Robot-based workpiece assembly method, device, equipment and readable storage medium

Abstract

This application discloses a robot-based workpiece assembly method, apparatus, device, and readable storage medium. The method includes: controlling a robot to move to a target area near a loading table; acquiring the workpiece pose and the pose of a first assembly position; controlling the robot's end effector to move according to the workpiece pose and the first assembly position pose, so that the first assembly part on the workpiece to be registered aligns with the first assembly position; switching the robot's end effector to a force feedback-based compliant control mode, and adaptively adjusting the robot's end effector pose according to the real-time force signal from the robot's six-dimensional force sensor to assemble the first assembly part with the first assembly position; using the assembled first assembly position as a positioning reference, controlling the robot to complete the assembly of the second assembly part of the workpiece to be registered with the second assembly position on the loading table. This method employs a step-by-step assembly strategy to achieve multi-assembly position assembly of the workpiece, significantly reducing the cumulative positioning error between multiple assembly positions and improving efficiency and assembly accuracy.

Description

Technical Field

[0001] This invention relates to the field of robot control technology, and in particular to a robot-based workpiece assembly method, apparatus, equipment, and readable storage medium. Background Technology

[0002] Currently, the industrial sector commonly uses fixed-base industrial robotic arms for insertion operations, employing a holistic approach of perception, planning, control, and execution. This approach utilizes a composite solution of visual initial positioning and force-controlled compliant insertion, but still requires significant manual assistance or adjustments. Unlike traditional fixed-base robotic arms, mobile chassis-based industrial insertion operations need to address the issue of excessive chassis positioning and navigation errors. Furthermore, traditional "visual initial positioning" solutions for industrial insertion tasks may encounter situations where visual obstruction of the hole position or workpiece occurs during operation, necessitating insertion algorithms with certain visual loss and error compensation mechanisms. End-effector compliant force control is a key technology for solving the "last few millimeters" problem, enabling the robotic arm to proactively adapt to uncertainty and achieve true adaptive assembly. However, when navigation positioning, visual recognition or obstruction, or errors in the robotic arm's own parameters exist, extremely precise force perception and complex compliant control strategies are required to avoid jamming or damage, often leading to a significant increase in development costs. Summary of the Invention

[0003] The purpose of this invention is to propose a robot-based workpiece assembly method, apparatus, equipment, and readable storage medium, aiming to solve the problem of low accuracy in multi-position assembly of workpieces in the existing step-by-step assembly strategy.

[0004] To address the aforementioned technical problems, a first aspect of this application provides a robot-based workpiece assembly method. The method includes: controlling a robot to move to a target area near a loading platform; acquiring the workpiece pose of the workpiece to be registered held by the robot, and the pose of a first assembly position on the loading platform; controlling the robot end effector to move according to the workpiece pose and the first assembly position pose, so that a first assembly portion on the workpiece to be registered aligns with the first assembly position; switching the robot end effector to a force feedback-based compliant control mode, and adaptively adjusting the robot end effector pose according to the real-time force signal from a six-dimensional force sensor at the robot end effector, to assemble the first assembly portion with the first assembly position; using the assembled first assembly position as a positioning reference, controlling the robot to complete the assembly of a second assembly portion of the workpiece to be registered with a second assembly position on the loading platform.

[0005] In one embodiment, controlling the robot end effector to move based on the workpiece pose and the first assembly position pose, so as to align the first assembly part on the workpiece to be registered with the first assembly position, includes: calculating a pose error based on the workpiece pose and the first assembly position pose, and constructing an interaction matrix for describing the mapping relationship between the rate of change of image features and the camera spatial velocity; controlling the robot end effector to move based on the interaction matrix and the pose error, so as to align the first assembly part on the workpiece to be registered with the first assembly position.

[0006] In one embodiment, switching the robot end effector to a force feedback-based compliant control mode includes: adjusting the controller parameters of the robot end effector to below a preset threshold; adaptively adjusting the robot end effector's pose based on the real-time force signal from the six-dimensional force sensor of the robot end effector includes: controlling the robot end effector to move the workpiece to be registered in a first predetermined direction, so that the edge of the first assembly part is in complete contact with the first assembly position, and detecting the first radial force of the robot end effector on the assembly plane; keeping the first radial force constant, performing force-guided search of the robot end effector along a second radial direction in the assembly plane, and during the force-guided search process, monitoring the axial force, the first radial force, and the second radial force fed back by the robot end effector in real time; if a combined force signal feature of decreasing axial force, decreasing first radial force, and increasing second radial force is detected, stopping the attitude adjustment of the robot end effector.

[0007] In one embodiment, the determination conditions for the combined force signal characteristics include: within a preset time window, the decrease in the axial force exceeds a preset axial force threshold, the decrease in the first radial force exceeds a preset first radial force threshold, and after the preset time window, the decrease in the second radial force exceeds a second radial force threshold.

[0008] In one embodiment, the force-guided search of the robot end effector along the second radial direction in the assembly plane includes: obtaining the velocity deviation value of the robot end effector at the previous moment, and integrating the velocity deviation at the previous moment to obtain the velocity deviation value at the current moment; integrating the velocity deviation at the current moment to obtain the position deviation value at the current moment; using the position deviation value at the current moment to perform motion control on the robot end effector; and using the velocity deviation value at the current moment as the velocity deviation value at the previous moment, returning to execute the step of "obtaining the velocity deviation value of the robot end effector at the previous moment and integrating the velocity deviation at the previous moment to obtain the velocity deviation value at the current moment".

[0009] In one embodiment, the step of using the first assembled position as a positioning reference to control the robot to complete the assembly of the second assembly part of the workpiece to be registered with the second assembly position of the unloading table includes: switching the control mode of the robot end effector to a constant force control mode, applying a constant downward pressure axially to the robot end effector to keep the first assembly part in contact with the first assembly position and maintain the axial force within a preset threshold range; using the first assembly position as a reference, controlling the robot end effector to perform force-guided search along a second predetermined direction in the assembly plane, and during the force-guided search process, monitoring the axial force, first radial force, and torque fed back by the robot end effector in real time; if the change amplitude of the axial force, radial force, and torque is detected to be less than the corresponding preset change amplitude, the assembly operation of the second assembly part and the second assembly position is completed.

[0010] In one embodiment, the method further includes: if the change in at least one of the axial force, radial force, and torque exceeds the preset change range, then triggering a pause, or triggering the re-execution of the step of "controlling the robot end effector to perform force-guided search in the assembly plane along a second predetermined direction".

[0011] To address the aforementioned technical problems, a second aspect of this application provides a robot-based workpiece assembly device. The device includes: a robot overall movement module for controlling the robot to move to a target area near a loading platform; a pose data acquisition module for acquiring the workpiece pose of the workpiece to be registered held by the robot, and the pose of a first assembly position on the loading platform; a visual servo module for controlling the robot end effector to move according to the workpiece pose and the first assembly position pose, so that a first assembly part on the workpiece to be registered aligns with the first assembly position; an assembly module for switching the robot end effector to a force feedback-based compliant control mode, adaptively adjusting the robot end effector pose according to the real-time force signal from a six-dimensional force sensor at the robot end effector, to assemble the first assembly part with the first assembly position; and controlling the robot to complete the assembly of a second assembly part of the workpiece to be registered with a second assembly position on the loading platform, using the assembled first assembly position as a positioning reference.

[0012] To address the aforementioned technical problems, a third aspect of this application provides a computer device including a processor and a memory coupled to each other; the memory stores a computer program, and the processor executes the computer program to implement the steps of the method provided in the first aspect above.

[0013] To address the aforementioned technical problems, a fourth aspect of this application provides a computer-readable storage medium storing program data, which, when executed by a processor, implements the steps of the method provided in the first aspect above.

[0014] The present invention provides the following advantages: Unlike existing technologies, this application controls a robot to move to a target area near the unloading platform; acquires the workpiece pose of the workpiece to be registered held by the robot, and the pose of the first assembly position of the first assembly position on the unloading platform; controls the robot end effector to move based on the workpiece pose and the first assembly position pose, so that the first assembly part on the workpiece to be registered is aligned with the first assembly position; switches the robot end effector to a compliant control mode based on force feedback, and adjusts the robot end effector position according to the real-time force signal from the six-dimensional force sensor of the robot end effector. The robot adaptively adjusts its posture to assemble the first assembly part with the first assembly position. Using the assembled first assembly position as a positioning reference, the robot is controlled to complete the assembly of the second assembly part of the workpiece to be registered with the second assembly position of the unloading table. This realizes a multi-assembly position step-by-step assembly strategy for the workpiece using the robot, which significantly reduces the cumulative positioning error between multiple assembly positions and improves the overall coaxiality and fitting accuracy of the assembly of multiple parts of complex workpieces. The entire process of movement, posture acquisition, alignment, force control assembly, and reference positioning secondary assembly is executed automatically without human intervention, improving the automation level and production efficiency of workpiece assembly. Attached Figure Description

[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0016] in: Figure 1 This is a schematic diagram illustrating the application scenario of the workpiece assembly method of this application; Figure 2 This is a schematic flowchart of an embodiment of the robot-based workpiece assembly method of this application; Figure 3 This is a flowchart illustrating an embodiment of step S13 of this application; Figure 4 This is a flowchart illustrating an embodiment of step S14 of this application; Figure 5 This is a schematic block diagram of an embodiment of the robot-based workpiece assembly device of this application; Figure 6This is a schematic block diagram of the structure of an embodiment of the computer device of this application; Figure 7 This is a schematic block diagram of an embodiment of a computer-readable storage medium of this application. Detailed Implementation

[0017] 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.

[0018] This application is based on a robot-based workpiece assembly method, which is applied to scenarios where robots assemble workpieces. Specifically, the loading table can be set with multiple assembly positions, such as assembly holes, for assembling workpieces. The assembly position is, for example, a pin. The robot is used to load the workpiece and assemble it into the assembly position in the correct posture. This application solves the problem of how to quickly and accurately assemble the workpiece into position in the case of multiple assembly positions.

[0019] Please see Figure 1 , Figure 1 This is a schematic diagram of the application scenario of the workpiece assembly method of this application. The image is acquired by an image acquisition device installed on the robot. As shown in the figure, the robot end effector 110 is used to load the workpiece 120 to be registered, and the unloading table 130 is provided with an assembly position for assembling the workpiece 120 to be registered.

[0020] In this paper, the coordinate system of the robot end effector is defined. The assembly direction between the assembly part and the assembly position is the z-axis direction / axial direction. The plane perpendicular to the z-axis is the XY plane. On the XY plane, the direction parallel to the line connecting the two assembly parts of the workpiece to be registered is the y-axis direction, and the direction perpendicular to the y-axis is the x-axis direction.

[0021] Please see Figure 2 , Figure 2 This is a schematic flowchart of an embodiment of the robot-based workpiece assembly method of this application. It should be noted that if substantially the same result is achieved, this embodiment does not necessarily reflect that outcome. Figure 1 The illustrated process sequence is limited. It includes the following steps: S11: Control the robot to move to the target area near the unloading platform; S12: Collect the workpiece pose of the workpiece to be registered held by the robot, and the first assembly pose of the first assembly position of the unloading platform. S13: Based on the workpiece pose and the first assembly position pose, control the robot end effector to move so that the first assembly part on the workpiece to be registered is aligned with the first assembly position. S14: Switch the robot end effector to a compliant control mode based on force feedback, and adaptively adjust the robot end effector pose according to the real-time force signal of the six-dimensional force sensor of the robot end effector to assemble the first assembly part and the first assembly position. S15: Using the first assembly position after assembly as the positioning reference, control the robot to complete the assembly of the second assembly part of the workpiece to be registered with the second assembly position of the unloading table.

[0022] The robot described in this application is, for example, a mobile manipulator used in automated CNC machine tool processing. This type of robot typically consists of a mobile chassis (such as an AGV or AMR) carrying a collaborative robotic arm. The robot includes a base and a robot end effector (i.e., the robotic arm). The base and the robot end effector are driven by independent drive systems. The drive system located on the base moves the entire robot, while the robot end effector is used to load the workpiece to be registered. The drive system located on the robot end effector allows for fine-tuning of the robot end effector. After the robot moves to the target area near the unloading table according to the drive commands, the base brakes to secure it.

[0023] Step S12 can be performed by using an image acquisition device (e.g., a camera) set on the robot to acquire the pose of the first assembly position corresponding to the first assembly position, and determining the workpiece pose based on the CAD model of the workpiece to be registered and the acquired image depth RGBD data of the workpiece to be registered.

[0024] Acquiring the workpiece pose may include the following steps: acquiring RGB and depth images of the CAD model of the workpiece to be registered in candidate poses; inputting the rendered images and the RGB and depth images acquired by the sensor into a pose refinement network to predict the rotation and translation matrices; correcting the RGB and depth images based on the rotation and translation matrices and rendering them again to obtain the corrected RGB and depth images; using a pose scoring encoder, predicting the workpiece pose based on the corrected RGB and depth images and providing corresponding confidence scores; and obtaining the final 6D workpiece pose by calculating a weighted average of the confidence scores. Specifically: selecting the top K workpiece poses with the highest confidence scores (e.g., K=5) and all workpiece poses with confidence scores exceeding a threshold (>0.8), calculating a weighted average of the translation and rotation matrices using the confidence scores, and combining the weighted averaged rotation and translation matrices to obtain the final workpiece pose, which includes pose information in a total of six dimensions: translation and rotation.

[0025] Step S13 can optimize the difference between the workpiece pose and the first assembly position pose to obtain the movement path of the robot end effector. After the robot end effector moves along the path, the alignment of the workpiece to be registered with the assembly position is achieved, including: the first assembly part is aligned with the first assembly position, and the second assembly part is aligned with the second assembly position.

[0026] Step S13 achieves coarse positioning of the workpiece and the assembly position. There is still a deviation between the assembly part and the assembly position of the workpiece that exceeds the assembly gap, usually between 1-3 cm. At this time, since there is no additional global camera to provide a global view, when relying solely on the robot camera, the unloading table will be completely blocked by the workpiece, and the visual information will be completely lost. Therefore, visual positioning cannot be used subsequently, and adaptive admittance control in compliant control mode is adopted instead.

[0027] Specifically, in step S14, under the compliant control mode based on force feedback, the controller parameters (especially stiffness K) are set to a low value so that the end effector can move flexibly when subjected to contact force. The robot end effector's pose is adaptively adjusted according to the real-time force signal from the robot end effector's six-dimensional force sensor, so that the first assembly part and the first assembly position are completely aligned and assembled in place.

[0028] Step S15: Using the first assembly position after assembly as the constraint center and the workpiece length as the motion radius, control the robot to move, align the second assembly part of the workpiece to be registered with the second assembly position, and transfer it into place.

[0029] In this embodiment, for robot-driven workpiece assembly tasks, the robot first performs coarse positioning of the assembly part based on the workpiece and assembly position poses through pose acquisition and alignment. Then, force feedback compliant control is switched to complete the assembly, balancing positioning efficiency and assembly accuracy, effectively avoiding workpiece collisions, jamming, or scratches caused by rigid assembly. Based on the completed assembly position, the robot is controlled to align and assemble another assembly part and position. A step-by-step assembly strategy is adopted, significantly reducing the cumulative positioning error between multiple assembly positions and improving the overall coaxiality and fitting accuracy of complex workpiece multi-part assembly. The entire process of movement, pose acquisition, alignment, force-controlled assembly, and secondary assembly with reference positioning is executed automatically without human intervention, improving the automation level and production efficiency of workpiece assembly. This method is suitable for scenarios where robots are used for multi-hole workpiece assembly, enabling fast and accurate alignment and assembly in such scenarios. Please see Figure 3 In one embodiment, step S13 may include the following steps: S21: Calculate the pose error based on the workpiece pose and the first assembly position pose, and construct an interaction matrix to describe the mapping relationship between the rate of change of image features and the camera spatial velocity.

[0030] Specifically, the pose error is expressed by the following formula: in, Indicates pose error. This represents the pose of the material platform, specifically a vector containing multiple visual features. Indicates the workpiece pose, and further, This represents the coordinate features of the material platform obtained through a series of image detections, which are used to calculate the platform pose. It is a set of parameters that represent the potential prior knowledge of the system, such as camera intrinsic parameters or a 3D model of the workpiece to be registered.

[0031] Let the spatial velocity of the camera be expressed as ,in Represents the instantaneous linear velocity at the origin of the camera coordinate system. This represents the instantaneous angular velocity in the camera coordinate system. The first derivative with respect to time and The functional relationship between them can be expressed as: ,in Is with The relevant interaction matrix. Combining the above two equations, we can obtain the functional relationship between the camera speed and the derivative of the error: ( = ),in To control the speed parameters of the robot's end effector, and to ensure that the error decays exponentially through decoupling, i.e. Combining the above formulas, we can obtain: in yes The Moore-Penrose pseudo-inverse, i.e. When k=6, if If the determinant of the matrix is ​​not equal to 0, then Reversible, with a final control rate of: The control rate, as the input parameter of the robot's end effector, drives the robot's end effector to move.

[0032] Consider the 3D coordinates of a point in the camera coordinate system as follows: The image then transforms these 3D points into a two-dimensional plane, using... To express, specifically: , The coordinates of image points, expressed in pixels, have been determined. For camera intrinsic parameters set, These are the coordinates of the principal point. Focal length This represents the pixel size ratio. In multi-hole assembly scenarios, it is taken as... That is, the image plane coordinates of that point.

[0033] The interaction matrix used to describe the mapping relationship between the rate of change of image features and the camera spatial velocity can be represented as: S22: Based on the interaction matrix and pose error, control the movement of the robot's end effector to align the first assembly part on the workpiece to be registered with the first assembly position.

[0034] Specifically, by controlling the robot's end effector movement based on the interaction matrix and pose error, the robot's end effector speed control parameters can be determined based on the pose error e. For details, please refer to the following formula: or The control rate is further calculated based on the interaction matrix and pose error. Please refer to the following formula for details: Finally, the control rate is used as the input parameter of the robot end effector to drive the robot end effector to move, so that the first assembly part is aligned with the first assembly position.

[0035] This embodiment optimizes the robot end effector based on the pose error between the workpiece pose and the first assembly position pose, calculates the robot end effector control rate, and then controls the robot end effector to move, aligning the first assembly part with the first assembly position. This reduces complex coordinate transformation and iterative calculation, lowers the computational load of the control system, improves the real-time performance of the alignment process, and achieves fast alignment with high accuracy.

[0036] Please see Figure 4 In one embodiment, step S14 may include the following steps: S31: Adjust the controller parameters of the robot end effector to below the preset threshold.

[0037] Specifically, adjusting the controller parameters of the robot end effector to below a preset threshold can be achieved by adjusting one or more of the virtual inertia, damping, and stiffness to below the preset threshold. In this case, the robot end effector and the unloading table can make flexible contact without damaging the workpiece or the unloading table.

[0038] S32: Control the robot end effector to move the workpiece to be registered in a first predetermined direction, so that the edge of the first assembly part is in complete contact with the first assembly position, and detect the first radial force of the robot end effector on the assembly plane.

[0039] The first predetermined direction is, for example, one of a first radial direction and a second radial direction. It is understood that the first and second radial directions of this application may correspond to the x and y axes of the robot's end effector coordinate system, or they may not correspond to the x and y axes of the coordinate system. The first and second radial directions may be perpendicular to each other, or they may not be perpendicular. For example, in this embodiment, the first and second radial directions are respectively directions parallel to the x-axis and parallel to the y-axis in the robot's end effector coordinate system. In this embodiment, the first radial direction is, for example, parallel to the y-axis direction, and the second radial direction is parallel to the x-axis direction. For example, Figure 1If both the first assembly position and the assembly part are on the right side, then in this step, the workpiece to be registered will be moved to the right so that the workpiece to be registered is placed at the rightmost end and the right edge of the workpiece to be registered completely touches the first assembly position. At this time, the first radial force Fy is detected to increase instantaneously.

[0040] S33: Keep the first radial force constant, perform force-guided search on the robot end effector along the second radial direction in the assembly plane, and monitor the axial force, first radial force, and second radial force fed back by the robot end effector in real time during the force-guided search process.

[0041] The first radial force is constant, which keeps the first assembly part of the workpiece to be registered in close contact with the first assembly position. In this state, the robot end effector is controlled to perform force-guided search along the second radial direction, that is, to perform a small-range grid scan in the XY plane along the x-axis direction to attempt to align the first assembly part with the first assembly position within a small range. A six-dimensional force sensor is used to monitor the axial force, the first radial force, and the second radial force fed back by the robot end effector in real time.

[0042] S34: If a combined force signal characteristic of decreasing axial force, decreasing first radial force, and increasing second radial force is detected, stop adjusting the attitude of the robot end effector.

[0043] Understandably, when the first assembly part enters the first assembly position, the axial force Fz will suddenly decrease, and at the same time, the instantaneous decrease of the first radial force Fy will be detected. Subsequently, the second radial force Fx will tend to infinity. Based on this characteristic, the combined force signal characteristic judgment conditions of axial force, first radial force, and second radial force can be set accordingly. If the corresponding judgment conditions are met, it can be determined that the first assembly part is assembled in place.

[0044] This embodiment uses constant force control to maintain tight contact between assembled parts, preventing loosening of the reference and causing secondary assembly deviations, and ensuring the consistency of the reference in step-by-step assembly; secondary force-guided search assembly is performed based on the reference of the first assembly position, eliminating the need for global repositioning, simplifying the control process, and improving the efficiency of continuous assembly of multiple assembly positions; by monitoring the multi-dimensional changes in axial force, radial force, and torque to determine the assembly status, it is possible to accurately identify whether the second assembly part is smoothly matched, ensuring the accuracy and reliability of secondary assembly.

[0045] In one embodiment, the determination criteria for the combined force signal characteristics may include: within a preset time window, the decrease in axial force exceeds a preset axial force threshold, the decrease in first radial force exceeds a preset first radial force threshold, and after the preset time window, the increase in second radial force exceeds a second radial force threshold.

[0046] It is understandable that before and after the first assembly part enters the first assembly position, the axial force and the first radial force decrease significantly. After that, the second radial force increases significantly. Based on this, a judgment threshold can be set for the change of force in the three directions. When the change of force in the three directions exceeds the corresponding threshold, it is determined that the first assembly part and the first assembly position are properly assembled.

[0047] In one embodiment, step S33 involves force-guided searching of the robot end effector along the second radial direction within the assembly plane, including the following steps: Obtain the velocity deviation value of the robot's end effector at the previous moment, and integrate the velocity deviation value at the previous moment to obtain the velocity deviation value at the current moment; Integrating the velocity deviation at the current moment yields the position deviation value at the current moment; The robot's end effector is motion-controlled using the current position deviation value. Using the current speed deviation value as the previous speed deviation value, return to execute the step of "obtaining the previous speed deviation value of the robot end effector and integrating the previous speed deviation value to obtain the current speed deviation value".

[0048] Specifically, the speed deviation value at the current moment Position deviation value from the previous time step The results are derived from the following formulas: in, This represents the velocity deviation of the robot's end effector at the previous moment. This indicates the cycle of motion control for the robot's end effector. This represents the acceleration deviation parameter at the robot's end effector. This represents the position deviation value of the robot's end effector at the previous moment.

[0049] Motion control of the robot's end effector is achieved using the current position deviation value. Specifically, for the equivalent axial vector in the pose component, the current position deviation value must first be converted into an attitude matrix for computation, and then converted into an equivalent axial vector. This is done by solving the rotation quaternion. Combine it with the robot's real-time pose quaternion Right multiplication yields the target quaternion The target quaternion is converted into a rotation vector and sent to the robot end effector to complete the compliant control of the robot end effector posture.

[0050] The above velocity deviation values ​​at the previous time step Position deviation value from the previous time step The calculation method can be derived from the following: The interaction between the robot and its environment can be viewed as a virtual second-order system composed of multi-degree-of-freedom mass-spring-damped elements. Based on this second-order system, the dynamic relationship between positional deviation and force deviation can be established. First, consider the system's dynamic expression: in, , and These represent virtual inertia, damping, and stiffness, respectively. For positional deviation, It is the absolute position of the robot's end effector in space. It is the desired position; for The first derivative with respect to time represents the velocity deviation. for The second derivative with respect to time represents the acceleration deviation; It's a force deviation. It is the actual contact force between the assembly part of the component to be registered and the surrounding materials, which is collected by a six-dimensional force / torque sensor. It is the expected force set, that is, the contact force that needs to be maintained between the expected assembly part and the assembly position.

[0051] Since the data measured by the six-dimensional force sensor includes the additional force from the sensor's zero-point drift and the gravity of the end load, zero-point compensation and gravity compensation must be performed on the sensor beforehand to ensure that the measured force value accurately reflects the actual contact force between the assembly part and the assembly position. Therefore, the above formula can be expressed in the following form: This embodiment achieves robot end-effector control by continuously integrating the velocity deviation to obtain the position deviation, ensuring a smooth and jitter-free force-guided search process and preventing damage to the workpiece from violent lurching of the robot end-effector. The current control quantity is calculated iteratively based on historical deviations, eliminating the need for complex real-time solutions, resulting in high computational efficiency. This adapts to the high-frequency motion control requirements of robots and ensures the real-time performance and continuity of the force-guided search. Integral control can effectively eliminate steady-state errors, improve the positioning accuracy of the force-guided search, and ensure that the assembly part smoothly falls into the corresponding assembly position.

[0052] In one embodiment, step S15 includes: Switch the control mode of the robot end effector to constant force control mode, apply a constant downward pressure in the axial direction of the robot end effector, keep the first assembly part in contact with the first assembly position and maintain the axial force within the preset threshold range. Using the first assembly position as a reference, the robot end effector is controlled to perform force-guided search in the assembly plane along the second predetermined direction, and the axial force, first radial force, and torque fed back by the robot end effector are monitored in real time during the force-guided search process; If the changes in axial force, radial force, and torque are detected to be less than the corresponding preset changes, the assembly operation of the second assembly part and the second assembly position is completed.

[0053] Specifically, in constant force control mode, the downward axial force of the robot end effector remains constant, maintaining the assembly state of the first assembly part and the first assembly position. With the first assembly position as the center and the workpiece to be registered as the radius, the robot end effector is controlled to perform force-guided search along a second predetermined direction within the assembly plane. This force-guided search can be performed in a manner similar to step S33, ensuring the second assembly part and the second assembly position are registered and aligned. During this process, a six-dimensional force sensor monitors the axial force, first radial force, and torque fed back by the robot end effector in real time. If the detected changes in the axial force, radial force, and torque are all less than their corresponding preset changes, it indicates that the force fed back by the robot end effector remains relatively stable, and the second assembly part and the second assembly position are properly assembled.

[0054] This embodiment uses constant force control to maintain tight contact between assembled parts, preventing loosening of the reference and causing secondary assembly deviations, thus ensuring the consistency of the reference in step-by-step assembly. Secondary force-guided search assembly is performed based on the reference of the first assembly position, eliminating the need for global repositioning, simplifying the control process, and improving the efficiency of continuous assembly at multiple assembly positions. By monitoring the multi-dimensional changes in axial force, radial force, and torque, the assembly status can be determined, accurately identifying whether the second assembly part is smoothly engaged, ensuring the accuracy and reliability of secondary assembly.

[0055] In one embodiment, if the change in at least one of the axial force, radial force, and torque exceeds the corresponding preset change range, a pause is triggered, or the step of "force-guided search of the robot end effector along the second predetermined direction in the assembly plane" is re-executed.

[0056] If the change in at least one of the axial force, radial force, and torque exceeds the corresponding preset change range, it indicates that there may be abnormal contact between the workpiece and the assembly position. This embodiment sets up an abnormal trigger mechanism for abnormal changes in force signals. When abnormal situations such as assembly jamming or misalignment occur, the system can be paused or re-searched in a timely manner to avoid damage to the workpiece, robot, or unloading table caused by forced assembly. It supports automatic retry of the search and assembly in abnormal situations without manual intervention, improving the continuous operation capability and anomaly resolution capability of the assembly system and reducing downtime.

[0057] Please see Figure 5 , Figure 5This is a schematic block diagram of an embodiment of the robot-based workpiece assembly device of this application. The device 200 includes: a robot overall movement module 210, a pose data acquisition module 220, a vision servo module 230, and an assembly module 240. The robot overall movement module 210 is used to control the robot to move to a target area near the unloading table; the pose data acquisition module 220 is used to acquire the workpiece pose of the workpiece to be registered held by the robot, and the first assembly position pose of the first assembly position of the unloading table; the vision servo module 230 is used to control the robot end effector to move according to the workpiece pose and the first assembly position pose, so that the first assembly part on the workpiece to be registered is aligned with the first assembly position; the assembly module 240 is used to switch the robot end effector to a compliant control mode based on force feedback, and adaptively adjust the robot end effector pose according to the real-time force signal of the six-dimensional force sensor of the robot end effector to assemble the first assembly part and the first assembly position; and control the robot to complete the assembly of the second assembly part of the workpiece to be registered with the second assembly position of the unloading table, using the assembled first assembly position as the positioning reference.

[0058] In one embodiment, the visual servo module 230 is further configured to: calculate the pose error based on the workpiece pose and the first assembly position pose, and construct an interaction matrix to describe the mapping relationship between the rate of change of image features and the camera spatial velocity; and control the robot end effector to move based on the interaction matrix and the pose error so that the first assembly part on the workpiece to be registered is aligned with the first assembly position.

[0059] In one embodiment, the assembly module 240 is further configured to: adjust the controller parameters of the robot end effector to below a preset threshold; control the robot end effector to move the workpiece to be registered in a first predetermined direction so that the edge of the first assembly part is in complete contact with the first assembly position, and detect the first radial force of the robot end effector on the assembly plane; keep the first radial force constant, perform force guidance search on the robot end effector along the second radial direction in the assembly plane, and monitor the axial force, the first radial force, and the second radial force fed back by the robot end effector in real time during the force guidance search process; if a combined force signal feature of decreasing axial force, decreasing first radial force, and increasing second radial force is detected, stop adjusting the posture of the robot end effector.

[0060] In one embodiment, the assembly module 240 is further configured to: obtain the speed deviation value of the robot end effector at the previous moment, and integrate the speed deviation at the previous moment to obtain the speed deviation value at the current moment; integrate the speed deviation at the current moment to obtain the position deviation value at the current moment; use the position deviation value at the current moment to perform motion control on the robot end effector; and use the speed deviation value at the current moment as the speed deviation value at the previous moment, and return to execute the step of "obtaining the speed deviation value of the robot end effector at the previous moment and integrating the speed deviation at the previous moment to obtain the speed deviation value at the current moment".

[0061] In one embodiment, the assembly module 240 is further configured to: switch the control mode of the robot end effector to a constant force control mode, apply a constant downward pressure on the robot end effector axially to keep the first assembly part in contact with the first assembly position and maintain the axial force within a preset threshold range; using the first assembly position as a reference, control the robot end effector to perform force-guided search in the assembly plane along a second predetermined direction, and monitor the axial force, first radial force, and torque fed back by the robot end effector in real time during the force-guided search process; if the changes in the axial force, radial force, and torque are detected to be less than the corresponding preset changes, the assembly operation of the second assembly part and the second assembly position is completed. If the changes in at least one of the axial force, radial force, and torque are detected to exceed the preset changes, a pause is triggered, or the step of "controlling the robot end effector to perform force-guided search in the assembly plane along a second predetermined direction" is re-executed.

[0062] For the functions performed by each module of device 200, please refer to the description of the various embodiments of the robot-based workpiece assembly method in this application above, which will not be repeated here.

[0063] Please see Figure 6 , Figure 6 This is a schematic block diagram of a computer device according to an embodiment of the present application. The computer device 900 includes a processor 910 and a memory 920 coupled to each other. The memory 920 stores a computer program, and the processor 910 executes the computer program to implement the robot-based workpiece assembly method described in the above embodiments.

[0064] For a description of each step of the processing, please refer to the description of each step in the above embodiment of the robot-based workpiece assembly method of this application, and it will not be repeated here.

[0065] The memory 920 can be used to store program data and modules. The processor 910 executes various functional applications and data processing by running the program data and modules stored in the memory 920. The memory 920 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function, etc.; the data storage area may store data created based on the use of the computer device 900, etc. In addition, the memory 920 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 920 may also include a memory controller to provide the processor 910 with access to the memory 920.

[0066] In the various embodiments of this application, the disclosed methods, apparatus, and devices can be implemented in other ways. For example, the embodiments of the computer devices described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection between devices or units through some interfaces, and may be electrical, mechanical, or other forms.

[0067] The units described as separation sections may or may not be physically separated. The sections shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment, depending on actual needs.

[0068] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0069] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium.

[0070] See Figure 7 , Figure 7 This is a schematic block diagram of a computer-readable storage medium according to an embodiment of the present application. The computer-readable storage medium 700 stores program data 710, which, when executed, implements the steps of the robot-based workpiece assembly method embodiments described above.

[0071] For a description of each step of the processing, please refer to the description of each step in the above embodiment of the robot-based workpiece assembly method of this application, and it will not be repeated here.

[0072] Any references to memory, storage, database, or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM), etc.

[0073] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0074] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0075] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A robot-based workpiece assembly method, characterized in that, The method includes: Control the robot to move to the target area near the unloading platform; The workpiece pose of the workpiece to be registered held by the robot and the first assembly pose of the first assembly position of the unloading table are collected. Based on the workpiece pose and the first assembly position pose, control the robot end effector to move so that the first assembly part on the workpiece to be registered is aligned with the first assembly position. The robot end effector is switched to a compliant control mode based on force feedback, and the robot end effector's pose is adaptively adjusted according to the real-time force signal from the six-dimensional force sensor of the robot end effector in order to assemble the first assembly part with the first assembly position. Using the first assembly position after assembly as the positioning reference, the robot is controlled to complete the assembly of the second assembly part of the workpiece to be registered with the second assembly position of the unloading table.

2. The method according to claim 1, characterized in that, The step of controlling the robot end effector to move according to the workpiece pose and the first assembly position pose, so as to align the first assembly part on the workpiece to be registered with the first assembly position, includes: Based on the workpiece pose and the first assembly pose, the pose error is calculated, and an interaction matrix is ​​constructed to describe the mapping relationship between the rate of change of image features and the camera spatial velocity. The robot end effector is controlled to move based on the interaction matrix and the pose error, so that the first assembly part on the workpiece to be registered is aligned with the first assembly position.

3. The method according to claim 1, characterized in that, Switching the robot end effector to a force feedback-based compliant control mode includes: Adjust the controller parameters of the robot end effector to below a preset threshold; The adaptive adjustment of the robot's end-effector pose based on the real-time force signal from the six-dimensional force sensor at the robot's end-effector includes: The robot end effector is controlled to move the workpiece to be registered in a first predetermined direction, so that the edge of the first assembly part is in complete contact with the first assembly position, and the first radial force of the robot end effector on the assembly plane is detected. Keeping the first radial force constant, the robot end effector performs a force-guided search along the second radial direction in the assembly plane, and during the force-guided search process, the axial force, the first radial force, and the second radial force fed back by the robot end effector are monitored in real time; If a combined force signal characteristic of decreased axial force, decreased first radial force, and increased second radial force is detected, the attitude adjustment of the robot end effector is stopped.

4. The method according to claim 3, characterized in that, The determination criteria for the combined force signal characteristics include: Within a preset time window, the decrease in axial force exceeds a preset axial force threshold, the decrease in first radial force exceeds a preset first radial force threshold, and after the preset time window, the decrease in second radial force exceeds a second radial force threshold.

5. The method according to claim 3, characterized in that, The force-guided search of the robot end effector along the second radial direction within the assembly plane includes: The velocity deviation value of the robot end effector at the previous moment is obtained, and the velocity deviation value at the current moment is obtained by integrating the velocity deviation at the previous moment. Integrating the velocity deviation at the current moment yields the position deviation value at the current moment; The robot end effector is motion controlled using the current position deviation value. Using the current speed deviation value as the previous speed deviation value, return to execute the step of "obtaining the previous speed deviation value of the robot end effector and integrating the previous speed deviation value to obtain the current speed deviation value".

6. The method according to claim 1, characterized in that, The step of using the first assembly position after assembly as a positioning reference to control the robot to complete the assembly of the second assembly part of the workpiece to be registered with the second assembly position of the unloading table includes: Switch the control mode of the robot end to constant force control mode, apply a constant downward pressure to the robot end in the axial direction, so that the first assembly part keeps in contact with the first assembly position and maintains the axial force within a preset threshold range. Using the first assembly position as a reference, the robot end effector is controlled to perform force-guided search in the assembly plane along the second predetermined direction, and during the force-guided search process, the axial force, the first radial force, and the torque fed back by the robot end effector are monitored in real time. If the changes in the axial force, radial force, and torque are detected to be less than the corresponding preset changes, the assembly operation of the second assembly part and the second assembly position is completed.

7. The method according to claim 6, characterized in that, The method further includes: If the change in at least one of the axial force, radial force, and torque exceeds the preset change range, a pause is triggered, or the step of "controlling the robot end effector to perform force-guided search in the assembly plane along the second predetermined direction" is re-executed.

8. A robot-based workpiece assembly device, characterized in that, The device includes: The robot's overall movement module is used to control the robot to move to the target area near the unloading platform; The pose data acquisition module is used to acquire the workpiece pose of the workpiece to be registered held by the robot, and the first assembly pose of the first assembly position of the unloading platform. A visual servo module is used to control the movement of the robot end effector based on the workpiece pose and the first assembly position pose, so that the first assembly part on the workpiece to be registered is aligned with the first assembly position. The assembly module is used to switch the robot end effector to a compliant control mode based on force feedback, adaptively adjust the robot end effector pose according to the real-time force signal of the six-dimensional force sensor of the robot end effector, so as to assemble the first assembly part and the first assembly position; and control the robot to complete the assembly of the second assembly part of the workpiece to be registered and the second assembly position of the unloading table, using the assembled first assembly position as the positioning reference.

9. A computer device, characterized in that, The device includes a processor and a memory coupled to each other; the memory stores a computer program, and the processor executes the computer program to implement the steps of the method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program data that, when executed by a processor, implements the steps of the method as described in any one of claims 1-7.

Images