Furnace front quick separation robot teaching method and related equipment

Abstract

The invention discloses a furnace front fast separation robot teaching method and related equipment, and relates to the technical field of industrial robot motion control, and the method comprises the steps: obtaining a target precision threshold value of a target process point; determining an adaptive step convergence algorithm according to the target precision threshold and the robot dynamic parameters; in response to the offset direction teaching signal, generating an iterative moving step length through an adaptive step length convergence algorithm; and controlling the stokehole fast separation robot to move according to the iterative movement step length until a preset convergence condition is met. According to the method, efficient and accurate positioning is realized through adaptive step size convergence control, the operation process can be simplified, the manual dependence is reduced, and the teaching consistency and universality are improved.

Description

Technical Field

[0001] The present application relates to the technical field of industrial robot motion control, and more specifically, to a furnace-front quick-splitting robot teaching method and related equipment. Background Art

[0002] With the continuous development of robotics and intelligent manufacturing, furnace-front quick-splitting robots are taking on increasingly critical tasks in metallurgical production scenarios involving high temperatures, high risks, and heavy loads. To ensure their operational accuracy and path reliability, the teaching process, a key pre-setting step before the furnace-front quick-splitting robots execute their tasks, plays an important role in guiding the robots to complete operational planning, including motion trajectories and key points. In practical applications, especially teaching tasks requiring high-precision positioning within ±0.5mm, higher requirements are placed on the teaching method's stability, efficiency, and ability to adapt to complex environmental changes.

[0003] However, in the existing technology, the teaching process mainly adopts a non-quantitative teaching method, which is not only inefficient, but also heavily dependent on the operator's experience, judgment and manual adjustment capabilities, and the operation process lacks an intelligent feedback mechanism. In high-temperature and highly vibrating work sites such as furnace environments, traditional teaching methods often face problems such as too many iterations, slow convergence, and unstable positioning. Especially in the initial stage of offset and the convergence process near the target point, it is difficult to achieve matching adjustment of step length and error, resulting in the robot being unable to efficiently achieve the accuracy target within a limited time, affecting the teaching quality and the reliability of automatic operation. In other words, the relevant technologies generally have technical problems such as low convergence efficiency of teaching step length, poor precision control, and heavy control burden. Summary of the Invention

[0004] The Summary of the Invention section of this application introduces a series of simplified concepts that will be further described in detail in the Detailed Description of the Invention section. The Summary of the Invention section of this application is not intended to limit the key features and essential technical features of the claimed technical solution, nor is it intended to determine the scope of protection of the claimed technical solution.

[0005] The teaching method and related equipment of the furnace-front quick-splitting robot provided in this application can achieve efficient and precise positioning through adaptive step-length convergence control, simplify the operating process, reduce manual dependence, and improve the consistency and universality of teaching.

[0006] In the first aspect, the present application provides a teaching method for a furnace-front fast-splitting robot, comprising: obtaining a target accuracy threshold of a target process point; determining an adaptive step-length convergence algorithm based on the target accuracy threshold and the robot dynamic parameters, wherein the adaptive step-length convergence algorithm includes an initial step-length parameter and a convergence ratio parameter; generating an iterative movement step-length through the adaptive step-length convergence algorithm in response to an offset direction teaching signal; and controlling the movement of the furnace-front fast-splitting robot according to the iterative movement step-length until a preset convergence condition is reached.

[0007] In some embodiments, the preset convergence condition is that the residual movement error of the furnace-front fast-splitting robot is less than half of the target accuracy threshold.

[0008] In some embodiments, obtaining the target accuracy threshold of the target process point includes: obtaining the application scenario of the target process point; when the application scenario is a sample transfer scenario or a processing placement scenario, determining the target accuracy threshold to be a first preset value; when the application scenario is a spectral excitation scene or a visual detection scene, determining the target accuracy threshold to be a second preset value, wherein the second preset value is less than the first preset value.

[0009] In some embodiments, the adaptive step size convergence algorithm is determined based on the target accuracy threshold and the robot dynamic parameters, including: when the target accuracy threshold is the first preset value, determining the initial step size parameter of the adaptive step size convergence algorithm to be the first step size value, and the convergence ratio parameter of the adaptive step size convergence algorithm to be the first ratio according to the robot dynamic parameters; when the target accuracy threshold is the second preset value, determining the initial step size parameter of the adaptive step size convergence algorithm to be the second step size value, and the convergence ratio parameter of the adaptive step size convergence algorithm to be the second ratio according to the robot dynamic parameters, wherein the second step size value is smaller than the first step size value, and the second ratio is smaller than the first ratio.

[0010] In some embodiments, the robot dynamic parameters include the gripper stiffness coefficient, the limit moving speed and the historical collision offset statistics of the furnace front fast separation robot.

[0011] In some embodiments, the teaching method of the furnace-front quick-splitting robot further includes: determining the misalignment direction of the excitation hole relative to the positioning hole by the coaxial state of the positioning hole of the teaching standard block and the target workstation excitation hole; and generating the offset direction teaching signal according to the misalignment direction.

[0012] In some embodiments, the teaching method of the furnace-front fast-splitting robot also includes: when the number of movement iterations of the furnace-front fast-splitting robot is greater than the preset number of iterations corresponding to the target accuracy threshold, controlling the furnace-front fast-splitting robot to switch to a quantitative movement mode for movement, and triggering a hardware abnormality warning signal, wherein the quantitative movement mode is a state of executing movement by referring to a pre-stored fixed step size parameter.

[0013] In the second aspect, the present application also provides a teaching device for a furnace-front fast-splitting robot, comprising: a data acquisition unit for acquiring a target accuracy threshold of a target process point; an algorithm determination unit for determining an adaptive step-length convergence algorithm based on the target accuracy threshold and the robot dynamic parameters, wherein the adaptive step-length convergence algorithm includes an initial step-length parameter and a convergence ratio parameter; a step-length generation unit for generating an iterative movement step length through the adaptive step-length convergence algorithm in response to an offset direction teaching signal; and a teaching movement unit for controlling the movement of the furnace-front fast-splitting robot according to the iterative movement step length until a preset convergence condition is reached.

[0014] In a third aspect, the present application further provides an electronic device comprising: a memory and a processor, wherein the processor is configured to implement the steps of the furnace front quick separation robot teaching method described in the first aspect when executing the computer program stored in the memory.

[0015] In a fourth aspect, the present application further provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the furnace front quick separation robot teaching method described in the first aspect.

[0016] In a fifth aspect, the present application also provides a computer program product, including a computer program or computer executable instructions. When the computer program or computer executable instructions are executed by a processor, the furnace front fast separation robot teaching method provided in the embodiment of the present application is implemented.

[0017] In summary, the present application realizes adaptive convergence control by dynamically adjusting the iteration step size according to the accuracy threshold of the target process point, that is, the furnace-front fast-splitting robot can automatically reduce the step size when approaching the target position to improve positioning accuracy; and when the initial offset is large, a larger step size is used to speed up the approach speed, which can improve the precision control capability and the overall teaching efficiency; the operator only needs to input the offset direction, and the algorithm can automatically calculate and control the movement to complete the positioning convergence, which can greatly reduce manual intervention in repeated fine-tuning and experience judgment, lower the operation threshold, and at the same time improve the consistency and reproducibility of the teaching results, and is suitable for operators with different skill levels. In summary, the furnace-front fast-splitting robot teaching method provided by the present application achieves efficient and precise positioning through adaptive step-size convergence control, which can simplify the operating process, reduce manual dependence, and improve the consistency and universality of teaching. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Various other advantages and benefits will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiment below. The accompanying drawings are for illustration purposes only and are not to be considered as limiting the present description. The same reference symbols are used throughout the drawings to represent the same components. In the drawings:

[0019] Figure 1 A flow chart of a teaching method for a furnace-front quick separation robot provided in an embodiment of the present application;

[0020] Figure 2 A schematic structural diagram of a teaching standard block provided in an embodiment of the present application;

[0021] Figure 3 A schematic diagram of the structure of a furnace-front quick separation robot teaching device provided in an embodiment of the present application;

[0022] Figure 4 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] Terms in the specification, claims, and drawings of this application, such as "first," "second," "third," "fourth," and the like (if any), are used to distinguish between similar objects, rather than to describe a particular order or precedence. Therefore, it is understood that these terms can be used interchangeably where appropriate, so that the embodiments described can be implemented in a different order, unless otherwise specified in the drawings or descriptions. In addition, the terms "is" and "has" and any variations thereof in this application are intended to cover all possible constituent elements on a non-exclusive basis. For example, a process, method, system, product, or apparatus that includes several steps or units is not necessarily limited to the steps or units that are explicitly listed, but may also include other steps or units that are not explicitly listed, or steps or units that are inherent to the process, method, product, or apparatus.

[0024] In this application, a "module" or "unit" refers to a computer program or part of a computer program that has a specific function and works in conjunction with other related parts to achieve a predetermined goal. These modules or units can be implemented by software, hardware (such as processing circuits or memories), or a combination of the two. One or more processors or memories can implement one or more modules or units. At the same time, each module or unit can also be part of a larger module or unit.

[0025] The technical solutions in this application will be described in detail below in conjunction with the accompanying drawings in the embodiments. It should be noted that the embodiments described are only part of this application, not all embodiments. In the following description, the "some embodiments" mentioned are only a subset of all possible embodiments, which may be the same or different subsets, and different embodiments can be combined with each other without conflict.

[0026] Figure 1 This is a flow chart of a method for teaching a furnace-front quick separation robot provided by an embodiment of the present application. Figure 1 The teaching method of the furnace front quick separation robot provided in the embodiment of the present application may include the following steps 101 to 104:

[0027] Step 101, obtaining a target accuracy threshold of a target process point;

[0028] In some examples, a furnace-front fast-splitting robot refers to an automated robot deployed in a furnace-front fast-splitting laboratory for performing tasks such as automatic sample distribution, precise sorting, and standardized processing; the furnace-front fast-splitting robot mainly performs directional processing on metallurgical samples of different forms, and has high-precision motion control, multi-axis collaborative operation, and rapid task response capabilities; the furnace-front fast-splitting robot of the embodiment of the present application can also complete tasks such as sampling, temperature measurement, slag and iron identification, and separation. The target process point refers to the position point that the furnace-front fast-splitting robot needs to accurately reach and perform specific tasks during the teaching process, such as the sample clamping point, sample placement point, spectral excitation point, visual inspection point, etc. The target process point is usually the key posture target for the subsequent repeated operation of the furnace-front fast-splitting robot, and is required to have clear spatial coordinates and functional attributes. The target accuracy threshold is the absolute value of the positioning accuracy tolerance range required for the target process point, which represents the maximum allowable deviation between the final positioning result of the furnace-front fast-splitting robot and the actual target point. For example, if it is a sample transfer or a simple placement operation, the target accuracy threshold can be set to 1mm. If it is a high-precision task such as spectral excitation or image alignment, the target accuracy threshold needs to be controlled at 0.5mm or even lower.

[0029] By implementing step 101, the accuracy threshold required for the target process point is clarified, providing a quantitative standard for the subsequent teaching convergence process, making the entire teaching process goal-oriented, and ensuring that the control strategy can be dynamically adjusted according to specific accuracy requirements during the teaching process, avoiding unified and extensive accuracy settings, thereby better adapting to tasks of different complexities and improving the accuracy and practicality of the final positioning.

[0030] Step 102: determining an adaptive step-length convergence algorithm based on the aforementioned target accuracy threshold and the robot dynamic parameters, wherein the adaptive step-length convergence algorithm may include an initial step-length parameter and a convergence ratio parameter;

[0031] In some examples, robot dynamic parameters refer to a set of physical and operational characteristics that influence the actual movement performance of the end-effector of a furnace-based fast-splitting robot. These parameters are used to dynamically adjust the control strategy, including but not limited to the gripper stiffness coefficient, maximum movement speed, and historical collision offset statistics. The adaptive step-size convergence algorithm dynamically adjusts the step size of each movement based on the current offset error, enabling the furnace-based fast-splitting robot to reduce its step size when approaching the target point and increase it when moving away from it, thereby achieving fast, stable, and precise positioning. The adaptive step-size convergence algorithm integrates the concept of gradient descent with the error convergence criterion, avoiding the slow convergence or unstable oscillations associated with fixed step-size strategies. The initial step size parameter is the starting step size for the adaptive step-size convergence algorithm iteration. It can be set to the maximum single movement distance within the maximum controllable range to rapidly approach the target point. The initial step size parameter can be dynamically calculated based on parameters such as the furnace-based fast-splitting robot's maximum movement speed and safe braking distance, or it can be set to a fixed recommended value based on the task type, such as a 3mm initial step size for sample placement and a 1mm initial step size for spectral excitation. The convergence ratio parameter controls the reduction ratio of the step size in each iteration. For example, it can be any value between 0.3 and 0.7. The closer to the target, the more significant the step size reduction, avoiding overshoot or repeated oscillation. The convergence ratio parameter can be set in conjunction with the target accuracy threshold, such that the higher the accuracy, the smaller the ratio. Alternatively, the dynamic convergence curve can be set based on the inertia or delay characteristics of the furnace-front fast-distribution robot to obtain the convergence ratio parameter. The target accuracy threshold can be used as the reference value for the convergence endpoint, and the robot's dynamic parameters can be used as constraints and adjustment factors. These can be input into the preset algorithm model to automatically calculate the most suitable initial step size and convergence ratio, thus constructing an adaptive step size convergence algorithm that meets the accuracy and safety requirements of the current task.

[0032] By implementing step 102, the dynamic response characteristics of the robot itself are combined with the process accuracy requirements to establish a step length control algorithm with adaptive adjustment capabilities, thereby achieving rapid approximation when the deviation is large, automatically reducing the step length when approaching the target point, and accurately positioning. This can significantly reduce the number of teaching iterations, improve convergence efficiency, and reduce the risk of repeated debugging caused by inappropriate step lengths.

[0033] Step 103 , generating an iterative moving step length by an adaptive step length convergence algorithm in response to the offset direction teaching signal;

[0034] In some examples, the offset direction teaching signal is a directional input provided by an operator or a sensing device to indicate the relative offset direction between the current target process point and the robot end position, such as "left", "forward", "down", etc. The offset direction teaching signal can be input through a human-machine interface, a joystick, a button, or other input device. The operator manually selects the offset direction after observing the misalignment between the target point and the teaching block. A visual sensor can also be used to determine the direction of the coaxial deviation between the target hole position and the teaching block. For example, if the visual sensor determines that the excitation hole is offset to the right relative to the positioning hole, an offset direction teaching signal of "right" is automatically issued. The iterative movement step size refers to the current round of movement distance value dynamically generated by the adaptive step size convergence algorithm based on the current offset error each time the offset direction teaching signal is received. The iterative movement step size gradually decreases with the number of iterations to approach the target position to avoid overshoot or insufficient precision. For example, the step size generated for the first time is 3mm, the second time is 1.5mm, and the third time is 0.8mm, until the preset convergence conditions are met.

[0035] Through the implementation of step 103, the received offset direction is used as a trigger signal, and combined with the convergence algorithm, the optimal step length required for each movement is automatically calculated. This can realize a human-machine collaboration mechanism in which the direction is determined by humans and the distance is determined by the algorithm. This can not only simplify the operation process, but also reduce the dependence on the operator's skills, ensuring that each movement is more reasonable and close to the target, thereby improving the intelligence and consistency of the teaching process.

[0036] Step 104, controlling the movement of the furnace front fast separation robot according to the iterative movement step length until a preset convergence condition is reached;

[0037] In some examples, based on the current iterative movement step value output by the adaptive step-size convergence algorithm, the end-point of the furnace-front quick-splitting robot is driven to perform precise small-range movement operations according to the received offset direction. By continuously reducing the step size, the furnace-front quick-splitting robot is controlled to "successively approach" the target process point position in space, thereby ensuring stable and controllable automatic positioning in scenarios with high precision requirements. The preset convergence condition is a threshold standard used to determine whether the iteration is complete, and can be used to determine whether the current positioning accuracy meets the task requirements; the preset convergence condition can be offset error < target accuracy threshold and / or the iteration reaches the maximum number of iterations, etc.

[0038] By implementing step 104, whether the preset accuracy requirements are met after each round of movement is determined, and self-closed-loop control of the process can be achieved. This can effectively avoid excessive movement and invalid iterations, improve the efficiency and stability of the entire teaching process, and trigger an early warning mechanism in special circumstances, which helps to ensure equipment safety and achieve maintainability management.

[0039] In summary, the embodiment of the present application realizes adaptive convergence control by dynamically adjusting the iteration step size according to the precision threshold of the target process point, that is, the furnace-front fast-splitting robot can automatically reduce the step size when approaching the target position to improve positioning accuracy; and when the initial offset is large, a larger step size is adopted to speed up the approach speed, which can improve the precision control capability and the overall efficiency of teaching; the operator only needs to input the offset direction, and the algorithm can automatically calculate and control the movement to complete the positioning convergence, which can greatly reduce the manual intervention in repeated fine-tuning and experience judgment, lower the operation threshold, and at the same time improve the consistency and reproducibility of the teaching results, and is suitable for operators with different skill levels. In summary, the furnace-front fast-splitting robot teaching method provided by the embodiment of the present application realizes efficient and precise positioning through adaptive step-size convergence control, which can simplify the operation process, reduce manual dependence, and improve the consistency and universality of teaching.

[0040] In some embodiments, the preset convergence condition is that the residual movement error of the furnace-front fast-splitting robot is less than half of the target accuracy threshold.

[0041] In some examples, the residual motion error refers to the real-time spatial offset distance between the current end execution point of the furnace-front fast-splitting robot, such as the gripper or tool center point, and the target process point, usually in millimeters (mm), and is used to quantify positioning accuracy. When the error distance between the furnace-front fast-splitting robot and the target point is less than half of the target accuracy threshold, the positioning is considered "close enough" and further iterations can be stopped in advance. The introduction of "half of the target accuracy" as the judgment threshold is to improve positioning stability, avoid frequent oscillations or over-adjustments, and ensure that the final error does not exceed the target accuracy threshold, even if there is a slight error in the control accuracy of the last step.

[0042] Through the implementation of the above embodiment, the convergence end point is set to an error less than 1 / 2 of the accuracy threshold, which can ensure that the final positioning point of the furnace-front quick-splitting robot has a higher redundant accuracy space. Therefore, even if there are minor disturbances in actual operation, such as mechanical jitter, thermal deformation, etc., it will not affect the operation accuracy, and can enhance the robustness and stability of the furnace-front quick-splitting robot.

[0043] In some embodiments, the aforementioned step 101 may include: obtaining an application scenario of the target process point; when the application scenario is a sample transfer scenario or a processing placement scenario, determining the target accuracy threshold to be a first preset value; when the application scenario is a spectral excitation scene or a visual detection scene, determining the target accuracy threshold to be a second preset value, wherein the second preset value is less than the first preset value.

[0044] In some examples, an application scenario refers to the actual usage scenario corresponding to a specific task that a furnace-based fast-sorting robot must complete. This scenario directly impacts the positioning accuracy requirements and can be determined through user input, parameter distribution from the task scheduling system, or automatic identification from the process management system. The sample transfer scenario refers to the process by which a furnace-based fast-sorting robot transfers samples from one workstation to another. This scenario primarily focuses on handling efficiency and clamping stability, with relatively low accuracy requirements. The processing placement scenario refers to the operation of a furnace-based fast-sorting robot precisely placing workpieces, components, or samples into processing equipment or positioning fixtures. Sample transfer and processing placement scenarios have relatively low positioning accuracy requirements, prioritizing efficiency and mechanical coordination. The first preset value can be set to 1.0mm to 2.0mm. The spectral excitation scenario refers to the process by which a furnace-based fast-sorting robot aligns and inserts samples into the excitation aperture of a spectrometer for elemental analysis. This scenario requires high positioning center and angle accuracy to avoid excitation offset or detection failure. The visual inspection scenario refers to the process by which a furnace-based fast-sorting robot collaborates with a vision system for alignment inspection, feature recognition, or image analysis. This scenario requires the fixture or component to be aligned with the camera imaging area, as positional deviations can affect detection accuracy. Scenarios such as spectral excitation scenarios and visual detection scenarios require higher positioning accuracy because they involve precise excitation or camera alignment. Any slight deviation may affect the detection effect or data collection accuracy. The first preset value can be set to 0.2mm~0.5mm.

[0045] By implementing the above-mentioned embodiments, differentiated accuracy standards can be set for different operational scenarios, achieving a flexible trade-off between accuracy and efficiency. For example, stricter standards can be automatically adopted for high-precision scenarios such as spectral excitation, improving operational adaptability and generalization capabilities, and enhancing the level of industrial intelligence.

[0046] In some embodiments, the aforementioned step 102 may include: when the target accuracy threshold is a first preset value, determining, based on the dynamic parameters of the robot, the initial step length parameter of the adaptive step length convergence algorithm to be the first step length value, and the convergence ratio parameter of the adaptive step length convergence algorithm to be the first ratio; when the target accuracy threshold is a second preset value, determining, based on the dynamic parameters of the robot, the initial step length parameter of the adaptive step length convergence algorithm to be the second step length value, and the convergence ratio parameter of the adaptive step length convergence algorithm to be the second ratio, wherein the second step length value is smaller than the first step length value, and the second ratio is smaller than the first ratio.

[0047] In some examples, the robot dynamic parameters are key parameters that reflect the motion state and operating performance of the furnace-front fast-splitting robot. They are used to dynamically adjust the control strategy to achieve smooth and precise operation. They may include the gripper stiffness coefficient, the maximum moving speed, and historical collision offset statistics. The first step length value is the initial step length parameter of the adaptive step length convergence algorithm when the target accuracy threshold is large (i.e., the accuracy requirement is low). Its purpose is to quickly approach the target in the stage of large deviation and improve positioning efficiency. A safe initial step length can be automatically calculated as the first step length value based on the robot dynamic parameters. Specifically, the product of the maximum moving speed and the gripper stiffness coefficient can be divided by a smaller accuracy weight coefficient to ensure that the target position is approached quickly and efficiently when the initial deviation is large. For example, if the gripper stiffness coefficient is 0.8, the maximum moving speed is 80 mm / s, and the accuracy weight factor is 30, then the first step length value = (0.8×80) / 30 = 2.13 mm. The second step size is the initial step size parameter of the adaptive step size convergence algorithm when the target accuracy threshold is small (i.e., high precision requirements). Its purpose is to more accurately approach the target position. The initial step size is small to prevent overshoot. A safe initial step size can also be automatically calculated as the second step size based on the robot's dynamic parameters. Specifically, it can be calculated by multiplying the maximum movement speed and the gripper stiffness coefficient by a smaller precision weight factor to reduce the risk of overshoot in the high-precision stage. For example, if the gripper stiffness coefficient is 0.6, the maximum movement speed is 60 mm / s, and the precision weight factor is 40, then the second step size = (0.6 × 60) / 40 = 0.9 mm. The first ratio is the convergence ratio parameter corresponding to the first step size value. It controls the degree of step size reduction in each iteration. It is used for tasks with low precision requirements and allows the step size to decrease more slowly, prioritizing efficiency. The second ratio is the convergence ratio parameter corresponding to the second step size value. It is used for tasks with high precision requirements. The convergence speed is relatively slow, but it is more stable and accurate. Both the first ratio and the second ratio can be automatically calculated based on the dynamic parameters of the robot. Specifically, the convergence ratio can be determined based on the fluctuation degree of the historical collision offset statistics and the rigidity coefficient. If the fluctuation is large or the rigidity is low, the ratio value will be automatically reduced to improve stability.

[0048] Exemplarily, Table 1 shows example data of the iterative convergence process of the adaptive step size convergence algorithm when the target accuracy threshold is the first preset value (δ1=1.5 mm):

[0049] Table 1 Example data of iterative convergence process 1

[0050]

[0051] Table 2 shows example data of the iterative convergence process of the adaptive step size convergence algorithm when the target accuracy threshold is the second preset value (δ2=0.5mm):

[0052] Table 2 Sample data of iterative convergence process 2

[0053]

[0054]

[0055] Through the implementation of the above embodiments, different convergence strategies are switched according to the accuracy threshold to achieve targeted adjustments. For high-precision requirements, the convergence process is more detailed; for general scenarios, the approximation speed is improved, which can enhance the intelligent adjustment capability and scene adaptation efficiency of the teaching method.

[0056] In some embodiments, the aforementioned robot dynamic parameters may include the gripper stiffness coefficient, the limit moving speed and the historical collision offset statistics of the furnace front fast separation robot.

[0057] In some examples, the gripper rigidity coefficient refers to the ability of the gripper of the furnace-front quick-splitting robot to resist deformation under stress, that is, the rigidity of the gripper end structure, which is usually expressed as the force required for unit deformation. It can be analyzed through finite element simulation of the gripper material and structure, or directly measured through a gripper force test. For example, a gripper rigidity coefficient of 200N / mm means that the gripper requires 200 Newtons of force for every millimeter of displacement. High rigidity is suitable for high-speed teaching, while low rigidity requires a small step size to ensure safety and accuracy. The maximum moving speed refers to the maximum moving speed that the furnace-front quick-splitting robot can reach without affecting operational stability and safety. The unit is usually mm / s. It can be set by the controller parameters of the furnace-front quick-splitting robot and can be adjusted in combination with safety regulations and usage experience. For example, a maximum moving speed of 100mm / s means that a larger initial step size can be used in the initial stage when the accuracy requirements are low or the path is long. Historical collision offset statistics refer to the offset data recorded when the furnace-front quick-dispensing robot encountered contact or collisions due to positioning errors or path deviations during past tasks. These offsets, such as maximum and average deviations, can be recorded and analyzed through the robot's operation log, sensor feedback, and collision detection module. For example, if statistics show that three out of the last ten teaching attempts exhibited offsets greater than 2mm, the step size can be automatically tightened or the convergence ratio increased to prevent repetition risk. The step size strategy employed can be automatically adjusted based on the gripper stiffness coefficient, maximum travel speed, and historical collision offset statistics.

[0058] By implementing the above-mentioned embodiments and taking the gripper stiffness, limit speed and historical deviation data as algorithm inputs, the motion response characteristics can be evaluated more accurately, thereby optimizing the step size strategy, avoiding convergence deviation caused by equipment performance differences, and improving the personalized adaptability and convergence stability of the method.

[0059] In some embodiments, the aforementioned furnace front quick separation robot teaching method may also include: determining the misalignment direction of the excitation hole relative to the positioning hole by teaching the coaxial state of the positioning hole of the standard block and the excitation hole of the target workstation; and generating an offset direction teaching signal according to the misalignment direction.

[0060] In some examples, the teaching standard block is a standardized device used to assist positioning and calibration. It is a structural component with high-precision machined holes or feature surfaces, which is used to provide a reference alignment benchmark during the teaching process. The positioning hole is a hole set on the teaching standard block for pins, needles or optical detection. It has high position accuracy and can be used as a reference point for judging spatial deviations. The structure of the teaching standard block is as follows: Figure 2 As shown, the teaching standard block is a disc-shaped structure with a diameter of φ48.00±0.04mm, a height of 8.00±0.02mm, and a surface flatness of no more than 0.05mm to ensure installation stability and measurement accuracy. A positioning hole with a diameter of φ13.00±0.02mm is located in the center of the disc, which is used for precise coaxial alignment with the excitation hole of the spectral excitation table. A truncated cone positioning base with a diameter of φ54.15±0.04mm and a height of 1.30±0.02mm is coaxially located at the bottom of the disc, which is used for reference positioning in the Z-axis direction. An open slot with a width of 0.70±0.02mm is provided on the upper surface of the disc, aligned with the centerline of the jaws of the furnace-front fast separation robot gripper to assist in gripper positioning and posture adjustment. The target station excitation hole is an operational positioning hole or functional hole on the target station, which can be a key location point for the equipment to complete specific process operations (such as excitation, processing, and gripping). The misalignment direction refers to the direction of the spatial deviation between the target workstation excitation hole and the positioning hole of the teaching standard block, which is usually divided into X-axis, Y-axis or Z-axis direction deviation. The coaxiality of the two holes can be detected by visual systems, pin alignment, laser rangefinders, etc.; based on the detection results of the misalignment direction, it can be automatically converted into an offset direction teaching signal that can be recognized by the robot to guide the robot to fine-tune the position; for example, if the detection result is a positive deviation in the Y-axis direction, a teaching signal of "moving in the positive direction of the Y-axis" is generated to enter the convergence algorithm iteration.

[0061] Through the implementation of the above embodiment, the coaxial error of the structured teaching block is used to determine the misalignment direction, combined with visual or mechanical alignment methods to replace traditional manual judgment, which can improve the accuracy and consistency of direction recognition, and at the same time provide a basis for the automatic generation of offset direction signals, effectively simplifying the operation process.

[0062] In some embodiments, the aforementioned teaching method for the furnace-front fast-splitting robot may further include: when the number of movement iterations of the furnace-front fast-splitting robot is greater than the preset number of iterations corresponding to the target accuracy threshold, controlling the furnace-front fast-splitting robot to switch to a quantitative movement mode for movement, and triggering a hardware abnormality warning signal, wherein the quantitative movement mode is a state in which movement is performed by referring to a pre-stored fixed step size parameter.

[0063] In some examples, the number of movement iterations refers to the number of step updates and displacements performed by a furnace-based fast-splitting robot according to the adaptive step-size convergence algorithm when controlling its position. This number can be automatically recorded by the control system during algorithm execution, with each completed iteration counted as one. For example, if a furnace-based fast-splitting robot performs 10 movements and step-size adjustments from its initial position to a position close to the target, the number of iterations is 10. The preset number of iterations corresponding to the target accuracy threshold refers to a reasonable maximum number of iterations set based on the target accuracy threshold required by the task. Exceeding this number typically indicates an abnormal convergence process or environmental interference. The preset number of iterations can be determined based on experience, historical data, or simulation models and varies with target accuracy. For example, for an excitation scenario with a target accuracy of 0.2 mm, the preset number of iterations might be set to 15. The quantitative movement mode is a backup, non-adaptive control method. When the number of movement iterations exceeds the preset number, a limited number of mechanical movements are performed using a pre-set fixed step size to prevent the furnace-based fast-splitting robot from entering an infinite loop or oscillation. For example, each movement is set to 0.5 mm, with a maximum of three movements, and no dynamic adjustment based on error. When the number of moving iterations exceeds the preset number of iterations, it will be automatically identified as a potential fault or abnormal state, and an alarm signal will be sent to the host computer or operator to prompt that manual intervention is required; for example, if the preset number of iterations is 15, and the preset convergence condition is still not reached after the 16th iteration, the fault information will be immediately displayed through a buzzer, indicator light or human-machine interface.

[0064] Through the implementation of the above embodiment, a maximum number of iterations is set, and when convergence is impossible, it automatically switches to quantitative movement and triggers an early warning, avoiding long-term invalid teaching, improving the safety of the method and maintenance efficiency, and providing abnormal prompts to the operator, which helps to quickly locate the source of the problem, such as hardware failure, positioning error, etc., and enhance the closed-loop control capability of the teaching process.

[0065] Furthermore, as an implementation of the aforementioned method embodiment, the present application also provides a furnace-front fast separation robot teaching device for implementing the aforementioned method embodiment. This device embodiment corresponds to the aforementioned method embodiment. For ease of reading, this furnace-front fast separation robot teaching device embodiment will no longer repeat the details of the aforementioned method embodiment one by one, but it should be clear that the device in the embodiment of the present application can correspond to and implement all the contents of the aforementioned method embodiment. Figure 2 As shown, the furnace-front fast-splitting robot teaching device 20 includes: a data acquisition unit 201, an algorithm determination unit 202, a step size generation unit 203 and a teaching movement unit 204, wherein the data acquisition unit 201 is used to obtain the target accuracy threshold of the target process point; the algorithm determination unit 202 is used to determine the adaptive step size convergence algorithm according to the aforementioned target accuracy threshold and the robot dynamic parameters, wherein the adaptive step size convergence algorithm may include an initial step size parameter and a convergence ratio parameter; the step size generation unit 203 is used to generate an iterative movement step size through the adaptive step size convergence algorithm in response to the offset direction teaching signal; the teaching movement unit 204 is used to control the movement of the furnace-front fast-splitting robot according to the iterative movement step size until a preset convergence condition is reached.

[0066] In some embodiments, the preset convergence condition is that the residual movement error of the furnace-front fast-splitting robot is less than half of the target accuracy threshold.

[0067] In some embodiments, the data acquisition unit 201 is also used to obtain the application scenario of the target process point; when the application scenario is a sample transfer scenario or a processing placement scenario, the target accuracy threshold is determined to be a first preset value; when the application scenario is a spectral excitation scene or a visual detection scene, the target accuracy threshold is determined to be a second preset value, wherein the second preset value is less than the first preset value.

[0068] In some embodiments, the algorithm determination unit 202 is also used to determine, based on the dynamic parameters of the robot, when the target accuracy threshold is a first preset value, the initial step length parameter of the adaptive step length convergence algorithm to be the first step length value, and the convergence ratio parameter of the adaptive step length convergence algorithm to be the first ratio; when the target accuracy threshold is a second preset value, based on the dynamic parameters of the robot, determine, based on the dynamic parameters of the robot, the initial step length parameter of the adaptive step length convergence algorithm to be the second step length value, and the convergence ratio parameter of the adaptive step length convergence algorithm to be the second ratio, wherein the second step length value is smaller than the first step length value, and the second ratio is smaller than the first ratio.

[0069] In some embodiments, the robot dynamic parameters include the gripper stiffness coefficient, the limit moving speed and the historical collision offset statistics of the furnace front fast separation robot.

[0070] In some embodiments, the step size generation unit 203 is further used to determine the misalignment direction of the excitation hole relative to the positioning hole by teaching the coaxial state of the positioning hole of the standard block and the excitation hole of the target workstation; and generate an offset direction teaching signal according to the misalignment direction.

[0071] In some embodiments, the teaching movement unit 204 is also used to control the furnace-front fast-splitting robot to switch to a quantitative movement mode for movement and trigger a hardware abnormality warning signal when the number of movement iterations of the furnace-front fast-splitting robot is greater than a preset number of iterations corresponding to a target accuracy threshold. The quantitative movement mode is a state in which movement is performed by referring to a pre-stored fixed step size parameter.

[0072] The present application also provides a computer-readable storage medium, which stores computer-executable instructions or computer programs. When the computer-executable instructions or computer programs are executed by a processor, the processor will execute any step of the furnace front fast separation robot teaching method provided in the present application.

[0073] In some embodiments, the computer-readable storage medium may be a random access memory (RAM), a read-only memory (ROM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); or it may be various devices including one or any combination of the above memories.

[0074] In some embodiments, computer-executable instructions may be in the form of a program, software, software module, script, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and may be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment.

[0075] In some embodiments, computer-executable instructions may, but need not, correspond to a file in a file system, may be stored as part of a file that stores other programs or data, for example, in one or more scripts in a HyperText Markup Language (HTML) document, in a single file dedicated to the program in question, or in multiple coordinating files (for example, files storing one or more modules, subroutines, or code portions).

[0076] In some embodiments, computer-executable instructions may be deployed to be executed on one electronic device, or on multiple electronic devices located at one site, or on multiple electronic devices distributed across multiple sites and interconnected by a communication network.

[0077] like Figure 3As shown, the present application also provides an electronic device 30, including a memory 310, a processor 320 and a computer program 311 stored in the memory 310 and executable on the processor. When the processor 320 executes the computer program 311, any step of the above-mentioned furnace front fast separation robot teaching method is implemented.

[0078] The present application also provides a computer program product, which includes a computer program or computer-executable instructions, wherein the computer program or computer-executable instructions are stored in a computer-readable storage medium. A processor of an electronic device reads the computer program or computer-executable instructions from the computer-readable storage medium, and the processor executes the computer program or computer-executable instructions, causing the electronic device to perform any step of the method for teaching a furnace-front fast-sorting robot described above.

[0079] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A teaching method for a furnace-front quick separation robot, characterized in that: include: Obtain target accuracy threshold of target process point; Determining an adaptive step-length convergence algorithm according to the target accuracy threshold and the robot dynamic parameters, wherein the adaptive step-length convergence algorithm includes an initial step-length parameter and a convergence ratio parameter; In response to the offset direction teaching signal, generating an iterative moving step size by the adaptive step size convergence algorithm; The movement of the furnace front fast separation robot is controlled according to the iterative movement step length until a preset convergence condition is reached.

2. The teaching method of the furnace front quick separation robot according to claim 1, characterized in that: The preset convergence condition is that the remaining movement error of the furnace-front fast-splitting robot is less than half of the target accuracy threshold.

3. The teaching method for the furnace front quick separation robot according to claim 1, characterized in that: The target accuracy threshold of the target process point is obtained, including: Obtaining an application scenario of the target process point; When the application scenario is a sample transfer scenario or a processing placement scenario, determining the target accuracy threshold to be a first preset value; When the application scenario is a spectral excitation scenario or a visual detection scenario, the target accuracy threshold is determined to be a second preset value, wherein the second preset value is smaller than the first preset value.

4. The teaching method for the furnace front quick separation robot according to claim 3, characterized in that: Determining the adaptive step size convergence algorithm according to the target accuracy threshold and the robot dynamic parameters includes: When the target accuracy threshold is the first preset value, determining, according to the robot dynamic parameters, an initial step length parameter of the adaptive step length convergence algorithm as the first step length value and a convergence ratio parameter of the adaptive step length convergence algorithm as the first ratio; When the target accuracy threshold is the second preset value, based on the robot dynamic parameters, the initial step length parameter of the adaptive step length convergence algorithm is determined to be a second step length value, and the convergence ratio parameter of the adaptive step length convergence algorithm is determined to be a second ratio, wherein the second step length value is smaller than the first step length value, and the second ratio is smaller than the first ratio.

5. The teaching method for the furnace-front quick separation robot according to claim 4, characterized in that: The robot dynamic parameters include the gripper rigidity coefficient, the limit moving speed and the historical collision offset statistics of the furnace front quick separation robot.

6. The teaching method for the furnace front quick separation robot according to claim 1, characterized in that: The furnace front quick separation robot teaching method also includes: Determine the misalignment direction of the excitation hole relative to the positioning hole by comparing the coaxial state of the positioning hole of the teaching standard block and the excitation hole of the target station; The offset direction teaching signal is generated according to the misalignment direction.

7. The teaching method for the furnace front quick separation robot according to claim 1, characterized in that: The furnace front quick separation robot teaching method also includes: When the number of movement iterations of the furnace-front quick-splitting robot is greater than the preset number of iterations corresponding to the target accuracy threshold, the furnace-front quick-splitting robot is controlled to switch to the quantitative movement mode for movement and trigger a hardware abnormality warning signal, wherein the quantitative movement mode is a state of executing movement by referring to a pre-stored fixed step size parameter.

8. A furnace-front quick separation robot teaching device, characterized in that: include: A data acquisition unit, used to obtain a target accuracy threshold of a target process point; an algorithm determination unit, configured to determine an adaptive step-length convergence algorithm according to the target accuracy threshold and the robot dynamic parameters, wherein the adaptive step-length convergence algorithm includes an initial step-length parameter and a convergence ratio parameter; a step size generating unit, configured to generate an iterative moving step size by using the adaptive step size convergence algorithm in response to the offset direction teaching signal; The teaching moving unit is used to control the movement of the furnace front fast separation robot according to the iterative movement step length until a preset convergence condition is reached.

9. An electronic device comprising: A memory and a processor, characterized in that the processor is used to implement the steps of the furnace front fast separation robot teaching method as described in any one of claims 1 to 7 when executing the computer program stored in the memory.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the furnace front quick separation robot teaching method according to any one of claims 1 to 7 are implemented.

Images