of the first collaborative robot 110, and the second 3D scan module 120 is fixed to a separate profile. For example, the first collaborative robot 110 may perform calibration by scanning the object with the second 3D scan module 120 while changing the robot pose while the calibration object is mounted, and the second collaborative robot 130 ) may perform calibration by scanning an object in the second 3D scan module 120 while changing a robot pose in a state in which the object is mounted.

7 is a diagram showing a calibration object in an embodiment of the present invention.

(a) of FIG. 7 shows the real object of the calibration, and (b) shows an image scanned by the scan module. 7(c) shows an image obtained by scanning an object with a 3D scan module while changing a pose of the robot holding the object.

The calibration object has a structure in which the diameter gradually widens as it goes to the outside like a pyramid structure, and has a structure in which four faces are combined. Here, all four faces are implemented with different inclination angles to increase calibration accuracy.

In this way, when a plurality of robot postures (poses) and object scan results are used, a relationship between the base coordinates of the robot (robot coordinate system) and the coordinates of the 3D scan module is created. Through this, all coordinate values obtained from the 3D scan module can be calibrated based on the robot coordinate system. In this way, calibration may be performed between the first and second cooperative robots and the first and second scan sensors to generate a common coordinate system. In addition, calibration is performed between cobots, and a common coordinate system can be obtained by using a coordinate transformation matrix obtained through calibration.

8 is a conceptual diagram of calibration according to an embodiment of the present invention. In FIG. 8, the 3D scan sensor is assumed to be a second 3D scan module fixed to a profile, and robot 1 and robot 2 are assumed to be first and second cooperative robots, respectively. H represents a matrix and corresponds to a transformation matrix according to the relationship between each component.

은 로봇1에서 3D 스캔 센서 간의 회전과 위치이동값을 나타내는 4×4 행렬이며 아래 수학식 1과 같으며, 이러한 회전과 위치이동 좌표를 로봇 포즈값으로

로 변환하여 표현할 수도 있다.

은 협동로봇2에서 3D 스캔 센서 간의 회전과 위치이동값을 나타내는 4×4 행렬이며 로봇 포즈값이다. 로봇1과 로봇2와의 관계는

로 나타낼 수 있다. 일련의 과정을 통하여 로봇간의 공통좌표계 생성이 가능하다.

Is a 4 × 4 matrix representing the rotation and displacement values between the 3D scan sensors in robot 1 and is shown in Equation 1 below, and these rotation and displacement coordinates are used as robot pose values.

It can also be expressed by converting to .

is a 4x4 matrix representing the rotation and position movement values between the 3D scan sensors in the cooperative robot 2 and is the robot pose value. The relationship between robot 1 and robot 2 is

can be expressed as Through a series of processes, it is possible to create a common coordinate system between robots.

Coordinate system transformation through calibration can utilize a pre-transformation matrix or Euler Angle Transform, which is an intuitive method of expressing the posture of a coordinate axis. Since these technologies correspond to previously disclosed technologies, a detailed description thereof will be omitted.

The following is a brief description of the collaborative robot interlocking algorithm. First, the configuration of bin picking work using the robot-scanner system is explained.

9 is a conceptual diagram of a picking operation algorithm in an embodiment of the present invention. 9 is the concept of the bin picking algorithm, which is divided into two paths, up and down, based on the robot. Indicates the situation of scanning and picking an object. That is, this FIG. 9 is related to the position deviation of the current scanning model with respect to the reference model.

In the figure, the robot is the cobot 1, the robot tool means a tool part of the cobot 1 on which the scanner is mounted, and the scanner represents the 3D scan sensor 1. An object (train) is an object at a reference position placed on a tray and corresponds to a reference model. The object (picking) shown below corresponds to the scanning model as an object placed at an arbitrary position on the current tray.

로 저장한다.

로 회전과 위치이동을 포함한

행렬이다. 이때 R은 회전(rotation), t는 위치이동(translation)을 나타낸다. 이를 로봇 좌표로 변환하면

의 6축 좌표로 나타낼 수 있다. First, the object (train) is scanned and the pose at this time is

Save as

Including rotation and positioning with

it is a matrix At this time, R represents rotation and t represents translation. Converting this to robot coordinates

It can be represented by the 6-axis coordinates of

는 다음 수학식 2로 표현될 수 있다.And poses related to the object (picking) to be worked on

Can be expressed as Equation 2 below.

는 3D스캐너와 협동로봇1의 엔드이팩터 간의 관계를 나타내는 행렬로 회전과 위치이동으로 구성된

행렬이다.

는 대상체의 이동 편차를 나타내는 회전과 위치이동으로 구성된 행렬이다.here

is a matrix representing the relationship between the 3D scanner and the end-effector of the cobot 1, consisting of rotation and positional movement.

it is a matrix

is a matrix composed of rotation and positional movement representing the movement deviation of the object.

10 is a conceptual diagram of a mounting operation algorithm in an embodiment of the present invention.

The upper figure of FIG. 10 is the concept of scanning an object gripped by the cobot 2 with the 3D scan sensor 2. At this time, the case of scanning with the reference model mounted on the robot tool based on the cobot 2 and the currently gripped object It is divided into the case where the rod end face is scanned, and there is a mutual deviation.

The lower figure of FIG. 10 is for the case of mounting the gripped object on the cobot 2. Here, in contrast to the reference model, the pose required for mounting the current object is expressed as H _install , which is defined as in Equation 3.

Here, each parameter mentioned in 2 and 3 in mathematics may be defined as in Equation 4 below.

All matrices used at this time have the form of a 4 × 4 matrix as shown in Equation 1.

11 is a view illustrating a process of installing a glass room mirror of a vehicle according to an embodiment of the present invention by way of example.

11 shows the inverted rear edge of the room mirror 10 placed on the tray 30 using the first 3D scan module 115 mounted on the first collaborative robot 110, and then the scanning result is displayed. Based on the corrected robot posture, it shows how the product is picked and moved with the gripper 111.

The following figures show how the first cooperative robot 110 delivers the room mirror 10 while facing the second cooperative robot 130. After completion of delivery, the first cooperative robot 110 returns to its initial position and stands by.

The lower figures show the scanning results after the second collaborative robot 130 approaches the front of the second 3D scan module 120 with the arm arm while gripping the room mirror 10 and precisely scans the end of the rod on the rear side of the room mirror. Based on the posture of the robot corrected according to Fig. 1, the room mirror 10 is mounted on the fixing piece of the vehicle glass and then the grip is released.

According to the present invention as described above, the end surface of the rod of the room mirror can be accurately coupled and mounted to the fastening portion of the room mirror of the vehicle glass using the collaborative robot and the 3D scan module.

In addition, based on the 3D scanning of the shape of the end of the rod of the room mirror, the current scanned model analyzes the positional deviation from the reference model to change the robot posture, so that the end of the rod can be precisely aimed at the fastening part of the glass and fastened. .

According to this, even if there is a positional deviation of the end of the rod with respect to the room mirror for each room mirror product, the posture of the robot can be adaptively changed during the installation process of the room mirror according to the distorted state of the product, thereby increasing the accuracy of the mounting operation.

Although the present invention has been described with reference to the embodiments shown in the drawings, this is merely exemplary, and those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Therefore, the true technical scope of protection of the present invention should be determined by the technical spirit of the appended claims.

100: vehicle glass room mirror mounting system
110: first collaborative robot 111: gripper
115: first 3D scan module 120: second 3D scan module
130: second collaborative robot 131: gripper
140: control module

Claims (9)

The first 3D scan module and the gripper are mounted on the arm of the articulated robot, and the arm is moved to scan the back of the room mirror waiting on the tray. a first cooperative robot that changes the position of the robot for product picking and controls the arm part with the changed position of the robot to hold the room mirror;
When the arm of the second collaborative robot holding the room mirror received from the first collaborative robot approaches the front side, scanning the rear surface of the room mirror, the end of the rod end of the room mirror mounted on the vehicle glass and rotating A second 3D scan module that precisely scans the surface;
A gripper is mounted on the arm of the articulated robot, and the arm holding the room mirror is approached with the second 3D scan module to be precisely scanned, and then the arm is controlled according to the scanning result to mount the room mirror on the vehicle glass, Based on the second position deviation of the current rod end surface scanning model compared to the rod end surface reference model, the preset product mounting robot posture is changed and the arm part is controlled with the changed robot posture, so that the rod end surface of the room mirror is of the vehicle glass. a second cooperative robot to be mounted on the fixing piece; and
It controls the first and second cooperative robots and the first and second 3D scan modules, and analyzes each positional deviation from the scanning results of the first and second 3D scan modules to determine the first and second cooperative robots. A vehicle glass room mirror mounting system including a control module that transmits The method of claim 1,
The first position deviation or the second position deviation,
A vehicle glass room mirror mounting system including the 3D position and rotational movement amount of the current scanning model compared to the reference position of the preset reference model. According to claim 1 or claim 3,
The control module,
Calculate the first position deviation by 3D matching a scanning model obtained by scanning a rear edge shape of the room mirror through the first 3D scan module and a pre-stored reference model using an Iterative Closest Point (ICP) algorithm;
A vehicle glass room that calculates the second position deviation by 3D matching the scanning model obtained by scanning the shape of the rod end surface on the back side of the room mirror through the second 3D scan module and the previously stored reference model of the rod end surface using an ICP algorithm. mirror mounting system. The method of claim 1,
The first collaborative robot,
Pre-store the posture of the product picking robot generated based on the reference model as a default posture,
The second collaborative robot,
A vehicle glass room mirror mounting system that pre-stores the product mounting robot posture generated based on the rod end surface reference model as a default posture The method of claim 1,
The first collaborative robot,
According to the scanning result, the dark part is controlled to grip the room mirror, and then transferred to the second cooperative robot disposed next to the side,
The second collaborative robot,
The vehicle glass room mirror mounting system for receiving the room mirror in a state in which the arm parts face each other with the first collaborative robot, and then moving the arm part toward the front of the second 3D scan module at a set position through a posture change. The method of claim 1,
The first cooperative robot and the second cooperative robot,
A vehicle glass room mirror mounting system in which each gripper grips an edge of the room mirror, but grips different edge portions of the room mirror so that there is no interference between the grippers in a transfer process of the room mirror. The method of claim 1,
The vehicle glass,
A vehicle glass room mirror mounting system prepared in advance by placing a concave surface facing upward on a cradle provided between the first and second cooperative robots. The method of claim 1,
Coordinate calibration between the first or second cooperative robot and the first 3D scan module or the second 3D scan module is performed in advance based on the coordinate system of the first or second cooperative robot, so that the 3D scan with the cooperative robot A vehicle glass room mirror mounting system that has a common coordinate system between modules. The method of claim 8,
The control module,
A plurality of robot poses and a plurality of object scans acquired by scanning the object through the first or second 3D scan module while changing the robot pose in a state where a three-dimensional calibration object is mounted on the first or second collaborative robot A vehicle glass room mirror mounting system that performs the calibration based on the result.

Images