Integrated wafer conveying manipulator

Abstract

The invention relates to the technical field of semiconductors, in particular to an integrated wafer conveying manipulator which is composed of a manipulator body, a data acquisition module and a fault diagnosis module. The data acquisition module is arranged, the visual sensor is used for detecting key operation nodes of the manipulator body, parameter data of multiple motors arranged in the manipulator body are acquired, and whether the operation state of the manipulator body is normal or not is comprehensively judged in combination with the fault diagnosis module, so that fault diagnosis of the manipulator body is achieved; and the accuracy and reliability of fault diagnosis are further improved.

Description

Technical Field

[0001] The present invention relates to the field of semiconductor technology, and particularly to an integrated wafer transfer manipulator. Background Art

[0002] A wafer is a silicon wafer used to fabricate silicon semiconductor circuits and is one of the most important basic materials in the field of semiconductor technology. In the handling, picking, placing, and storage of wafers, a manipulator is required to transfer wafers or wafer cassettes.

[0003] Traditional manipulators can only transfer wafers or wafer cassettes separately and cannot achieve the simultaneous transfer of both, resulting in low transfer efficiency of wafers and wafer cassettes in a wafer storage and transportation device with a narrow space and unable to ensure the high efficiency of the production process. In addition, in a working environment with high precision and high efficiency, the manipulator needs to maintain a high degree of stability and reliability. However, due to the long-term operation of the manipulator, various faults may occur, such as wear of mechanical components and malfunction of motors. Traditional fault diagnosis methods usually rely on manual inspection and maintenance, which are inefficient and difficult to detect potential problems in a timely manner. Although existing fault diagnosis methods utilize visual sensors and motor parameter monitoring, there are still some technical deficiencies in actual applications. For example, the detection results of visual sensors are easily affected by changes in environmental light, resulting in a high false positive rate. At the same time, although motor parameter monitoring can provide relatively accurate data, in a complex working environment, single motor parameter monitoring may not be able to comprehensively reflect the operating state of the manipulator, easily missing potential faults and resulting in low accuracy and reliability of fault diagnosis.

[0004] Therefore, the present invention proposes an integrated wafer transfer manipulator suitable for use inside a wafer storage and transportation device. Summary of the Invention

[0005] Based on this, it is necessary to provide an integrated wafer transfer manipulator for the above technical problems to solve the problems raised in the above background art.

[0006] To achieve the above object, the present invention provides the following technical solution: An integrated wafer transfer manipulator, which is applied to a wafer storage and transportation device, includes: a manipulator main body for synchronously transferring a wafer cassette and wafers; a data acquisition module for real-time collecting operation state images of key operation nodes of the manipulator main body and real-time collecting motor parameter data built in the manipulator main body; the motor parameter data includes motor temperature data, motor torque data, and motor travel data; a fault diagnosis module for processing and analyzing the operation state images by using an image processing algorithm to generate a mechanical component recognition result, and combining a convolutional neural network model to obtain a mechanical component abnormality recognition result; processing the motor parameter data by using a moving average filtering algorithm to generate a motor parameter abnormality result; generating a comprehensive fault report based on the mechanical component abnormality recognition result and the motor parameter abnormality result to determine whether the operation state of the manipulator main body is normal, and when the operation state of the manipulator main body is abnormal, triggering an alarm system to push the comprehensive fault report through a human-machine interface or a mobile terminal to ensure timely response and processing.

[0007] Optionally, the manipulator main body includes: a mounting part for integrally mounting a wafer clamping part, a wafer cassette clamping part, and an electrical control module, and the wafer clamping part and the wafer cassette clamping part are respectively arranged on both sides of the mounting part; a wafer clamping part, the wafer clamping part includes an adsorption-type wafer clamping end and a wafer clamping control structure, the wafer clamping control structure is arranged inside the mounting part, the adsorption-type wafer clamping end is fixedly installed on the side of the mounting part away from the wafer cassette clamping part, and the wafer clamping control structure is connected to the adsorption-type wafer clamping end through an air pipe for controlling the adsorption, clamping or release of the wafer by the adsorption-type wafer clamping end, and controlling the gas flow rate by using the wafer clamping control structure to adjust the strength of the adsorption force of the adsorption-type wafer clamping end on the wafer; a wafer cassette clamping part, the wafer cassette clamping part includes a wafer cassette clamping end for compatibly clamping wafer cassettes of multiple sizes and a wafer cassette clamping control structure, the wafer cassette clamping control structure is fixedly installed on one side of the mounting part, and the wafer cassette clamping end penetrates through the wafer cassette clamping control structure and is slidably connected to the wafer cassette clamping control structure to realize the wafer cassette clamping control structure driving the wafer cassette clamping end to move towards or away from each other; an electrical control module, the electrical control module is electrically connected to the wafer clamping part and the wafer cassette clamping part respectively for realizing unified control of the wafer clamping part and the wafer cassette clamping part and synchronous transfer of wafers and wafer cassettes.

[0008] Optionally, the installation part includes a bottom plate housing and a joint shaft disposed on one side of the bottom plate housing. The bottom plate housing is a hollow housing structure for integrating and installing a wafer clamping control structure inside the bottom plate housing and reducing the space occupancy rate of the wafer clamping control structure. The joint shaft is disposed on the robotic arm of the wafer handling robot and drives the integrated wafer transfer manipulator to perform a horizontal rotation movement under the drive of the robotic arm of the wafer handling robot.

[0009] Optionally, the adsorption type wafer clamping end portion includes a wafer chuck housing fixedly installed on one side of the bottom plate housing and a ceramic chuck fixedly installed on the side of the wafer chuck housing away from the bottom plate housing. A first air passage communicating with an air pipe is provided inside the wafer chuck housing. A second air passage communicating with the first air passage is provided inside the ceramic chuck. A plurality of vacuum adsorption ports communicating with the second air passage are opened at the upper end of the ceramic chuck, and the vacuum adsorption ports are uniformly arranged on the upper surface of the ceramic chuck. The wafer clamping control structure includes a solenoid valve bracket fixedly installed inside the bottom plate housing, a solenoid valve fixedly installed on the solenoid valve bracket and communicating with the air pipe, and a speed regulating valve installed on the air pipe.

[0010] Optionally, the wafer cassette clamping control structure includes a back plate fixedly installed on one side of the bottom plate housing, a motor fixedly installed inside the back plate, a transmission block fixedly installed at the output shaft end of the motor, a transmission structure rotatably connected to the transmission block and used to drive the wafer cassette clamping end portion to move towards or away from each other, a first guide body fixedly installed inside the back plate and used to guide the wafer cassette clamping end portion, and a second guide body fixedly installed inside the back plate and used to guide the transmission block. The motor and the second guide body are both arranged on the vertical center line inside the back plate. The transmission block is slidably connected to the second guide body, and the wafer cassette clamping end portion is slidably connected to the first guide body.

[0011] Optionally, the wafer cassette clamping end portion includes a pair of jaw brackets, a pair of clamping blocks A, a pair of clamping blocks B, a pair of support blocks A, a pair of support blocks B, and a wafer stop. The wafer stop is fixedly installed on the side of the back plate facing the wafer cassette to be clamped. The pair of jaw brackets are symmetrically distributed on both sides of the back plate. One end of the jaw bracket close to the back plate penetrates the back plate and is slidably connected to the back plate. One end of the jaw bracket penetrating the inside of the back plate is fixedly connected to the transmission structure. One end of the jaw bracket penetrating the inside of the back plate is sleeved outside the first guide body and is slidably connected to the first guide body. The clamping blocks A and B are fixedly installed on the jaw bracket located outside the back plate in sequence from outside to inside. The support blocks A and B are both fixedly installed on the jaw bracket, and the support blocks A and B are respectively arranged corresponding to the lower ends of the clamping blocks A and B.

[0012] Optionally, the transmission structure includes a pair of link transmission parts, and the pair of link transmission parts are symmetrically distributed on both sides of the motor. The link transmission part includes a main link and a rotating link. One end of the main link is rotatably connected to the transmission block, and the other end is rotatably connected to the middle of the rotating link. The end of the rotating link is fixedly connected to one end of the claw bracket penetrating through the inside of the back plate.

[0013] Optionally, the wafer cassette clamping control structure further includes a in-place detection structure for judging whether different-sized wafer cassettes to be transferred are successfully clamped. The in-place detection structure is arranged inside the back plate. The in-place detection structure includes a motor retraction position in-place sensor, a first motor extension position in-place sensor, and a second motor extension position in-place sensor. The motor retraction position in-place sensor is arranged at the starting working position of the transmission block, and the first motor extension position in-place sensor and the second motor extension position in-place sensor are arranged on the movement track of the transmission block in sequence from top to bottom.

[0014] Optionally, the data acquisition module includes a vision sensor installed at the key operation nodes of the robot body, and a temperature sensor, a torque sensor, and a displacement sensor installed on the motor.

[0015] Optionally, the fault diagnosis module includes: an image data processing and analysis sub-module, which is used to preprocess the collected operation state images by using an ambient light compensation algorithm to generate preprocessed operation state images; use the SIFT algorithm to perform feature retrieval and description on the preprocessed operation state images to generate feature descriptors, use the watershed algorithm to perform image segmentation on the preprocessed operation state images to obtain a labeled image and the result of separated mechanical components. In the result of mechanical components, use the feature descriptors to perform matching processing to generate the mechanical component recognition result, and input the mechanical component recognition result into the trained convolutional neural network model to output the mechanical component abnormality recognition result; a motor data processing and analysis sub-module, which is used to process the motor parameter data by using a moving average filtering algorithm to generate a motor parameter abnormality result; a report generation and alarm sub-module, which is used to generate a comprehensive fault report based on the mechanical component abnormality recognition result and the motor parameter abnormality result to judge whether the operation state of the robot body is normal, and when the operation state of the robot body is abnormal, trigger the alarm system to push the comprehensive fault report through the human-machine interface or the mobile terminal to ensure timely response and processing.

[0016] An integrated wafer transfer robot provided by the present invention has the following beneficial effects.

[0017] 1. By integrating and installing a wafer clamping part and a wafer cassette clamping part, not only can the unified control of the two be achieved, but also the synchronous transfer of the wafer and the wafer cassette can be realized. When the integrated wafer transfer manipulator needs to synchronously transfer the wafer cassette and the wafer, it can first clamp the wafer cassette and then the wafer, and then execute the synchronous transfer instruction. Compared with the prior art in which two manipulators separately perform the operations of clamping and transferring the wafer cassette and the wafer, it can reduce the round-trip operation steps and effectively improve the work efficiency. At the same time, the wafer clamping part and the wafer cassette clamping part adopt an integrated structure design, which can effectively reduce the space occupancy rate. For two separate wafer cassette grippers and wafer grippers, it is easy to cause collisions between the wafer cassette gripper and the wafer gripper in a limited space, and additional work such as motion planning is required. However, the integrated wafer transfer manipulator provided in this application can significantly reduce the equipment volume and complexity, and can operate more flexibly in a wafer storage and transportation device with a narrow space, and after clamping the wafer or the wafer cassette, place it at the required position to improve the transfer efficiency of the wafer and the wafer cassette, thereby ensuring the high efficiency of the production process.

[0018] 2. By setting an adsorption-type wafer clamping end and a wafer clamping control structure, the adsorption clamping or release of the wafer can be achieved, and the strength of the adsorption force of the adsorption-type wafer clamping end on the wafer can be adjusted by controlling the gas flow rate, so as to avoid the wafer falling off or being damaged due to too weak or too strong adsorption force.

[0019] 3. By setting a wafer cassette clamping end, a wafer cassette clamping control structure and a in-place detection structure, not only can the clamping or release of the wafer cassette be achieved, the compatible clamping of wafer cassettes of different sizes can be realized, but also whether the wafer cassette of different sizes is successfully clamped can be detected, so as to improve the accuracy of the clamping detection of the corresponding size wafer cassette and further improve the transfer efficiency of the production process.

[0020] 4. By setting a data acquisition module, using a vision sensor to detect the key operation nodes of the manipulator body, and collecting the parameter data of multiple motors built in the manipulator body, combined with a fault diagnosis module, comprehensively judge whether the operation state of the manipulator body is normal, so as to realize the fault diagnosis of the manipulator body and further improve the accuracy and reliability of the fault diagnosis. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a schematic diagram of the overall structure of the present invention.

[0022] Figure 2 It is a front view of the integrated wafer transfer manipulator of the present invention.

[0023] Figure 3 It is a top view of the integrated wafer transfer manipulator of the present invention.

[0024] Figure 4 Schematic diagram of the transmission structure of the present invention.

[0025] Figure 5 Schematic diagram of the wafer cassette clamping end of the present invention.

[0026] Figure 6 Partial enlarged view of the in-place detection structure of the present invention.

[0027] Figure 7 Schematic diagram of the installation position of the vacuum detection sensor of the present invention.

[0028] Figure 8 Block diagram of the fault diagnosis module of the present invention.

[0029] Reference numerals: ceramic chuck 1, bottom plate housing 2, back plate 3, wafer gripper housing 4, joint shaft 5, gripper bracket 6, clamping block A 7, clamping block B 8, support block A 9, support block B 10, drive block 11, solenoid valve 12, main connecting rod 13, rotating connecting rod 14, solenoid valve bracket 15, wafer stopper 16, motor 17, motor retracted position in-place sensor 18, first motor extended position in-place sensor 19, vacuum detection sensor 20, speed control valve 21, electrical control module 22, vacuum suction port 23, air pipe 24, first guide body 25, second guide body 26, robotic arm 27, main body structure 28, second motor extended position in-place sensor 29. Detailed implementation manners

[0030] The embodiments of the present application will be described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary and are only used to explain the present application and should not be construed as a limitation to the present application. And without conflict, the features in the following embodiments and the embodiments can be combined with each other. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without making creative efforts belong to the scope protected by the present application.

[0031] In the description of the present application, it should be noted that unless otherwise clearly defined and limited, the terms "installation", "connection", and "connection" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and it can be the communication inside two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific situations.

[0032] Refer to the attached Figures 1-6, an integrated wafer transfer manipulator, which is applied to a wafer storage and transportation device and is composed of a manipulator main body, a data acquisition module, and a fault diagnosis module.

[0033] In some alternative implementation manners of the present application, the manipulator main body is composed of a mounting part, a wafer clamping part, a wafer cassette clamping part, and an electrical control module.

[0034] In some alternative implementation manners of the present application, referring to the attached Figures 1-4 , the wafer clamping part and the wafer cassette clamping part are respectively arranged on both sides of the mounting part, and the electrical control module 22 is electrically connected to the wafer clamping part and the wafer cassette clamping part respectively, for realizing the unified control of the wafer clamping part and the wafer cassette clamping part, and the synchronous transfer of the wafer and the wafer cassette. In addition, it should be noted that the electrical control module 22 (i.e., the controller) can be integrally installed inside the mounting part (i.e., the bottom plate housing 2) to centrally control the wafer clamping part and the wafer cassette clamping part.

[0035] In some alternative implementation manners of the present application, during the unified control of the wafer clamping part and the wafer cassette clamping part by the electrical control module 22, when the integrated wafer transfer manipulator needs to synchronously transfer the wafer cassette and the wafer, by successively controlling the wafer cassette clamping part to clamp the wafer cassette or the wafer clamping part to clamp the wafer, and then executing the synchronous transfer instruction, so as to reduce the round-trip operation steps.

[0036] It should be noted that during the unified control of the wafer clamping part and the wafer cassette clamping part by the electrical control module 22, the electrical control module 22 will limit the clamping to only one direction, that is, when clamping the wafer cassette, it will not clamp the wafer, and when clamping the wafer, it will not clamp the wafer cassette, and after executing the single-direction clamping operation, it will execute the clamping operation in the other direction, so as to realize the synchronous transfer of the wafer cassette and the wafer.

[0037] In some alternative implementation manners of the present application, in order to avoid the problem that the wafer clamping part and the wafer cassette clamping part interfere with each other in a limited space, the clamping directions of the wafer clamping part and the wafer cassette clamping part of the present application are arranged back to back.

[0038] In some alternative implementation manners of the present application, referring to the attached Figures 1-2, the installation part includes a bottom plate housing 2 and a joint shaft 5 provided on one side of the bottom plate housing 2. The bottom plate housing 2 is a hollow shell structure. Specifically, the bottom plate housing 2 is in a cuboid structure, and the wafer clamping control structure and the electrical control module 22 can be integrally installed inside it to reduce the space occupancy rate of the wafer clamping control structure and the electrical control module 22. The joint shaft 5 is provided on the robotic arm 27 of the wafer handling robot. Driven by the robotic arm 27 of the wafer handling robot, it drives the integrated wafer transfer manipulator to perform a horizontal rotation movement. In addition, it should be noted that the integrated wafer transfer manipulator involved in this application can be integrally installed on the robotic arm 27 of the wafer handling robot through the joint shaft 5 to realize the handling process of the wafer cassette and the wafer.

[0039] In some alternative implementation manners of the present application, the wafer clamping part includes an adsorption type wafer clamping end and a wafer clamping control structure. The wafer clamping control structure is arranged inside the installation part. The adsorption type wafer clamping end is fixedly installed on the side of the installation part away from the wafer cassette clamping part. The wafer clamping control structure is communicated with the adsorption type wafer clamping end through an air pipe 24, and is used to control the adsorption, clamping or release of the wafer by the adsorption type wafer clamping end, and the gas flow rate is controlled by the wafer clamping control structure to adjust the strength of the adsorption force of the adsorption type wafer clamping end on the wafer.

[0040] In some alternative implementation manners of the present application, refer to the attached Figures 1-3 , the adsorption type wafer clamping end includes a wafer jaw housing 4 fixedly installed on one side of the bottom plate housing 2 and a ceramic chuck 1 fixedly installed on the side of the wafer jaw housing 4 away from the bottom plate housing 2. A first air passage communicated with the air pipe 24 is arranged inside the wafer jaw housing 4. A second air passage communicated with the first air passage is arranged inside the ceramic chuck 1. A plurality of vacuum adsorption ports 23 communicated with the second air passage are opened at the upper end of the ceramic chuck 1. The vacuum adsorption ports 23 are evenly arranged on the upper surface of the ceramic chuck 1. By providing the first air passage communicated with the air pipe 24, the second air passage communicated with the first air passage, and the vacuum adsorption ports 23 communicated with the second air passage, the adsorption and clamping of the wafer can be realized. At the same time, in order to generate a uniform adsorption force on the upper surface of the ceramic chuck 1, the plurality of vacuum adsorption ports 23 in this application need to be evenly arranged on the upper surface of the ceramic chuck 1. And the number of the vacuum adsorption ports 23 can be flexibly set according to actual production requirements and will not be limited here. As shown in the attached Figure 3 , the number of the vacuum adsorption ports 23 is set to 3.

[0041] In some alternative implementation manners of the present application, refer to the attached Figure 3, the wafer clamping control structure includes a solenoid valve bracket 15 fixedly installed inside the bottom plate housing 2, a solenoid valve 12 fixedly installed on the solenoid valve bracket 15 and connected to the air pipe 24, and a speed regulator valve 21 installed on the air pipe 24. By integrally installing the solenoid valve bracket 15, the solenoid valve 12, the air pipe 24, and the speed regulator valve 21 inside the bottom plate housing 2, it can not only effectively reduce the space occupancy rate, but also prevent the solenoid valve bracket 15, the solenoid valve 12, the air pipe 24, and the speed regulator valve 21 from being exposed to the external environment, playing a role in protecting and accommodating the wafer clamping control structure, and delaying the aging speed of the solenoid valve 12, the air pipe 24, and the speed regulator valve 21. Additionally, during the process of the solenoid valve 12 controlling the ceramic chuck 1 to adsorb and clamp the wafer, the gas flow rate is regulated by the speed regulator valve 21 to adjust the strength of the adsorption force of the ceramic chuck 1 on the wafer. Among them, by controlling the speed regulator valve 21, the gas flow rate through the air pipe 24 can be effectively controlled, and thus the adsorption force of the vacuum adsorption port 23 can be effectively controlled, thereby avoiding damage or detachment of the wafer caused by too large or too small adsorption force of the vacuum adsorption port 23; Refer to the appendix Figure 7 , in order to monitor the magnitude of the adsorption force in real time, a vacuum detection sensor 20 can also be installed on the side of the main body structure 28 of the wafer handling robot. The input end of the vacuum detection sensor 20 is connected to the air pipe 24 and is arranged in front of the gas flowing through the speed regulator valve 21 to monitor the pressure change of the gas in the air pipe 24 before flowing through the speed regulator valve 21 in real time.

[0042] In some optional implementation manners of the present application, the working principle of the wafer clamping part includes: through the provided solenoid valve 12, when the ceramic chuck 1 needs to clamp the wafer, the solenoid valve 12 starts to operate, and the air pipe 24 connected to it starts to inhale, forming a negative pressure at the vacuum adsorption port 23 on the ceramic chuck 1, thereby realizing vacuum adsorption of the wafer. And through the provided speed regulator valve 21, the gas flow rate can be controlled, so as to control the strength of the adsorption force of the ceramic chuck 1 on the wafer. At the same time, through the provided vacuum detection sensor 20, the pressure change of the gas in the air pipe 24 before flowing through the speed regulator valve 21 can be monitored in real time to monitor the magnitude of the adsorption force in real time.

[0043] In some optional implementation manners of the present application, refer to the appendix Figures 1-5, the wafer cassette clamping part includes a wafer cassette clamping end for compatibly clamping wafer cassettes of multiple sizes and a wafer cassette clamping control structure. The wafer cassette clamping control structure is fixedly installed on one side of the installation part. The wafer cassette clamping end penetrates through the wafer cassette clamping control structure and is slidably connected to the wafer cassette clamping control structure. An in-place detection structure for judging whether a wafer cassette to be transferred of different sizes is successfully clamped is arranged inside the wafer cassette clamping control structure. The wafer cassette clamping control structure drives the wafer cassette clamping end to move towards or away from each other. During the process of the wafer cassette clamping control structure driving the wafer cassette clamping end to move towards each other, the in-place detection structure detects whether the wafer cassette clamping end successfully clamps the wafer cassette to be transferred according to the size of the wafer cassette to be transferred, so that the wafer cassette clamping end successfully clamps the wafer cassette to be transferred of different sizes. And during the process of the wafer cassette clamping control structure driving the wafer cassette clamping end to move away from each other, the in-place detection structure detects whether the wafer cassette clamping end successfully releases the wafer cassette to be transferred according to the size of the wafer cassette to be transferred, so that the wafer cassette clamping end successfully releases the wafer cassette to be transferred.

[0044] In some alternative implementation manners of the present application, referring to the attached Figures 1-4 , the wafer cassette clamping control structure includes a back plate 3 fixedly installed on one side of the bottom plate housing 2, a motor 17 fixedly installed inside the back plate 3, a transmission block 11 fixedly installed at the output shaft end of the motor 17, a transmission structure rotatably connected to the transmission block 11 and used to drive the wafer cassette clamping end to move towards or away from each other, a first guide body 25 fixedly installed inside the back plate 3 and used to guide the wafer cassette clamping end, and a second guide body 26 fixedly installed inside the back plate 3 and used to guide the transmission block 11. The back plate 3 is a hollow shell structure, and its structure and function are similar to those of the bottom plate housing 2, so no further description will be given here. The motor 17 and the second guide body 26 are evenly arranged on the vertical center line inside the back plate 3. It should be noted that in order to improve the force balance of the link transmission parts and the wafer cassette clamping end distributed on both sides of the motor 17, the motor 17 and the second guide body 26 involved in the present application are both arranged on the vertical center line of the back plate 3. The transmission block 11 is slidably connected to the second guide body 26, and the wafer cassette clamping end is slidably connected to the first guide body 25. Driven by the motor 17, the transmission block 11 moves vertically up and down along the second guide body 26, and drives the transmission structure to perform reciprocating translational motion inside the back plate 3, so as to drive the wafer cassette clamping end to move towards or away from each other. During the movement of the wafer cassette clamping end, the wafer cassette clamping end is slidably matched with the first guide body 25, which is used to improve the stability of the cooperation between the wafer cassette clamping end and the wafer cassette clamping control structure and reduce the space occupancy rate of the wafer cassette clamping end and the wafer cassette clamping control structure.

[0045] In some alternative implementation manners of the present application, referring to the attached Figures 1-4 , the wafer cassette clamping end portion includes a pair of jaw brackets 6, a pair of clamping blocks A 7, a pair of clamping blocks B 8, a pair of support blocks A 9, and a pair of support blocks B 10. The pair of jaw brackets 6 are symmetrically distributed on both sides of the back plate 3. The jaw bracket 6 has a "concave" structure, and the two "protruding" end portions of the "concave" structure respectively penetrate through the back plate 3 and are slidably connected to the back plate 3. At the same time, these two "protruding" end portions are respectively fixedly connected to the two end portions of the rotating link 14. By adopting the jaw bracket 6 with this "concave" structure, compared with the jaw bracket 6 with a rectangular structure or the jaw bracket 6 with one end portion (similar to a convex structure), it can not only improve the stability of the overall structure of the wafer cassette clamping end portion during reciprocating translational motion, but also reduce the weight and save material costs. At the same time, the "protruding" end portion of the jaw bracket 6 is sleeved outside the first guide body 25 and is slidably connected to the first guide body 25. In order to further improve the stability of the movement of the jaw bracket 6, the number of the first guide bodies 25 can be set to 2, that is, the two end portions of the jaw bracket 6 are respectively slidably matched with the first guide body 25. With the driving effect of the transmission structure and the guiding effect of the first guide body 25, the jaw bracket 6 performs reciprocating translational motion along the horizontal direction of the back plate 3. The clamping blocks A 7 and the clamping blocks B 8 are sequentially fixedly installed on the jaw bracket 6 located outside the back plate 3 from outside to inside. The support blocks A 9 and the support blocks B 10 are both fixedly installed on the jaw bracket 6, and the support blocks A 9 and the support blocks B 10 are respectively arranged corresponding to the lower ends of the clamping blocks A 7 and the clamping blocks B 8. When the transmission structure performs reciprocating translational motion inside the back plate 3, the clamping blocks A 7 and the clamping blocks B 8 are respectively used to clamp the wafer cassette to be transferred with corresponding sizes, and under the cooperative action of the corresponding support blocks A 9 and support blocks B 10, they are used to prevent the wafer cassette to be transferred with corresponding sizes from falling off during the clamping process.

[0046] It should be noted that the wafer cassette clamping end portion involved in the present application can clamp two sizes of wafer cassettes, that is, the two clamping blocks A 7 and the two support blocks A 9 are used to clamp a wafer cassette with one size specification (such as an 8-inch wafer cassette), and the two clamping blocks B 8 and the two support blocks B 10 are used to clamp a wafer cassette with another size specification (such as a 6-inch wafer cassette). If it is necessary to clamp more size specifications of wafer cassettes (such as a 12-inch wafer cassette), corresponding clamping blocks and support blocks for wafer cassettes with corresponding sizes can be arranged on the jaw bracket 6 and outside the support block A 9 according to actual production needs, which will not be limited here.

[0047] In some alternative implementation manners of the present application, referring to the attached Figure 4, the transmission structure includes a pair of connecting rod transmission parts, and the pair of connecting rod transmission parts are symmetrically distributed on both sides of the motor 17. The connecting rod transmission part includes a main connecting rod 13 and a rotating connecting rod 14. One end of the main connecting rod 13 is rotatably connected to the transmission block 11, and the other end is rotatably connected to the middle of the rotating connecting rod 14. The two ends of the rotating connecting rod 14 are respectively fixedly connected to the two ends of the jaw bracket 6, so as to stably drive the wafer cassette clamping end to perform reciprocating translational motion.

[0048] In some alternative implementation manners of the present application, refer to the attached Figure 5 , the wafer cassette clamping end further includes a wafer baffle 16, and the wafer baffle 16 is fixedly installed on the side of the back plate 3 facing the wafer cassette to be clamped. The wafer baffle 16 is used to protect the wafers in the wafer cassette.

[0049] In some alternative implementation manners of the present application, refer to the attached Figure 6 , the in-place detection structure is arranged inside the back plate 3. The in-place detection structure includes a motor retracted position in-place sensor 18, a first motor extended position in-place sensor 19, and a second motor extended position in-place sensor 29. Among them, the two motor extended position in-place sensors involved in the present application cooperate with the clamping block A7 and the clamping block B8 respectively. The motor retracted position in-place sensor 18 is arranged at the starting working position of the transmission block 11. The starting working position of the transmission block 11 represents that the wafer cassette clamping end is in the maximum open state. The first motor extended position in-place sensor 19 and the second motor extended position in-place sensor 29 are arranged on the movement track of the transmission block 11 from top to bottom in sequence. As the clamping block A7 or the clamping block B8 clamps the wafer cassette to be transferred with the corresponding size, the first motor extended position in-place sensor 19 and the second motor extended position in-place sensor 29 are respectively used to judge whether the wafer cassette to be transferred with the corresponding size is successfully clamped. For example, when using the clamping block A7 and the support block A9 to clamp the wafer cassette to be transferred with the corresponding size, when the motor 17 drives the transmission block 11 to move upward along the second guide body 26, if the clamping block A7 and the support block A9 correctly clamp the wafer cassette to be transferred with the corresponding size, the first motor extended position in-place sensor 19 can sense the transmission block 11, then it is judged that the clamping is successful. If the clamping block A7 and the support block A9 do not correctly clamp the wafer cassette to be transferred with the corresponding size, the position where the transmission block 11 moves upward at this time may exceed the clamping in-place position or not reach the clamping in-place position, resulting in the first motor extended position in-place sensor 19 not sensing the transmission block 11, then it is judged that the clamping fails; Subsequently, after completing the wafer cassette transfer process, the wafer cassette to be transferred with the corresponding size can be released with the clamping block A7 or the clamping block B8, and the motor retracted position in-place sensor 18 is used to judge whether the wafer cassette to be transferred with the corresponding size is successfully released.

[0050] In some alternative implementation manners of the present application, the data acquisition module includes a vision sensor (not shown) installed at key operating nodes of the robot arm main body, and a temperature sensor, a torque sensor, and a displacement sensor (not shown) installed on the motor. Among them, the vision sensor is used to collect operation state images of key operating nodes of the robot arm main body in real time, the temperature sensor is used to collect motor temperature data in real time, the torque sensor is used to collect motor torque data in real time, and the displacement sensor is used to collect motor travel data in real time.

[0051] It should be noted that the key operating nodes include machine components that make up the robot arm main body, such as installation parts, wafer clamping parts, wafer cassette clamping parts, etc. By identifying the operation states of the mechanical components of the key operating nodes, it is used to identify and classify abnormal conditions of each mechanical component included in the machine components, such as wear, looseness, deformation, etc.

[0052] In some alternative implementation manners of the present application, the fault diagnosis module is used to process and analyze the operation state image by using an image processing algorithm to generate a mechanical component recognition result, and combine it with a convolutional neural network model to obtain a mechanical component abnormality recognition result; use a moving average filtering algorithm to process the motor parameter data to generate a motor parameter abnormality result; based on the mechanical component abnormality recognition result and the motor parameter abnormality result, generate a comprehensive fault report to determine whether the operation state of the robot arm main body is normal, and when the operation state of the robot arm main body is abnormal, trigger an alarm system to push the comprehensive fault report through a human-machine interface or a mobile terminal to ensure timely response and handling.

[0053] In some alternative implementation manners of the present application, refer to the appendix Figure 8 , the fault diagnosis module is composed of an image data processing and analysis sub-module, a motor data processing and analysis sub-module, and a report generation and alarm sub-module.

[0054] In some alternative implementation manners of the present application, the image data processing and analysis sub-module is used to preprocess the collected operation state image by using an ambient light compensation algorithm to generate a preprocessed operation state image; use the SIFT algorithm to perform feature retrieval and description on the preprocessed operation state image to generate a feature descriptor, use the watershed algorithm to perform image segmentation on the preprocessed operation state image to obtain a labeled image and the separated mechanical component result, in the mechanical component result, perform matching processing by using the feature descriptor to generate a mechanical component recognition result, and input the mechanical component recognition result into the trained convolutional neural network model to output a mechanical component abnormality recognition result.

[0055] In some alternative implementation manners of the present application, an environmental light compensation algorithm is adopted to preprocess the acquired operation status image to generate a preprocessed operation status image, including: dividing the acquired operation status image into multiple small tiles, for example, small tiles of 8×8 pixels; calculating local illumination model parameters for each small tile, including local average brightness and local contrast, where the calculation formula for local average brightness is: and the calculation formula for local contrast is: In the formula, N represents the number of pixels in the small tile, represents the local average brightness, represents the local contrast, represents the operation status image; adjusting the pixel values of each small tile so that its brightness and contrast conform to the global illumination model, and the calculation formula for parameter adjustment is: In the formula, represents the result of parameter adjustment of the pixel values of each small tile, α and β respectively represent the adjustment parameters of local average brightness and local contrast, and can be adjusted according to actual needs. Integrating all small tiles into one image to generate a preprocessed operation status image. Through the above improvement scheme, it can better adapt to light changes, avoid the acquired operation status image from being too bright or too dark, and thus improve the accuracy and reliability of image preprocessing.

[0056] In some alternative implementation manners of the present application, the SIFT algorithm is adopted to perform feature retrieval and description on the preprocessed operation status image to generate a feature descriptor, including: Image preprocessing: converting the preprocessed operation status image into a grayscale image and removing noise, for example, using Gaussian filtering or median filtering; Constructing a scale space: constructing a Gaussian pyramid, and smoothing the grayscale image through Gaussian functions of different scales, and the formula is: In the formula, represents the scale space, represents the Gaussian kernel function, σ represents the scale parameter of the Gaussian pyramid, represents the grayscale image; Extreme point detection: detecting extreme points in the constructed Gaussian pyramid, and these points are local extremes in both the scale space and the spatial position. Comparing the values of each pixel point and its adjacent points through a preset neighborhood window (for example, 3×3×3) to determine the extreme points, and the formula is: In the formula, represents the pixel value of the local extreme point, represents the pixel value of the adjacent point, and k represents the offset of the scale parameter between each pixel point and its adjacent points; Precise positioning of extreme points: precisely positioning the extreme points by using Taylor expansion to improve the boundary response, and the formula is: In the formula, x represents the initial position of the extreme point, H is the Hessian matrix, and D(x) represents the gradient vector of the extreme point. Indicates the corrected position of the extreme point; Feature point direction assignment: Assign a main direction to each feature point to improve the rotational invariance of the feature point. The formula is: , where, Indicates the direction cumulative histogram, Indicates the direction gradient, Indicates the gradient direction, r and Both represent the direction gradient parameters, r and Both take values of 1. θ represents the main direction of the statistical histogram, and δ represents the scale parameter of the gradient direction; Generating feature descriptors: Describe each feature point using a target-dimensional vector. For example: Usually, a 128-dimensional vector is used to describe each extreme point. The formula is: , where, Represents an element in the feature descriptor, Represents a preset small tile, for example: an 8x8 small tile.

[0057] In some alternative implementation manners of the present application, taking a 256x256 pixel image collected as an example for illustration, this image is a high-resolution photo of a machine part, which contains multiple different mechanical parts. Key feature points need to be extracted from it, and different parts in the image need to be separated for subsequent part recognition and analysis. Specifically, it includes the following steps.

[0058] 1. Image preprocessing: Convert the image into a grayscale image and perform noise removal. Here, a Gaussian filter can be used to remove noise. For example: The noise is evenly distributed on the image and needs to be reduced to 10% of the average value of the image pixels.

[0059] 2. Scale space construction: Smooth the image through Gaussian filtering to construct a Gaussian pyramid. For example: The scale parameter σ of the Gaussian pyramid is 1.6, and the Gaussian kernel function is as follows: , where, x and y represent the spatial position parameters, σ represents the scale parameter of the Gaussian pyramid, and the image is smoothed at multiple scales to obtain the scale space .

[0060] 3. Extreme point detection: Detect each pixel point of the image in the Gaussian pyramid and perform extreme point detection in the scale space. For example: When the scale parameter σ of the Gaussian pyramid is 2, the pixel value at the image position (100, 150) is 128, and the values of adjacent pixels are 120, 130, and 118 respectively. According to the following formula: , where, k = 1.6, and the calculation result is as follows: D(100, 150, 2) = 128 - 125 = 3. Since the current pixel value is a local extreme point, it is marked as an extreme point candidate.

[0061] 4. Precise localization of extreme points: Use Taylor expansion to precisely locate extreme points. For example, at the extreme point x = (125, 125) near the center of the image, according to the following formula: , where , D(x) = 5, and the calculation results are as follows: .

[0062] 5. Feature point direction assignment: Assign directions to extreme points and use gradient direction statistics. For example, the gradient direction distribution at the pixel point (100, 150) is shown in Table 1 below. Calculate the direction cumulative histogram according to the following formula: , where r and both take the value of 1, and the main direction of the statistical histogram is 90°.

[0063] Table 1 Gradient direction distribution at the pixel point (100, 150)

[0064] 6. Generate feature descriptors: Generate descriptors for the extreme point (100, 150). For example, the descriptor calculation uses a 16x16 neighborhood block. In the descriptor calculation, the direction statistics of each small block (such as 4x4) are shown in Table 2 below. Calculate the feature descriptor according to the following formula: , and after calculation, a 128-dimensional descriptor vector is generated.

[0065] Table 2 Direction statistics of each small block (such as 4x4)

[0066] In some alternative implementation manners of the present application, the watershed algorithm is used to perform image segmentation on the preprocessed image of the operating state to obtain a labeled image and the result of separated mechanical components. In the result of mechanical components, feature descriptors are used for matching processing to generate the mechanical component recognition result, including: calculating the gradient image based on the grayscale image after noise removal: using the Sobel algorithm to calculate the image gradient, and the formula of the Sobel algorithm is: , where and represent the gradients in the horizontal and vertical directions, represents the gradient magnitude; labeled image: label the image, and the distance transformation or manual labeling method can be used to label areas such as the background or foreground; watershed transformation: use the watershed transformation algorithm to separate different mechanical components in the image, and the formula is: , where represents the gradient image, represents the labeled image, Indicates the labeled image after watershed transformation; Feature point matching and integration: In the labeled image after watershed transformation, use feature descriptors for matching (such as nearest neighbor matching or FLANN matching algorithm) to generate feature point matching results, and combine the feature point matching results with the segmented mechanical components to generate mechanical component recognition results.

[0067] In some alternative implementation manners of the present application, input the mechanical component recognition result into a trained convolutional neural network model to output a mechanical component anomaly recognition result, including: constructing and training a convolutional neural network model, which is used to recognize and classify anomalies of each mechanical component included in the machine components, such as wear, looseness, deformation, etc.; input the mechanical component recognition result into the trained convolutional neural network model to output a mechanical component anomaly recognition result.

[0068] In some alternative implementation manners of the present application, the motor data processing and analysis sub-module is used to process the motor parameter data by using a moving average filtering algorithm to generate a motor parameter anomaly result.

[0069] In some alternative implementation manners of the present application, the formula of the moving average filtering algorithm is: , where represents the motor parameter data, specifically the motor temperature data or the motor torque data or the motor travel data, m represents the size of the moving window, and the value can be 10, represents the average value of the current motor parameter, specifically the average value of the motor temperature data or the motor torque data or the motor travel data. After processing the motor parameter data by using the moving average filtering algorithm, generate the average value of the current motor parameter, and compare it with a preset motor parameter threshold to determine whether the deviation of each motor parameter exceeds the allowable range. If it exceeds, mark it as an anomaly and generate a motor parameter anomaly result, such as too high temperature, too large torque, abnormal travel, etc.

[0070] In some alternative implementation manners of the present application, the report generation and alarm sub-module is used to generate a comprehensive fault report based on the mechanical component anomaly recognition result and the motor parameter anomaly result to determine whether the operation state of the robot main body is normal. Among them, the comprehensive fault report may include the fault type, location, degree, and recommended maintenance measures. The recommended maintenance measures can be pre-stored in the constructed database. During the process of generating the comprehensive fault report, based on the fault type, location, and degree, the corresponding recommended maintenance measures in the database can be called; when the operation state of the robot main body is abnormal, the alarm system can be triggered to push the comprehensive fault report through the human-machine interface or the mobile terminal to ensure timely response and processing.

[0071] The above content is a further detailed description of the present invention in combination with specific embodiments, and it cannot be determined that the specific implementation of the present invention is only limited to these descriptions. For those of ordinary skill in the technical field to which the present invention pertains, without departing from the concept of the present invention, several simple deductions or substitutions can still be made, and all should be regarded as belonging to the protection scope of the present invention.

Claims

1. An integrated wafer transfer manipulator is applied to a wafer storage and transportation device, and is characterized in that, Including: A manipulator body for synchronously transporting a wafer cassette and wafers; A data acquisition module for real-time acquisition of the operating state images of the key operating nodes of the manipulator body, and real-time acquisition of the motor parameter data built in the manipulator body; the motor parameter data includes motor temperature data, motor torque data, and motor travel data; A fault diagnosis module for processing and analyzing the operating state images using image processing algorithms to generate mechanical component recognition results, and combining with a convolutional neural network model to obtain mechanical component anomaly recognition results; Using a moving average filtering algorithm to process the motor parameter data to generate motor parameter anomaly results; Based on the mechanical component anomaly recognition results and the motor parameter anomaly results, generate a comprehensive fault report to determine whether the operating state of the manipulator body is normal, and when the operating state of the manipulator body is abnormal, trigger an alarm system to push the comprehensive fault report through a human-machine interface or a mobile terminal to ensure timely response and handling.

2. The integrated wafer transfer manipulator according to claim 1, wherein The manipulator body includes: An installation part for integrally installing a wafer clamping part, a wafer cassette clamping part, and an electrical control module, and the wafer clamping part and the wafer cassette clamping part are respectively arranged on both sides of the installation part; A wafer clamping part, which includes an adsorption type wafer clamping end and a wafer clamping control structure. The wafer clamping control structure is arranged inside the installation part, and the adsorption type wafer clamping end is fixedly installed on the side of the installation part away from the wafer cassette clamping part. The wafer clamping control structure is connected to the adsorption type wafer clamping end through an air pipe, and is used to control the adsorption, clamping, or release of the wafer by the adsorption type wafer clamping end, and use the wafer clamping control structure to control the gas flow rate to adjust the strength of the adsorption force of the adsorption type wafer clamping end on the wafer; A wafer cassette clamping part, which includes a wafer cassette clamping end for compatibly clamping wafer cassettes of multiple sizes and a wafer cassette clamping control structure. The wafer cassette clamping control structure is fixedly installed on one side of the installation part, and the wafer cassette clamping end penetrates through the wafer cassette clamping control structure and is slidably connected to the wafer cassette clamping control structure to realize the wafer cassette clamping control structure driving the wafer cassette clamping end to move towards or away from each other; An electrical control module, which is electrically connected to the wafer clamping part and the wafer cassette clamping part respectively, and is used to realize the unified control of the wafer clamping part and the wafer cassette clamping part, and the synchronous transportation of wafers and wafer cassettes.

3. The one-piece wafer transfer manipulator according to claim 2, wherein The installation part includes a bottom plate housing and a joint shaft arranged on one side of the bottom plate housing. The bottom plate housing is a hollow shell structure, which is used to integrally install the wafer clamping control structure inside the bottom plate housing and reduce the space occupancy rate of the wafer clamping control structure. The joint shaft is arranged on the robotic arm of the wafer handling robot and drives the integrated wafer transfer manipulator to perform a horizontal rotational motion under the drive of the robotic arm of the wafer handling robot.

4. The one-piece wafer transfer manipulator according to claim 3, characterized in that, The adsorption - type wafer clamping end includes a wafer chuck housing fixedly installed on one side of the bottom plate housing and a ceramic chuck fixedly installed on the side of the wafer chuck housing away from the bottom plate housing. A first air passage communicating with an air pipe is provided inside the wafer chuck housing. A second air passage communicating with the first air passage is provided inside the ceramic chuck. A plurality of vacuum adsorption ports communicating with the second air passage are opened at the upper end of the ceramic chuck, and the vacuum adsorption ports are evenly arranged on the upper surface of the ceramic chuck. The wafer clamping control structure includes a solenoid valve bracket fixedly installed inside the bottom plate housing, a solenoid valve fixedly installed on the solenoid valve bracket and communicating with the air pipe, and a speed regulator installed on the air pipe.

5. The one-piece wafer transfer manipulator according to claim 3, characterized in that, The wafer cassette clamping control structure includes a back plate fixedly installed on one side of the bottom plate housing, a motor fixedly installed inside the back plate, a transmission block fixedly installed at the output shaft end of the motor, a transmission structure rotatably connected to the transmission block and used to drive the wafer cassette clamping end to move towards or away from each other, a first guide body fixedly installed inside the back plate and used to guide the wafer cassette clamping end, and a second guide body fixedly installed inside the back plate and used to guide the transmission block. The motor and the second guide body are both evenly arranged on the vertical center line inside the back plate. The transmission block is slidably connected to the second guide body, and the wafer cassette clamping end is slidably connected to the first guide body.

6. The one-piece wafer transfer manipulator according to claim 5, wherein The wafer cassette clamping end includes a pair of jaw brackets, a pair of clamping blocks A, a pair of clamping blocks B, a pair of support blocks A, a pair of support blocks B, and a wafer stop. The wafer stop is fixedly installed on the side of the back plate facing the wafer cassette to be clamped. The pair of jaw brackets are symmetrically distributed on both sides of the back plate. One end of the jaw bracket close to the back plate penetrates through the back plate and is slidably connected to the back plate. One end of the jaw bracket penetrating through the inside of the back plate is fixedly connected to the transmission structure. One end of the jaw bracket penetrating through the inside of the back plate is sleeved outside the first guide body and is slidably connected to the first guide body. The clamping blocks A and B are fixedly installed on the jaw bracket outside the back plate in sequence from outside to inside. The support blocks A and B are both fixedly installed on the jaw bracket, and the support blocks A and B are respectively arranged corresponding to the lower ends of the clamping blocks A and B.

7. An integrated wafer transfer manipulator according to claim 6, wherein The transmission structure includes a pair of link transmission parts. The pair of link transmission parts are symmetrically distributed on both sides of the motor. The link transmission part includes a main link and a rotating link. One end of the main link is rotatably connected to the transmission block, and the other end is rotatably connected to the middle of the rotating link. The end of the rotating link is fixedly connected to one end of the jaw bracket penetrating through the inside of the back plate.

8. An integrated wafer transfer robot according to claim 5, characterized in that, The wafer cassette clamping control structure further includes a in-place detection structure for determining whether different-sized wafer cassettes to be transferred are successfully clamped. The in-place detection structure is disposed inside the backplane. The in-place detection structure includes a motor retraction position in-place sensor, a first motor extension position in-place sensor, and a second motor extension position in-place sensor. The motor retraction position in-place sensor is arranged at the starting working position of the transmission block. The first motor extension position in-place sensor and the second motor extension position in-place sensor are sequentially arranged from top to bottom on the movement trajectory of the transmission block.

9. The one-piece wafer transfer manipulator according to claim 5, characterized in that, The data acquisition module includes a vision sensor installed at the key operating nodes of the manipulator body, as well as a temperature sensor, a torque sensor, and a displacement sensor installed on the motor.

10. An integrated wafer transfer robot according to claim 1, characterized in that, The fault diagnosis module includes: An image data processing and analysis sub-module, which is used to preprocess the acquired operation state image by using an ambient light compensation algorithm to generate a preprocessed operation state image; perform feature retrieval and description on the preprocessed operation state image by using the SIFT algorithm to generate a feature descriptor, perform image segmentation on the preprocessed operation state image by using the watershed algorithm to obtain a labeled image and the result of separated mechanical components. In the result of mechanical components, perform matching processing by using the feature descriptor to generate a mechanical component recognition result, and input the mechanical component recognition result into the trained convolutional neural network model to output a mechanical component abnormality recognition result; A motor data processing and analysis sub-module, which is used to process the motor parameter data by using a moving average filtering algorithm to generate a motor parameter abnormality result; A report generation and alarm sub-module, which is used to generate a comprehensive fault report based on the mechanical component abnormality recognition result and the motor parameter abnormality result to determine whether the operation state of the manipulator body is normal, and when the operation state of the manipulator body is abnormal, trigger an alarm system to push the comprehensive fault report through a human-machine interface or a mobile terminal to ensure timely response and processing.

Images