A soldering method, system, device and storage medium for patch components

Abstract

The present invention relates to the field of computer technology, and discloses a soldering method, system, device and storage medium for patch components. The soldering method is applied to a computer, and the method comprises: obtaining at least one image of a board to be soldered, determining the size and position of each module according to reference information and position information, and marking and generating a first annotation map; inputting the first annotation map and the photo of the board to be soldered into a telemetry target detection model for processing, obtaining a second annotation map containing the module to be soldered, and determining the position coordinates of the module to be soldered according to the second annotation map; sending the position coordinates of the module to be soldered to an integrated control chip, so that the integrated control chip controls the double tweezers system and the laser welding assembly to solder the patch component to the corresponding position of the module to be soldered according to the position coordinates. Thereby greatly improving the accuracy of soldering, and solving the problem of resistor cold soldering and solder joint loss caused by personal mistakes of hardware R&D engineers.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a soldering method, system, equipment and storage medium for patch components. Background Art

[0002] In the process of soldering of SMT machines, such as the process of mass production of printed circuit boards (PCBs) in factories, after the machine is equipped with a dispensing machine and a screen printer, the surface mount components are quickly and accurately placed on the PCB pads by moving the mounting head. Then, the soldering of the SMT components is completed through soldering technologies such as wave soldering and reflow soldering. These technologies are mature in large-scale board production, but are not suitable for the replacement of SMT components during signal debugging after the board is completed.

[0003] In the actual debugging process, hardware R&D engineers need to adjust the resistors or capacitors and other chip components on the fault signal link according to the type and link of the fault signal to meet the requirements of line impedance matching and eliminate signal reflection. However, hardware R&D engineers often cannot obtain a good signal by adjusting the resistance value or capacitance value once. They may even need to adjust other resistors on the link where the fault signal is located to successfully complete the debugging, which increases the adjustment time and labor costs. In addition, when re-soldering the circuit, it is also possible that the resistors are poorly soldered or even the solder joints are lost due to the personal mistakes of the engineers, thus causing the fault signal. Summary of the invention

[0004] The present application provides a soldering method, system, device and storage medium for patch components, so as to at least solve the problems of resistance cold soldering generated during the debugging process causing fault signals, and increasing labor costs and R&D cycles.

[0005] The present application provides a soldering method for patch components, which is applied to computers. The method comprises:

[0006] Acquire at least one image of a board to be welded, the board to be welded is placed on a reference board of a welding station, and the image includes reference information on the reference board and position information of multiple modules on the board to be welded;

[0007] Determine the size and position of each module according to the reference information and the position information, and mark and generate a first marked image;

[0008] Inputting the first annotated image and the photo of the board to be welded into the telemetry target detection model for processing, obtaining a second annotated image including the module to be welded, and determining the position coordinates of the module to be welded according to the second annotated image;

[0009] The position coordinates of the module to be welded are sent to the integrated control chip, so that the integrated control chip controls the double tweezers system and the laser welding assembly to weld the patch component to the corresponding position of the module to be welded according to the position coordinates.

[0010] Optionally, in a possible implementation manner, the method further includes:

[0011] Inputting at least one preset characteristic module into the computer, the at least one characteristic module comprising: a power connector module, a silk screen module, an electrolytic capacitor module, a heat sink module and a chip capacitor and resistor module;

[0012] Inputting the first annotated image and the photo of the board to be welded into the telemetry target detection model for processing, obtaining a second annotated image including the module to be welded, and determining the position coordinates of the module to be welded according to the second annotated image, including:

[0013] Inputting the first labeled image and the photo of the board to be welded into the telemetry target detection model, performing global average pooling and maximum pooling layer processing on the first labeled image, and obtaining global features and local features that match at least one preset feature module;

[0014] According to the global features and the local features, a second annotation graph including the module to be welded is obtained;

[0015] The position coordinates of the module to be welded are determined according to the second marked image and the first marked image.

[0016] Optionally, in another possible implementation, before performing global average pooling and maximum pooling layer processing on the first labeled image, the method further includes:

[0017] Get training samples;

[0018] Determine the loss function based on the training samples and the method of joint optimization of classification and regression;

[0019] Determine the total loss function of the training sample based on the loss function and the stochastic gradient descent algorithm;

[0020] The model parameters are trained according to the total loss function to obtain the telemetry target detection model.

[0021] Optionally, in another possible implementation manner, the at least one image includes a first group of images and a second group of images; acquiring at least one image of the board to be welded, including:

[0022] Acquire a first group of images taken by a first camera from a first perspective on the board to be welded, and acquire a second group of images taken by a second camera from a second perspective on the board to be welded;

[0023] The first camera is located directly above the board to be welded and forms a first angle with each corner of the reference board on the z-axis; the second camera is located flush with the board to be welded.

[0024] Optionally, in another possible implementation manner, before acquiring at least one image of the board to be welded, the method further includes:

[0025] Get the status of the dual tweezers system;

[0026] According to the status, check whether the dual tweezers system has completed the pick-up operation. The pick-up operation means that the tweezers head of the dual tweezers system has clamped the patch component and is located at the specified position;

[0027] If not, a control instruction is sent to the integrated control chip so that the integrated control chip controls the double-tweezers system to complete the object picking operation.

[0028] The present application also provides another method for soldering patch components, which is applied to an integrated control chip. The method comprises:

[0029] Receiving position coordinates of the module to be welded, where the position coordinates are generated by a computer according to at least one image on the board to be welded;

[0030] Parsing the position coordinates to determine the actual position of the module to be welded on the board to be welded, and generating at least one set of pulse signals;

[0031] At least one group of pulse signals is sent to the dual tweezers system and the laser welding assembly, and the pulse signals are used to control the dual tweezers system and the laser welding assembly to weld the patch components to the corresponding positions of the modules to be welded.

[0032] Optionally, in a possible implementation manner, at least one group of pulse signals includes a first group of pulse signals and a second group of pulse signals; sending at least one group of pulse signals to the dual-tweezers system and the laser welding assembly includes:

[0033] Sending a first group of pulse signals to a servo motor of the dual-tweezers system to control the servo motor to drive the tweezers head of the dual-tweezers system to perform a welding task;

[0034] The second group of pulse signals is sent to the servo motor of the laser welding component to control the servo motor to drive the laser head of the laser welding component to perform the welding task.

[0035] The present application also provides a patch component welding system, which includes: a computer, an integrated control chip, a double tweezers system, a laser welding assembly, at least two cameras, a board to be welded, a patch component and a welding table;

[0036] The welding table includes a reference board on which the board to be welded is placed;

[0037] The double tweezers system and the laser welding assembly are both movably connected to the welding table;

[0038] A first camera of the at least two cameras is located directly above the board to be welded and forms a first angle with each corner of the reference board on the z-axis, and a second camera is located flush with the board to be welded;

[0039] The computer is connected to the integrated control chip, and the integrated control chip is connected to the double tweezers system and the laser welding component;

[0040] The first camera and the second camera are used to capture at least one image on the board to be welded and send it to the computer;

[0041] The computer is used to execute the aforementioned soldering method for a patch component; the integrated control chip is used to execute the aforementioned soldering method for a patch component;

[0042] The double tweezers system and the laser welding assembly are used to receive at least one set of pulse signals from the integrated control chip, and weld the patch components clamped by the tweezers head of the double tweezers system to the corresponding position of the module to be welded, and the module to be welded is the module on the board to be welded.

[0043] The present application also provides a soldering device for patch components, the device comprising:

[0044] An acquisition module is used to acquire at least one image of a board to be welded, the board to be welded is placed on a reference board of a welding station, and the image includes reference information on the reference board and position information of multiple modules on the board to be welded;

[0045] A marking module, used to determine the size and position of each module according to the reference information and the position information, and mark and generate a first marking map;

[0046] A first processing module is used to input the first annotated image and the photo of the board to be welded into the telemetry target detection model for processing, obtain a second annotated image containing the module to be welded, and determine the position coordinates of the module to be welded according to the second annotated image;

[0047] The first sending module is used to send the position coordinates of the module to be welded to the integrated control chip, so that the integrated control chip controls the double tweezers system and the laser welding assembly to weld the patch component to the corresponding position of the module to be welded according to the position coordinates.

[0048] The present application also provides another soldering device for patch components, the device comprising:

[0049] A receiving module, used for receiving the position coordinates of the module to be welded, where the position coordinates are generated by a computer according to at least one image on the board to be welded;

[0050] A second processing module is used to analyze the position coordinates to determine the actual position of the module to be welded on the board to be welded, and generate at least one group of pulse signals;

[0051] The second sending module is used to send at least one group of pulse signals to the double tweezers system and the laser welding assembly, and the pulse signals are used to control the double tweezers system and the laser welding assembly to weld the patch components to the corresponding positions of the modules to be welded.

[0052] The present application also provides an electronic device, comprising: a memory for storing a computer program; and a processor for implementing the steps of any of the above-mentioned methods for soldering patch components when executing the computer program.

[0053] The present application also provides a computer-readable storage medium, in which a computer program is stored, wherein when the computer program is executed by a processor, the steps of any of the above-mentioned methods for soldering patch components are implemented.

[0054] The present application also provides a computer program product, including a computer program, which implements the steps of any of the above-mentioned soldering methods for patch components when the computer program is executed by a processor.

[0055] The present application provides a method for soldering patch components. By taking an image of a board to be soldered and using reference information and position information to determine the size and position of each module, and then processing it through a telemetry target detection model, the position of the module to be soldered can be accurately marked, thereby greatly improving the soldering accuracy and solving the problem of resistor cold soldering and solder joint loss caused by personal errors of hardware R&D engineers.

[0056] In addition, this method realizes automatic identification and positioning of the modules to be welded, reduces the steps of manual intervention, thus speeding up the welding speed, improving production efficiency and saving labor costs. At the same time, the integrated control chip directly controls the dual tweezers system and laser welding components to weld according to the position coordinates, further shortening the welding cycle.

[0057] The welding system provided in the present application includes a laser welding assembly, which can be applied to laboratory signal testing and tuning, and has low cost and high reliability. In addition, it can also quickly replace small-scale patch resistors and capacitors. At the same time, the telemetry target detection model can detect and replace the spatial position of the assembly, generate a second annotation map containing the module to be welded, and determine the position coordinates of the module to be welded based on the second annotation map, and finally send the position coordinates to the integrated control chip to control the dual tweezers system and the laser welding assembly to complete the position movement and welding tasks of the patch components.

[0058] Since the above-mentioned laser welding assembly is a laser welding assembly that can realize 360° rotation on the x-plane. The laser welding assembly is internally configured with a temperature sensor, a distance sensor, etc. After the processor in the laser welding assembly receives the returned data, it controls the power transmission to realize the temperature change of the focal point. It has a safer temperature range than the industrial grade and a simpler and more comprehensive mobile structure. The method is applied to a portable automatic spot welding machine to complete the welding work of the resistor and capacitor solder joints, and has the beneficial effects of firm solder joints, less pollution, and fast speed.

[0059] The above system of the present application also includes a dual tweezers system, which includes telescopic rods and servo motors in the x, y, and z axis directions. The servo motors of the telescopic rods in each direction are controlled by an integrated control chip, and combined with the photo upload function of at least two cameras, the system as a whole has a mobile architecture for mobile feedback monitoring. The servo motor receives fine pulses to achieve precise movement, eliminating the error limit of the DC motor. At the same time, the telemetry target detection model is used to detect the moving distance, combined with the servo motor feedback system, to achieve precise position control. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0061] Figure 1 A structural diagram of a soldering system for a patch component provided in an embodiment of the present application;

[0062] Figure 2 A structural diagram of a welding system including patch components of a board to be welded provided in an embodiment of the present application;

[0063] Figure 3 A schematic diagram of a process for soldering a patch component provided in an embodiment of the present application;

[0064] Figure 4 A schematic diagram of the structure of a double tweezers system provided in an embodiment of the present application;

[0065] Figure 5 A schematic flow chart of another method for soldering a patch component provided in an embodiment of the present application;

[0066] Figure 6 A schematic flow chart of another method for soldering a patch component provided in an embodiment of the present application;

[0067] Figure 7A schematic diagram of a remote sensing target detection model processing process provided in an embodiment of the present application;

[0068] Figure 8 A structural block diagram of a welding device provided in an embodiment of the present application;

[0069] Fig. 9 A schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present application;

[0070] Among them, there are a computer 10, an integrated control chip 20, a dual tweezers system 30, a tweezers head 31, servo motors G1 to G3 on the dual tweezers system 30, telescopic rods L1 and L2, a laser welding assembly 40, a welding head 41, servo motors G4 to G6 on the laser welding assembly 40, a first camera 51, a second camera 52, a welding table 60, a reference board 61, a receiving box 70 and a board 100 to be welded. DETAILED DESCRIPTION

[0071] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application.

[0072] It should be noted that, in the description of this application, the terms "include", "comprise" or any other variant thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. The terms "first", "second", etc. in this application are used to distinguish similar objects, and are not used to describe a specific order or sequence.

[0073] The technical solution of the present application is applied to the signal debug test stage of PCB boards / boards in the process of hardware development. In the relevant process of hardware development, the board signal test results intuitively reflect the quality of each signal in data. The quality of these signals is directly related to the quality of the product and the trust of customers, so the signals that fail in the board test results are debugged. Usually when performing signal debugging, hardware development engineers need to adjust the resistance or capacitance and other patch components on the fault signal link according to the type and link of the fault signal to meet the line impedance matching and eliminate signal reflection. After adjusting the resistance, the board is assembled, and the signal quality of the test point is retested, and then the waveform of the fault signal is obtained for further analysis.

[0074] In the actual debugging process, hardware R&D engineers often cannot adjust the resistance or capacitance of a resistor once to obtain a good signal. They may even need to adjust other resistors on the link where the fault signal is located to successfully complete the debugging. In addition, when re-soldering a circuit, such as a resistor in a 0402 package, it is also possible that the resistor will be poorly soldered or even the solder joint will be lost due to the engineer's personal error, thereby obtaining a fault signal. Therefore, during the debugging process, engineers not only need to consider how to optimize the fault signal, but also need to be patient with the resistor welding. Sometimes, progress will stagnate due to poorly soldered resistors, which will limit the accuracy of signal debugging and even delay product release.

[0075] In order to solve the above problems, the present application provides a soldering system for patch components, which can be applied to portable automatic spot welding machines, and can also be applied to laboratory signal testing and tuning, etc., and has low cost and high reliability. At the same time, it also has the function of quickly completing the replacement of small-scale patch resistors and capacitors.

[0076] Based on the above-mentioned patch component welding system, the present application also provides a welding method for remote sensing image targets, which can detect the spatial position coordinates of the module to be welded, and send the spatial position coordinates to the integrated control chip, so that the integrated control chip controls the double tweezers system and the laser welding assembly to weld the patch component to the corresponding position of the module to be welded according to the position coordinates, thereby automatically completing the movement and patch component welding operations.

[0077] In order to enable those skilled in the art to better understand the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0078] In conjunction with the specific application environment architecture or specific hardware architecture on which the execution of the soldering method for patch components depends, the specific application environment architecture or the specific hardware architecture is described herein.

[0079] This embodiment is applied to an automatic patch component welding system, such as Figure 1 As shown, the system includes: a computer 10, an integrated control chip 20, a dual tweezers system 30, a tweezers head 31, servo motors G1-G3 installed on the dual tweezers system 30, a laser welding assembly 40, a welding head 41, servo motors G4-G6 installed on the laser welding assembly 40, a first camera 51, a second camera 52, a welding table 60, a reference board 61, a receiving box 70 and a board to be welded 100. In addition, the system may also include other more or fewer devices or elements, which are not limited in this embodiment.

[0080] The connection relationship between each component or device is as follows Figure 1As shown, the computer 10 includes but is not limited to a personal computer (PC), such as a desktop computer, a laptop computer, etc. In addition, the computer 10 can also be a server, a server cluster, a data center, and other devices.

[0081] The integrated control chip 20 is a circuit that integrates multiple functional modules and interfaces, and is mainly used to implement various control functions. These functional modules may include a processor, a memory, an input / output interface, and a specific control algorithm. By integrating these modules, the integrated control chip 20 can run programs autonomously and complete data processing and control tasks.

[0082] Optionally, the integrated control chip 20 in the embodiment of the present application may be a microcontroller unit (MCU). The computer 10 and the integrated control chip 20 may be connected by wire or wirelessly. In the present embodiment, the two are connected wirelessly, for example, via a wireless local area network (WLAN).

[0083] At least one servo motor is pre-installed on the double tweezers system 30, such as Figure 1 The system includes three servo motors, namely G1 to G3. Each servo motor can receive a pulse signal sent from the integrated control chip 20. For example, when the servo motor G1, G2 or G3 receives a pulse signal, it will rotate by a pulse angle, thereby realizing the movement of the tweezer head 31 in the double tweezer system 30. At the same time, each time the servo motor G1 to G3 rotates by an angle, a pulse signal will be fed back to the integrated control chip 20. Therefore, the integrated control chip 20 will forward the pulse signal to the computer 10. The pulse signals received and sent by the integrated control chip 20 will form a closed-loop negative feedback, and the system can accurately control the rotation of the servo motors G1 to G3, thereby realizing precise positioning. The control accuracy can reach 0.01mm (millimeter) after experimental verification.

[0084] This embodiment takes the servo motor G1 as an example. The main working principle of the servo motor G1 is: the integrated control chip 20, as the uplink of G1, processes the actual position coordinates of the chip resistor and capacitor output from the computer 10, and the actual position coordinates contain the data of the coordinate system. The servo motor G1 mainly uses the data of the y-axis. The integrated control chip 20 interprets this data and compiles it into a pulse current to drive G1. After receiving the pulse current, G1 controls the motor to rotate a specified angle. In addition, the servo motor G1 can control the displacement of the roller through the conveyor belt so that the double tweezers system 30 finally reaches the specified position.

[0085] In addition, the servo motors G1 to G3 also have a feedback function. The integrated control chip 20 can determine the waveform of the next pulse signal only after receiving the real-time position feedback of the servo motor, such as the servo motor G1. Generally, the servo motors G1 to G3 can use photoelectric encoders to achieve the following functions:

[0086] Function 1: Provide the relative position data between the stator and rotor of the motor; Function 2: Provide speed feedback signal through the angular encoder to measure speed; Function 3: Provide the angular displacement signal of the transmission system as the position feedback signal. Through the feedback system, the displacement of the motor can be controlled more accurately.

[0087] Similarly, servo motors G4 to G6 may be pre-installed on the laser welding assembly 40. Each servo motor may drive the laser welding assembly 40 to move at different locations, but the working principles and functions are the same as those of the aforementioned servo motors G1 to G3, and will not be described in detail in this embodiment.

[0088] In addition, the double-tweezers system 30 includes a fixing frame and a telescopic rod, etc. This embodiment does not limit the shape and structure of the double-tweezers system 30.

[0089] Furthermore, the laser welding assembly 40 also includes: a temperature sensor, a distance sensor, a transceiver, a processor, a welding switch, etc. These components or device structures are Figure 1 It should be understood that the number of temperature sensors and distance sensors in this embodiment can be set as needed.

[0090] The transceiver is connected to the integrated control chip 20 for receiving control instructions or control signals sent from the integrated control chip 20 and forwarding them to the processor, which is used to parse the control instructions or control signals and start the welding switch. When the welding switch is turned on (Turn ON), the welding head 41 of the laser welding assembly 40 discharges welding liquid, which can be used to weld the patch component to a specific module on the PCB board; when the welding switch is turned off (Turn OFF), the welding head 41 can be controlled to stop discharging liquid.

[0091] It should be noted that, in this embodiment, the chip component is a chip resistor and capacitor component or module.

[0092] An embodiment of the present invention provides an embodiment of a welding method for patch components. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0093] In this embodiment, a welding method is provided for welding a patch component to a target module on a board 100 (such as a PCB board) to be welded. Figure 2 As shown, it is a structural diagram of a welding system including patch components of a board to be welded provided in an embodiment of the present application.

[0094] Before performing the above welding method, the following preparation procedures need to be completed. Specifically:

[0095] Preparation stage 1: Check whether the tweezers head 31 of the dual tweezers system 30 and the welding head 41 of the laser welding assembly 40 are at the first preset position P1 and the second preset position P2 (position), for example, P1 and P2 are two position points on the reference plate.

[0096] If yes, the first step preparation stage 1 is completed; if no, the computer 10 controls the dual tweezers system 30 and the laser welding assembly 40 through the integrated control chip 20 to move the tweezers head 31 to the first preset position P1 and the welding head 41 to the second preset position P2.

[0097] Preparation stage 2: The computer 10 detects whether the tweezers head 31 of the double tweezers system 30 has clamped a patch component, which is pre-placed in the receiving box 70, and the number can be one or more. If the tweezers head 31 has not clamped a patch component, the integrated control chip 20 is used to control the tweezers head 31 of the double tweezers system 30 to clamp a patch component from the receiving box 70, which is the patch component to be soldered on the board 100 to be soldered.

[0098] If it is detected through the photos taken by at least one camera that a patch component is clamped on the tweezers head 31, it is confirmed that the preparation stage 2 is completed.

[0099] The welding method provided in this embodiment can be performed after the aforementioned "preparation stage 1" and "preparation stage 2" are completed, such as Figure 3 FIG. 1 is a flow chart of a soldering method for a patch component provided in this embodiment. The method can be performed by Figure 1 or Figure 2 The computer 10 shown is executed. Specifically, the method includes:

[0100] Step S101: acquiring at least one image of a board to be welded.

[0101] Among them, the board to be welded, such as a PCB board, is placed on a reference board 61 of a welding table 60. Since the reference board 61 has reference size, data and other information, at least one image taken includes: reference information on the reference board 61 and position information of multiple modules on the board to be welded 100.

[0102] The board 100 to be welded includes at least one module to be welded (or called a rework module), such as a connector module, an interface module, a grounding module, and the like.

[0103] The reference information is some information such as dimensions and sizes pre-marked on the reference board 61. The position information of multiple modules on the board 100 to be welded includes but is not limited to the size, position, etc. of the connector module, interface module, and grounding module. All this information can be reflected through images or photos taken by the camera.

[0104] In a possible implementation, the at least one image includes a first group of images and a second group of images. Each group of images includes at least one image. Specifically, one implementation of this step is:

[0105] The computer 10 obtains a first set of images taken by the first camera 51 from a first perspective on the board 100 to be welded, and obtains a second set of images taken by the second camera 52 from a second perspective on the board 100 to be welded. The first camera 51 is located directly above the overall structure of the board to be welded, and forms a first angle with each corner of the reference board on the z-axis, such as a 45° angle with each corner of the 40 cm background board on the z-axis, which can clearly present the content of the reference board. The second camera 52 is positioned flush with the board to be welded, and the vertical distance between the first camera 51 and the board is, as shown in FIG. Figure 1 or Figure 2 shown.

[0106] In this embodiment, the at least one image may also be referred to as a bird's-eye view of the board.

[0107] Optionally, in order to clearly reflect the contents of each module of the board, the number of pixels of the first camera 51 and the second camera 52 is selected to be 16 million pixels.

[0108] Step S102: Determine the size and position of each module according to the reference information and the position information, and annotate and generate a first annotated diagram.

[0109] The first annotation map may be understood as a feature map that annotates the initial module position and size of the image captured by the camera.

[0110] One implementation of this step is: first, use image processing technology, such as template matching, edge detection, etc., to identify reference information in the image. The reference information usually includes the edge of the reference plate, specific marking points or lines, etc. Then use image processing technology to identify the position information of each module in the image. This may include the outline of the module, the coordinates of the center point, etc.

[0111] Establish a coordinate system. According to the identified reference information, establish a unified coordinate system in the image. This coordinate system will be used to determine the size and position of each module in the future. Finally, use the extracted position information and the established coordinate system to calculate the size (such as width, height) and specific position (such as the coordinates of the center point) of each module.

[0112] Step S103: input the first annotated image and the photo of the board to be welded into the telemetry target detection model for processing, obtain a second annotated image including the module to be welded, and determine the position coordinates of the module to be welded according to the second annotated image.

[0113] Before step S103, the method further includes: inputting at least one preset feature module into the computer, wherein the at least one feature module includes: one or more of a power connector module, a silk screen module, an electrolytic capacitor module, a heat sink module, and a chip capacitor and resistor module. These feature modules are identical or partially identical to at least one module to be welded included in the aforementioned board 100 to be welded, and are used for subsequent feature comparison.

[0114] Step S103 specifically includes: inputting the first annotated image and the photo of the board to be welded into the telemetry target detection model, performing global average pooling and maximum pooling layer processing on the first annotated image to obtain global features and local features that match at least one preset feature module; obtaining a second annotated image containing the module to be welded based on the global features and the local features; and determining the position coordinates of the module to be welded based on the second annotated image and the first annotated image.

[0115] In this embodiment, the board 100 to be welded is placed on the reference board 61 and aligned with the dotted board as the reference system. The camera obtains the remote sensing image by taking pictures, and the remote sensing image is processed by the global average pooling and maximum pooling layers to obtain the global features and local features (including the module to be welded) that match the preset feature module. Then, after the two are feature spliced, the feature annotation map of the module to be tested is obtained, that is, the second annotation map.

[0116] Step S104: sending the position coordinates of the module to be welded to the integrated control chip.

[0117] The integrated control chip 20 controls the dual-tweezers system 30 and the laser welding assembly 40 to weld the patch component to the position corresponding to the module to be welded according to the position coordinates. The module to be welded is a module on the board 100 to be welded.

[0118] Specific control process such as Figure 4 As shown, the double forceps system 30 includes servo motors G1-G3 and telescopic rods L1 and L2. Figure 4The left side of the system (side view in the y-axis direction) is shown, which shows the process of the dual tweezers system 30 grabbing the patch component and placing it on the board. The dual tweezers system 30 is equipped with three micro servo DC motors (such as G1, G2 and G3), which are used to drive the sliding wheel, the telescopic rod (L1 and L2) and the tweezers head 31 respectively.

[0119] First, after the servo motors G1 to G3 receive the pulse signal from the integrated control chip 20, the servo motors G1 to G3 drive the sliding wheel, telescopic rods L1 and L2, and the tweezers head 31 to rotate or move. Each time the servo motors G1 to G3 move / rotate an angle, a pulse is fed back to the system. In this way, the pulses received and sent by the system form a closed-loop negative feedback, and the system can accurately control the rotation of the motor, thereby achieving precise positioning. Optionally, after testing, the control accuracy of this method can reach 0.01mm.

[0120] Similarly, the driving principle and performance of servo motor G2 are the same as those of G1. The function of servo motor G2 is to drive the telescopic rod L1 to move on the x-axis, and the moving distance is determined by the computer. The driving principle and performance of servo motor G3 are the same as those of G1. The function of servo motor G3 is to drive the telescopic rod L2 to move on the z-axis, and the moving distance is also determined by the computer.

[0121] The bottom of the telescopic rod L2 is externally connected to a tweezer piece controlled by two excitation coils. This embodiment adopts this tweezer piece design, which can clamp the patch component more stably. In the natural state, the two tweezer pieces are fixed to both sides by springs and will not move arbitrarily. The spring force formula applies F=kx, F is the elastic force, and k is the stiffness coefficient. Optionally, k=0.001N / mm.

[0122] In addition, a magnetic induction coil is arranged inside the tweezer head 31 of the double tweezer system 30. After a certain current passes through, the magnetic induction coil generates a stable magnetic field. When the tweezer head 31 is required to clamp the patch component, the integrated control chip 20 supplies power to the magnetic induction coil to generate a magnetic field, such as providing current to the magnetic induction coil, generating a magnetic field according to the electromagnetic induction effect, and attracting the C1 and C2 permanent magnets. At this time, the tweezers clamp the patch component until it is fixed to the solder joint and then released.

[0123] Three servo motors, such as G4 to G6, are also pre-installed in the laser welding assembly 40. The displacement method of the servo motors G4 to G6 is the same as the method of controlling the servo motors G1 to G3 in the double tweezers system 30.

[0124] The working principle of the laser welding assembly 40 is to utilize the excellent directivity and high power density of the laser beam to focus the light on a very small area through an optical device, so as to form a heat source area with highly concentrated energy at the welded part in a very short time.

[0125] The module to be welded is at a 45° angle to the horizontal plane and is controlled by a bottom micro motor to achieve 360° rotation. Two reflectors are placed inside the laser component to focus the scattered light beams toward the emission port. The laser welding endpoint temperature can be calculated based on the preset formula to set the power and distance.

[0126] Specifically, assuming that the illumination duration is finite and the distance (in cm) between the focal point and the surface is not zero, the maximum temperature on the central axis of the circular laser spot is:

[0127]

[0128] Among them, T represents the maximum temperature on the central axis, A represents the absolute value of air humidity, and q s0 represents the power density at the center of the spot, ω r represents the radius of the spot surface, λ t represents the escape density at time t, and π is the circumference of a circle.

[0129] When the laser power is too high, the surface of the PCB board may be vaporized, causing other layers of circuits to break. Therefore, when welding, the appropriate laser density is calculated based on the PCB material and the laser radius. Then, the integrated control chip 20 instructs the laser welding component 40 on the illumination duration and power through signals or control instructions, thereby controlling the welding head 41 to perform welding operations on the patch components at a safe laser density.

[0130] In addition, the computer collects pulse data from the integrated control chip, obtains the set power density and distance parameters, and then compiles the feedback signal of the sensor to complete the welding. According to the actual welding situation in the laboratory, the focal point temperature is set to about 400°C, and the action time does not exceed 5 seconds. In this embodiment, a distance sensor and a temperature sensor are added to the laser welding assembly 40. When the focal point penetrates the tin wire and lasts for a period of time, the focal point temperature data is collected and compared with the set value temperature. If it is greater than the set value, the laser density is reduced; otherwise, heating continues until the set temperature is reached.

[0131] Figure 1 or Figure 2 The design scheme of the platform lifting is shown in FIG. 6 , in which a rectangular parallelepiped with a length, width and height of 50 cm, 50 cm and 2 cm respectively is set as the welding platform 60. Figure 1 or Figure 2 As shown, a gray flat layer of 40 cm in length and width is placed above the welding table 60 as a reference plate 61, which has evenly distributed visible points, the purpose of which is to provide a reference for image recognition board coordinates and component information. In order to save costs, the welding table 60 can be cast by 3D printing technology.

[0132] The present application provides a method for soldering patch components. By taking an image of a board to be soldered and using reference information and position information to determine the size and position of each module, and then processing it through a telemetry target detection model, the position of the module to be soldered can be accurately marked, thereby greatly improving the soldering accuracy and solving the problem of resistor cold soldering and solder joint loss caused by personal errors of hardware R&D engineers.

[0133] In addition, this method realizes automatic identification and positioning of the modules to be welded, reduces the steps of manual intervention, thus speeding up the welding speed, improving production efficiency and saving labor costs. At the same time, the integrated control chip directly controls the dual tweezers system and laser welding components to weld according to the position coordinates, further shortening the welding cycle.

[0134] In addition, since the above-mentioned laser welding component is a laser welding component that can achieve 360° rotation on the x-plane. The laser welding component is internally configured with a temperature sensor, a distance sensor, etc. After the processor in the laser welding component receives the returned data, it controls the power transmission to achieve the temperature change of the focal point. It has a safer temperature range than the industrial grade and a simpler and more comprehensive mobile structure. The method is applied to a portable automatic spot welding machine to complete the welding work of the resistor and capacitor solder joints, and has the beneficial effects of firm solder joints, less pollution, and fast speed.

[0135] The above system of the present application also includes a dual tweezers system, which includes telescopic rods and servo motors in the x, y, and z axis directions. The servo motors of the telescopic rods in each direction are controlled by an integrated control chip, and combined with the photo upload function of at least two cameras, the system as a whole has a mobile architecture for mobile feedback monitoring. The servo motor receives fine pulses to achieve precise movement, eliminating the error limit of the DC motor. At the same time, the telemetry target detection model is used to detect the moving distance, combined with the servo motor feedback system, to achieve precise position control.

[0136] In addition, the method provided in this embodiment also includes: during the welding process, the first camera 51 and the second camera 52 take pictures in real time to check whether the welding is completed; if so, the picture pool or the image pool is cleared and the system position is initialized. If not, the integrated control chip sends a pulse signal to continue to control the dual tweezers system and the laser welding assembly to collaboratively complete the welding task.

[0137] like Figure 5 As shown, this embodiment also provides another soldering method for patch components, which can be applied to the integrated control chip 20. The method can be performed after the aforementioned steps S101 to S104, and specifically includes:

[0138] Step S105: receiving the position coordinates of the module to be welded.

[0139] The position coordinates are generated by a computer according to at least one image on the board to be welded. The specific generation process is described in steps S101 to S103 of the aforementioned embodiment and will not be described in detail here.

[0140] Optionally, the location coordinates may be sent by transmitting via messages / instructions / messages.

[0141] In addition, the computer can determine the position coordinates of the module to be welded in real time and send them to the integrated control chip 20 in real time.

[0142] Step S106: parsing the position coordinates to determine the actual position of the module to be welded on the board to be welded, and generating at least one set of pulse signals.

[0143] The integrated control chip 20 interprets the position coordinates through the encoder to generate the spatial coordinate pulse waveform of the module to be welded. These pulse waveforms are used to drive the servo motor to achieve the spatial displacement of the double tweezers system and the laser welding assembly.

[0144] Step S107: sending at least one set of pulse signals to the dual tweezers system and the laser welding assembly. The pulse signals are used to control the dual tweezers system 30 and the laser welding assembly 40 to weld the patch components to the corresponding positions of the modules to be welded.

[0145] Specifically, the at least one group of pulse signals includes a first group of pulse signals and a second group of pulse signals. One implementation of step S107 is: sending the first group of pulse signals to the servo motors G1 to G3 of the dual-tweezers system 30 to control the servo motors G1 to G3 to drive the tweezers head 31 of the dual-tweezers system to perform the welding task; sending the second group of pulse signals to the servo motors G4 to G6 of the laser welding assembly 40 to control the servo motors G4 to G6 to drive the welding head 41 of the laser welding assembly 40 to perform the welding task. The above welding task includes welding the patch component to the position corresponding to the module to be welded.

[0146] In addition, in this embodiment, the integrated control chip 20 is also used to provide feedback to the laser welding component 40 to complete the temperature and power settings based on the actual position of the module to be tested, so that the dual tweezers system 30 and the laser welding component 40 can work in parallel to complete the high-temperature welding and welding work of the components.

[0147] Specifically, the temperature and power setting process can refer to formula (1) of the aforementioned embodiment, which will not be described in detail in this embodiment.

[0148] In another implementation of this embodiment, Figure 6 As shown, in the above steps S101 to S104, or before step S101, it also includes:

[0149] Step S201: Obtain the status of the dual-tweezers system.

[0150] The state of the double tweezers system can be reflected by a picture or image taken by a camera. The state of the double tweezers system includes whether the patch component is clamped on the tweezers head 31. The steps of this embodiment can be executed by the computer 10.

[0151] Step S202: Check whether the dual-tweezers system has completed the pick-up operation according to the status. The pick-up operation means that the tweezers head of the dual-tweezers system has clamped the patch component and is located at a specified position.

[0152] The computer analyzes the image content based on at least one image captured by the first camera 51 and the second camera 52 to determine whether the patch component is clamped on the tweezers head 31, and determines whether the current position of the tweezers head 31 is at a specified position, such as position P1.

[0153] Specifically, the process of determining the designated position is the same as the method of determining the position coordinates of the module to be welded in the aforementioned embodiment, which will not be described in detail here. This step corresponds to the "preparation stage 2" in the aforementioned embodiment.

[0154] Step S203: If not, a control instruction is sent to the integrated control chip so that the integrated control chip controls the dual-tweezers system to complete the object-picking operation.

[0155] If it is determined through image analysis that the patch component is not clamped on the tweezers head 31, the computer 10 sends a control instruction to the integrated control chip 20. After receiving the control instruction, the integrated control chip 20 generates a new pulse signal, such as a group of pulse signals, and sends it to the servo motors G1~G3 of the dual tweezers system 30, so that the servo motors G1~G3 clamp a patch component from the receiving box 70 and send it to the designated position, completing the operation process of "preparation stage 2".

[0156] The intermediate control process can be implemented multiple times. At the same time, the computer 10 determines the status of the dual-tweezers system 30 in real time based on the feedback of the dual-tweezers system 30.

[0157] If the detection result is yes, the aforementioned steps S101 to S104 are executed.

[0158] In addition, this embodiment also provides a training process of a telemetry target detection model, which is used to generate a second annotation map in step S103 and determine the actual position coordinates of the module to be welded.

[0159] Specifically, Figure 7As shown, in the above step S103, before performing global average pooling and maximum pooling layer processing on the first labeled image, it also includes: obtaining training samples; determining the loss function based on the training samples and the classification and regression joint optimization method; determining the total loss function of the training samples based on the loss function and the stochastic gradient descent algorithm; training model parameters according to the total loss function to obtain a telemetry target detection model.

[0160] In the above step S101, the computer receives at least one remote sensing image captured by the camera, which uses a bird's-eye view to obtain information about ground buildings, transportation, mountains, rivers, lakes, etc. through remote sensing satellites. In this embodiment, the bird's-eye view of the board to be welded is taken, and different modules are evenly arranged, which conforms to the characteristics of remote sensing images, so it is more appropriate to use a remote sensing image target detection method.

[0161] There are various components on the bird's-eye view of the board. In order to better distinguish the chip resistor and capacitor components from other components, the module attention mechanism is used, combined with the convolutional neural network, to learn and optimize the weight of each module in the feature map. The convolutional neural network focuses on and extracts important feature information through different weights to improve the performance and generalization ability of the model. Because the module attention mechanism (Attention Mechanism, AM) has a good robust response capability to time-varying features and dynamically adjusts the module weights. Based on this, this embodiment provides a module channel attention mechanism suitable for remote sensing image target detection tasks.

[0162] like Figure 7 As shown in the figure, a 1×1 convolution layer is used to fuse features and perform normalization operations. The module dimension of the convolution output is C / R, where R is the scaling factor. The purpose is to reduce the amount of computation of the BN operation and the ReLu nonlinear activation operation and improve the detection efficiency. The activated features are again restored to the module weight dimension through a 1×1 convolution kernel, and the output is a 1×1×C feature dimension map. Finally, the attention weight vector is obtained by processing the sigmoid function, and the vector is multiplied by each module of the output feature map. Finally, the channel attention-weighted labeled feature map is obtained, i.e., the second labeled map. The weight value of the module to be welded in the second labeled map is the largest and needs to be processed first.

[0163] In this embodiment, the resistor and capacitor components should have a higher weight value. Assume that the local feature of each resistor and capacitor module in the feature graph is X i , and activate it with the ReLu function. The convolutional neural network predicts the corresponding proposed box of the area to be welded

[0164]

[0165] in, Indicates the coordinates of the upper left corner of the prediction area. Indicates the width and height of the prediction area. The training sample includes various modules of the board / PCB to be soldered. The training sample is defined as N represents the number of samples. i represents similar module labels (need to be further identified by convolutional learning), W represents the parameters of the network, and Y i represents the possible category label of sample i, p i (X) represents the probability that sample i is predicted as the target, and Lcls represents the loss value calculated last time. This method uses the classification and regression joint optimization method to define the loss function, which is as follows:

[0166] L(X,Y,G|W)=L cls (p i (X),Y)+μSign(Y)L bbr (G * ,G) (2)

[0167] Among them, L represents the loss function, μ is the penalty parameter, the value range of μ is between 0 and 1, G represents the suggested box of the welding area, G * Indicates the region proposal box with higher weight value, L bbr Represents the regional difference interval function. According to the above definition, the Sign(Y) function is defined as the regression of the region proposal box of the negative sample, and the total loss function of the sample is:

[0168]

[0169] A specific implementation of the telemetry target detection model obtained by training the model parameters according to the total loss function includes: using the stochastic gradient descent (SGD) algorithm to obtain L RNG The smallest optimal parameter W * . Among them, all convolutional layer parameters of each random number generator (RNG) are randomly initialized using a zero-mean Gaussian distribution with a standard deviation of 0.01. In each iteration, the network inputs a batch of training samples and then updates the parameters of the model. When the training is completed, the most effective region proposal box with all the same as the comparison module is output. Select the chip resistor and capacitor in the region proposal box and output the actual spatial coordinate position through the computer.

[0170] The model training method provided in this embodiment can quickly identify the SMD RC component module after training with the convolutional neural network, thereby quickly analyzing the actual coordinate position of the module to be welded, and the dual tweezers system and the laser welding assembly work together to complete the welding. It can greatly reduce the problems of SMD RC component welding failure due to human factors, including cold soldering, solder joint drop, and tinning in dense areas. In addition, this method can also allow hardware R&D engineers to focus more on signal debugging work, reduce welding work time, improve work efficiency, reduce toxic gas inhalation, and ensure the cleanliness of the board.

[0171] This method uses image recognition and reinforcement learning technology to package the trained telemetry target detection model directly into the computer, which can complete module recognition offline and reduce welding costs.

[0172] In addition, in this embodiment, the above-mentioned components, systems, and devices can be added to a mobile control console. When the experimenter needs it, the mobile console automatically moves to the required end to achieve end-to-end distance.

[0173] The present application also provides a patch component welding system, which includes: a computer 10, an integrated control chip 20, a double tweezers system 30, a laser welding assembly 40, at least two cameras, such as a first camera 51 and a second camera 52, a board 100 to be welded, a patch component and a welding table 60; its structure and connection relationship are as described above Figure 1 or Figure 2 shown.

[0174] The welding platform 60 includes a reference plate 61, on which the board 100 to be welded is placed. Figure 2 As shown. The dual tweezers system 30 and the laser welding assembly 40 are both movably connected to the welding table; the first camera 51 of the at least two cameras is located directly above the board to be welded, and forms a first angle with each corner on the reference board 61 on the z-axis, and the position of the second camera 52 is flush with the board to be welded. The computer 10 is connected to the integrated control chip 20, and the integrated control chip 20 is connected to the dual tweezers system 30 and the laser welding assembly 40.

[0175] The first camera 51 and the second camera 52 are used to capture at least one image of the board to be welded and send it to the computer 10. The computer is used to receive and execute the above Figure 3 and Figure 5 The integrated control chip 20 is used to execute the above-mentioned Figure 5 The soldering method and steps of the SMD components are shown.

[0176] The dual tweezers system 30 and the laser welding assembly 40 are used to receive at least one set of pulse signals from the integrated control chip 20, and weld the patch components clamped by the tweezers head 31 of the dual tweezers system 30 to the corresponding position of the module to be welded, which is the module on the board to be welded.

[0177] The specific process is described in the above method embodiment, which will not be repeated here in this embodiment.

[0178] The welding system provided in the present application includes a laser welding assembly, which can be applied to laboratory signal testing and tuning, and has low cost and high reliability. In addition, it can also quickly replace small-scale patch resistors and capacitors. At the same time, the telemetry target detection model can detect and replace the spatial position of the assembly, generate a second annotation map containing the module to be welded, and determine the position coordinates of the module to be welded based on the second annotation map, and finally send the position coordinates to the integrated control chip to control the dual tweezers system and the laser welding assembly to complete the position movement and welding tasks of the patch components.

[0179] In addition, a soldering device for patch components is also provided in this embodiment, which is used to implement the above-mentioned embodiments and preferred embodiments, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0180] This embodiment provides a soldering device for patch components, such as Figure 8 As shown, a welding device includes: an acquisition module 810, a marking module 820, a first processing module 830 and a first sending module 840. In addition, the device may also include other more or fewer modules, such as a storage module, etc., which is not limited in this embodiment.

[0181] When the welding device is used as the computer 10 in the aforementioned embodiment, the functions of each module include the following:

[0182] The acquisition module 810 is used to acquire at least one image of the board to be welded, where the board to be welded is placed on a reference board of a welding station, and the image includes reference information on the reference board and position information of multiple modules on the board to be welded.

[0183] The marking module 820 is used to determine the size and position of each module according to the reference information and the position information, and mark and generate a first marking map.

[0184] The first processing module 830 is used to input the first annotated image and the photo of the board to be welded into the telemetry target detection model for processing, obtain a second annotated image containing the module to be welded, and determine the position coordinates of the module to be welded according to the second annotated image.

[0185] The first sending module 840 is used to send the position coordinates of the module to be welded to the integrated control chip, so that the integrated control chip controls the double tweezers system and the laser welding assembly to weld the patch component to the corresponding position of the module to be welded according to the position coordinates.

[0186] In another possible implementation, the welding device includes: a receiving module 910, a second processing module 920 and a second sending module 930. In addition, the device may also include other more or fewer modules.

[0187] When the welding device is used as the integrated control chip 20 of the aforementioned embodiment, the functions of each module include the following:

[0188] The receiving module 910 is used to receive the position coordinates of the module to be welded, and the position coordinates are generated by a computer according to at least one image on the board to be welded.

[0189] The second processing module 920 is used to analyze the position coordinates to determine the actual position of the module to be welded on the board to be welded, and generate at least one group of pulse signals.

[0190] The second sending module 930 is used to send at least one group of pulse signals to the dual tweezers system and the laser welding assembly, and the pulse signals are used to control the dual tweezers system and the laser welding assembly to weld the patch components to the corresponding positions of the modules to be welded.

[0191] The soldering device for patch components in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0192] It should be noted that the description of the features in the embodiment corresponding to the welding device can refer to the relevant description of the embodiment corresponding to the welding method, and will not be repeated here.

[0193] The embodiment of the present invention also provides an electronic device having the above Figure 8 The welding device shown.

[0194] The embodiment of the present application also provides an electronic device, such as Fig. 9As shown, the electronic device includes: a processor 901 and a memory 902, wherein the memory 902 stores a computer program, and the processor 901 is configured to run the computer program to perform the above-mentioned Figure 3 , Figure 5 or Figure 6 The steps in an embodiment of a method for soldering a patch component are shown.

[0195] Specifically, the processor 901 may be a central processing unit (CPU), a network processor or a combination thereof. The processor 901 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0196] The memory 902 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 902 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 902 may include a memory remotely arranged relative to the processor 901, and these remote memories may be connected to the electronic device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0197] The memory 902 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 902 may also include a combination of the above types of memory.

[0198] The electronic device further includes an input device 903 and an output device 904. The processor 901, the memory 902, the input device 903 and the output device 904 may be connected via a bus or other means.

[0199] The input device 903 can receive input digital or character information and generate key signal input related to the user settings and function control of the electronic device, such as a touch screen, a keypad, a mouse, a track pad, a touch pad, an indicator bar, one or more mouse buttons, a trackball, a joystick, etc. The output device 904 may include a display device, an auxiliary lighting device (e.g., an LED), a tactile feedback device (e.g., a vibration motor), etc. The above display device includes but is not limited to a liquid crystal display, a light emitting diode, a display, and a plasma display.

[0200] In addition, the electronic device also includes one or more communication interfaces for the electronic device to communicate with other devices or communication networks.

[0201] In this embodiment, the electronic device may be the computer 10 of the aforementioned embodiment, or may also be the integrated control chip 20. Alternatively, it may be other devices or equipment, such as a camera, a servo motor, and the like.

[0202] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above-mentioned embodiments of the soldering method for patch components when running.

[0203] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0204] An embodiment of the present application further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the steps in any of the above-mentioned embodiments of the soldering method for patch components are implemented.

[0205] An embodiment of the present application also provides another computer program product, including a non-volatile computer-readable storage medium, the non-volatile computer-readable storage medium storing a computer program, and when the computer program is executed by a processor, the steps in any of the above-mentioned embodiments of the soldering method for patch components are implemented.

[0206] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to function. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0207] The above is a detailed introduction to a soldering method and system for patch components provided by the present application. This article uses specific examples to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method and core ideas of the present application. It should be pointed out that for ordinary technicians in this technical field, without departing from the principles of the present application, several improvements and modifications can be made to the present application, and these improvements and modifications also fall within the scope of protection of the present application.

Claims

1. A soldering method for patch components, characterized in that: Applied to a computer, the method comprises: Acquire at least one image of a board to be welded, the board to be welded is placed on a reference board of a welding station, and the image includes reference information on the reference board and position information of multiple modules on the board to be welded; Determine the size and position of each module according to the reference information and the position information, and mark and generate a first marked image; Inputting the first annotated image and the photo of the board to be welded into the telemetry target detection model for processing, obtaining a second annotated image including the module to be welded, and determining the position coordinates of the module to be welded according to the second annotated image; The position coordinates of the module to be welded are sent to the integrated control chip, so that the integrated control chip controls the double tweezers system and the laser welding assembly to weld the patch component to the corresponding position of the module to be welded according to the position coordinates.

2. The method according to claim 1, characterized in that: The method further comprises: Inputting at least one preset feature module into the computer, wherein the at least one feature module includes: one or more of a power connector module, a silk screen module, an electrolytic capacitor module, a heat sink module, and a chip capacitor and resistor module; Inputting the first annotated image and the photo of the board to be welded into the telemetry target detection model for processing to obtain a second annotated image including the module to be welded, and determining the position coordinates of the module to be welded according to the second annotated image, including: Inputting the first labeled image and the photo of the board to be welded into the telemetry target detection model, performing global average pooling and maximum pooling layer processing on the first labeled image, and obtaining global features and local features that match the at least one preset feature module; Obtaining a second annotated graph including the module to be welded according to the global features and the local features; The position coordinates of the module to be welded are determined according to the second marking map and the first marking map.

3. The method according to claim 2, characterized in that Before performing global average pooling and maximum pooling layer processing on the first labeled image, the method further includes: Get training samples; Determine a loss function based on the training samples and the classification and regression joint optimization method; Determine the total loss function of the training sample according to the loss function and the stochastic gradient descent algorithm; The model parameters are trained according to the total loss function to obtain the telemetry target detection model.

4. The method according to claim 1, characterized in that: The at least one image includes a first group of images and a second group of images; the at least one image captured on the board to be welded includes: Acquire a first group of images taken by a first camera at a first viewing angle on the board to be welded, and acquire a second group of images taken by a second camera at a second viewing angle on the board to be welded; The first camera is located directly above the board to be welded and forms a first angle with each corner of the reference board on the z-axis; the second camera is located flush with the board to be welded.

5. The method according to any one of claims 1 to 4, characterized in that: Before acquiring at least one image of the board to be welded, the method further includes: Get the status of the dual tweezers system; According to the state, checking whether the double-tweezers system completes the picking operation, wherein the picking operation refers to that the tweezers head of the double-tweezers system clamps the patch component and is located at a specified position; If not, a control instruction is sent to the integrated control chip so that the integrated control chip controls the double-tweezers system to complete the object picking operation.

6. A method for soldering a patch component, characterized in that: Applied to an integrated control chip, the method comprises: Receiving position coordinates of a module to be welded, wherein the position coordinates are generated by a computer according to at least one image on a board to be welded; Parsing the position coordinates to determine the actual position of the module to be welded on the board to be welded, and generating at least one set of pulse signals; The at least one group of pulse signals is sent to the dual-tweezers system and the laser welding assembly, and the pulse signals are used to control the dual-tweezers system and the laser welding assembly to weld the patch components to the positions corresponding to the modules to be welded.

7. The method according to claim 6, characterized in that The at least one group of pulse signals includes a first group of pulse signals and a second group of pulse signals; The step of sending the at least one set of pulse signals to the dual tweezers system and the laser welding assembly comprises: Sending the first group of pulse signals to the servo motor of the double-tweezers system to control the servo motor to drive the tweezers head of the double-tweezers system to perform a welding task; The second group of pulse signals is sent to the servo motor of the laser welding assembly to control the servo motor to drive the laser head of the laser welding assembly to perform the welding task.

8. A soldering system for patch components, characterized in that: The system comprises: a computer, an integrated control chip, a double tweezers system, a laser welding assembly, at least two cameras, a board to be welded, a patch component and a welding table; Wherein, the welding table comprises a reference plate, on which the board to be welded is placed; The double tweezers system and the laser welding assembly are both movably connected to the welding table; The first camera of the at least two cameras is located directly above the board to be welded and forms a first angle with each corner of the reference board on the z-axis, and the second camera is located flush with the board to be welded; The computer is connected to the integrated control chip, and the integrated control chip is connected to the double-tweezers system and the laser welding assembly; The first camera and the second camera are used to capture at least one image on the board to be welded and send it to the computer; The computer is used to execute the soldering method of the patch component according to any one of claims 1 to 5; The integrated control chip is used to perform the soldering method of the patch component as claimed in claim 6 or 7; The dual-tweezers system and the laser welding assembly are used to receive at least one group of pulse signals from the integrated control chip, and weld the patch components clamped by the tweezers heads of the dual-tweezers system to the corresponding positions of the modules to be welded, and the modules to be welded are modules on the boards to be welded.

9. An electronic device, characterized in that: comprising a memory and a processor, wherein the memory and the processor are connected; The memory stores computer instructions, and the processor executes the soldering method of patch components according to any one of claims 1 to 5, or claim 6 or 7, by executing the computer instructions.

10. A computer-readable storage medium, characterized in that: The computer readable storage medium has computer instructions stored thereon; The computer instructions are used to enable a computer to execute any one of claims 1 to 5, or the soldering method for patch components described in claim 6 or 7.

Images