Intelligent longitudinal beam assembling system

Abstract

The invention relates to the technical field of intelligent manufacturing, in particular to an intelligent longitudinal beam assembling system which comprises a workpiece conveying module used for conveying a longitudinal beam to be assembled to a designated position; the positioning and clamping module is used for accurately positioning and clamping the longitudinal beams conveyed in place; the nut conveying module is used for conveying the to-be-assembled nut to a specified position; the tightening and assembling module is used for tightening a nut to a specified position on the longitudinal beam and collecting a fourth image; the photographing and identifying module is used for photographing and identifying the installation condition of the nut on the longitudinal beam; the photographing recognition module at least respectively acquires a first image, a second image and a third image from a longitudinal beam assembly area according to a set frequency; adopting a target detection algorithm based on deep learning to identify the nut position and the tightening state in the image; and establishing a nut position model according to the preset assembly position of the nut on the longitudinal beam, and judging the installation omission and tightening state of the nut. According to the technical scheme, the longitudinal beam assembly precision and reliability can be improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to an intelligent assembly system for longitudinal beams. Background Art

[0002] In the automotive manufacturing, machinery manufacturing and other industries, the longitudinal beam is an important structural component, and its assembly quality and efficiency are directly related to the overall performance and production cost of the product. The traditional longitudinal beam assembly process mainly relies on manual operation, which is not only inefficient, but also easily affected by human factors, making it difficult to ensure assembly accuracy and consistency. With the development of industrial automation and intelligent technology, how to realize the automation and intelligence of the longitudinal beam assembly process has become an urgent problem to be solved in the industry.

[0003] Although the existing longitudinal beam assembly systems have achieved automation to a certain extent, they still have many shortcomings. For example, some systems lack precise workpiece conveying and positioning clamping mechanisms, which leads to unstable position of the longitudinal beam during assembly and affects the tightening accuracy of the nuts. In addition, the traditional nut assembly inspection method mainly relies on manual visual inspection or the use of simple sensors. This method is not only inefficient, but also prone to missed inspections or misjudgments, and cannot guarantee assembly quality.

[0004] In terms of image recognition technology, although there are some computer vision-based nut assembly detection systems, most of these systems use traditional image processing algorithms, which have poor recognition effects on complex and changeable nut assembly situations and are easily affected by environmental factors such as lighting and shadows. At the same time, these systems often require special algorithm training and parameter adjustment for each longitudinal beam model, lacking flexibility and versatility. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent assembly system for longitudinal beams, and the technical solution can improve the assembly accuracy and reliability of longitudinal beams.

[0006] To achieve the above object, the present invention provides a longitudinal beam intelligent assembly system, comprising: A workpiece conveying module is used to convey the longitudinal beam to be assembled to a designated position; Positioning and clamping module, used to accurately position and clamp the longitudinal beams delivered to the place; A nut conveying module is used to convey the nuts to be assembled to the position specified by the tightening assembly module; A tightening assembly module, used for tightening the nut to a specified position on the longitudinal beam and acquiring a fourth image; A photographing and identifying module is used to photograph and identify the installation status of nuts on the longitudinal beam; the photographing and identifying module collects at least a first image, a second image and a third image of the longitudinal beam assembly area at a set frequency; a target detection algorithm based on deep learning is used to extract features from the preprocessed image to identify the position and tightening status of the nut in the image; A nut position model is established according to the preset assembly position of the nut on the longitudinal beam; during the image recognition process, the recognized nut position is compared with the position model to determine the nut installation omission and tightening status.

[0007] Beneficial effects of the basic solution: This system integrates multiple functional modules, from longitudinal beam conveying, positioning clamping, nut conveying to tightening assembly, realizing the automation of the longitudinal beam assembly process, reducing manual intervention, improving production efficiency and reducing labor costs.

[0008] The workpiece conveying module can accurately convey the longitudinal beam to the specified position, and the positioning and clamping module can further accurately position and clamp the longitudinal beam, providing a stable and reliable foundation for subsequent nut assembly, ensuring the accuracy of the nut tightening position, and improving the overall assembly accuracy.

[0009] The camera recognition module can fully obtain the nut assembly status information by collecting images of different stages (first image, second image and third image) at a set frequency. The target detection algorithm based on deep learning can accurately identify the nut position and tightening status. By comparing with the preset nut position model, it can accurately determine whether the nut is installed or not. Combined with the fourth image, it can further determine whether the tightening status meets the requirements, and timely discover deviations in the assembly process to ensure assembly accuracy.

[0010] The camera recognition module adopts a target detection algorithm based on deep learning to intelligently identify the position and status of the nut. Compared with traditional detection methods, it has higher accuracy and reliability and can automatically identify complex nut assembly situations.

[0011] By establishing a nut position model and comparing it during the image recognition process, errors such as missing nut installation can be discovered in a timely manner, realizing an intelligent error-proofing function, reducing the product defect rate caused by assembly errors, and improving the reliability of the production process. The nut position model can be replaced according to the specific processing object. Since the processing objects are often improved and modified during the production process, the traditional method requires the use of improved products and corresponding nut and bolt positions to train and iterate the algorithm. This technical solution avoids frequent optimization and new training of the target detection algorithm. The target detection algorithm only needs to focus on extracting the nut position, reducing the bloat of the algorithm and saving computing resources.

[0012] And because the system is modularly designed, when it is necessary to adapt to the assembly of longitudinal beams of different models or specifications, the parameters of each module can be flexibly adjusted, or some modules can be appropriately adjusted or upgraded, which reduces the maintenance cost and upgrade difficulty of the system.

[0013] As an implementable preferred solution, the first image acquisition timing is before the assembly process starts, the second image acquisition timing is during the assembly process, and the third image acquisition timing is after the assembly process is completed. The acquisition frequency is adjusted according to the assembly speed.

[0014] As an implementable preferred solution, the photographing and identifying module includes a main visual camera arranged near the tightening assembly module for capturing the overall image of the longitudinal beam. The photographing and identifying module is also provided with a light source device to fill in the shooting area.

[0015] As an implementable preferred solution, control point coordinate calibration is performed, including the following contents: it also includes an error-proofing detection module that sets the execution conditions and sequence of each assembly step according to the assembly process requirements, and collects data from each sensor in real time during the assembly process, and verifies and compares it with the preset standard data.

[0016] As an implementable preferred scheme, the workpiece conveying module adopts a conveyor belt to convey the longitudinal beam, and a positioning block is provided on the conveyor belt; the positioning and clamping module adopts a mechanical positioning and clamping device, including a positioning pin, a positioning block, a clamping cylinder and a clamping block; the nut conveying module adopts a vibration plate to convey the nuts, and a photoelectric sensor is provided at the outlet of the vibration plate; the balance arm provides stable support through a cantilever structure design; the collaborative robot is installed at the mobile end of the balance arm.

[0017] As an implementable preferred solution, the tightening assembly module includes an electric wrench, which has a built-in torque sensor and an angle sensor for real-time monitoring of torque and rotation angle data during the tightening process, and transmitting the data to a control module to control the tightening process and perform error prevention detection; by monitoring the data change curves of the torque sensor and the angle sensor, the abnormality of the nut or bolt is judged.

[0018] As an implementable preferred solution, the front end of the electric wrench is integrated with an auxiliary visual camera for capturing assembly details of nuts and bolts at close range, such as the tightening degree of the nuts and the meshing condition of the threads, and collecting a fourth image.

[0019] As an implementable preferred solution, the workflow of the system includes: During the initialization phase, the control module performs initialization operations on each device, the collaborative robot performs origin return, the electric wrench performs torque calibration and zeroing operations, the camera recognition module performs self-test and parameter setting, each sensor performs initialization configuration, and the control module reads the assembly process parameters and historical data in the database; In the loading stage, the operator places the longitudinal beam to be assembled on the product fixture, and the fixture positions and clamps the longitudinal beam. The nut loading mechanism grabs the nut and transports it to the preset loading position according to the instruction of the control module, and the camera recognition module monitors the nut loading situation in real time; During the assembly phase, the collaborative robot grabs the electric wrench and moves to the nut assembly position. The camera recognition module monitors the robot's position and posture in real time. The auxiliary visual camera at the front end of the electric wrench takes close-up photos of the nuts and bolts. The electric wrench tightens the nuts according to the set parameters. The torque sensor and angle sensor monitor the tightening data in real time, and the camera recognition module continuously captures images of the nut tightening process. During the inspection phase, after the nuts are tightened, the visual recognition module conducts a comprehensive inspection of the assembly status of the nuts. The error-proofing detection module re-inspects the assembly results based on the sensor data and the preset error-proofing logic. After completing the assembly of one longitudinal beam, the system automatically enters the assembly cycle of the next longitudinal beam.

[0020] As an implementable preferred solution, the tightening assembly module is also used to optimize the motion trajectory planning of the collaborative robot. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is the logical diagram of the intelligent assembly system for longitudinal beams.

[0022] Figure 2 Schematic diagram of the structure of an electronic device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] In order to make the technical solution and advantages of the present application clearer, the technical solution of the present invention will be further described in detail below in conjunction with the accompanying drawings. It will be understood that the specific embodiments described herein are only partial embodiments of the present invention, which are only used to explain the present application, rather than to limit the present application. It should be noted that the technical features or combinations of technical features described in the following embodiments should not be considered isolated, and they can be combined with each other to achieve better technical effects. The same reference numerals appearing in the drawings of the following embodiments represent the same features or components, which can be applied to different embodiments.

[0024] In addition, unless otherwise defined, the technical terms or scientific terms used in the description of the present invention should have the common meanings understood by those skilled in the art in the art to which the present invention belongs.

[0025] Reference numerals: electronic device 500 includes processor 501 , communication interface 502 , memory 503 , and bus 504 .

[0026] The present invention is further described in detail below in conjunction with the accompanying drawings: Embodiment 1 Reference Figure 1 The longitudinal beam intelligent assembly system includes a workpiece conveying module, a positioning and clamping module, a nut conveying module, a tightening and assembly module, a camera recognition module, a control module and an anti-error detection module.

[0027] The workpiece conveying module is used to convey the longitudinal beam to be assembled to the specified position for subsequent positioning, clamping and nut assembly operations. It includes conveying equipment such as conveyor belts, conveyor rollers, conveyor chains, and corresponding drive devices and control devices.

[0028] In this embodiment, the workpiece conveying module uses a conveyor belt to convey the longitudinal beam. The conveyor belt is driven by a motor, and the conveying speed is controlled by a variable frequency speed regulator to meet the conveying requirements of different longitudinal beams. A positioning block is provided on the conveyor belt to perform preliminary positioning of the longitudinal beam to ensure that the longitudinal beam does not deviate or shake during the conveying process.

[0029] The positioning and clamping module is used to accurately position and clamp the longitudinal beam that has been delivered to the position for the subsequent nut assembly operation. It includes a positioning device, a clamping device, and corresponding driving and control devices.

[0030] In this embodiment, the positioning and clamping module adopts a mechanical positioning and clamping device. The positioning device includes a positioning pin and a positioning block. The positioning pin is inserted into the positioning hole on the longitudinal beam to achieve accurate positioning of the longitudinal beam; the clamping device includes a clamping cylinder and a clamping block. The clamping cylinder drives the clamping block to clamp the longitudinal beam to ensure that the longitudinal beam will not move or deform during the assembly process.

[0031] The nut conveying module is used to convey the nuts to be assembled to the position specified by the tightening assembly module for subsequent tightening operations. The module may include a nut storage device, a nut conveying device, and corresponding driving devices and control devices.

[0032] In this embodiment, the nut conveying module uses a vibration plate to convey nuts. A nut track is provided in the vibration plate, and the nuts are conveyed along the track under the action of the vibration plate. The vibration frequency and amplitude of the vibration plate can be adjusted by a controller to meet the conveying requirements of different nuts. In addition, a photoelectric sensor is provided at the outlet of the vibration plate to detect whether the nut has reached the specified position and transmit the signal to the control system.

[0033] The tightening assembly module is used to tighten the nuts to the specified position on the longitudinal beam. It includes a balance arm, a collaborative robot and an electric wrench, as well as the corresponding drive device and control device.

[0034] The balance arm is made of high-strength aluminum alloy, and the optimized cantilever structure design ensures good stability when bearing heavy loads. It integrates a high-precision guide rail and slider system to achieve smooth horizontal sliding, and has an automatic balancing function. It can adjust the posture in real time according to the load changes of the collaborative robot, reducing the inertia force and vibration of the robot during movement. In this embodiment, the collaborative robot uses a multi-joint industrial robot with more than 6 degrees of freedom and a repeatability accuracy of ±0.05mm. The robot's joints are driven by high-performance servo motors and equipped with harmonic reducers. While ensuring high torque output, it effectively reduces the transmission gap and improves motion accuracy.

[0035] The balancing arm works in conjunction with the collaborative robot, which is installed at the mobile end of the balancing arm. During the assembly process, the balancing arm provides stable support and flexible movement range for the collaborative robot, enabling the robot to perform nut assembly operations at different positions and angles. According to the instructions, the collaborative robot accurately grasps the electric wrench and moves to the longitudinal beam nut assembly position according to the preset trajectory to tighten the nut. At the same time, the robot can also carry other assembly tools, such as nut feeding suction cups, to complete the feeding and preliminary positioning of the nuts.

[0036] The high-precision electric wrench is used, and the torque range can be flexibly adjusted according to the assembly requirements of the longitudinal beam nuts, with the accuracy controlled within ±3%. The wrench has the functions of fast tightening and precise control. When approaching the final torque value, it automatically switches to the low-speed high-precision tightening mode to ensure the tightening quality of the nut.

[0037] The electric wrench has built-in torque sensor and angle sensor, which can monitor the torque and rotation angle data in the tightening process in real time and transmit the data to the control module. It is not only used to control the tightening process, but also as an important means of error prevention detection. During the tightening process, the sensor monitors the torque and angle change curves in real time. If the curve fluctuates abnormally, such as a sudden drop in torque or a rapid increase in angle, it indicates that there may be problems such as nut stripping and bolt deformation. The control module analyzes and judges based on the sensor data. Once an abnormality is found, the tightening operation is stopped immediately and an alarm is issued.

[0038] The electric wrench is electrically connected to the control module, which sends instructions such as tightening torque and rotation angle to the electric wrench according to the assembly process requirements. During the tightening process, the electric wrench adjusts the motor output torque in real time according to the data fed back by the sensor to ensure that the tightening process meets the process standards. At the same time, the electric wrench also supports manual control mode. During equipment debugging or special circumstances, the operator can perform manual operation through the handheld controller The camera recognition module is used to take a camera to recognize the installation status of the nuts on the longitudinal beam in the shooting area and transmit the recognition result to the control system. The module includes a camera, an image processor and a corresponding control device.

[0039] In this embodiment, the photo recognition module uses a high-resolution industrial camera for photo recognition. The camera is installed near the tightening assembly module, and the shooting range covers all nut installation positions on the longitudinal beam. The camera transmits the captured image to the image processor for processing. The image processor uses advanced image processing algorithms to recognize and analyze the image to determine whether the nut is installed in place and whether there is any omission. The recognition result is transmitted to the control module through the communication interface for subsequent processing.

[0040] The high-resolution industrial camera has a pixel of no less than 5 million and has high frame rate shooting capability, which can meet the image acquisition needs in the rapid assembly process. The camera is equipped with a large aperture lens, which can obtain clear images under different lighting conditions, and has an automatic focus function, which can quickly adjust the focal length to ensure that the nut and assembly position are always in a clear imaging state. In addition, the camera supports a Gigabit Ethernet interface to ensure high-speed transmission of image data.

[0041] Several photo recognition modules are set up. Specifically, a main visual camera is installed above the longitudinal beam assembly station to monitor the assembly status of the longitudinal beam as a whole, and the shooting range covers the entire longitudinal beam assembly area. A small auxiliary visual camera is integrated at the front end of the electric wrench, which is specially used to shoot the assembly details of nuts and bolts at close range, such as the tightening degree of nuts, the meshing condition of threads, etc. A visual camera is also set up in the nut feeding area to detect the feeding quantity and position accuracy of nuts.

[0042] In order to improve the recognition accuracy and reliability, the photo recognition module is also provided with a light source device to fill in the shooting area. The light source device can select a suitable type and installation method according to the actual situation to ensure that the captured image is clear, bright and easy to recognize.

[0043] The control module is used to receive signals from each module and process and control them to realize the automatic operation of the entire longitudinal beam intelligent assembly system. The module can include a controller, a communication interface, an input and output device, and a corresponding control program.

[0044] In this embodiment, the control module uses a programmable logic controller (PLC) as a controller. PLC has the advantages of high reliability, flexible programming, and easy expansion, and is suitable for complex industrial automation control systems. PLC connects and exchanges data with each module through a communication interface, receives signals from each module, and processes and controls them. At the same time, PLC can also communicate with the host computer to realize functions such as remote monitoring and fault diagnosis.

[0045] The control module is written with modular design ideas, including workpiece conveying control program, positioning and clamping control program, nut conveying control program, tightening assembly control program, and photo recognition control program. The modules work together through function calls and data exchange to complete the intelligent assembly task of multiple nuts on the longitudinal beam.

[0046] The camera recognition module collects at least the first image, the second image and the third image of the longitudinal beam assembly area at a set frequency. The first image is collected before the assembly process starts, the second image is collected during the assembly process, and the third image is collected after the assembly process is completed. For example, image collection is performed before tightening the nut after loading, during tightening, and after tightening to ensure that the nut assembly status information can be fully obtained. The acquisition frequency is adjusted according to the assembly speed, up to 10 frames per second, to ensure that key images can be accurately captured during rapid assembly. By analyzing the first image and the third image, the nut position installed by the tightening assembly module can be detected; further analysis of the second image and the close-range image collected by the electric wrench can obtain the assembly status of the nut.

[0047] Preprocess the collected images to improve image quality and provide a basis for subsequent image analysis. Preprocessing algorithms include grayscale processing, filtering and denoising, and image enhancement. Grayscale processing converts color images into grayscale images, reduces the amount of data, and increases processing speed. Median filtering or Gaussian filtering is used to remove noise interference in the image to make the image clearer. Image enhancement uses methods such as histogram equalization and contrast stretching to enhance the contrast between the nut and the background in the image, which facilitates subsequent feature extraction.

[0048] The deep learning-based target detection algorithm is used to extract features from the preprocessed image and identify the position and state of the nut in the image. The target detection algorithm is trained with a large number of nut sample images and can accurately learn the shape, texture and other features of the nut.

[0049] According to the preset assembly position of the nut on the longitudinal beam, a nut position model is established. During the image recognition process, the recognized nut position is compared with the position model. If a nut is not detected at a preset position, it is determined that the nut is missing. At the same time, combined with the assembly process requirements, the tightening state of the nut is judged. If the nut does not reach the specified tightening degree, it is also regarded as an assembly abnormality, and an alarm signal is issued in time to notify the operator to handle it.

[0050] The error-proofing detection module is used to perform strict logical control on the longitudinal beam nut assembly process. The error-proofing program sets the execution conditions and sequence of each assembly step according to the assembly process requirements. For example, only after it is detected that the nut has been correctly loaded and positioned accurately, the collaborative robot is allowed to grab the electric wrench for tightening; after tightening, it must pass the visual recognition module inspection before the next nut can be assembled. If a step does not meet the execution conditions, the system will automatically stop running and issue an alarm to prevent incorrect assembly.

[0051] During the assembly process, data from each sensor is collected in real time, such as torque data and angle data of the electric wrench, image data of the camera recognition module, etc., and verified and compared with the preset standard data. For example, the actual tightening torque of the electric wrench is compared with the torque range required by the process. If it exceeds the range, it is judged as a tightening abnormality; the nut position detected by the visual recognition module is compared with the preset position. If the deviation exceeds the allowable range, it is considered an assembly error. Through data verification and comparison, errors in the assembly process can be discovered and corrected in a timely manner.

[0052] The system workflow is as follows: In the initialization phase, after the system is started, the control module first performs initialization operations on each device. The collaborative robot performs origin return action to determine its initial position; the electric wrench performs torque calibration and zeroing operations to ensure tightening accuracy; the camera recognition module performs self-test and parameter setting, adjusts the focal length and exposure parameters; each sensor performs initialization configuration and prepares to collect data. At the same time, the control module reads the assembly process parameters and historical data in the database to prepare for subsequent assembly operations.

[0053] During the loading stage, the operator places the longitudinal beam to be assembled on the product fixture, and the positioning pin and clamping mechanism of the fixture quickly position and clamp the longitudinal beam. According to the instructions of the control module, the nut feeding mechanism grabs the nuts from the nut silo and transports them to the preset feeding position. During the loading process, the camera recognition module monitors the nut loading situation in real time. If the nut loading error or omission is found, the feeding mechanism is notified in time to make adjustments.

[0054] During the assembly phase, the collaborative robot grabs the electric wrench and moves to the nut assembly position according to the instructions of the control module. During the movement, the camera recognition module monitors the position and posture of the robot in real time. If a deviation is found, the control module adjusts the robot's motion trajectory in time to ensure that the electric wrench is accurately aligned with the nut. At the same time, the auxiliary visual camera at the front end of the electric wrench takes close-up photos of the nut and bolt, further accurately adjusts the position of the nut, and ensures that the nut and bolt threads are accurately engaged.

[0055] The electric wrench starts tightening the nut according to the tightening torque and rotation angle parameters set by the control module. During the tightening process, the torque sensor and angle sensor monitor the tightening data in real time and transmit the data to the control module. The control module adjusts the motor output torque of the electric wrench in real time based on the data feedback to ensure a smooth and accurate tightening process. At the same time, the camera recognition module continuously captures images of the nut tightening process to monitor the tightening status of the nut, such as whether there is thread slippage and whether the nut is tightened in place.

[0056] After the nuts are tightened, the visual recognition module conducts a comprehensive inspection of the assembly status of the nuts. The main visual camera captures the overall image of the longitudinal beam to detect whether the nut is installed in the correct position and whether there is any omission; the auxiliary visual camera captures close-up images of the nuts and bolts to detect the tightening degree of the nuts, thread engagement, etc. If an abnormality is detected, the visual recognition module transmits the image data to the control module and issues an alarm signal.

[0057] The error-proofing detection module re-tests the assembly results based on the sensor data and the preset error-proofing logic. It checks whether the torque data and angle data of the electric wrench are within the qualified range, and whether the position sensor detects the position of each component is correct. If an error is found, the error-proofing detection module immediately notifies the control module to stop the assembly operation and perform corresponding processing.

[0058] After completing the assembly of one longitudinal beam, the system automatically enters the assembly cycle of the next longitudinal beam.

[0059] Embodiment 2 The technical feature that distinguishes this embodiment from the above-mentioned embodiments is that the motion trajectory planning of the collaborative robot is optimized to improve assembly efficiency and accuracy.

[0060] This embodiment uses an improved genetic algorithm to optimize the motion trajectory of the collaborative robot.

[0061] Initialize the population and randomly generate a certain number of individuals as the initial population. Each individual represents a motion trajectory of the collaborative robot.

[0062] The fitness value of each individual is calculated according to the fitness function. The fitness function is as follows:

[0063] in, T(x) It represents the time required for the collaborative robot to complete the nut assembly task; E(x) Indicates the energy consumption during the movement of the collaborative robot; A(x) Indicates assembly accuracy, measured by the deviation between the torque value after the nut is tightened and the preset torque value. The smaller the deviation, the higher the assembly accuracy. , , are the weight coefficients of time, energy consumption and accuracy respectively.

[0064] Use roulette wheel selection or tournament selection to select several individuals from the current population as parents.

[0065] Perform crossover operations on selected parent individuals to generate new offspring individuals; perform mutation operations on offspring individuals to increase the diversity of the population. Dynamically adjust the crossover and mutation probabilities according to the fitness values ​​of the individuals. For individuals with higher fitness, use smaller crossover and mutation probabilities to retain their excellent characteristics; for individuals with lower fitness, use larger crossover and mutation probabilities to increase their diversity.

[0066] In each generation, several individuals with the highest fitness are selected and retained directly to the next generation to ensure that the genetic information of excellent individuals is not lost.

[0067] Perform local search for individuals with higher fitness to further improve their fitness value. This application uses simulated annealing algorithm for optimization. The specific steps are as follows: Set initial temperature , this temperature determines the algorithm's exploration ability in the early stage of the search. The higher the initial temperature, the more the algorithm tends to search globally. Generally, the appropriate initial temperature can be selected according to the scale and complexity of the actual problem.

[0068] Set end temperature , when the temperature drops below this value, the algorithm stops searching. Usually the termination temperature is a positive number close to zero.

[0069] Set the cooling rate , used to control the temperature drop rate, generally ranges from (0,1). In this embodiment .

[0070] Select individuals with higher fitness from the current population and use them as the initial solution of the simulated annealing algorithm. The initial solution represents a motion trajectory of the collaborative robot.

[0071] Generate neighborhood solutions and calculate new solutions and current solution The fitness difference is as follows:

[0072] in, is the solution calculated according to the fitness function The fitness value of .

[0073] if , indicating that the new solution has higher fitness, so we directly accept the new solution. .

[0074] if , then the new solution is accepted with a certain probability, and the probability formula is expressed as follows:

[0075] Reduce the temperature according to the cooling formula. The cooling formula is as follows:

[0076] in, is the current temperature, is the cooling rate.

[0077] In the current solution A new solution is randomly generated in the neighborhood of For the collaborative robot’s motion trajectory, the neighborhood solution can be generated by making small adjustments to certain path points in the trajectory. For example, for a certain path point in the trajectory ), a new point can be randomly selected in a small range near it ) to replace the point, thus obtaining a new trajectory .

[0078] Determine whether the current temperature is lower than the termination temperature. If , then stop searching, the current solution This is the optimal solution obtained by local search; otherwise, continue searching.

[0079] The individuals with the highest fitness are directly retained to the next generation.

[0080] Repeat steps until a preset number of iterations is reached or other stopping criteria are met.

[0081] The individual with the highest fitness is selected from the final population as the optimal solution, that is, the optimal motion trajectory of the collaborative robot.

[0082] The disclosed embodiment also provides a method for intelligently assembling a longitudinal beam, which utilizes the intelligently assembling system for longitudinal beams.

[0083] The embodiment of the present disclosure further provides a storage medium, in which a computer program is stored. When the computer program is executed by a processor, all steps of the above-mentioned longitudinal beam intelligent assembly method can be implemented.

[0084] Those skilled in the art can understand that all or part of the processes in the method for intelligently assembling a longitudinal beam can be implemented by instructing related hardware through a computer program, and the program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it can include the processes of various embodiments of the method for intelligently assembling a longitudinal beam. Among them, any reference to memory, storage, database or other media used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0085] The embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above-mentioned longitudinal beam intelligent assembly method when executing the program. In the embodiment of the present application, the processor is the control center of the computer system, which can be a processor of a physical machine or a processor of a virtual machine.

[0086] Reference Figure 2 The electronic device 500 includes: at least one processor 501, at least one communication interface 502, at least one memory 503 and at least one bus 504. The bus 504 is used to realize the connection and communication between these components, the communication interface 502 is used to communicate with other node devices for signaling or data, and the memory 503 stores machine-readable instructions executable by the processor 501. When the electronic device 500 is running, the processor 501 communicates with the memory 503 through the bus 504, and when the machine-readable instructions are called by the processor 501, the steps of the above-mentioned longitudinal beam intelligent assembly method are executed.

[0087] The above contents are only embodiments of the present invention. The common sense such as the known specific structures and characteristics in the scheme is not described in detail here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field to which the invention belongs before the application date or the priority date, can obtain all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for the ordinary technicians in the relevant field to implement this application. It should be pointed out that for the technicians in this field, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to explain the content of the claims.

Claims

1. A longitudinal beam intelligent assembly system, characterized in that: include: A workpiece conveying module is used to convey the longitudinal beam to be assembled to a designated position; Positioning and clamping module, used to accurately position and clamp the longitudinal beams delivered to the place; A nut conveying module is used to convey the nuts to be assembled to the position specified by the tightening assembly module; A tightening assembly module, used for tightening the nut to a specified position on the longitudinal beam and acquiring a fourth image; A photographing and identifying module is used to photograph and identify the installation status of nuts on the longitudinal beam; the photographing and identifying module collects at least a first image, a second image and a third image of the longitudinal beam assembly area at a set frequency; a target detection algorithm based on deep learning is used to extract features from the preprocessed image to identify the position and tightening status of the nut in the image; A nut position model is established according to the preset assembly position of the nut on the longitudinal beam; during the image recognition process, the recognized nut position is compared with the position model to determine the nut installation omission and tightening status.

2. The longitudinal beam intelligent assembly system according to claim 1, characterized in that: The first image acquisition timing is before the assembly process starts, the second image acquisition timing is during the assembly process, and the third image acquisition timing is after the assembly process is completed. The acquisition frequency is adjusted according to the assembly speed.

3. The longitudinal beam intelligent assembly system according to claim 2, characterized in that: The photographing and identifying module includes a main visual camera arranged near the tightening assembly module, which is used to capture the overall image of the longitudinal beam. The photographing and identifying module is also provided with a light source device to fill in the light of the shooting area.

4. The intelligent assembly system for longitudinal beams according to claim 1, characterized in that: Calibrate the coordinates of the control points, including the following: It also includes an error-proofing detection module that sets the execution conditions and sequence of each assembly step according to the assembly process requirements, and collects data from each sensor in real time during the assembly process, and verifies and compares it with preset standard data.

5. The intelligent assembly system for longitudinal beams according to claim 1, characterized in that: The workpiece conveying module adopts a conveyor belt to convey the longitudinal beam, and a positioning block is provided on the conveyor belt; the positioning and clamping module adopts a mechanical positioning and clamping device, including a positioning pin, a positioning block, a clamping cylinder and a clamping block; the nut conveying module adopts a vibration plate to convey the nuts, and a photoelectric sensor is provided at the outlet of the vibration plate; the tightening assembly module includes a balance arm and a collaborative robot; the balance arm provides stable support through a cantilever structure design; the collaborative robot is installed at the mobile end of the balance arm.

6. The longitudinal beam intelligent assembly system according to claim 1, characterized in that: The tightening assembly module includes an electric wrench, which has a built-in torque sensor and an angle sensor for real-time monitoring of torque and rotation angle data during the tightening process, and transmitting the data to a control module to control the tightening process and perform error prevention detection; by monitoring the data change curves of the torque sensor and the angle sensor, abnormalities of the nut or bolt can be judged.

7. The intelligent assembly system for longitudinal beams according to claim 6, characterized in that: The front end of the electric wrench is integrated with an auxiliary visual camera for photographing the assembly details of the nut and the bolt at close range, such as the tightening degree of the nut and the meshing condition of the thread, and collecting the fourth image.

8. The intelligent assembly system for longitudinal beams according to claim 7, characterized in that: The workflow of the system includes: During the initialization phase, the control module performs initialization operations on each device, the collaborative robot performs origin return, the electric wrench performs torque calibration and zeroing operations, the camera recognition module performs self-test and parameter setting, each sensor performs initialization configuration, and the control module reads the assembly process parameters and historical data in the database; In the loading stage, the operator places the longitudinal beam to be assembled on the product fixture, and the fixture positions and clamps the longitudinal beam. The nut loading mechanism grabs the nut and transports it to the preset loading position according to the instruction of the control module, and the camera recognition module monitors the nut loading situation in real time; During the assembly phase, the collaborative robot grabs the electric wrench and moves to the nut assembly position. The camera recognition module monitors the robot's position and posture in real time. The auxiliary visual camera at the front end of the electric wrench takes close-up photos of the nuts and bolts. The electric wrench tightens the nuts according to the set parameters. The torque sensor and angle sensor monitor the tightening data in real time, and the camera recognition module continuously captures images of the nut tightening process. During the inspection phase, after the nuts are tightened, the visual recognition module conducts a comprehensive inspection of the assembly status of the nuts. The error-proofing detection module re-inspects the assembly results based on the sensor data and the preset error-proofing logic. After completing the assembly of one longitudinal beam, the system automatically enters the assembly cycle of the next longitudinal beam.

9. The intelligent assembly system for longitudinal beams according to claim 1, characterized in that: The tightening assembly module is also used to optimize the motion trajectory planning of the collaborative robot.

Images