A bearing positioning system and method for a rail vehicle door hanger assembly workstation

Abstract

The present invention discloses a bearing positioning system and method for a rail vehicle door hanger assembly workstation. The industrial robot is controlled by a robot controller, and an end effector is provided at the free end of the industrial robot. An industrial camera, lens, and light source are mounted on the end effector. The industrial camera and lens are connected, and the industrial camera is connected to a vision controller to transmit captured images to the vision controller. The light source is connected to a light source controller to control the on / off function, exposure time, and exposure intensity of the light source. The vision controller and light source controller are both connected to the robot controller. A positioning plate is provided in the material box, and the positioning plate is provided with positioning marks and grooves arranged in an array for placing materials. The present invention replaces manual repetitive operations. The machine vision system can identify the positioning of all hanger bearing parts in a single inspection, not only ensuring product accuracy but also reducing personnel requirements while improving production efficiency.

Description

Technical Field

[0001] The invention relates to a bearing positioning system and method for a rail vehicle door hanger assembly workstation, belonging to the technical field of automatic assembly and manufacturing. Background Art

[0002] Rail vehicle doors are the interface for people entering and exiting the vehicle. They are one of the most frequently used components and also the most prone to failure. Doors are critical to passenger safety and operational order. As a key component of urban rail vehicle doors, the pylon bearing is crucial for the doors to open and close properly. The manufacturing of the pylon bearing is particularly critical, as its process precision and performance directly impact the overall performance of the vehicle door.

[0003] Currently, bracket bearings are all assembled semi-automatically, requiring 3-4 operators to work simultaneously. The low production capacity makes it difficult to meet market demand, and the quality consistency of the products is difficult to guarantee. Existing equipment using small-field camera machine vision requires the robot to use the visual system to locate the bracket bearing parts when grabbing them for assembly. Each time the system locates and photographs the parts, it can only recognize one or a few parts, and cannot recognize dozens of parts at once. As a result, when the robot grabs the bracket bearing parts for assembly, it needs to use the visual system to take photos and locate them every time it grabs one or several parts. Since the vehicle door bracket assembly workstation equipment has been automatically and repeatedly performing assembly operations, the number of visual photo inspections increases, the processing time is long, and the equipment cycle is slow, resulting in the existing bearing positioning system being unable to meet production capacity requirements.

[0004] Therefore, those skilled in the art are in urgent need of solving the following technical problems:

[0005] 1. The existing key components of rail vehicle doors are all based on manual semi-automatic assembly, with low production capacity, making it difficult to meet market demand, and the quality consistency and stability of manually assembled products are difficult to guarantee.

[0006] 2. The materials of the hanger bearing parts are stored in 5-layer material boxes, with 20 hanger bearing parts placed in each material box. Due to the large gap between the material box and the positioning plate, the position deviation of a single bearing can reach up to ±30mm, and the angle deviation can reach ±3°. When only the robot is used for grasping without using the visual positioning system, it is difficult to guarantee the grasping position deviation accuracy requirements. If the robot grasps inaccurately, it will cause damage to the bearing parts.

[0007] 3. Existing machine vision systems can only identify one or a few parts at a time when using a camera with a small field of view to locate parts. The limited field of view makes it impossible to identify dozens of parts at once, resulting in numerous visual inspections, long processing times, and slow equipment cycles. Using a larger field of view or multiple cameras for simultaneous capture significantly increases the cost of the vision system. Summary of the Invention

[0008] Purpose: To overcome the deficiencies in the prior art, the present invention provides a bearing positioning system and method for a rail vehicle door hanger assembly workstation to replace manual repetitive operations. A machine vision system can identify the positioning of all hanger bearing parts in a single inspection, which not only ensures product accuracy but also reduces personnel requirements while improving production efficiency.

[0009] Technical solution: To solve the above technical problems, the technical solution adopted by the present invention is:

[0010] In the first aspect, a bearing positioning system for a rail vehicle door hanger assembly workstation includes: an industrial robot, a vision system, a material box, and a positioning plate. The industrial robot is controlled by a robot controller, and an end effector is provided at the free end of the industrial robot for gripping materials. The vision system includes: an industrial camera, a lens, and a light source. The industrial camera, lens, and light source are mounted on the end effector. The industrial camera and lens are connected. The industrial camera is connected to the vision controller and transmits the captured images to the vision controller. The light source is connected to the light source controller to control the opening and closing of the light source, the exposure time, and the exposure intensity. The vision controller and the light source controller are both connected to the robot controller. A positioning plate is provided in the material box, and the positioning plate is provided with positioning marks and grooves arranged in an array. The grooves are used to place materials.

[0011] As a preferred solution, the grooves are in an array of 5x4 or 5x8.

[0012] As a preferred solution, the light source is a white LED ring light source.

[0013] As a preferred solution, the vision controller communicates with the robot controller via Socket, the vision controller serves as a Socket communication server, and the robot controller serves as a Socket communication client.

[0014] As a preferred solution, the light source controller and the robot controller communicate through a digital IO port.

[0015] As a preferred solution, the industrial camera adopts a CCD camera, the lens adopts a fixed-focus lens, and the lens and the industrial camera are connected via a C interface.

[0016] In a second aspect, a bearing positioning method for a rail vehicle door hanger assembly workstation includes the following steps:

[0017] Step 1: Move the industrial robot's end-effector vision system over an empty positioning plate to acquire the robot's current reference image pose data. Acquire an image of the empty positioning plate with positioning markers as the reference template image. Move the industrial robot's end effector to the center of each groove on the empty positioning plate and record the reference grasping pose data for each groove.

[0018] Step 2: According to the conversion matrix equation between the camera coordinate system and the industrial robot coordinate system, set the center coordinates of the positioning mark in the reference template image in the industrial robot coordinate system to X0=0, Y0=0, and the rotation angle θ0=0.

[0019] Step 3: Based on the pose data of the industrial robot's current reference shooting position, move the visual system at the end of the industrial robot to above the positioning plate in the material box to obtain an image of the material on the positioning plate as the image to be processed.

[0020] Step 4: Preprocess the image to be processed to obtain edge information in the image to be processed, perform hand-eye calibration based on the edge information in the image to be processed, obtain the center coordinates of the positioning mark of the image to be processed in the robot coordinate system, compare the center coordinates of the image to be processed with the center coordinates of the reference template image, and obtain the center coordinate difference (Δx, Δy) and rotation angle difference Δθ of the positioning plate of the image to be processed.

[0021] Step 5: If the center coordinate difference (Δx, Δy) and the rotation angle difference Δθ of the positioning plate of the image to be processed are within the set threshold, proceed to step 6.

[0022] Step 6: Based on the position information of each groove in the reference template image, calculate the center position coordinate X of all materials in the reference image template in the robot coordinate system. n , Y n , n represents the quantity of materials.

[0023] Step 7: Based on the center coordinate difference (Δx, Δy) of the positioning plate of the image to be processed and the rotation angle difference Δθ, the center position coordinates X of all materials are obtained. n ,Y n , calculate the position deviation ΔX of the center position of all materials in the image to be processed and all materials in the reference image template n , ΔY n ,n represents the quantity of materials.

[0024] Step 8: Based on the industrial robot benchmark above each groove, grab the pose data and find the pose deviation ΔX of the corresponding center position.n , ΔY n , adjust the end effector of the industrial robot, and after the adjustment, the end effector of the industrial robot grabs all materials in sequence.

[0025] As a preferred solution, the threshold value of the center coordinate difference Δx is ±15 mm, the threshold value of Δy is ±15 mm, and the threshold value of the rotation angle difference Δθ is ±1.5°.

[0026] As a preferred solution, X n =P 中心距 *Cosα+(j 列 -1)*D 列距

[0027] Y n =P 中心距 *Sinα+(i 行 -1)*D 行距

[0028] Where α is the angle between the line connecting the center of the first row and first column groove of the positioning plate and the center of the positioning mark of the positioning plate and the X axis of the robot coordinate system, D 行距 D is the distance between the centers of adjacent grooves in each row. 列距 is the distance between the centers of adjacent grooves in each row, i 行 is the row number of the groove where the nth material is located, j 列 is the number of columns of the groove where the nth material is located, P 中心距 It is the distance between the center position of the groove in the first row and first column of the positioning plate and the center of the positioning mark of the positioning plate.

[0029] As a preferred solution, ΔX n =(-X n *(1-CosΔθ)-Y n *SinΔθ)+Δx

[0030] ΔY n =(-Y n *(1-CosΔθ)+X n *SinΔθ)+Δy.

[0031] Beneficial Effects: The present invention provides a bearing positioning system and method for a rail vehicle door hanger assembly workstation. Using a narrow-field visual camera, a single image can identify the Cartesian position coordinates and rotation angles of all hanger bearing components in a material bin, replacing the need for high-cost, wide-field cameras. Furthermore, each layer of material bins can be fully inspected and identified with a single image, reducing the number of images, increasing equipment cycle time, and improving production capacity. Furthermore, the visual algorithm employed during implementation is adaptable to various operating conditions, such as those involving heavy oil pollution and strong lighting. It exhibits strong robustness and can accommodate different models of similar linear bearing components, enhancing the processing capabilities of machine vision. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The figure is a structural diagram of a bearing positioning system for a rail vehicle door hanger assembly workstation.

[0033] Figure 2 Schematic diagram of the structure of the industrial robot system.

[0034] Figure 3 Schematic diagram of the process of the present invention.

[0035] Figure 4 It is a structural diagram of the positioning plate. DETAILED DESCRIPTION

[0036] The present invention will be further described below with reference to specific embodiments.

[0037] like Figure 1 As shown, the first embodiment is a bearing positioning system for a rail vehicle door hanger assembly workstation, comprising: an industrial robot 1, a set of visual controller 2, a set of industrial camera and lens 3, a set of LED ring light source 4 and light source controller 5, 5 sets of hanger bearing rough positioning material boxes 6 and positioning plates 7, and all the hanger bearings 01 to be tested are stored on the positioning plates in the material boxes.

[0038] Twenty hanger bearings 01 to be tested are stored on the positioning plate 7 in each layer of the material box.

[0039] The industrial camera, lens and white LED ring light source are all mounted on the end effector 8 of the industrial robot and follow the movement of the industrial robot, that is, forming a hand-eye system with the industrial robot.

[0040] like Figure 2 As shown, the light source controller 5 and the robot controller 9 communicate through the digital IO port. The digital IO port of the robot controller controls the opening and closing of the white LED ring light source of the light source controller 5 each time a photo is taken, thereby controlling the light source exposure time when taking a photo. The exposure intensity of the white LED ring light source can be PWM-regulated by the light source controller.

[0041] The industrial camera adopts a CCD camera and a fixed-focus lens. The lens and the industrial camera are connected via a C interface to complete the base image acquisition and current image acquisition of the center position of the bracket bearing positioning plate.

[0042] The industrial camera is connected to the vision controller 2 via an Ethernet interface, and the captured image is transmitted to the vision controller via the TCP / IP protocol. The software algorithm in the vision controller processes the image, including distortion correction and image edge correction, that is, pre-processing is performed through grayscale processing, smoothing filtering and other algorithms, and template matching is performed with the calibrated reference template image. After the matching is successful, the vision system will automatically calculate the center coordinate difference (Δx, Δy) and rotation angle difference Δθ between the positioning plate and the reference image based on the results of the hand-eye calibration, and send the data to the robot controller 9 via the Socket protocol.

[0043] The visual controller 2 is connected to the robot controller 9 through an Ethernet interface for Socket communication. The visual controller acts as a Socket communication server and the robot controller acts as a Socket communication client. The robot controller controls the visual controller to take pictures of the bearing positioning plate and accepts and processes the bearing positioning plate posture detection data sent by the visual controller.

[0044] The robot controller pre-processes the bearing positioning plate's posture detection data to determine whether it is within a reasonable range. It then analyzes and calculates the posture of all 20 bearing components on the positioning plate. The posture data array for all bearing components is stored in an internal register within the robot controller, ultimately achieving bearing component positioning.

[0045] like Figure 3 As shown, the second embodiment of a bearing positioning method for a rail vehicle door hanger assembly workstation includes the following steps:

[0046] Step 1: Manually teach the robot to move to a position above an empty positioning plate with a clear image of the empty positioning plate in the vision system. Save and record the robot's pose data for the current reference position. Simultaneously, manually take a photo with the vision system to capture an image of the empty positioning plate with positioning mark 701. This image will serve as the reference template image for subsequent processing. Then, manually move the robot to 20 part-grabbing positions on the empty positioning plate, and teach and record these 20 sets of robot reference grasping pose data.

[0047] Step 2: By establishing the conversion matrix equation between the camera coordinate system and the industrial robot coordinate system, the center coordinates of the positioning mark in the reference template image in the robot coordinate system are obtained as X0=0, Y0=0, and the rotation angle θ0=0.

[0048] Step 3: The robot controller and the vision controller establish Socket communication through a three-way handshake to ensure successful communication. The first handshake: the client (robot controller) tries to connect to the server (vision controller) and sends a syn packet to the server, syn=j, and the client enters the SYN_SEND state and waits for the server to confirm; the second handshake: the server (vision controller) receives the client (robot controller) syn packet and confirms it (ack=j+1), and at the same time sends a SYN packet (syn=k) to the client, that is, a SYN+ACK packet. At this time, the server enters the SYN_RECV state; the third handshake: the client (robot controller) receives the SYN+ACK packet from the server (vision controller) and sends an acknowledgment packet ACK (ack=k+1) to the server. After this packet is sent, the client and server enter the ESTABLISHED state, completing the three-way handshake.

[0049] Step 4: Based on the pose data of the current reference photographing position, the industrial robot moves to the position directly above the positioning plate in the material box. The industrial robot controls the light source controller to turn on the visual system light source. The camera synchronously takes a photo of the positioning plate in the material box, obtains the photo image taken of the current material box positioning plate, and sends the image to the visual controller to complete the acquisition of the image to be processed. After the photo is taken, the system turns off the light source.

[0050] Step 5: The visual controller performs edge correction on the acquired image to be processed, and preprocesses it through grayscale processing, smoothing filtering, and edge upper and lower limit algorithms. The image is blurred and noise is reduced through preprocessing to obtain edge information in the image to be processed, and other trivial details in the image are removed, thereby improving the computing speed of the visual system. Hand-eye calibration is performed based on the edge information in the image to be processed to obtain the center coordinates of the positioning mark of the image to be processed in the robot coordinate system. The image to be processed is matched with the center coordinates of the reference template image to obtain the center coordinate difference (Δx, Δy) and rotation angle difference Δθ of the positioning plate of the image to be processed, and the data is sent to the industrial robot controller via the Socket protocol.

[0051] Step 6: After the industrial robot controller obtains the center coordinate difference (Δx, Δy) and rotation angle difference Δθ of the positioning plate of the image to be processed, it makes a preliminary judgment. If the position and angle deviations are within a certain range (i.e., |Δx| ≤ 15mm, |Δy| ≤ 15mm, |Δθ| ≤ 1.5°), the system algorithm determines that the vision controller has processed the image to be processed and proceeds to step 7. If it fails, step 4 is repeated. If the failure is repeated for more than three consecutive times, the robot system will issue an alarm prompting manual intervention.

[0052] like Figure 4As shown, step 7: Set the vector position relationship between the center of the reference image template of the positioning plate and the 20 bracket bearing parts in the system algorithm, that is, the angle α formed by the line formed by the center position of the bracket bearing parts in the first row and first column of the positioning plate and the center of the positioning mark of the positioning plate and the X-axis of the robot coordinate system. It is known that the row center distance D of the center holes of adjacent bracket bearing parts in each row and column of the positioning plate 行距 / Column center distance D 列距 , and the distance between the center of the part in the first row and the first column and the center of the positioning mark on the positioning plate is P 中心距 , the center position coordinates (X n =P 中心距 *Cosα+(j 列 -1)*D 列距 ,Y n =P 中心距 *Sinα+(i 行 -1)*D 行距 ), n=1,2···20.

[0053] Step 8: The "center coordinate difference (Δx, Δy) and rotation angle difference Δθ of the positioning plate identified in step 6 are fused with the "position coordinates of the 20 pylon bearings in the reference image template" calculated in step 7. According to the relationship between the sine and cosine trigonometric functions, the position deviation between the measured pylon bearing and the center position of the pylon bearing in the reference image module can be obtained as (ΔX n =(-X n *(1-CosΔθ)-Y n *SinΔθ)+Δx,ΔY n =(-Y n *(1-CosΔθ)+X n *SinΔθ)+Δy).

[0054] Step 9: The robot controller caches the pose data arrays of all 20 parts calculated in step 8 in the array register of the controller. The system performs linear offset based on the robot pose data of the rack bearing parts in the same row and column under the reference template (i.e., the robot reference grasping pose data in step 1) according to the part position grasped by the robot each time, so as to adapt to the current position deviation and gradually guide the industrial robot end effector to achieve accurate and seamless grasping of the rack bearing parts.

[0055] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A bearing positioning method for a rail vehicle door hanger assembly workstation, characterized by: The following steps are involved: Step 1: Move the vision system at the end of the industrial robot to the top of an empty positioning plate, obtain the pose data of the industrial robot's current reference shooting position, and obtain an image of the empty positioning plate with positioning marks as the reference template image; move the end effector of the industrial robot to the center of each groove of the empty positioning plate, and record the industrial robot's reference grasping pose data above each groove; Step 2: According to the conversion matrix equation between the camera coordinate system and the industrial robot coordinate system, set the center coordinates of the positioning mark in the reference template image in the industrial robot coordinate system to X0=0, Y0=0, and the rotation angle θ0=0; Step 3: Based on the pose data of the industrial robot's current reference shooting position, move the visual system at the end of the industrial robot to above the positioning plate in the material box to obtain an image of the material on the positioning plate as the image to be processed; Step 4: Preprocess the image to be processed to obtain edge information in the image to be processed. Perform hand-eye calibration based on the edge information in the image to be processed to obtain the center coordinates of the positioning marker of the image to be processed in the robot coordinate system. Compare the center coordinates of the image to be processed with the center coordinates of the reference template image to obtain the center coordinate difference (Δx, Δy) and rotation angle difference Δθ of the positioning plate of the image to be processed. Step 5: If the center coordinate difference (Δx, Δy) and the rotation angle difference Δθ of the positioning plate of the image to be processed are within the set threshold, proceed to step 6; Step 6: Based on the position information of each groove in the reference template image, calculate the center position coordinates (X n , Y n ), n represents the quantity of materials; Step 7: Based on the center coordinate difference (Δx, Δy) of the positioning plate of the image to be processed and the rotation angle difference Δθ, the center position coordinates (X n , Y n ), calculate the pose deviation of all materials in the image to be processed and the center position of all materials in the reference image template ( , ), n represents the quantity of materials; Step 8: According to the industrial robot benchmark above each groove, grab the pose data and find the pose deviation of the corresponding center position ( , ), adjust the end effector of the industrial robot, and after the adjustment, the end effector of the industrial robot grabs all materials in sequence.

2. A bearing positioning method for a rail vehicle door hanger assembly workstation according to claim 1, characterized in that: The threshold for the center coordinate difference Δx is ±15 mm, the threshold for Δy is ±15 mm, and the threshold for the rotation angle difference Δθ is ±1.5°.

3. The method for positioning bearings in a rail vehicle door hanger assembly workstation according to claim 1, wherein: X n =P 中心距 *Cosα+(j 列 -1)* D 列距 Y n =P 中心距 *Sinα+(i 行 -1)*D 行距 Where α is the angle between the line connecting the center of the first row and first column groove of the positioning plate and the center of the positioning mark of the positioning plate and the X axis of the robot coordinate system, D 行距 D is the distance between the centers of adjacent grooves in each row. 列距 is the distance between the centers of adjacent grooves in each row, i 行 is the row number of the groove where the nth material is located, j 列 is the number of columns of the groove where the nth material is located, P 中心距 It is the distance between the center position of the groove in the first row and first column of the positioning plate and the center of the positioning mark of the positioning plate.

4. A bearing positioning method for a rail vehicle door hanger assembly workstation according to claim 3, characterized in that: ； 。

Images