A computer-controlled multi-station transfer robot collaborative control system

Abstract

This invention discloses a computer-controlled multi-station transfer robot collaborative control system, belonging to the field of industrial automation and robot collaborative control technology. It includes a computer control unit, a coordinate positioning module, a text acquisition module, a pressure regulation module, a displacement drive module, a difference analysis module, a secondary adjustment module, a defect observation module, a parameter recording module, and a power management module. The coordinate positioning module integrates laser positioning and inertial measurement with temperature compensation; the text acquisition module uses dual-redundant acquisition and cross-validation; the pressure regulation module dynamically adjusts the pressure based on a PID algorithm; the secondary adjustment module uses an FPGA chip to perform linear calculations to correct deviations; the defect observation module quantifies workpiece defects; the parameter recording module implements closed-loop correction and data optimization; and the power management module uses dual-redundant power supply to ensure stability. This invention can improve multi-station transfer efficiency, reduce workpiece deformation defect rate, and meet the needs of precision manufacturing scenarios.

Description

Technical Field

[0001] This invention relates to the field of industrial automation and robotic arm collaborative control technology, and in particular to a computer-controlled multi-station transfer robotic arm collaborative control system. Background Technology

[0002] In the field of industrial automation, multi-station transfer robots are core equipment for achieving precise workpiece transfer and improving production efficiency, and are widely used in high-precision manufacturing scenarios such as precision electronic components and automotive parts. Existing collaborative control systems for multi-station transfer robots mainly collect position information through a single sensor, combine it with simple control algorithms to achieve motion scheduling, and rely on manually preset parameters to adjust pressure and displacement to meet the workpiece transfer needs between multiple stations. However, as the manufacturing industry's requirements for product precision, production stability, and intelligence levels continue to increase, existing systems are gradually revealing their insufficient adaptability.

[0003] The existing technologies have the following problems: They use a single sensor for positioning, failing to consider the influence of environmental factors such as temperature and vibration, lacking targeted compensation mechanisms, resulting in large coordinate calculation errors and making it difficult to meet the positioning requirements of precision manufacturing; existing pressure control technologies mostly use fixed parameter modes, not linked to position deviations or workpiece defects, and displacement adjustment lacks a secondary correction mechanism, easily leading to unstable workpiece clamping or deformation; existing technologies lack dedicated defect detection and parameter feedback correction modules, making it impossible to dynamically adjust operating parameters according to the actual state of the workpiece, and historical data is not used for algorithm coefficient optimization; traditional communication methods have high latency, making it difficult to guarantee data transmission integrity; the power supply system lacks redundancy design, and single-point failures can easily lead to system shutdown; using a single data source to collect position information lacks a cross-validation mechanism, making it difficult to identify abnormal data and affecting the accuracy of control commands.

[0004] Therefore, it is essential to invent a computer-controlled multi-station transfer robot collaborative control system to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a computer-controlled multi-station transfer robot collaborative control system to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a computer-controlled multi-station transfer robot collaborative control system, specifically comprising:

[0007] The system includes a computer control unit, coordinate positioning module, text acquisition module, pressure regulation module, displacement drive module, difference analysis module, secondary adjustment module, defect observation module, parameter recording module, and power management module. The computer control unit communicates with each module via the Profinet industrial Ethernet bus with a communication delay of ≤5ms, and coordinates centralized data processing, command issuance, and overall module scheduling.

[0008] The coordinate positioning module uses a dual-axis laser positioning sensor and an inertial measurement unit for fusion positioning, and outputs the target's three-dimensional coordinates to the displacement driving module; the text acquisition module uses a high-speed vision sensor and an absolute encoder for dual redundancy to acquire the initial position coordinate text and the position coordinate text after movement.

[0009] The pressure regulation module adjusts the output pressure of the robot arm based on the proportional-integral-derivative (PID) algorithm, and works with the displacement drive module to achieve smooth movement; the difference analysis module uses a real-time operating system (RTOS) to schedule, calculate the coordinate difference between the secondary text and the primary text, and output it to the secondary adjustment module;

[0010] The secondary adjustment module has a built-in FPGA chip that performs linear calculations to generate adjustment instructions for pressure adjustment, displacement compensation, and motion speed correction, which are then sent to the pressure adjustment module and the displacement drive module.

[0011] The defect observation module uses a high-resolution industrial camera and machine vision algorithms to output quantitative information on the external deformation defects of the workpiece to the parameter recording module.

[0012] The parameter recording module uses a non-volatile memory chip to record data and performs closed-loop numerical correction on the output pressure and displacement parameters of the robot arm based on defect quantification information.

[0013] The power management module adopts a dual-redundant power supply design to provide stable power to each module;

[0014] The linear calculation process includes linear calculation of target coordinate positioning, linear calculation of position difference, and linear calculation of parameter correction based on defects. The calculation process substitutes parameters under actual working conditions for compensation calculation.

[0015] The technical effects and advantages of this invention are as follows:

[0016] 1. This invention uses a dual-axis laser positioning sensor and an inertial measurement unit (IMU) for fusion positioning, combined with linear interpolation of the spatial rectangular coordinate system and temperature compensation algorithm. The coordinate calculation error is ≤ ±0.005mm, and the final movement accuracy reaches ±0.01mm, which meets the requirements of precision manufacturing.

[0017] 2. The pressure regulation of this invention is based on the PID algorithm, combined with the linear operation of the secondary adjustment module, which can dynamically generate the adjustment amount according to the position difference value; the displacement drive supports segmented compensation, and the linkage pressure regulation achieves smooth movement, avoiding workpiece clamping failure or deformation.

[0018] 3. This invention quantifies workpiece deformation defects through a defect observation module, and the parameter recording module performs closed-loop parameter correction based on defect data; every 1000 cumulative runs, it automatically fits historical data, optimizes correction coefficients, and continuously improves operational accuracy.

[0019] 4. This invention adopts the Profinet industrial Ethernet bus, with communication latency ≤5ms and data accuracy ≥99.9%; the dual-redundant power supply design can seamlessly switch faulty power supplies within 10ms, and the system's mean time between failures is ≥12,000 hours.

[0020] 5. The text acquisition module of this invention adopts a dual-redundancy design of visual sensor and absolute encoder, combined with Kalman filtering and cross-validation mechanism, and automatically re-acquires data when the deviation exceeds the threshold, so as to ensure that the location data is accurate and reliable. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the module connections of the collaborative control system of the present invention;

[0022] Figure 2 This is a schematic diagram of the linear calculation process of the present invention;

[0023] Figure 3 This is a schematic diagram of the position difference analysis and secondary adjustment process of the present invention;

[0024] Figure 4 This is a schematic diagram of the defect observation and parameter correction process of the present invention. Detailed Implementation

[0025] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0026] This invention provides, for example Figure 1-4The diagram illustrates a computer-controlled multi-station transfer robot collaborative control system, comprising a computer control unit, a coordinate positioning module, a text acquisition module, a pressure regulation module, a displacement drive module, a difference analysis module, a secondary adjustment module, a defect observation module, a parameter recording module, and a power management module. Each module communicates with the computer control unit via a Profinet industrial Ethernet bus to collaboratively achieve precise control of the robot. Specific technical details are as follows:

[0027] The computer control unit adopts the UNO-2484G industrial-grade embedded computer, equipped with a Linux real-time operating system, a main frequency of 2.0GHz, 4GB of memory, 64GB of storage, supports the Profinet bus protocol, and has a communication latency of ≤5ms. Its core functions include centralized data processing, instruction issuance, module collaborative scheduling, and status monitoring. The built-in linear calculation algorithm library includes three types of algorithms: target coordinate interpolation, position difference calculation, and defect parameter correction.

[0028] The computer control unit has a built-in fault diagnosis module that monitors the communication status, power supply status, and data transmission integrity of each module in real time. When a communication interruption, data loss, or power supply abnormality is detected, a graded alarm (audible and visual alarm + upper computer pop-up alarm) is immediately triggered and the fault code is recorded. At the same time, a safe shutdown mode (high-priority fault, such as communication interruption of core modules) or a degraded operation mode (medium-low priority fault, such as abnormal data of non-core sensors) is initiated according to the fault level to ensure the safety of the equipment and workpiece.

[0029] This unit supports remote monitoring and parameter configuration by a host computer. The host computer software is compatible with Windows 10 / 11 and Linux operating systems and can display the system operating status, module working parameters, positioning accuracy data and defect statistics in real time. It supports remote modification of core configurations such as preset thresholds, adjustment coefficients and motion parameters. The modified data is synchronized to the computer control unit through AES encryption to prevent data tampering.

[0030] The coordinate positioning module consists of a Keyence LK-G5000 dual-axis laser positioning sensor, an MPU-9250 inertial measurement unit (IMU), and a PT100 temperature sensor. The laser positioning sensor has a measurement range of 0~500mm and a resolution of 0.001mm, and is used to acquire the current position coordinates of the robot arm. The IMU has a sampling rate of 100Hz and is used to acquire vibration data of the robot arm to compensate for positioning errors. The PT100 temperature sensor has a measurement range of -10℃~60℃ and an accuracy of ±0.1℃, and is used to acquire ambient temperature.

[0031] During operation, the coordinates of two ceramic reference points (P1, P2) are first determined using a calibration tool with an accuracy of ±0.002mm. The temperature coefficient of this ceramic calibration block is ≤1.5×10⁻⁶. -6 / ℃, ensuring the stability of the reference point coordinates, the calibrated coordinates are stored in the parameter recording module; then, the laser sensor collects the current position coordinates of the robot in real time, the IMU collects vibration data to compensate for coordinate errors, and the temperature sensor collects the ambient temperature T; the computer control unit calls the data collected by this module to calculate the straight-line distance L2 from the reference point P1 to P2, and combines it with the straight-line distance L1 from the current position of the robot to the target position measured by the laser sensor to obtain the ratio k=L1 / L2 (0≤k≤1); then according to the linear expansion coefficient α of the robot body steel, α takes a value of 10×10 -6 / ℃-13×10 -6 / ℃, calculate the temperature compensation amount Δx T Δy T Δz T Finally, the k value and temperature compensation amount are substituted into the linear calculation formula of the target coordinates to obtain the three-dimensional coordinates of the target. The coordinate calculation error is ≤ ±0.005mm.

[0032] The linear formula for calculating the target coordinates is: x = x1 + k(x2 - x1) + Δx T y = y1 + k(y2 - y1) + Δy T z = z1 + k(z2 - z1) + Δz T k = L1 / L2;

[0033] The temperature compensation amounts are ΔxT=α・(T-T0)・x1, ΔyT=α・(T-T0)・y1, and ΔzT=α・(T-T0)・z1, where α is the coefficient of linear expansion of the steel of the robot body; T is the current ambient temperature measured by the PT100 temperature sensor, which has a measurement range of -10℃ to 60℃ and a measurement accuracy of ±0.1℃; and T0 is the calibration ambient temperature.

[0034] The text acquisition module consists of a Baslerac A2500-14gm high-speed vision sensor, a Heidenhain ROD426 1024-line absolute encoder, and an FPGA preprocessing chip. The vision sensor has a pixel count of 2592×1944 and a frame rate of 14fps. It captures workpiece positioning marks through a 16mm industrial lens and extracts the initial position coordinate text and the position coordinate text after movement. The absolute encoder outputs an RS485 signal to synchronously acquire robot joint position data as a redundancy verification of coordinate data. The FPGA preprocessing chip is used to perform Kalman filtering noise reduction on the acquired coordinate data to remove outliers with deviations exceeding three times the standard deviation.

[0035] The acquired coordinate data is generated as text in the format "x,y,z,t", where x, y, and z coordinate values ​​are rounded to four decimal places in mm, and t is a timestamp with a precision of 1ms. The text data is transmitted to the computer control unit in real time via the Profinet bus with a transmission delay of ≤10ms and a data accuracy of ≥99.9%. This module employs a cross-validation mechanism; if the coordinate deviation between the vision sensor and the encoder exceeds 0.02mm, it is considered an acquisition anomaly, immediately triggering a re-acquisition and sending an alarm signal to the computer control unit.

[0036] The pressure regulation module consists of a FestoMPPE-3-1 / 8-6-010-B type proportional pressure valve, a SICKPBT-RB040SG1SS type pressure sensor, and a PID control circuit. The proportional pressure valve has an adjustment range of 0.1~0.8MPa and a response time of ≤10ms, and is used to adjust the output pressure of the robotic arm. The pressure sensor has a measurement accuracy of ±0.5%FS and is used to provide real-time feedback on the output pressure. The PID control circuit has built-in preset PID parameters (proportional coefficient Kp=2.5, integral coefficient Ki=0.3, derivative coefficient Kd=0.1), which can be finely adjusted via a host computer.

[0037] This module communicates with the computer control unit in real time, receiving pressure control commands and dynamically adjusting the output pressure: when the robot's moving speed is >0.3m / s, the pressure automatically increases by 10% to enhance gripping stability; when receiving a pressure adjustment command from the secondary adjustment module, the pressure adjustment is completed within 5ms. The module has built-in overpressure, underpressure, and pressure relief protection circuits. When the pressure sensor detects that the pressure exceeds 0.85MPa (overpressure threshold) or falls below 0.08MPa (underpressure threshold), it automatically performs pressure relief or pressure replenishment actions, while locking the robot's movement function until the pressure returns to the normal range of 0.1~0.8MPa, preventing workpiece damage or gripping failure.

[0038] The displacement drive module consists of a Panasonic MSMA042A1G servo motor, a Panasonic MDDDT3530 servo driver, a C3-grade ball screw (5mm lead), a THKSR30 linear guide, and a position feedback sensor. The servo motor has a rated power of 400W and a rated speed of 3000rpm, providing power for the robot's movement. The parallelism error of the linear guide is ≤0.01mm / m to ensure smooth movement. The position feedback sensor is used to collect the robot's position information in real time.

[0039] Based on the target coordinate commands output by the computer control unit, the servo driver adopts a pulse + direction control mode (pulse equivalent 0.001mm / pulse) to drive the robot arm to move along the X, Y, and Z axes with a movement accuracy of ±0.01mm, a maximum movement speed of 0.5m / s, and an acceleration time of ≤0.2s. This module supports three motion modes: jog control (manual single-axis movement), continuous trajectory control (movement along a preset path), and teach-and-playback control (repeated execution after recording the teach path). Motion trajectory parameters can be preset and stored in the computer control unit via upper-level software, adapting to different production scenarios. The module collects the position feedback signals and current signals of the servo motor in real time and uploads them to the computer control unit, linking with the pressure regulation module to ensure smooth and precise movement.

[0040] The difference analysis module uses an STM32H743VI embedded processor with a main frequency of 480MHz and runs on the FreeRTOS real-time operating system. It is used for rapid processing of position difference calculations. After receiving the primary and secondary text forwarded by the computer control unit, it first parses the initial position A(x) to obtain the result. a ,y a ,z a ), the position coordinates B(x) after the movement B ,y B ,z B The data includes the corresponding timestamps t1 and t2, and the time difference Δt = t2 - t1 is calculated. If Δt > 50ms, it is determined to be an abnormal acquisition and a re-acquisition is triggered. If the data is valid, it is substituted into the Euclidean distance formula with weighted coefficients to perform linear calculation of the position difference to obtain the position difference value ΔS.

[0041] The linear calculation formula for the position difference value ΔS is: ΔS = √[w X ・(x B -x a ) 2 +w Y ・(y B -y a ) 2 +w z ・(z B -z a ) 2 ];

[0042] Wherein, the weighting coefficient w X w Y w Z The robot arm's load distribution, workpiece center of gravity position, or process requirements are dynamically adjusted. For example, when the workpiece has a large load in the X-axis direction, w X It can be set to 1.2 to enhance the sensitivity to differences in this direction.

[0043] The computer control unit calls the preset allowable difference threshold ΔS0 (range 0.01mm~0.1mm, default value 0.05mm) and compares ΔS with ΔS0: when ΔS≤ΔS0, a qualified signal is sent to the parameter recording module; when ΔS>ΔS0, priority is marked according to the size of ΔS (ΔS>0.1mm is high priority, 0.05mm<ΔS≤0.1mm is medium priority), and the difference data is sent to the secondary adjustment module.

[0044] The secondary adjustment module uses a Xilinx Artix-7 FPGA chip with a logic operation speed of ≥100MHz for fast linear calculation. After receiving ΔS and priority from the difference analysis module, high-priority data is processed first, with an adjustment command issuance delay of ≤3ms. Based on the preset adjustment coefficients and initial offset, the pressure adjustment ΔP and the X, Y, and Z-axis displacement compensation (ΔL) are calculated using the adjustment calculation formula and compensation calculation formula, respectively. X ΔL Y ΔL Z ).

[0045] The formula for calculating the adjustment amount is ΔP = k1・ΔS + b1;

[0046] The compensation amount is calculated using the formula ΔL. X =k X ・ΔS+b X ΔL Y =k Y ・ΔS+b Y ΔL Z =k Z ・ΔS+b Z Where k1 is the pressure adjustment coefficient, ranging from 0.3 MPa / mm to 0.7 MPa / mm, with a default value of 0.5 MPa / mm; b1 is the initial pressure adjustment offset, ranging from 0.01 MPa to 0.03 MPa, with a default value of 0.02 MPa; k X k Y k Z The displacement adjustment coefficients for each axis range from 0.4 mm / mm to 0.6 mm / mm, with a default value of 0.5 mm / mm; b X b Y b Z The initial offset for each axis displacement ranges from 0.002mm to 0.008mm, with a default value of 0.005mm.

[0047] This module adopts a segmented adjustment strategy, first quickly compensating for 80% of the deviation, and then finely compensating for the remaining 20%, avoiding over-adjustment that could lead to secondary deviations. After the adjustment command is reviewed by the computer control unit, it is simultaneously sent to the pressure regulation module and the displacement drive module to achieve precise correction of the position deviation.

[0048] The defect observation module consists of a Canon MS-500 5-megapixel industrial camera, a telecentric lens (50mm focal length, 100~200mm shooting distance), a CCSRL-100 ring LED light source, and a machine vision algorithm module (developed based on OpenCV). The brightness adjustment range of the LED light source is 30%~100%, which can be adaptively adjusted according to the workpiece material (metal, plastic, glass, etc.) and ambient light intensity to ensure the clarity and stability of defect image acquisition.

[0049] During operation, an industrial camera captures images of the workpiece after it has been gripped, and transmits these images to a machine vision algorithm module. Defect regions are extracted using image segmentation and edge detection techniques, and the area S of the defect region is calculated. defect The maximum depth d of the defect was measured using focused stacking technology. depth Substituting into the defect-based parameter correction linear calculation formula D=(S defect / S total )×ω+(d depth / d max )×(1-ω)(S total d represents the total area of ​​the workpiece clamping region. max =0.5mm, ω is the area weighting coefficient, with a value range of 0.4~0.6 and a default value of 0.5. ω is preset according to the workpiece material, defect type, or process requirements), to obtain the defect degree quantification value D (D∈[0,1]); generate "D,S defect ,d depth Defect information in the format ",t" is output to the computer control unit; when D≥0.8, an emergency stop signal is immediately sent to the computer control unit to avoid the generation of batches of defective products.

[0050] The parameter recording module consists of a 32GB non-volatile SD card (read / write speed ≥10MB / s), a real-time clock (RTC, accuracy ±1min / month), and a data management unit. Its core functions include data storage, parameter correction, and optimization: real-time storage of target coordinates, initial pressure, displacement parameters, ΔS, D value, adjustment amount, fault codes, etc., for each run. Each record is timestamped, and it can store ≥100,000 running data entries with a data retention period ≥1 year. After receiving the D value from the defect observation module, the computer control unit calls the correction coefficients and baseline values ​​stored in this module, substitutes them into the linear calculation formula to generate the pressure correction amount ΔF and the triaxial displacement correction amount (ΔL). X '、ΔL Y'、ΔL Z Feedback is sent to the pressure regulation module and displacement drive module; every 1000 cumulative runs, the least squares method is used to fit historical data and optimize k2 and k 3X The default values ​​of equal coefficients continuously improve the accuracy of system operation.

[0051] This module supports data export, allowing historical operating data, fault logs, and parameter optimization reports to be exported as CSV files via USB or Ethernet, facilitating subsequent data analysis, troubleshooting, and system maintenance.

[0052] The power management module consists of a MeanWell RSP-300-24 switching power supply, a MeanWell RSP-100-5 switching power supply, and a PhoenixContactQUINT-ORING24DC / 20 redundant switching module. The input voltage is AC220V±10%, and the output voltage is DC24V (accuracy ±0.1V) and DC5V (accuracy ±0.05V). The DC24V supply powers high-power devices such as servo motors and proportional pressure valves, while the DC5V supply powers low-power devices such as sensors and embedded processors.

[0053] The module incorporates overvoltage protection (DC24V output threshold 28V, DC5V output threshold 5.6V), overcurrent protection (DC24V output threshold 10A, DC5V output threshold 8A), short-circuit protection (response time ≤1ms), and surge protection (2kV) circuits. Upon detecting a power supply abnormality, it immediately disconnects the faulty circuit and triggers an alarm. Employing a dual-redundant power supply design, when one power source fails, the redundancy switching module can seamlessly switch to the other power source within 10ms, ensuring continuous system operation with a mean time between failures (MTBF) of ≥12,000 hours.

[0054] It should be further noted that the value ranges of various parameters involved in this invention (including but not limited to the linear expansion coefficient α, weighting coefficient w) are as follows: X / w Y / w Z Adjustment coefficients k1 / b1 / k X / b X Defect weighting coefficient ω, correction coefficient k2 / b2 / k 3X / b 3XThe method is based on more than 1,000 systematic tests and data fitting optimizations of various typical materials (such as aluminum alloys, stainless steel, and engineering plastics), different sizes (such as 5kg, 10kg, and 20kg loads), and different process requirements. Those skilled in the art can make adaptation adjustments within the range or optimize it using the self-learning function of the parameter recording module according to the actual physical characteristics (mass, stiffness, and coefficient of friction), environmental conditions, and precision requirements of the workpiece in specific application scenarios.

[0055] The invention will now be described in further detail using the application scenario of a precision electronic component production line (producing mobile phone camera modules).

[0056] Equipment parameter settings:

[0057] Robot model: XYZ-300, maximum load 5kg, X / Y / Z axis travel 500mm, moving speed range 0.1-0.5m / s;

[0058] System core parameters: Preset allowable difference threshold ΔS0 = 0.05 mm; pressure adjustment coefficient k1 = 0.5 MPa / mm, initial pressure adjustment offset b1 = 0.02 MPa; displacement adjustment coefficients for each axis k X =k Y =k Z =0.5mm / mm, initial offset b of each axis displacement X =b Y =b Z =0.005mm; Pressure defect correction coefficient k2=0.2MPa, pressure correction reference value b2=0.01MPa; Displacement defect correction coefficient for each shaft k 3X =k 3Y =k 3Z =0.03mm, correction reference value b for each axis displacement 3X =b 3Y =b 3Z =0.005mm; area weighting coefficient ω=0.5, maximum allowable defect depth d max =0.5mm;

[0059] Defect quantification value D judgment criteria: No deformation D=0, slight deformation (dent depth < 0.1mm) D=0.3, moderate deformation (0.1mm ≤ dent depth < 0.3mm) D=0.6, severe deformation (dent depth ≥ 0.3mm) D=0.9.

[0060] Target coordinate positioning:

[0061] Using a calibration tool with an accuracy of ±0.002mm, the initial workstation reference point P1 (0,0,0)mm and the assembly workstation reference point P2 (80,80,40)mm were determined, and the coordinate data was stored in the parameter recording module. The temperature sensor measured the current ambient temperature as T=30℃, and the linear expansion coefficient α of the robot arm body steel was 11.5×10⁻⁶. -6 / ℃, calibration temperature T0=25℃; calculate the straight-line distance L2 from reference point P1 to P2 = √(80 2 +80 2 +40 2 =120mm; The straight-line distance L1 from the current position (15,15,8)mm of the robot arm to the target position is measured by the laser rangefinder, which is 75mm. Calculate k = 75 / 120 = 0.625; Calculate the temperature compensation Δx. T =11.5×10 -6 ×(30-25)×15≈0.00086mm, Δy T =0.00086mm, Δz T =11.5×10 -6 ×(30-25)×8≈0.00046mm; Substituting into the linear interpolation formula to calculate the target coordinates: x=0+0.625×80+0.00086=50.00086mm y=0+0.625×80+0.00086=50.00086mm z=0+0.625×40+0.00046=25.00046mm

[0062] The final target coordinates are (50.0009, 50.0009, 25.0005) mm (rounded to four decimal places), which are sent from the computer control unit to the displacement drive module.

[0063] Text capture and movement control in one operation:

[0064] The text acquisition module's vision sensor captures the positioning mark of the mobile phone camera module at the loading station, extracts the initial position coordinates, and generates the text "50.0009,50.0009,25.0005,200.3" (timestamp t1=200.3ms). After the joint position data acquired by the absolute encoder is verified to be correct, it is filtered by the FPGA and uploaded to the computer control unit. The computer control unit sends an initial pressure command of 0.3MPa to the pressure regulation module, which uses a PID algorithm to stabilize the output pressure at 0.3MPa±0.005MPa. At the same time, the displacement drive module receives the target coordinate command and drives the robot to move to the position (50.0009,50.0009,25.0005)mm at a speed of 0.3m / s.

[0065] Secondary text acquisition and difference analysis:

[0066] After the robotic arm moves into position, the text acquisition module collects the coordinates of its new position and generates secondary text "50.0709,50.0409,25.0505,248.6" (timestamp t2=248.6ms). The time difference Δt = 248.6 - 200.3 = 48.3ms < 50ms, indicating the data is valid. The difference analysis module parses the coordinate data and calculates the position difference value.

[0067] ΔS=√[1.0×(50.0709-50.0009) 2 +1.0×(50.0409-50.0009) 2 +1.0×(25.0505-25.0005) 2 ]=√(0.0049+0.0016+0.0025)=√0.009≈0.0949mm>0.05mm

[0068] Data classified as medium-priority difference is sent to the secondary adjustment module.

[0069] Second adjustment:

[0070] After receiving the difference data, the secondary adjustment module calculates the adjustment amount:

[0071] ΔP=0.5×0.0949+0.02≈0.0674MPa;

[0072] ΔL X =0.5×0.07+0.005=0.04mm;

[0073] ΔL Y =0.5×0.04+0.005=0.025mm;

[0074] ΔL Z =0.5×0.05+0.005=0.03mm;

[0075] After the adjustment command is reviewed by the computer control unit, it is sent to the pressure regulation module and the displacement drive module. The pressure regulation module increases the output pressure to 0.3 + 0.0674 = 0.3674 MPa. The displacement drive module adopts a segmented adjustment strategy, first quickly compensating for 80% of the deviation (X-axis 0.032 mm, Y-axis 0.02 mm, Z-axis 0.024 mm), and then finely compensating for the remaining 20%. After adjustment, the secondary text is updated to "50.0009, 50.0009, 25.0005, 251.8", ΔS≈0 mm, which meets the accuracy requirements.

[0076] Defect observation and parameter correction:

[0077] The defect observation module uses an industrial camera to capture images of the clamped mobile phone camera module. The brightness of the ring LED light source is adjusted to 80%. After image acquisition, defect information is extracted using machine vision algorithms, and the defect area S is measured. defect =2mm 2 Maximum depth of defect d depth =0.08mm; Calculate the defect quantification value: D=(2 / 100)×0.5+(0.08 / 0.5)×0.5=0.01+0.08=0.09;

[0078] The computer control unit calls the correction coefficient from the parameter recording module to calculate the correction amount:

[0079] ΔF=0.2×0.09+0.01=0.028MPa;

[0080] ΔL X =0.03×0.09+0.005≈0.0077mm

[0081] The pressure regulation module adjusts the output pressure to 0.3674-0.028=0.3394MPa; the displacement drive module fine-tunes the X-axis displacement by 0.0077mm to ensure that subsequent clamping is defect-free.

[0082] Data recording and optimization:

[0083] The parameter recording module records data such as target coordinates, initial pressure, ΔS, D value, and adjustment amount for this operation, generating a record:

[0084] "50.0009, 50.0009, 25.0005, 0.3MPa, 0.0949mm, 0.09, 0.0674MPa, 0.04mm, 2024-05-2014:32:15"; After 1000 runs, the system automatically uses the least squares method to fit the historical data, optimizing k2 to 0.19MPa, k 3X The accuracy of defect correction in subsequent operations has been improved to 0.029mm.

[0085] Through the above implementation process, the positioning accuracy of the robotic arm is stabilized within ±0.01mm, the multi-station transfer efficiency is improved by 35%, and the deformation defect rate of the mobile phone camera module is controlled below 0.8%, meeting the high-precision and high-efficiency production requirements of the production line.

[0086] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A computer-controlled multi-station transfer robot collaborative control system, characterized in that, It includes a computer control unit, coordinate positioning module, text acquisition module, pressure regulation module, displacement drive module, difference analysis module, secondary adjustment module, defect observation module, parameter recording module, and power management module; the computer control unit communicates with each module via Profinet industrial Ethernet bus with a communication delay of ≤5ms, and coordinates centralized data processing, command issuance, and full module collaborative scheduling. The coordinate positioning module uses a dual-axis laser positioning sensor and an inertial measurement unit for fusion positioning, and outputs the target's three-dimensional coordinates to the displacement driving module; the text acquisition module uses a high-speed vision sensor and an absolute encoder for dual redundancy to acquire the initial position coordinate text and the position coordinate text after movement. The pressure adjustment module adjusts the output pressure of the robot arm based on the proportional-integral-differential algorithm, and works with the displacement drive module to achieve smooth movement; the difference analysis module uses a real-time operating system to calculate the coordinate difference between the secondary text and the primary text and outputs it to the secondary adjustment module. The secondary adjustment module has a built-in FPGA chip that performs linear calculations to generate adjustment instructions for pressure adjustment, displacement compensation, and motion speed correction, which are then sent to the pressure adjustment module and the displacement drive module. The defect observation module uses a high-resolution industrial camera and machine vision algorithms to output quantitative information on the external deformation defects of the workpiece to the parameter recording module. The parameter recording module uses a non-volatile memory chip to record data and performs closed-loop numerical correction on the output pressure and displacement parameters of the robot arm based on defect quantification information. The power management module adopts a dual-redundant power supply design to provide stable power to each module; The linear calculation process includes linear calculation of target coordinate positioning, linear calculation of position difference, and linear calculation of parameter correction based on defects. The calculation process substitutes parameters under actual working conditions for compensation calculation. The linear calculation of the target's three-dimensional coordinates is performed by a computer control unit, using a target coordinate linear calculation formula combined with a temperature compensation algorithm. Specifically: Let the coordinates of two reference points be P1(x1,y1,z1) and P2(x2,y2,z2); the target point P(x,y,z) lies on the line connecting P1 and P2. The ratio of the distance from P1 to P to the distance from P1 to P2 is k, where k ranges from 0 to 1. The target coordinate linear calculation formula is: x = x1 + k(x2 - x1) + Δx T y = y1 + k(y2 - y1) + Δy T z = z1 + k(z2 - z1) + Δz T k = L1 / L2, where L1 is the straight-line distance from the current position of the robot to the target position measured by the laser rangefinder, and the measurement accuracy of the laser rangefinder is ±0.001mm; L2 is the straight-line distance from the reference point P1 to P2 preset and stored in the parameter recording module of the calibration tool; temperature compensation Δx T =α・(T-T0)・x1、Δy T =α・(T-T0)・y1、Δz T =α・(T-T0)・z1, where α is the coefficient of linear expansion of the steel of the robot body; T is the current ambient temperature measured by the PT100 temperature sensor, which has a measurement range of -10℃ to 60℃ and a measurement accuracy of ±0.1℃; T0 is the calibration ambient temperature; The linear calculation of the positional difference is completed by the computer control unit calling data from the difference analysis module, using the Euclidean distance formula combined with weighting coefficients. Specifically, let the initial position coordinates corresponding to a single text be A(x). a ,y a ,z a The timestamp of the second text is t1; the coordinates of the moved position of the second text are B(x). B ,y B ,z B The timestamp is t2; the position weight coefficients w for the X, Y, and Z axes are... X w Y w Z The values ​​range for all values ​​are [0.8, 1.2], with a default value of 1.0; the linear calculation formula for the positional difference value ΔS is: ΔS = √[w X ・(x B -x a ) 2 +w Y ・(y B -y a ) 2 +w z ・(z B -z a ) 2 When ΔS > ΔS0 and Δt = t2 - t1 ≤ 50 ms, the secondary adjustment module performs linear adjustment based on ΔS, where ΔS0 is the preset allowable difference threshold of the computer control unit, with a value range of 0.01 mm to 0.1 mm and a default value of 0.05 mm; the linear relationship of the adjustment amount is: ΔP = k1・ΔS + b1, ΔL X =k X ・ΔS+b X ΔL Y =k Y ・ΔS+b Y ΔL Z =k Z ・ΔS+b Z Where k1 is the pressure adjustment coefficient, ranging from 0.3 MPa / mm to 0.7 MPa / mm, with a default value of 0.5 MPa / mm; b1 is the initial pressure adjustment offset, ranging from 0.01 MPa to 0.03 MPa, with a default value of 0.02 MPa; k X k Y k Z The displacement adjustment coefficients for each axis range from 0.4 mm / mm to 0.6 mm / mm, with a default value of 0.5 mm / mm; b X b Y b Z The initial offset for each axis displacement ranges from 0.002mm to 0.008mm, with a default value of 0.005mm. The defect-based parameter correction linear calculation is performed by the computer control unit, specifically: the formula for calculating the defect severity quantification value D is D=(S defect / S total )×ω+(d depth / d max )×(1-ω); where S defect S represents the area of ​​the defect region. total d represents the total area of ​​the workpiece clamping region. depth To determine the maximum depth of the defect measured using focused stacking technology, d max ω is the preset maximum allowable defect depth; ω is the area weighting coefficient, ranging from 0.4 to 0.6, with a default value of 0.5; D ranges from [0,1], where D=0 indicates no defect and D=1 indicates a severe defect; the linear relationship between the robot's output pressure correction and displacement correction is: ΔF=k2・D+b2、ΔL X '=k 3X ・D+b 3X ΔL Y '=k 3Y ・D+b 3Y ΔL Z '=k 3Z ・D+b 3Z Where k2 is the pressure defect correction coefficient, ranging from 0.1MPa to 0.3MPa, with a default value of 0.2MPa; b2 is the pressure correction reference value, ranging from 0.005MPa to 0.015MPa, with a default value of 0.01MPa; k 3X k 3Y k 3Z These are correction factors for displacement defects on each axis, with values ​​ranging from 0.02mm to 0.04mm and a default value of 0.03mm; b 3X b 3Y b 3Z The reference values ​​for displacement correction of each axis are set, with a range of 0.003mm to 0.007mm and a default value of 0.005mm. The parameter recording module automatically updates k2 and b every 1000 runs by fitting historical data using the least squares method. 3X b 3Y b 3Z The default value.

2. The computer-controlled multi-station transfer robot collaborative control system according to claim 1, characterized in that, The computer control unit is a UNO-2484G industrial-grade embedded computer, equipped with a Linux real-time operating system, with a main frequency of ≥2.0GHz, memory ≥4GB, and storage ≥64GB. The computer control unit has a built-in fault diagnosis module that monitors the communication status, power supply status, and data transmission integrity of each module in real time. When a fault is detected, it triggers a graded alarm and records the fault code, while simultaneously initiating a safe shutdown or degraded operation mode. The computer control unit supports remote monitoring and parameter configuration by a host computer. The host computer software is compatible with Windows 10 / 11 and Linux operating systems. Modified parameters are synchronized to the computer control unit via AES encrypted transmission.

3. The computer-controlled multi-station transfer robot collaborative control system according to claim 1, characterized in that, The text acquisition module includes a Baslerac A2500-14gm 2-megapixel vision sensor and a Heidenhain ROD426 1024-line absolute encoder; the text format is "x,y,z,t", where t is a timestamp with a precision of 1ms; the acquired data is uploaded to the computer control unit after Kalman filtering; when the coordinate deviation between the vision sensor and the encoder exceeds 0.02mm, it is determined to be an acquisition anomaly, triggering re-acquisition and sending an alarm signal to the computer control unit.

4. The computer-controlled multi-station transfer robot collaborative control system according to claim 1, characterized in that, The pressure regulation module includes a FestoMPPE-3-1 / 8-6-010-B type proportional pressure valve and a SICKPBT-RB040SG1SS type pressure sensor; the pressure regulation range is 0.1MPa~0.8MPa, and the pressure adjustment response time is ≤5ms; the pressure regulation module has built-in overpressure, underpressure, and pressure relief protection circuits. When the pressure is detected to exceed 0.85MPa or fall below 0.08MPa, it automatically performs pressure relief or pressure replenishment actions, and locks the robot arm movement function until the pressure returns to the normal range of 0.1MPa~0.8MPa.

5. The computer-controlled multi-station transfer robot collaborative control system according to claim 1, characterized in that, The displacement drive module includes a Panasonic MSMA042A1G servo motor, a C3-grade ball screw, and a THKSR30 linear guide. The ball screw has a lead of 5mm. The displacement drive module has a movement accuracy of ±0.01mm and a maximum movement speed of 0.5m / s. The displacement drive module supports three motion modes: inching control, continuous trajectory control, and teach-and-playback control. The motion trajectory parameters are preset by the host computer software and stored in the computer control unit.

6. The computer-controlled multi-station transfer robot collaborative control system according to claim 1, characterized in that, The defect observation module includes a Canon MS-500 5-megapixel industrial camera and a CCSRL-100 ring LED light source; the defect detection accuracy is 0.01mm; the brightness adjustment range of the LED light source is 30%~100%, which is adaptively adjusted according to the workpiece material and ambient light intensity; when the defect quantification value D≥0.8, the defect observation module immediately sends an emergency stop signal to the computer control unit.

7. The computer-controlled multi-station transfer robot collaborative control system according to claim 1, characterized in that, The power management module includes a Mean-Well RSP-300-24 switching power supply, a MeanWell RSP-100-5 switching power supply, and a Phoenix-ContactQUINT-ORING24DC / 20 redundant switching module. The input voltage of the power management module is AC220V±10%, and the output voltage is DC24V and DC5V, with the output accuracy of DC24V being ±0.1V and DC5V being ±0.05V. The power management module has built-in overvoltage, overcurrent, short circuit, and surge protection circuits, with protection thresholds of DC28V, 10A, ≤500mA, and 2kV, respectively. When one power supply fails, the other power supply seamlessly switches within 10ms through the redundant switching module, and the system's mean time between failures (MTBF) is ≥12,000 hours.

Images