Self-lubricating bearing gasket automatic nondestructive testing system based on visual mechanical arm

Abstract

The invention provides a self-lubricating bearing gasket automatic nondestructive testing system based on a visual mechanical arm, which is used for terahertz wave detection of a workpiece, and comprises a visual mechanical arm control system for moving the workpiece, and a terahertz system for performing terahertz imaging and analysis on the workpiece, the detected surface of the workpiece is larger than a scanning imaging window of the terahertz system; when a workpiece is detected, a mechanical arm of the visual mechanical arm control system clamps a workpiece sample, so that a detected surface of the sample intermittently moves at a scanning imaging window of the terahertz system, the detected surface is locally scanned and imaged by the terahertz system after each movement, and then each local image is spliced; a detected surface integral image capable of being used for terahertz analysis is formed; according to the invention, the detection efficiency can be improved, positioning errors caused by manual operation are greatly avoided, the image splicing effect is improved, the safety of detection personnel is guaranteed, and automatic detection of complex parts is realized.

Description

Technical Field

[0001] The invention relates to the technical field of nondestructive testing, in particular to an automatic nondestructive testing system for a self-lubricating bearing liner based on a visual mechanical arm. Background Art

[0002] Self-lubricating bearings are bearings that do not require the addition of lubricants. They have many advantages such as maintenance-free, low noise, strong load-bearing and impact resistance, and have become an important basic moving component widely used in my country's high-end equipment manufacturing industry. At present, self-lubricating bearings in high-precision fields such as aerospace and military equipment mainly use polytetrafluoroethylene polymer alloy liners to achieve self-lubrication. It is a homogeneous composite material formed by mixing a matrix resin with a large amount of PTFE fibers, high-performance reinforcing fibers, etc. It is a key component of self-lubricating bearings and directly determines the performance level and service life of the bearings. Therefore, conducting non-destructive testing and evaluation research on self-lubricating bearing liners has become the key to improving the quality of self-lubricating bearings and ensuring the safe operation of equipment.

[0003] In recent years, with the development of a series of new technologies and new materials, terahertz nondestructive testing has become a hot topic and emerging frontier in global research. Terahertz refers to electromagnetic waves with a frequency range of 0.1 to 10 THz. As a frequency band between microwaves and infrared, terahertz waves have the characteristics of wide frequency domain, low radiation, and high transmittance. In particular, the vibration and rotation energy level transitions of most polar and organic molecules are in the terahertz frequency band, showing strong characteristic absorption and resonance in this frequency band, and have strong spectral analysis. Compared with traditional nondestructive testing methods, terahertz nondestructive testing technology is an emerging detection method that can be realized online, non-contact, and quantifiable, and is one of the important research directions in the field of nondestructive testing in the future. However, due to the limited size of the scanning imaging window, the traditional terahertz time-domain spectroscopy system cannot realize the overall scanning imaging of the end face liner of a large-size self-lubricating bearing. Using the characteristics of the high positioning accuracy of the robotic arm to realize the self-use parallel movement of the sample to realize the local scanning of the sample can greatly avoid the positioning error caused by manual operation and improve the image stitching effect. Therefore, it is of great significance to design a visual robotic arm suitable for universities, scientific research institutions, enterprises and other units to use according to their teaching, research and production needs, which has the characteristics of low cost, simple structure, convenient debugging, easy operation and meeting the experimental precision requirements. Summary of the invention

[0004] The present invention proposes an automatic nondestructive testing system for self-lubricating bearing pads based on a visual manipulator, which has strong scalability, is easy to debug, and can meet the needs of automatic loading and unloading and high-precision positioning of samples during testing in conjunction with terahertz spectrum detection equipment. The system is based on a six-degree-of-freedom visual manipulator, and realizes motion control of the manipulator through command communication between a PC host computer and a slave computer and communication between the slave computer and a visual module. The manipulator control system is used to realize automatic loading and unloading of samples and monitoring of the testing process during nondestructive testing, which can improve testing efficiency, greatly avoid positioning errors caused by manual operation, improve image stitching effects, ensure the safety of testing personnel, and realize automated testing of complex parts.

[0005] The present invention adopts the following technical solutions.

[0006] A self-lubricating bearing pad automatic nondestructive testing system based on a visual mechanical arm is used for terahertz wave testing of workpieces. The testing system includes a visual mechanical arm control system for moving the workpiece, and also includes a terahertz system for performing terahertz imaging and analysis on the workpiece; the detected surface of the workpiece is larger than the scanning imaging window of the terahertz system; when the workpiece is tested, the mechanical arm of the visual mechanical arm control system clamps the workpiece sample, so that the detected surface of the sample moves intermittently at the scanning imaging window of the terahertz system, and after each movement, the detected surface is locally scanned and imaged by the terahertz system, and each local image is spliced ​​to form an overall image of the detected surface that can be used for terahertz analysis.

[0007] The visual robotic arm control system is a LabVIEW-based visual robotic arm control system, including a six-degree-of-freedom articulated visual clamping robotic arm driven by a servo, a power adapter, an STM32 slave microcontroller, a PC host computer, and a USB data transmission cable;

[0008] The PC host computer adopts LabVIEW to build the host computer interface, and the interface adopts a modular layout architecture, including a steering gear control module, a jog control module, an instruction storage and execution module, a real-time position display module, an image transmission module and an image stitching module;

[0009] When the visual robotic arm control system is working, the motion control of the robotic arm is realized through the command communication between the PC host computer and the slave computer, and the communication between the slave computer and the visual module of the robotic arm, so as to realize the automatic loading and unloading of samples and the monitoring of the detection process during non-destructive testing.

[0010] The servo-driven six-degree-of-freedom visual gripping robot arm is composed of a robot arm gimbal, a joint servo, a bracket, a joint connector, a gripper and an OpenMV visual module, wherein the joint servo adopts a bus servo, and different types of bus servos are used at different joint positions of the robot arm, namely, ZX15S bus servo, ZX15D bus servo and ZX20D bus servo, respectively. The OpenMV visual module includes OpenMV4Cam H7 Plus, an adapter board and an OpenMVWIFI expansion board;

[0011] The model selected by the described STM32 lower computer microcontroller is the STM32C8T6 minimum system board. The lower computer realizes data transmission with the host computer and the OpenMV visual module through a USB data cable, realizes command control through pre-designed communication instructions, realizes steering gear control by outputting PWM signals, and realizes inching control through a kinematic analysis program. The STM32 lower computer microcontroller can store instructions in advance at runtime and subsequently call the stored action instructions.

[0012] When the workpiece is a self-lubricating bearing and the inspected surface of the workpiece is a self-lubricating sleeve gasket, the working process of the non-destructive testing system includes the following steps:

[0013] Step S1: Place the robot arm at the selected position and press the bottom plate to make the suction cup on the bottom plate adhere to the workbench to fix the robot arm. Use the power adapter to power the lower computer. If the initial PWM pulse width in the lower computer program is 1500μs and the pulse width range of the servo is 500-2500μs, then when the robot arm is powered on, all servos will return to the middle position, so that the robot arm will be in a vertical state.

[0014] Step S2: Use a data cable to connect the PC host computer and the STM32 slave computer. The slave computer and the OpenMV visual module are also connected by a data cable for data transmission.

[0015] Step S3: Open the upper computer program interface, start running, and select the corresponding port for data transmission with the lower computer;

[0016] Step S4: according to the current position of the robot arm and the position of the sample to be clamped, the action group of the robot arm is set, that is, the robot arm is moved from the current position until it clamps the sample, and then the sample is placed on the detection workbench. Assuming that the sample needs to be moved four times on the workbench to complete the entire scan of the detected surface due to the limitation of the scanning window, after each movement, a quarter of the pad is scanned, and finally the sample is released and the robot arm is reset. In this step, the action of the robot arm is decomposed into multiple action groups to be completed, so as to avoid the robot arm hitting obstacles during the movement and causing damage to the experimental equipment;

[0017] Step S5: Enter the image transmission module in the host computer interface and start transmission. After the PC is connected to the local area network generated by the OpenMV vision module, the working image of the robotic arm can be observed through the host computer interface;

[0018] Step S6: After the robotic arm clamps the bearing part to the inspection position, it translates the sample four times on the workbench, and obtains four local scan images of the inspected surface of the liner through window scanning. The four local scan images are stitched together through image stitching technology to obtain a complete liner scan image. Finally, the terahertz time-domain spectrometer is used to perform imaging analysis and determine whether the self-lubricating sleeve liner sample has defects.

[0019] In step S4, the action group of the robot arm is set according to the current position of the robot arm and the position of the sample to be clamped. The specific process is: according to the current position of the robot arm, under the premise that there is no collision with the experimental equipment, the angles of the servos of each joint of the robot arm are adjusted one by one until a satisfactory position is reached, and then the angles of the servos of each joint are recorded as the action group. The number of action groups is set according to the operation requirements;

[0020] When adjusting the angles of the joint servos one by one, the conversion formula between the joint servo angle and the PWM pulse width of the lower computer is:

[0021] PWM=1500-1000*a / 270

[0022] Where PWM refers to the output signal pulse width, and a refers to the rotation angle input by the host computer interface;

[0023] The plus and minus signs in the conversion formula are adjusted according to the different installation methods of the servo. If the servo is installed in reverse, a minus sign is used in the formula.

[0024] The specific implementation method of the step S5, observing the image screen on the upper computer interface is: the upper computer sends instructions to the lower computer through the serial port, and the lower computer sends instructions to the OpenMV visual module after receiving the instructions. Through the WIFI expansion board of the OpenMV visual module, the OpenMV visual module generates a new local area network. When the computer is connected to the local area network, enter the specified URL on the PC computer, and you can see the transmitted image on the computer web page. Finally, localize the OpenMV visual module camera through the network camera adapter, and you can find the camera port at the monitoring-related interface of the upper computer interface. Select the port for monitoring, and you can observe the movement process of the sample in the image transmission module and visually monitor the detection process.

[0025] In step S6, the specific implementation method of combining the image stitching technology to obtain a complete pad scan image is: first, the obtained partial scan image is cut to obtain a sector with a central angle of 90°, and then the cut partial scan images are stitched using image processing software to obtain a complete pad scan image.

[0026] The specific implementation method of the step S6, in which the bearing workpiece is accurately positioned during the four movements, is that the mechanical arm has a jog function, that is, after inputting the point coordinates, the mechanical arm can start to move until the end gripper of the mechanical arm reaches the point coordinates;

[0027] The origin of the spatial coordinate system is located at the center of the robot arm base, the front of the robot arm is the positive direction of the y-axis, according to the Cartesian coordinate system, the right side of the robot arm is the positive direction of the x-axis, and the top of the origin is the positive direction of the z-axis. In this step, the coordinates of the sample in the robot arm coordinate system are first measured, and then the point coordinates are input to move the robot arm, so as to obtain the rotation angle of each joint of the robot arm, and then the action group is set. By setting the action group, it is avoided that the point coordinates directly input during the movement of the robot arm will collide with the experimental equipment of the self-lubricating bearing pad automatic non-destructive testing system and cause losses.

[0028] The specific implementation method of the robot arm jog function is: after determining the spatial coordinate system of the robot arm, when the target point coordinates and the structural dimensions of the robot arm are known, the required rotation angle of each joint can be obtained through inverse kinematics calculation, and the angle is converted into the PWM pulse width value to be set and then the PWM signal is output to the steering gear to make the robot arm move. The calculation formula of the joint angle is:

[0029]

[0030] z-L0-L3*sin(Alpha)=L5

[0031] arctan(x / y)*180 / π=θ6

[0032]

[0033] 180-arccos[(L1 2 +L2 2 -L4 2 -L5 2 ) / (2*L1*L2)]*180 / π=θ4θ5-θ4-Alpha=θ3

[0034] Among them, x, y, z refer to the coordinates of the target point, L1, L2, L3, L4, L5 refer to the structural dimensions of the robot arm, and θ3, θ4, θ5, θ6 in the formula refer to the rotation angles of the joints.

[0035] The support, connectors, grippers and gimbal of the robotic arm are all made of aluminum alloy.

[0036] The significant advantages of the present invention are:

[0037] (1) The robotic arm is used to realize automatic loading and unloading during nondestructive testing, which effectively overcomes the limitation of the traditional terahertz time-domain spectroscopy system that cannot realize the overall scanning imaging of the end face gasket of large-size self-lubricating bearings due to the limited size of the scanning imaging window. It also greatly avoids the positioning error caused by manual operation, improves the image stitching effect, improves the efficiency of detection, and reduces the position error caused by manual loading.

[0038] (2) Through wireless image transmission, the detection process can be monitored in real time, allowing researchers to remotely monitor the monitoring process. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention is further described in detail below with reference to the accompanying drawings and specific embodiments:

[0040] Attached Figure 1 A schematic diagram of the system structure of the invention;

[0041] Attached Figure 2 A cross-sectional view of a self-lubricating bearing component of an embodiment of the invention;

[0042] Attached Figure 3 This is a schematic diagram of the robot arm holding parts;

[0043] Attached Figure 4 It is a schematic diagram of the operation of the embodiment of the invention;

[0044] Attached Figure 5 Partial imaging and image stitching effect of self-lubricating bearing end face liner;

[0045] In the figure: 1- mechanical arm; 2- self-lubricating bearing parts; 3- detection platform. DETAILED DESCRIPTION

[0046] As shown in the figure, a self-lubricating bearing pad automatic non-destructive testing system based on a visual robotic arm is used for terahertz wave detection of workpieces. The detection system includes a visual robotic arm control system for moving the workpiece, and also includes a terahertz system for terahertz imaging and analysis of the workpiece; the inspected surface of the workpiece is larger than the scanning imaging window of the terahertz system; when the workpiece is inspected, the robotic arm of the visual robotic arm control system clamps the workpiece sample so that the inspected surface of the sample moves intermittently at the scanning imaging window of the terahertz system, and after each movement, the inspected surface is locally scanned and imaged by the terahertz system, and then the local images are spliced ​​to form an overall image of the inspected surface that can be used for terahertz analysis.

[0047] The visual robotic arm control system is a LabVIEW-based visual robotic arm control system, including a six-degree-of-freedom articulated visual clamping robotic arm driven by a servo, a power adapter, an STM32 slave microcontroller, a PC host computer, and a USB data transmission cable;

[0048] The PC host computer adopts LabVIEW to build the host computer interface, and the interface adopts a modular layout architecture, including a steering gear control module, a jog control module, an instruction storage and execution module, a real-time position display module, an image transmission module and an image stitching module;

[0049] When the visual robotic arm control system is working, the motion control of the robotic arm is realized through the command communication between the PC host computer and the slave computer, and the communication between the slave computer and the visual module of the robotic arm, so as to realize the automatic loading and unloading of samples and the monitoring of the detection process during non-destructive testing.

[0050] The servo-driven six-degree-of-freedom visual gripping robot arm is composed of a robot arm gimbal, a joint servo, a bracket, a joint connector, a gripper and an OpenMV visual module, wherein the joint servo adopts a bus servo, and different types of bus servos are used at different joint positions of the robot arm, namely, ZX15S bus servo, ZX15D bus servo and ZX20D bus servo, respectively. The OpenMV visual module includes OpenMV4Cam H7 Plus, an adapter board and an OpenMVWIFI expansion board;

[0051] The model selected by the described STM32 lower computer microcontroller is the STM32C8T6 minimum system board. The lower computer realizes data transmission with the host computer and the OpenMV visual module through a USB data cable, realizes command control through pre-designed communication instructions, realizes steering gear control by outputting PWM signals, and realizes inching control through a kinematic analysis program. The STM32 lower computer microcontroller can store instructions in advance at runtime and subsequently call the stored action instructions.

[0052] When the workpiece is a self-lubricating bearing and the inspected surface of the workpiece is a self-lubricating sleeve gasket, the working process of the non-destructive testing system includes the following steps:

[0053] Step S1: Place the robot arm at the selected position and press the bottom plate to make the suction cup on the bottom plate adhere to the workbench to fix the robot arm. Use the power adapter to power the lower computer. If the initial PWM pulse width in the lower computer program is 1500μs and the pulse width range of the servo is 500-2500μs, then when the robot arm is powered on, all servos will return to the middle position, so that the robot arm will be in a vertical state.

[0054] Step S2: Use a data cable to connect the PC host computer and the STM32 slave computer. The slave computer and the OpenMV visual module are also connected by a data cable for data transmission.

[0055] Step S3: Open the upper computer program interface, start running, and select the corresponding port for data transmission with the lower computer;

[0056] Step S4: according to the current position of the robot arm and the position of the sample to be clamped, the action group of the robot arm is set, that is, the robot arm is moved from the current position until it clamps the sample, and then the sample is placed on the detection workbench. Assuming that the sample needs to be moved four times on the workbench to complete the entire scan of the detected surface due to the limitation of the scanning window, after each movement, a quarter of the pad is scanned, and finally the sample is released and the robot arm is reset. In this step, the action of the robot arm is decomposed into multiple action groups to be completed, so as to avoid the robot arm hitting obstacles during the movement and causing damage to the experimental equipment;

[0057] Step S5: Enter the image transmission module in the host computer interface and start transmission. After the PC is connected to the local area network generated by the OpenMV vision module, the working image of the robotic arm can be observed through the host computer interface;

[0058] Step S6: After the robotic arm clamps the bearing part to the inspection position, it translates the sample four times on the workbench, and obtains four local scan images of the inspected surface of the liner through window scanning. The four local scan images are stitched together through image stitching technology to obtain a complete liner scan image. Finally, the terahertz time-domain spectrometer is used to perform imaging analysis and determine whether the self-lubricating sleeve liner sample has defects.

[0059] In step S4, the action group of the robot arm is set according to the current position of the robot arm and the position of the sample to be clamped. The specific process is: according to the current position of the robot arm, under the premise that there is no collision with the experimental equipment, the angles of the servos of each joint of the robot arm are adjusted one by one until a satisfactory position is reached, and then the angles of the servos of each joint are recorded as the action group. The number of action groups is set according to the operation requirements;

[0060] When adjusting the angles of the joint servos one by one, the conversion formula between the joint servo angle and the PWM pulse width of the lower computer is:

[0061] PWM=1500-1000*a / 270

[0062] Where PWM refers to the output signal pulse width, and a refers to the rotation angle input by the host computer interface;

[0063] The plus and minus signs in the conversion formula are adjusted according to the different installation methods of the servo. If the servo is installed in reverse, a minus sign is used in the formula.

[0064] The specific implementation method of the step S5, observing the image screen on the upper computer interface is: the upper computer sends instructions to the lower computer through the serial port, and the lower computer sends instructions to the OpenMV visual module after receiving the instructions. Through the WIFI expansion board of the OpenMV visual module, the OpenMV visual module generates a new local area network. When the computer is connected to the local area network, enter the specified URL on the PC computer, and you can see the transmitted image on the computer web page. Finally, localize the OpenMV visual module camera through the network camera adapter, and you can find the camera port at the monitoring-related interface of the upper computer interface. Select the port for monitoring, and you can observe the movement process of the sample in the image transmission module and visually monitor the detection process.

[0065] In step S6, the specific implementation method of combining the image stitching technology to obtain a complete pad scan image is: first, the obtained partial scan image is cut to obtain a sector with a central angle of 90°, and then the cut partial scan images are stitched using image processing software to obtain a complete pad scan image.

[0066] The specific implementation method of the step S6, in which the bearing workpiece is accurately positioned during the four movements, is that the mechanical arm has a jog function, that is, after inputting the point coordinates, the mechanical arm can start to move until the end gripper of the mechanical arm reaches the point coordinates;

[0067] The origin of the spatial coordinate system is located at the center of the robot arm base, the front of the robot arm is the positive direction of the y-axis, according to the Cartesian coordinate system, the right side of the robot arm is the positive direction of the x-axis, and the top of the origin is the positive direction of the z-axis. In this step, the coordinates of the sample in the robot arm coordinate system are first measured, and then the point coordinates are input to move the robot arm, so as to obtain the rotation angle of each joint of the robot arm, and then the action group is set. By setting the action group, it is avoided that the point coordinates directly input during the movement of the robot arm will collide with the experimental equipment of the self-lubricating bearing pad automatic non-destructive testing system and cause losses.

[0068] The specific implementation method of the robot arm jog function is: after determining the spatial coordinate system of the robot arm, when the target point coordinates and the structural dimensions of the robot arm are known, the required rotation angle of each joint can be obtained through inverse kinematics calculation, and the angle is converted into the PWM pulse width value to be set and then the PWM signal is output to the steering gear to make the robot arm move. The calculation formula of the joint angle is:

[0069]

[0070] z-L0-L3*sin(Alpha)=L5

[0071] arctan(x / y)*180 / π=θ6

[0072]

[0073] 180-arccos[(L1 2 +L2 2 -L4 2 -L5 2 ) / (2*L1*L2)]*180 / π=θ4

[0074] θ5-θ4-Alpha=θ3

[0075] Among them, x, y, z refer to the coordinates of the target point, L1, L2, L3, L4, L5 refer to the structural dimensions of the robot arm, and θ3, θ4, θ5, θ6 in the formula refer to the rotation angles of the joints.

[0076] The support, connectors, grippers and gimbal of the robotic arm are all made of aluminum alloy.

[0077] Example:

[0078] like Figure 1 As shown, a self-lubricating bearing pad automatic nondestructive testing system based on a visual robot arm includes a PC host computer 1, an OpenMV visual module 2, an STM32 microcontroller 3, a six-degree-of-freedom visual clamping robot arm 4, and a terahertz spectrum analysis and detection device 5. The six-degree-of-freedom visual clamping robot arm 4 is composed of a robot arm gimbal, a joint servo, a bracket, a joint connector, and a gripper, wherein the joint servo adopts a bus servo, and three types of bus servos are used according to different joint positions, namely, ZX15S bus servo, ZX15D bus servo, and ZX20OD bus servo. In addition, the OpenMV, visual module includes OpenMV4CamH7Plus, an adapter board, and an OpenMVWIFI expansion board, and the bracket, connector, gripper, and gimbal of the robot arm are all made of aluminum alloy. The model of the STM32 lower computer microcontroller is the STM32C8T6 minimum system board. The lower computer transmits data with the upper computer and OpenMV through the USB data cable, and the visual module. The command control is realized through the designed communication command, the cabin machine control is realized by outputting PWM signals, and the jog control is realized through the kinematic analysis program. In addition, the instructions can be stored in advance and the stored action instructions can be called later. The PC upper computer uses LabVIEW to build the upper computer interface. The interface adopts a modular construction idea, including the energy machine control module, the jog control module, the instruction storage and execution module, the real-time position display module, and the image transmission module.

[0079] When performing automatic nondestructive testing of self-lubricating bearing pads, the robot arm is placed at the selected position according to the initial position of the sample and the bottom plate is pressed to make the suction cup on the bottom plate adsorbed to the workbench to fix the robot arm. The power adapter is used to power the lower computer. Since the initial PWM pulse width in the lower computer program is 1500μs and the pulse width range of the servo is 500-2500μs, when the robot hip is powered on, all servos will return to the middle position. From the overall point of view, the robot arm will be in a vertical state. After setting up the action group, the robot arm clamps the part and moves the part to the detection platform for local scanning and detection. After completing the local scanning and detection in four times, the obtained local scanning images are spliced ​​to obtain a complete scanning image of the self-lubricating bearing end face pad.

[0080] Performance evaluation of an automatic nondestructive testing system for self-lubricating bearing pads based on a visual robotic arm:

[0081] The performance evaluation of a self-lubricating bearing pad automatic nondestructive testing system developed based on a visual robot arm is carried out by placing the robot arm in a fixed position at a known starting position of the sample part and moving the sample from the starting position to the specified testing position.

[0082] According to the distance between the starting position and the designated detection position and the obstacles during the movement, the action group of the robot arm is reasonably set to place the sample smoothly to the target detection position while avoiding accidental collision with the experimental instrument.

[0083] Self-lubricating bearing parts cross-section diagram Figure 2 As shown, it consists of a metal base and a self-lubricating bearing liner.

[0084] The robot arm picks up the parts, such as Figure 3 As shown in Figure 2, after the robot arm places the sample at the detection position, Figure 4 As shown, Figure 4 b is a schematic diagram of the detection platform. Figure 4 Schematic diagram of sample movement of a. Due to the limitation of the scanning window, the sample needs to be moved four times on the workbench, namely Figure 4 ①②③④ in a, scan one quarter of the pad each time. The high-precision characteristics of the robotic arm can greatly avoid the positioning error caused by manual operation. By controlling the image overlap area of ​​each scan, the image stitching effect and stitching efficiency are improved. The detection equipment can be used to conduct defect imaging research on the self-lubricating bearing sample pad, such as Figure 5 As shown, it can be observed in the imaging that the graphics corresponding to the defects formed by the laser marking on the self-lubricating bearing sample are accurate in position. Figure 5 A real picture of the laser marking defect of a self-lubricating bearing liner. Figure 5 b is the scanned image obtained by a single local scan. Figure 5c is the complete end face scan result obtained by stitching the results of four local scans. Through higher resolution imaging, small defects other than laser marking on the test sample can be observed, and further qualitative and quantitative analysis can be performed by obtaining the defect shape and position.

[0085] Although the present invention has been described in detail through specific implementation methods, a person skilled in the art should understand that any changes in form and detail made on this basis that do not exceed the scope of protection of the claims shall fall within the scope of protection of the present invention.

Claims

1. An automatic nondestructive testing system for self-lubricating bearing pads based on a visual robot arm, used for terahertz wave testing of workpieces, characterized by: The detection system includes a visual robotic arm control system for moving the workpiece, and also includes a terahertz system for performing terahertz imaging and analysis on the workpiece; the detected surface of the workpiece is larger than the scanning imaging window of the terahertz system; when the workpiece is detected, the robotic arm of the visual robotic arm control system clamps the workpiece sample, so that the detected surface of the sample moves intermittently at the scanning imaging window of the terahertz system, and after each movement, the detected surface is locally scanned and imaged by the terahertz system, and then the local images are spliced ​​to form an overall image of the detected surface that can be used for terahertz analysis.

2. The automatic nondestructive testing system for self-lubricating bearing pads based on a visual robot arm according to claim 1, characterized in that: The visual robotic arm control system is a LabVIEW-based visual robotic arm control system, including a six-degree-of-freedom articulated visual clamping robotic arm driven by a servo, a power adapter, an STM32 slave microcontroller, a PC host computer, and a USB data transmission cable; The PC host computer adopts LabVIEW to build the host computer interface, and the interface adopts a modular layout architecture, including a steering gear control module, a jog control module, an instruction storage and execution module, a real-time position display module, an image transmission module and an image stitching module; When the visual robotic arm control system is working, the motion control of the robotic arm is realized through the command communication between the PC host computer and the slave computer, and the communication between the slave computer and the visual module of the robotic arm, so as to realize the automatic loading and unloading of samples and the monitoring of the detection process during non-destructive testing.

3. The automatic nondestructive testing system for self-lubricating bearing pads based on a visual robot arm according to claim 2, characterized in that: The six-degree-of-freedom visual gripping robot arm driven by the servo is composed of a robot arm gimbal, a joint servo, a bracket, a joint connector, a gripper and an OpenMV visual module, wherein the joint servo adopts a bus servo, and different types of bus servos are respectively used at different joint positions of the robot arm, namely, ZX15S bus servo, ZX15D bus servo and ZX20D bus servo, and the OpenMV visual module includes OpenMV4 Cam H7 Plus, an adapter board and an OpenMVWIFI expansion board; The model selected by the described STM32 lower computer microcontroller is the STM32C8T6 minimum system board. The lower computer realizes data transmission with the host computer and the OpenMV visual module through a USB data cable, realizes command control through pre-designed communication instructions, realizes steering gear control by outputting PWM signals, and realizes inching control through a kinematic analysis program. The STM32 lower computer microcontroller can store instructions in advance at runtime and subsequently call the stored action instructions.

4. The automatic nondestructive testing system for self-lubricating bearing pads based on a visual robot arm according to claim 2, characterized in that: When the workpiece is a self-lubricating bearing and the inspected surface of the workpiece is a self-lubricating sleeve gasket, the working process of the non-destructive testing system includes the following steps: Step S1: Place the robot arm at the selected position and press the bottom plate to make the suction cup on the bottom plate adhere to the workbench to fix the robot arm. Use the power adapter to power the lower computer. If the initial PWM pulse width in the lower computer program is 1500μs and the pulse width range of the servo is 500-2500μs, then when the robot arm is powered on, all servos will return to the middle position, so that the robot arm will be in a vertical state. Step S2: Use a data cable to connect the PC host computer and the STM32 slave computer. The slave computer and the OpenMV visual module are also connected by a data cable for data transmission. Step S3: Open the upper computer program interface, start running, and select the corresponding port for data transmission with the lower computer; Step S4: according to the current position of the robot arm and the position of the sample to be clamped, the action group of the robot arm is set, that is, the robot arm is moved from the current position until it clamps the sample, and then the sample is placed on the detection workbench. Assuming that the sample needs to be moved four times on the workbench to complete the entire scan of the detected surface due to the limitation of the scanning window, after each movement, a quarter of the pad is scanned, and finally the sample is released and the robot arm is reset. In this step, the action of the robot arm is decomposed into multiple action groups to be completed, so as to avoid the robot arm hitting obstacles during the movement and causing damage to the experimental equipment; Step S5: Enter the image transmission module in the host computer interface and start transmission. After the PC is connected to the local area network generated by the OpenMV vision module, the working image of the robotic arm can be observed through the host computer interface; Step S6: After the robotic arm clamps the bearing part to the inspection position, it translates the sample four times on the workbench, and obtains four local scan images of the inspected surface of the liner through window scanning. The four local scan images are stitched together through image stitching technology to obtain a complete liner scan image. Finally, the terahertz time-domain spectrometer is used to perform imaging analysis and determine whether the self-lubricating sleeve liner sample has defects.

5. The automatic nondestructive testing system for self-lubricating bearing pads based on a visual robot arm according to claim 4, characterized in that: In step S4, the action group of the robot arm is set according to the current position of the robot arm and the position of the sample to be clamped. The specific process is: according to the current position of the robot arm, under the premise that there is no collision with the experimental equipment, the angles of the servos of each joint of the robot arm are adjusted one by one until a satisfactory position is reached, and then the angles of the servos of each joint are recorded as the action group. The number of action groups is set according to the operation requirements; When adjusting the angles of the joint servos one by one, the conversion formula between the joint servo angle and the PWM pulse width of the lower computer is: PWM=1500-1000*a / 270 Where PWM refers to the output signal pulse width, and a refers to the rotation angle input by the host computer interface; The plus and minus signs in the conversion formula are adjusted according to the different installation methods of the servo. If the servo is installed in reverse, a minus sign is used in the formula.

6. The automatic nondestructive testing system for self-lubricating bearing pads based on a visual robot arm according to claim 4, characterized in that: The specific implementation method of the step S5, observing the image screen on the upper computer interface is: the upper computer sends instructions to the lower computer through the serial port, and the lower computer sends instructions to the OpenMV visual module after receiving the instructions. Through the WIFI expansion board of the OpenMV visual module, the OpenMV visual module generates a new local area network. When the computer is connected to the local area network, enter the specified URL on the PC computer, and you can see the transmitted image on the computer web page. Finally, localize the OpenMV visual module camera through the network camera adapter, and you can find the camera port at the monitoring-related interface of the upper computer interface. Select the port for monitoring, and you can observe the movement process of the sample in the image transmission module and visually monitor the detection process.

7. The automatic nondestructive testing system for self-lubricating bearing pads based on a visual robot arm according to claim 4, characterized in that: In step S6, the specific implementation method of combining the image stitching technology to obtain a complete pad scan image is: first, the obtained partial scan image is cut to obtain a sector with a central angle of 90°, and then the cut partial scan images are stitched using image processing software to obtain a complete pad scan image.

8. The automatic nondestructive testing system for self-lubricating bearing pads based on a visual robot arm according to claim 4, characterized in that: The specific implementation method of the step S6, in which the bearing workpiece is accurately positioned during the four movements, is that the mechanical arm has a jog function, that is, after inputting the point coordinates, the mechanical arm can start to move until the end gripper of the mechanical arm reaches the point coordinates; The origin of the spatial coordinate system is located at the center of the robot arm base, the front of the robot arm is the positive direction of the y-axis, according to the Cartesian coordinate system, the right side of the robot arm is the positive direction of the x-axis, and the top of the origin is the positive direction of the z-axis. In this step, the coordinates of the sample in the robot arm coordinate system are first measured, and then the point coordinates are input to move the robot arm, so as to obtain the rotation angle of each joint of the robot arm, and then the action group is set. By setting the action group, it is avoided that the point coordinates directly input during the movement of the robot arm will collide with the experimental equipment of the self-lubricating bearing pad automatic non-destructive testing system and cause losses.

9. The automatic nondestructive testing system for self-lubricating bearing pads based on a visual robot arm according to claim 8, characterized in that: The specific implementation method of the robot arm jog function is: after determining the spatial coordinate system of the robot arm, when the target point coordinates and the structural dimensions of the robot arm are known, the required rotation angle of each joint can be obtained through inverse kinematics calculation, and the angle is converted into the PWM pulse width value to be set and then the PWM signal is output to the steering gear to make the robot arm move. The calculation formula of the joint angle is: z-L0-L3*sin(Alpha)=L5 arctan(x / y)*180 / π=θ6 180-arccos[(L1 2 +L2 2 -L4 2 -L5 2 ) / (2*L1*L2)]*180 / π=θ4 θ5-θ4-Alpha=θ3 Among them, x, y, z refer to the coordinates of the target point, L1, L2, L3, L4, L5 refer to the structural dimensions of the robot arm, and θ3, θ4, θ5, θ6 in the formula refer to the rotation angles of the joints.

10. The automatic nondestructive testing system for self-lubricating bearing pads based on a visual robot arm according to claim 3, characterized in that: The support, connectors, grippers and gimbal of the robotic arm are all made of aluminum alloy.

Images