Intelligent control upper pressing and lower jacking positioning pin auxiliary device and method

Abstract

The invention discloses an intelligent-control upper-pressing and lower-ejecting positioning pin auxiliary device and method, and relates to the related field of automatic assembling.The intelligent-control upper-pressing and lower-ejecting positioning pin auxiliary device comprises a track construction module, an upper-pressing and lower-ejecting positioning pin auxiliary method comprises the steps that after a workpiece is grabbed in a coarse positioning area, a collision-free track is constructed through pose deviation; the deviation analysis module is used for triggering microcosmic deviation analysis to output a track compensation action and a positioning pin compensation action after the workpiece is transferred into a visual working area; the positioning pin extending module is used for pre-extending a positioning pin on a jig base by adopting a positioning pin compensation action in the process of clamping a workpiece to a force sense guide area; the pressure parameter acquisition module is used for activating a force sensor to acquire embedding pressure parameters after the workpiece and the positioning pin are pre-embedded; and the auxiliary assembly module is used for performing three-force closed-loop optimization according to the embedding pressure parameters, outputting three-force dynamic control parameters and performing jig auxiliary assembly of the workpiece. The technical problem of low assembly precision existing in auxiliary assembly of an existing workpiece jig is solved, and the technical effect of improving the assembly precision is achieved.

Description

Technical Field

[0001] This application relates to the field of automated assembly, and in particular to an upper-pressing and lower-lifting positioning pin auxiliary device and method with intelligent control. Background Art

[0002] In the field of modern industrial automated assembly, high-precision and high-efficiency fixture-assisted assembly of workpieces is crucial for improving product quality and production efficiency. Currently, to solve the problem of workpiece fixture-assisted assembly, the traditional positioning pin-assisted assembly method based on preset trajectories and simple force feedback is mainly adopted, that is, first plan the motion trajectory of the robotic arm, and sense the force during the assembly process through simple force feedback to achieve the assembly of the positioning pin and the workpiece. Since this method does not fully consider the collision risk during the motion of the robotic arm, the planned trajectory may collide with the surrounding environment during actual operation, affecting the assembly safety and efficiency; at the same time, it is difficult to accurately obtain the microscopic deviation between the workpiece and the positioning pin through simple force feedback, resulting in low assembly accuracy and being unable to meet the increasingly high assembly requirements.

[0003] In the related technologies at the present stage, there is a technical problem of low assembly accuracy in workpiece fixture-assisted assembly. Summary of the Invention

[0004] This application provides an upper-pressing and lower-lifting positioning pin auxiliary device and method with intelligent control. After the robotic arm grabs the target workpiece in the rough positioning area, a collision-free trajectory is constructed through pose deviation analysis, and it is transferred to the vision working area to trigger microscopic deviation analysis, output the trajectory and positioning pin compensation actions. When driving the robotic arm to the force sense guidance area according to the corrected trajectory, the electric positioning pin is pre-extended synchronously with the positioning pin compensation action. After the robotic arm and the positioning pin are pre-fitted, the fitting pressure parameter is collected, and based on this parameter, three-force closed-loop optimization is performed to output dynamic control parameters, assisting the upper-pressing mechanism, the lower-lifting structure, and the electric positioning pin to perform linkage displacement to complete fixture-assisted assembly and other technical means, achieving the technical effect of improving assembly accuracy.

[0005] The present application provides an intelligent control upper pressing and lower jacking positioning pin auxiliary device, including: a trajectory construction module, configured to construct a collision-free trajectory through pose deviation analysis after the robotic arm grabs a target workpiece in the rough positioning area; a deviation analysis module, configured to trigger microscopic deviation analysis after driving the robotic arm to transfer the target workpiece into the vision working area by using the collision-free trajectory, and output a trajectory compensation action and a positioning pin compensation action; a positioning pin extension module, configured to synchronously perform pre-extension of K electric positioning pins on the fixture base by using the positioning pin compensation action during the process of driving the robotic arm to clamp the target workpiece to the force sense guiding area by using the collision-free trajectory corrected by the trajectory compensation action; a pressure parameter acquisition module, configured to activate a force sensor to acquire fitting pressure parameters after the robotic arm clamps the target workpiece into the force sense guiding area and pre-fits with the K electric positioning pins; an auxiliary assembly module, configured to perform three-force closed-loop optimization according to the fitting pressure parameters, output three-force dynamic control parameters, and assist the linkage displacement of the upper pressing mechanism, the lower jacking structure, and the K electric positioning pins to perform fixture auxiliary assembly of the target workpiece.

[0006] In a possible implementation manner, the trajectory construction module includes: a macroscopic position point cloud data acquisition module, configured to trigger a line laser scanner to acquire macroscopic position point cloud data according to the grasping behavior of the robotic arm in the rough positioning area, where the macroscopic position point cloud data includes H groups of non-coplanar feature points of the workpiece and positioning pin point clouds; a rough positioning analysis module, configured to perform rough positioning analysis according to the macroscopic position point cloud data and output a 6-degree-of-freedom pose deviation; a collision-free trajectory construction module, configured to construct the collision-free trajectory according to the 6-degree-of-freedom pose deviation, and then start the robotic arm to clamp and displace the workpiece.

[0007] In a possible implementation manner, the rough positioning analysis module includes: a workpiece CAD model calling module, configured to locally call the workpiece CAD model according to the workpiece ID of the target workpiece; a 6-degree-of-freedom pose deviation calculation module, configured to register the macroscopic position point cloud with the workpiece CAD model through the ICP algorithm, and calculate the 6-degree-of-freedom pose deviation of the target workpiece relative to the fixture, where the 6-degree-of-freedom pose deviation is composed of a translation component and a rotation component.

[0008] In a possible implementation, the deviation analysis module includes: an image acquisition module, configured to synchronously activate the main-view camera and the auxiliary-view camera after the target workpiece enters the vision working area, and acquire the workpiece image and the positioning pin image; a feature matching module, configured to perform edge detection and feature matching on the workpiece image and the positioning pin image respectively, and identify the spatial positions of K target positioning holes on the target workpiece and the spatial positions of W electric positioning pins on the fixture, where W is a positive integer greater than K; a neighboring positioning pin matching module, configured to perform neighboring positioning pin matching for the K target positioning holes according to the spatial positions of the K positioning holes and the spatial positions of the W positioning pins, and obtain K hole-pin matching groups; a 6-degree-of-freedom micro deviation calculation module, configured to extract the K electric positioning pins from the K hole-pin matching groups, and calculate and output a 6-degree-of-freedom micro deviation according to the spatial positions of the K positioning holes and the spatial positions of the K electric positioning pins of the K electric positioning pins; an associated compensation and correction module, configured to perform associated compensation and correction according to the 6-degree-of-freedom micro deviation, and output the trajectory compensation action and the positioning pin compensation action.

[0009] In a possible implementation, the associated compensation and correction module includes: a 6-degree-of-freedom micro deviation decomposition module, configured to decompose the 6-degree-of-freedom micro deviation to obtain a translation deviation and a rotation deviation; a fitting end correction amount calculation module, configured to calculate the fitting end correction amount of the robotic arm according to the translation deviation; a gripping pose correction amount calculation module, configured to calculate the gripping pose correction amount of the robotic arm according to the rotation deviation, where the fitting end correction amount and the gripping pose correction amount constitute the trajectory compensation action; a positioning pin compensation calculation module, configured to perform positioning pin compensation calculation after correcting the 6-degree-of-freedom micro deviation according to the trajectory compensation action, and obtain the positioning pin compensation action.

[0010] In a possible implementation, the positioning pin compensation calculation module includes: a 6-degree-of-freedom residual deviation output module, configured to perform micro deviation correction fitting of the 6-degree-of-freedom micro deviation according to the trajectory compensation action, and output a 6-degree-of-freedom residual deviation; a positioning pin extension amount calculation module, configured to extract a planar deviation from the 6-degree-of-freedom residual deviation to calculate the positioning pin extension amount; a biasing pin rotation angle calculation module, configured to extract an angular deviation from the 6-degree-of-freedom residual deviation to calculate the biasing pin rotation angle, where the positioning pin extension amount and the biasing pin rotation angle constitute the positioning pin compensation action.

[0011] In a possible implementation manner, the auxiliary assembly module includes: a three-force balance condition calling module for locally calling the three-force balance condition; a three-force deviation solving module for solving the three-force deviation of the fitting pressure parameter according to the three-force balance condition and outputting initial three-force control parameters, where the initial three-force control parameters include the upper pressing force, the lower top supporting force, and the positioning pin contact force; a PID parameter tuning module for performing PID tuning of the initial three-force control parameters according to the real-time fitting pressure and the three-force deviation of the three-force balance condition during the process of driving the linkage displacement of the upper pressing mechanism, the lower top structure, and the K electric positioning pins by using the initial three-force control parameters.

[0012] In a possible implementation manner, the collision-free trajectory construction module includes: an obstacle three-dimensional grid map construction module for presetting a grid resolution and constructing an obstacle three-dimensional grid map according to the 6-degree-of-freedom pose deviation with the grid resolution as a constraint; a gripping behavior simulation module for simulating the gripping behaviors of the K alternative trajectories and outputting K joint acceleration sequences and K gripping end velocity sequences after searching for the K alternative trajectories in the obstacle three-dimensional grid map with collision-free as the search condition; a gripping stability fusion evaluation module for performing gripping stability fusion evaluation based on the K joint acceleration sequences and the K gripping end velocity sequences and screening and positioning the collision-free trajectory according to the evaluation results.

[0013] In a possible implementation manner, the gripping stability fusion evaluation module includes: a gripping end velocity constraint matching module for matching the gripping end velocity constraint according to the mass characteristics of the target workpiece; a joint acceleration limit extraction module for retrieving and extracting the joint acceleration limit according to the robotic arm ID; an acceleration deviation rate acquisition module for traversing the first joint acceleration sequence by using the joint acceleration limit to obtain Q acceleration deviation rates of Q acceleration deviation nodes; a weighted fusion module for weighted-fusing the Q acceleration deviation rates of the Q acceleration deviation nodes according to the adjacent time differences between the Q acceleration deviation nodes and outputting a first deviation feature; a second deviation feature output module for, by analogy, evaluating the first gripping end velocity sequence by using the gripping end velocity and outputting a second deviation feature; a first stability coefficient output module for weighted-fusing the first deviation feature and the second deviation feature by using a preset stability weight and outputting a first stability coefficient; a collision-free trajectory screening module for, by analogy, performing gripping stability fusion evaluation on the K alternative trajectories, outputting K stability coefficients, and then sorting the K alternative trajectories in ascending order according to the K stability coefficients to screen the collision-free trajectory.

[0014] The present application also provides an intelligent control method for upper pressing and lower jacking positioning pins, including: after the robotic arm grabs the target workpiece in the rough positioning area, constructing a collision-free trajectory through pose deviation analysis; after using the collision-free trajectory to drive the robotic arm to transfer the target workpiece into the vision working area, triggering micro deviation analysis and outputting trajectory compensation actions and positioning pin compensation actions; during the process of using the collision-free trajectory corrected by the trajectory compensation actions to drive the robotic arm to clamp the target workpiece and move towards the force sensing guiding area, synchronously using the positioning pin compensation actions to pre-extend K electric positioning pins on the fixture base; after the robotic arm clamps the target workpiece and enters the force sensing guiding area and pre-engages with the K electric positioning pins, activating the force sensor to collect the engagement pressure parameters; based on the engagement pressure parameters, performing three-force closed-loop optimization, outputting three-force dynamic control parameters, assisting the linkage displacement of the upper pressing mechanism, the lower jacking structure, and the K electric positioning pins, and performing fixture-assisted assembly of the target workpiece.

[0015] It is intended to propose an intelligent control device and method for upper pressing and lower jacking positioning pins through the present application. The trajectory construction module is used to construct a collision-free trajectory through pose deviation analysis after the robotic arm grabs the target workpiece in the rough positioning area. The deviation analysis module is used to trigger micro deviation analysis after using the collision-free trajectory to drive the robotic arm to transfer the target workpiece into the vision working area and output trajectory compensation actions and positioning pin compensation actions. The positioning pin extension module is used to synchronously use the positioning pin compensation actions to pre-extend K electric positioning pins on the fixture base during the process of using the collision-free trajectory corrected by the trajectory compensation actions to drive the robotic arm to clamp the target workpiece and move towards the force sensing guiding area. The pressure parameter acquisition module is used to activate the force sensor to collect the engagement pressure parameters after the robotic arm clamps the target workpiece and enters the force sensing guiding area and pre-engages with the K electric positioning pins. The assisted assembly module is used to perform three-force closed-loop optimization based on the engagement pressure parameters, output three-force dynamic control parameters, assist the linkage displacement of the upper pressing mechanism, the lower jacking structure, and the K electric positioning pins, and perform fixture-assisted assembly of the target workpiece. The technical effect of improving the assembly accuracy is achieved. Brief Description of the Drawings

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings of the embodiments of the present invention will be briefly introduced below. Flowcharts are used in the present application to illustrate the operations performed by the devices according to the embodiments of the present application. It should be understood that the operations before or below do not necessarily need to be precisely executed in sequence. On the contrary, according to needs, they can be executed in reverse order or simultaneously. At the same time, other operations can also be added to these processes, or one or several operations can be removed from these processes.

[0017] Figure 1 It is a schematic structural diagram of an intelligent control device for upper pressing and lower jacking positioning pins provided by an embodiment of the present application.

[0018] Figure 2 It is a schematic flowchart of the intelligent control method for the upper pressing and lower jacking positioning pin assistance provided by the embodiment of the present application.

[0019] Explanation of reference numerals: Trajectory construction module 10, deviation analysis module 20, positioning pin extension module 30, pressure parameter acquisition module 40, auxiliary assembly module 50. Detailed implementation manners

[0020] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the following specifically gives the detailed implementation manners of the present application.

[0021] In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limitations of the present application. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of the present application.

[0022] In the following description, "some embodiments" are involved, which describe a subset of all possible embodiments. However, it can be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments, and can be combined with each other without conflict. The terms "first\second" involved are only used to distinguish similar objects and do not represent a specific order for the objects. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, device, product or server including a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or modules not clearly listed or inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. The terms used herein are only for the purpose of describing the embodiments of the present application.

[0023] The embodiment of the present application provides an intelligent control device for upper pressing and lower jacking positioning pin assistance, as Figure 1 shown. The device includes: A trajectory construction module 10, configured to construct a collision-free trajectory through pose deviation analysis after the robotic arm grasps the target workpiece in the rough positioning area.

[0024] Specifically, first, the robotic arm grasps the target workpiece in the rough positioning area (where the distance between the workpiece and the fixture is greater than 50 mm but less than or equal to 100 mm). This area is to ensure that there is sufficient distance between the workpiece and the fixture when the robotic arm grasps the workpiece, avoiding collisions during the grasping process. After grasping the target workpiece, through pose deviation analysis, the deviation between the actual position and the preset position of the workpiece is calculated. For example, the current position and pose information of the workpiece are obtained through a line laser scanner and compared with the preset standard position and pose to calculate the deviation value.

[0025] According to the results of the pose deviation analysis, combined with the kinematic model of the robotic arm (a mathematical model that describes the relationship between the positions and poses of the joints of the robotic arm and the position and pose of the end effector, used to calculate the motion parameters of the robotic arm), a collision-free trajectory from the current position of the robotic arm to the vision working area is constructed using a path planning algorithm (such as the sampling-based PRM algorithm or the optimization-based CHOMP algorithm). This trajectory is used to enable the robotic arm to avoid all obstacles during movement and safely transfer the target workpiece to the vision working area to prepare for micro deviation analysis.

[0026] For example, assume that the position coordinates of the target workpiece in the rough positioning area are (x 1 , y 1 , z 1 ), the pose is (α 1 , β 1 , γ 1 ), the preset standard position coordinates are (x 0 , y 0 , z 0 ), and the pose is (α 0 , β 0 , γ 0 ). After obtaining the feature point coordinates of the workpiece through the vision sensor, the pose deviation is calculated as Δx = x 1 - x 0 , Δy = y 1 - y 0 , Δz = z 1 - z 0 , Δα = α 1 - α 0 , Δβ = β 1 - β 0 , Δγ = γ 1 - γ 0Using the PRM algorithm, the configuration space of the robotic arm is divided into multiple regions, 1000 sample points are randomly sampled and generated, and a connectivity graph is constructed. A path from the initial configuration to the target configuration of the robotic arm is searched in the connectivity graph. This path passes through 10 key nodes, and each node corresponds to a specific position and posture of the robotic arm. The path is decomposed into 10 motion segments, and the motion parameters (such as speed and acceleration) of each motion segment are calculated by the motion controller to ensure that the robotic arm does not collide with the surrounding environment during the motion process.

[0027] In a possible implementation manner, the trajectory construction module 10 includes: a macro position point cloud data acquisition module, configured to trigger a line laser scanner to acquire macro position point cloud data according to the grasping behavior of the robotic arm in the rough positioning area, where the macro position point cloud data includes H groups of non-coplanar feature points of the workpiece and the positioning pin point cloud; a rough positioning analysis module, configured to perform rough positioning analysis according to the macro position point cloud data and output a 6-degree-of-freedom pose deviation; a collision-free trajectory construction module, configured to construct the collision-free trajectory according to the 6-degree-of-freedom pose deviation and then start the robotic arm to clamp and displace the workpiece.

[0028] Specifically, when the robotic arm performs a grasping behavior in the rough positioning area (where the workpiece is more than 50 mm but less than or equal to 100 mm away from the fixture), the line laser scanner is triggered to acquire macro position point cloud data. The line laser scanner generates point cloud data of the workpiece surface and the positioning pins on the fixture by emitting laser beams and receiving reflected light. These point cloud data include H groups of non-coplanar feature points of the workpiece and the point cloud information of the positioning pins. The non-coplanar feature points are used to determine the spatial position and posture of the workpiece, and the point cloud of the positioning pins is used to determine the preset assembly position of the workpiece. The acquired point cloud data is transmitted to the data processing unit for preliminary filtering and noise reduction processing to remove possible interference points and noise.

[0029] For example, assume that the point cloud data acquired by the line laser scanner in the rough positioning area includes H = 5 groups of non-coplanar feature points of the workpiece and the point cloud information of the positioning pins on the fixture. These feature points are distributed in different parts of the workpiece, and the point cloud of the positioning pins reflects the preset assembly position of the workpiece. After the acquired point cloud data is filtered, the noise points generated due to environmental light interference or the error of the scanner itself are removed, and the effective feature points and the positioning pin point cloud are retained.

[0030] Match the collected macroscopic position point cloud data (including the point cloud of workpiece feature points and positioning pins on the fixture) with a preset standard point cloud model. The standard point cloud model is pre-generated according to the design parameters of the workpiece and the positions of the positioning pins on the fixture, and contains the ideal feature points of the workpiece and the preset positions of the positioning pins. Through a point cloud matching algorithm (such as the Iterative Closest Point algorithm ICP), calculate the deviation between the actual position and orientation of the workpiece and the preset positions of the positioning pins on the fixture. The output result is a 6-degree-of-freedom pose deviation, including translational deviations (Δx, Δy, Δz) in three directions and rotational deviations (Δα, Δβ, Δγ) in three directions. Analyze the calculated 6-degree-of-freedom pose deviation to determine whether the deviation is within the allowable range. If the deviation is too large, it is necessary to re-adjust the grasping position of the robotic arm or re-collect the point cloud data.

[0031] For example, assume that the pose deviation calculated by the point cloud matching algorithm is: Δx = 10mm, Δy = 5mm, Δz = 8mm, Δα = 0.5°, Δβ = 0.3°, Δγ = 0.2°. These deviation values reflect the difference between the actual position and orientation of the workpiece and the preset positions of the positioning pins on the fixture. If the preset allowable deviation range is that the translational deviation does not exceed 15mm and the rotational deviation does not exceed 1°, then the currently calculated deviation is within the allowable range, and the subsequent trajectory construction process can continue.

[0032] According to the 6-degree-of-freedom pose deviation output by the rough positioning analysis module, combined with the kinematic model of the robotic arm, use a path planning algorithm (such as the PRM or CHOMP algorithm) to construct a collision-free trajectory from the current position of the robotic arm to the preset position of the positioning pins on the fixture, so as to safely move the workpiece to the preset position of the positioning pins on the fixture and prepare for subsequent microscopic deviation analysis and precise positioning. Optimize the generated trajectory to ensure that the robotic arm avoids all obstacles during movement and moves smoothly and efficiently. The optimization process includes adjusting parameters such as the smoothness, speed, and acceleration of the trajectory. After the collision-free trajectory is constructed, start the robotic arm to clamp the workpiece and move it to the preset position of the positioning pins on the fixture according to the planned trajectory.

[0033] In a possible implementation, the rough positioning analysis module includes: a workpiece CAD model calling module for locally calling the workpiece CAD model according to the workpiece ID of the target workpiece; a 6-degree-of-freedom pose deviation calculation module for registering the macroscopic position point cloud with the workpiece CAD model through the ICP algorithm and calculating the 6-degree-of-freedom pose deviation of the target workpiece relative to the fixture, where the 6-degree-of-freedom pose deviation is composed of a translational component and a rotational component.

[0034] Specifically, using the workpiece ID as an index, the CAD model of the corresponding workpiece is retrieved from the locally stored CAD model library. For example, if the workpiece ID is "Part_001", the system will call the CAD model file associated with this ID. The CAD model contains the ideal geometric shape and feature point information of the workpiece, which is the basis for pose deviation calculation.

[0035] The collected macroscopic position point cloud data is downsampled and filtered to reduce the computational amount and improve the registration accuracy. The processed point cloud data is registered with the CAD model. The ICP algorithm (point cloud matching algorithm) iteratively calculates the nearest point pairs between the point cloud data and the CAD model and minimizes the distance between them to obtain the best match. The final transformation matrix is obtained through the ICP algorithm, and the translation component and rotation component are extracted from the final transformation matrix as the 6-degree-of-freedom pose deviation. These deviation values reflect the difference between the actual position and orientation of the workpiece and the preset position of the positioning pins on the fixture.

[0036] In a possible implementation, the collision-free trajectory construction module includes: an obstacle three-dimensional grid map construction module for presetting a grid resolution and constructing an obstacle three-dimensional grid map according to the 6-degree-of-freedom pose deviation with the grid resolution as a constraint; a gripping behavior simulation module for simulating the gripping behaviors of the K alternative trajectories and outputting K joint acceleration sequences and K gripping end velocity sequences after searching for K alternative trajectories in the obstacle three-dimensional grid map with collision-free as the search condition; and a gripping stability fusion evaluation module for performing a gripping stability fusion evaluation based on the K joint acceleration sequences and K gripping end velocity sequences and screening and positioning the collision-free trajectory according to the evaluation results.

[0037] Specifically, a predefined grid resolution is set, for example, 1 cm³, to divide the workspace into small three-dimensional grid cells. According to the 6-degree-of-freedom pose deviation of the workpiece and the surrounding environment information (including other devices, tools, and possible obstacles), the positions and shapes of the obstacles in the workspace are mapped into the three-dimensional grid map. Each grid cell is marked as "obstacle-free" or "obstacle", thus forming a detailed obstacle distribution map. With the grid resolution as a constraint, a balance between the accuracy of the map and the computational efficiency is ensured. A higher resolution can provide more accurate obstacle information but will increase the computational amount; a lower resolution is the opposite.

[0038] For example, assume that the preset grid resolution is 1 cm³ and the working space is 1 m³ (100 cm × 100 cm × 100 cm). Then the entire working space is divided into 1,000,000 grid cells. According to the 6-degree-of-freedom pose deviation of the workpiece (Δx = 10 mm, Δy = 5 mm, Δz = 8 mm, Δα = 0.5°, Δβ = 0.3°, Δγ = 0.2°) and the surrounding environment information, the positions and shapes of the obstacles are mapped into the grid map. For example, the devices and tools around the robotic arm are marked as grid cells with "obstacles".

[0039] With collision-free as the search condition, search for possible collision-free trajectories in the 3D grid map of obstacles. Use path planning algorithms (such as the A* algorithm, RRT algorithm, etc.) to search for multiple alternative trajectories from the current position of the robotic arm to the target position (the positioning pin on the fixture). Assume that K alternative trajectories are found. Conduct a gripping behavior simulation for each alternative trajectory, and calculate the joint acceleration sequence and the gripping end velocity sequence of the robotic arm on each trajectory. These sequences reflect the dynamic characteristics of the robotic arm when performing the gripping task. Output K joint acceleration sequences and K gripping end velocity sequences to provide data support for the evaluation of gripping stability.

[0040] For example, assume that K = 5 alternative trajectories are found. Conduct a gripping behavior simulation for each trajectory, and obtain the following results: Trajectory 1: Joint acceleration sequence [a 1 , a 2 ,..., a N , gripping end velocity sequence [v 1 , v 2 ,..., v N ; Trajectory 2: Joint acceleration sequence [a 1 ', a 2 ',..., a N '], gripping end velocity sequence [v 1 , v 2 ,..., v N '];... Trajectory 5: Joint acceleration sequence [a 1 ''''], a 2 ''''],..., a N ''''], gripping end velocity sequence [v 1 ''''], v 2 ''''],..., v N ''''].

[0041] Define the evaluation criteria for clamping stability, such as the smoothness of joint acceleration, the uniformity of the clamping end velocity, the smoothness of the trajectory, etc. These criteria can be adjusted and optimized according to the actual application scenario. Conduct a comprehensive evaluation of the joint acceleration sequence and the clamping end velocity sequence of the K alternative trajectories. Through methods such as weighted average and fuzzy logic, fuse multiple evaluation indicators into a comprehensive score. According to the comprehensive score, select the trajectory with the highest score as the final collision-free trajectory for the motion control of the robotic arm.

[0042] For example, assume the following evaluation criteria are defined: smoothness of joint acceleration (weight 0.4), uniformity of the clamping end velocity (weight 0.3), smoothness of the trajectory (weight 0.3). Conduct a comprehensive evaluation of 5 alternative trajectories and obtain the following comprehensive scores: Trajectory 1: comprehensive score 0.75; Trajectory 2: comprehensive score 0.80; Trajectory 3: comprehensive score 0.65; Trajectory 4: comprehensive score 0.70; Trajectory 5: comprehensive score 0.85; According to the comprehensive score, select Trajectory 5 with the highest score as the final collision-free trajectory.

[0043] In a possible implementation, the clamping stability fusion evaluation module includes: a clamping end velocity constraint matching module for matching the clamping end velocity constraint according to the mass characteristics of the target workpiece; a joint acceleration limit extraction module for retrieving and extracting the joint acceleration limit according to the robotic arm ID; an acceleration deviation rate acquisition module for traversing the first joint acceleration sequence using the joint acceleration limit to obtain Q acceleration deviation rates of Q acceleration deviation nodes; a weighted fusion module for weighted fusing the Q acceleration deviation rates according to the adjacent time differences of the Q acceleration deviation nodes and outputting a first deviation feature; a second deviation feature output module for, by analogy, evaluating the first clamping end velocity sequence using the clamping end velocity and outputting a second deviation feature; a first stability coefficient output module for weighted fusing the first deviation feature and the second deviation feature using a preset stability weight and outputting a first stability coefficient; a collision-free trajectory screening module for, by analogy, performing the clamping stability fusion evaluation of the K alternative trajectories, outputting K stability coefficients, and then sorting the K alternative trajectories in ascending order according to the K stability coefficients to screen the collision-free trajectory.

[0044] Specifically, according to the mass characteristics of the target workpiece (such as weight, size, etc.), match the appropriate clamping end velocity constraint. Different workpiece mass characteristics require different clamping speeds to ensure the stability and safety of the clamping process. Using the mass characteristic parameters of the workpiece (such as weight, size, etc.) as input, retrieve the matching clamping end velocity constraint from a preset database. For example, for a heavier workpiece, a lower clamping speed is required to ensure stability.

[0045] For example, assume that the quality characteristics of the target workpiece are: weight 10 kg, and dimensions 500 mm × 300 mm × 200 mm. Based on these parameters, the system retrieves from the database the matching clamping end velocity constraint: the maximum velocity does not exceed 100 mm / s.

[0046] According to the ID of the robotic arm, retrieve and extract the joint acceleration limits of the robotic arm. Different robotic arm models have different joint acceleration limits to ensure the motion safety and stability of the robotic arm. Using the robotic arm ID as an index, retrieve from the locally stored robotic arm parameter database the joint acceleration limits corresponding to the robotic arm.

[0047] For example, assume that the ID of the robotic arm is "Robot_001", and the system retrieves from the database the joint acceleration limits of this robotic arm as: the maximum accelerations of joints 1 to 6 are 50 rad / s², 40 rad / s², 30 rad / s², 20 rad / s², 15 rad / s², and 10 rad / s² respectively.

[0048] Using the retrieved joint acceleration limits, traverse the first joint acceleration sequence (corresponding to the first alternative trajectory) to calculate Q acceleration deviation rates for Q acceleration deviation nodes. The acceleration deviation rate reflects the deviation degree between the actual acceleration and the limited acceleration. Specifically, traverse the first joint acceleration sequence, and for each time point, calculate the difference between the actual acceleration and the joint acceleration limit, that is, the acceleration deviation rate. The formula is: acceleration deviation rate = (actual acceleration - limited acceleration) / limited acceleration. Record the nodes where the acceleration deviation rate is greater than or equal to the preset threshold as acceleration deviation nodes.

[0049] For example, assume that the first joint acceleration sequence is: [30 rad / s², 45 rad / s², 55 rad / s², 60 rad / s², 40 rad / s²], and the joint acceleration limit is 50 rad / s². Traverse this sequence to calculate the acceleration deviation rate: at the 1st time point: (30 - 50) / 50 = -0.4; at the 2nd time point: (45 - 50) / 50 = -0.1; at the 3rd time point: (55 - 50) / 50 = 0.1; at the 4th time point: (60 - 50) / 50 = 0.2; at the 5th time point: (40 - 50) / 50 = -0.2. Assume that the preset acceleration deviation rate threshold is 0.1, then the 3rd and 4th time points are acceleration deviation nodes, and the corresponding acceleration deviation rates are 0.1 and 0.2 respectively.

[0050] According to the adjacent time differences of Q acceleration deviation nodes, Q acceleration deviation rates are weighted and fused to output the first deviation feature. The weighted fusion takes into account the time factor to more accurately reflect the influence of acceleration deviation. Specifically, calculate the adjacent time differences of each acceleration deviation node (i.e., the time interval between adjacent deviation nodes). Weight the acceleration deviation rates according to the adjacent time differences, and the formula is: weighted deviation rate = acceleration deviation rate × (1 / adjacent time difference). Add up all the weighted deviation rates to obtain the first deviation feature.

[0051] For example, assume Q = 2 acceleration deviation nodes, the acceleration deviation rates are 0.1 and 0.2 respectively, and the adjacent time differences are 0.5 s and 1.0 s respectively. Calculate the weighted deviation rates: The first deviation node: 0.1×(1 / 0.5) = 0.2; The second deviation node: 0.2×(1 / 1.0) = 0.2; The first deviation feature = 0.2 + 0.2 = 0.4.

[0052] Adopt the clamping end velocity constraint to evaluate the first clamping end velocity sequence (corresponding to the first alternative trajectory), and output the second deviation feature. The second deviation feature reflects the deviation degree between the actual value and the constraint value of the clamping end velocity. Specifically, traverse the first clamping end velocity sequence, and for each time point, calculate the difference between the actual velocity and the clamping end velocity constraint, that is, the velocity deviation rate, and the formula is: velocity deviation rate = (actual velocity - velocity constraint) / velocity constraint. Weight and fuse all the velocity deviation rates to obtain the second deviation feature.

[0053] Adopt a preset stability weight to weight and fuse the first deviation feature and the second deviation feature, and output the first stability coefficient. The stability coefficient reflects the clamping stability of the trajectory. For example, the weight of the acceleration deviation feature is 0.6, and the weight of the velocity deviation feature is 0.4. Calculate the stability coefficient, and the formula is: stability coefficient = first deviation feature × weight 1 + second deviation feature × weight 2.

[0054] Repeat the above evaluation process for each alternative trajectory to obtain K stability coefficients. Arrange the K stability coefficients in ascending order, and select the trajectory with the smallest stability coefficient as the final collision-free trajectory.

[0055] The deviation analysis module 20 is used to trigger microscopic deviation analysis after using the collision-free trajectory to drive the robotic arm to transfer the target workpiece into the vision working area, and output a trajectory compensation action and a positioning pin compensation action.

[0056] Specifically, the vision working area refers to the area used for high-precision vision inspection, equipped with a high-precision vision system, such as a binocular stereo vision camera or a 3D laser scanner. After the robotic arm transfers the target workpiece to the vision working area, a high-precision vision system (such as a binocular stereo vision camera or a 3D laser scanner) is used to perform microscopic deviation detection on the workpiece. For example, a binocular stereo vision camera captures images of the workpiece from different angles through two cameras, and uses the parallax principle to calculate the three-dimensional point cloud data on the surface of the workpiece.

[0057] The actual position and pose of the detected workpiece are compared with the preset standard position and pose to calculate the microscopic deviation. For example, through a point cloud matching algorithm, the actual point cloud is aligned with the standard point cloud, and the deviation value of each point is calculated to obtain the overall deviation of the workpiece.

[0058] Based on the microscopic deviation, the inverse kinematics algorithm is used to calculate the trajectory compensation action of the robotic arm and the compensation action of the positioning pin. For example, if the workpiece is offset by Δx in the x direction, the end of the robotic arm needs to move -Δx in the x direction for compensation; at the same time, according to the offset of the workpiece, the length compensation value that the positioning pin needs to extend is calculated.

[0059] For example, assume that the vision system in the vision working area detects that the target workpiece is offset by 2 mm in the x direction, 1 mm in the y direction, 0.5 mm in the z direction, and rotates 0.1 degrees around the x axis, 0.2 degrees around the y axis, and 0.3 degrees around the z axis in terms of pose. The deviation analysis module 20 calculates through the inverse kinematics algorithm that the end of the robotic arm needs to move -2 mm in the x direction, -1 mm in the y direction, -0.5 mm in the z direction, and rotate -0.1 degrees around the x axis, -0.2 degrees around the y axis, and -0.3 degrees around the z axis as the trajectory compensation action. For the positioning pin compensation action, assume that the initial extended length of the positioning pin is L 0 , and according to the offset of the workpiece, the additional extended length ΔL of the positioning pin is calculated. For example, if the workpiece is offset by 0.5 mm in the z direction, the positioning pin needs to extend an additional 0.5 mm, that is, ΔL = 0.5 mm.

[0060] In a possible implementation, the deviation analysis module 20 includes: an image acquisition module, configured to synchronously activate a main perspective camera and an auxiliary perspective camera after the target workpiece enters the vision working area, and acquire a workpiece image and a positioning pin image; a feature matching module, configured to perform edge detection and feature matching on the workpiece image and the positioning pin image respectively, and identify the spatial positions of K target positioning holes on the target workpiece and the spatial positions of W electric positioning pins on the fixture, where W is a positive integer greater than K; a neighboring positioning pin matching module, configured to perform neighboring positioning pin matching for the K target positioning holes according to the spatial positions of the K positioning holes and the spatial positions of the W positioning pins, and obtain K hole-pin matching groups; a 6-degree-of-freedom micro deviation calculation module, configured to extract the K electric positioning pins from the K hole-pin matching groups, and calculate and output a 6-degree-of-freedom micro deviation according to the spatial positions of the K positioning holes and the spatial positions of the K electric positioning pins; and an associated compensation and correction module, configured to perform associated compensation and correction according to the 6-degree-of-freedom micro deviation, and output the trajectory compensation action and the positioning pin compensation action.

[0061] Specifically, after the target workpiece enters the vision working area, the main perspective camera and the auxiliary perspective camera are synchronously activated to acquire a workpiece image and a positioning pin image. The main perspective camera and the auxiliary perspective camera are respectively installed at different positions in the vision working area to obtain multi-perspective images of the workpiece and the positioning pin. When the robotic arm moves the target workpiece to the vision working area, the two cameras are triggered to synchronously acquire images. The main perspective camera is mainly responsible for acquiring the image of the workpiece, and the auxiliary perspective camera is mainly responsible for acquiring the image of the positioning pin on the fixture. For example, acquire a main perspective image facing the positioning hole of the workpiece and an auxiliary perspective image of the fixture positioning pin at a 45-degree skew.

[0062] Perform edge detection and feature matching on the workpiece image and the positioning pin image respectively, and identify the spatial positions of the K target positioning holes on the target workpiece and the spatial positions of the W electric positioning pins on the fixture. Among them, use an edge detection algorithm, such as the Canny edge detection, to extract the edge information in the image. Use a feature matching algorithm, such as SIFT, SURF or ORB, to identify and match the positioning holes on the workpiece and the positioning pins on the fixture. Determine the spatial position of each positioning hole and positioning pin through the matching result.

[0063] Perform neighboring positioning pin matching for the K target positioning holes according to the spatial positions of the K positioning holes and the spatial positions of the W positioning pins, and obtain K hole-pin matching groups. Specifically, calculate the distance between each positioning hole and all the positioning pins. Select the nearest positioning pin for each positioning hole to form a hole-pin matching group. If there are multiple positioning pins with similar distances to a certain positioning hole, the nearest positioning pin can be selected, or other rules (such as preferentially selecting the pre-extended positioning pin) can be used for selection.

[0064] For example, assume that the distance between the first positioning hole and the first positioning pin is the closest, the distance between the second positioning hole and the second positioning pin is the closest, and the distance between the third positioning hole and the third positioning pin is the closest. The matching results are: hole-pin matching group 1: the first positioning hole and the first positioning pin; hole-pin matching group 2: the second positioning hole and the second positioning pin; hole-pin matching group 3: the third positioning hole and the third positioning pin.

[0065] After extracting K electric positioning pins from K hole-pin matching groups, based on the spatial positions of the K positioning holes and the spatial positions of the K electric positioning pins, the 6-degree-of-freedom micro deviation is calculated and output. Specifically, for each hole-pin matching group, the spatial position deviation between the positioning hole and the corresponding positioning pin is calculated. Using the least squares method or other optimization algorithms, the deviations of multiple hole-pin matching groups are combined to calculate the 6-degree-of-freedom micro deviation of the target workpiece relative to the fixture (including translational deviations in three directions and rotational deviations in three directions).

[0066] In addition, the 6-degree-of-freedom micro deviation can also be calculated using SVD decomposition. Specifically, taking the first group of hole-pin matching groups (for example, the first positioning hole and the first positioning pin) as a reference, a local coordinate system is established. Assume that the spatial position of the first positioning hole is P 1 , and the spatial position of the first positioning pin is Q 1 , then the origin of the local coordinate system can be set at P 1 , and the direction vector can be determined by Q 1 −P 1 . For each group of hole-pin matching groups, the relative pose transformation matrix between the positioning hole and the positioning pin is calculated. Convert all relative pose transformation matrices into homogeneous coordinate forms and construct an overall transformation matrix. Perform SVD decomposition on the overall transformation matrix. Through the results of the SVD decomposition, calculate the optimal pose transformation matrix. If the importance of different matching groups is considered, different weights can be assigned to each relative pose transformation matrix, and then the weighted average transformation matrix is calculated. The final 6-degree-of-freedom micro deviation can be calculated through the weighted average transformation matrix.

[0067] Based on the 6-degree-of-freedom micro deviation, correlation compensation and correction are performed, and the trajectory compensation action and the positioning pin compensation action are output. Specifically, according to the calculated 6-degree-of-freedom micro deviation, the trajectory compensation action of the robotic arm is calculated to adjust the movement trajectory of the robotic arm so that the target workpiece can be more accurately aligned with the positioning pin. At the same time, the compensation action of the positioning pin is calculated to adjust the position and posture of the positioning pin to better match the positioning hole of the target workpiece. Output the compensated trajectory and positioning pin action instructions for use by subsequent modules.

[0068] For example, suppose the compensation actions calculated based on micro-deviations are as follows: Trajectory compensation action: The robot moves 0.5 mm in the x-direction, 0.3 mm in the y-direction, and 0.2 mm in the z-direction; rotates 0.1° around the x-axis, 0.05° around the y-axis, and 0.08° around the z-axis. Positioning pin compensation action: The first positioning pin extends 0.2 mm in the z-direction; the second positioning pin extends 0.1 mm in the y-direction; and the third positioning pin extends 0.15 mm in the x-direction.

[0069] In one possible implementation, the associated compensation correction module includes: a 6-degree-of-freedom micro-deviation decomposition module, which is used to decompose the 6-degree-of-freedom micro-deviation to obtain translation deviation and rotation deviation; a fitting end correction amount calculation module, which is used to calculate the fitting end correction amount of the robot arm according to the translation deviation; a clamping posture correction amount calculation module, which is used to calculate the clamping posture correction amount of the robot arm according to the rotation deviation, wherein the fitting end correction amount and the clamping posture correction amount constitute the trajectory compensation action; a locating pin compensation calculation module, which is used to correct the 6-degree-of-freedom micro-deviation according to the trajectory compensation action, and then perform locating pin compensation calculation to obtain the locating pin compensation action.

[0070] Specifically, the translation component (Δx, Δy, Δz) and the rotation component (Δα, Δβ, Δγ) are extracted from the 6-DOF micro-deviation. The translation component represents the translation deviation of the target workpiece in space, and the rotation component represents the rotation deviation of the target workpiece in space.

[0071] The translation deviation is converted into the correction amount of the end effector of the robot arm. The correction amount indicates the distance that the end of the robot arm needs to move in space to compensate for the translation deviation. The calculation formula of the correction amount is: The end correction amount of the joint end = (Δx×safety factor, Δy×safety factor, Δz+anti-collision margin). Among them, the introduction of the safety factor (less than 1) is used to reduce the correction amount to prevent the robot arm from moving too fast or too violently; the increase of the anti-collision margin (a positive value) is used to ensure that the robot arm has enough space in the Z direction to avoid collision. The rotation deviation is converted into the clamping posture correction amount of the end effector of the robot arm. The correction amount indicates the angle that the end of the robot arm needs to rotate in space to compensate for the rotation deviation. The calculation formula of the correction amount is: The clamping posture correction amount = (Δα×rotation attenuation coefficient, Δβ×rotation attenuation coefficient, Δγ×rotation attenuation coefficient). Among them, the introduction of the rotation attenuation coefficient (less than 1) is used to reduce the rotation correction amount to avoid excessive rotation of the robot arm, thereby improving the stability and reliability of the correction.

[0072] Correct the position and posture of the target workpiece according to the correction amount of the end of the fitting and the correction amount of the clamping posture. Recalculate the deviation between the corrected target workpiece and the positioning pin. Calculate the compensation action of the positioning pin according to the corrected deviation, including the extension length and direction adjustment of the positioning pin.

[0073] In one possible implementation, the positioning pin compensation calculation module includes: a 6-degree-of-freedom residual deviation output module, which is used to perform micro-deviation correction fitting of the 6-degree-of-freedom micro-deviation based on the trajectory compensation action, and output the 6-degree-of-freedom residual deviation; a positioning pin extension calculation module, which is used to extract the plane deviation from the 6-degree-of-freedom residual deviation to calculate the positioning pin extension; a deflection pin angle calculation module, which is used to extract the angle deviation from the 6-degree-of-freedom residual deviation to calculate the deflection pin angle, wherein the positioning pin extension and the deflection pin angle constitute the positioning pin compensation action.

[0074] Specifically, after the robot arm performs the trajectory compensation action, the deviation between the target workpiece and the positioning pin is recalculated. The corrected deviation is fitted using the least squares method, SVD decomposition or other optimization algorithms to obtain the 6-DOF residual deviation (including the translation residual deviation in three directions and the rotation residual deviation in three directions).

[0075] Extract the translation residual deviation from the 6-DOF residual deviation. Calculate the extension of each locating pin based on the translation residual deviation. The extension indicates the distance the locating pin needs to extend to compensate for the translation residual deviation. Extract the rotation residual deviation from the 6-DOF residual deviation. Calculate the rotation angle of the deflection pin based on the rotation residual deviation. The rotation angle indicates the angle that the locating pin needs to rotate to compensate for the rotation residual deviation.

[0076] The positioning pin extension module 30 is used to pre-extend K electric positioning pins on the fixture base by using the positioning pin compensation action while driving the robot arm to clamp the target workpiece to the force guide area using the collision-free trajectory corrected by the trajectory compensation action.

[0077] Specifically, according to the positioning pin compensation action generated by the deviation analysis module 20, the K electric positioning pins on the fixture base are controlled by the positioning pin drive controller (such as a servo motor controller) to pre-extend. For example, the extension length of each positioning pin is determined by the rotation angle of the servo motor, and the rotation of the servo motor is accurately controlled by the pulse control signal, thereby realizing the extension action of the positioning pin.

[0078] While the robot arm holds the target workpiece and moves along the corrected collision-free trajectory, the positioning pin extension module 30 is synchronously controlled with the movement of the robot arm. For example, through PLC (Programmable Logic Controller) or industrial bus communication technology, the movement state of the robot arm and the extension state of the positioning pin are synchronized in real time to ensure that the two actions are coordinated and consistent.

[0079] For example, assuming that there are 4 electric positioning pins (K=4) on the fixture base, after the positioning pin extension module 30 receives the compensation action instruction from the deviation analysis module 20, it controls the extension of the 4 positioning pins respectively through the servo motor controller. If the initial extension length of the positioning pin is 10mm, according to the compensation action instruction, the first positioning pin needs to be extended by an additional 0.5mm, the second positioning pin needs to be extended by an additional 0.3mm, the third positioning pin needs to be extended by an additional 0.4mm, and the fourth positioning pin needs to be extended by an additional 0.2mm. The servo motor controller calculates the angle that each servo motor needs to rotate according to these compensation values, and drives the servo motor through a pulse signal to make the positioning pin reach the pre-extended state. When the robot arm clamps the target workpiece and moves along the corrected collision-free trajectory, the positioning pin extension module 30 communicates with the motion controller of the robot arm through the PLC. For example, every time the robot arm moves a preset distance (such as 10mm), the positioning pin extension module 30 adjusts the extension speed and length of the positioning pin according to the current position and speed of the robot arm to ensure that the positioning pin has completed the pre-extended action when the robot arm reaches the force sense guide area.

[0080] The pressure parameter acquisition module 40 is used to activate the force sensor to collect the engagement pressure parameters after the robot arm clamps the target workpiece into the force guide area and pre-engages with the K electric positioning pins.

[0081] Specifically, the force guide area refers to the area where the robot moves the target workpiece to make initial contact and fit with the positioning pin. In this area, the force sensor starts to work and collects the pressure parameters of the fitting process in real time. A high-precision force sensor (such as a six-axis force sensor) is installed at the end of the robot arm or on the positioning pin to collect the pressure parameters of the fitting process in real time. The six-axis force sensor can simultaneously measure the three components of force (F x , F y , F z ) and the three components of the moment (M x , M y , M z ). This sensor provides comprehensive force and torque information for precise control of assembly processes.

[0082] When the robot arm holds the target workpiece and enters the force guide area and pre-engages with K electric positioning pins, the force sensor starts to collect the engagement pressure parameters. At this time, the movement speed of the robot arm will be reduced to ensure the accuracy and safety of the engagement process. The force sensor will collect the pressure parameters (F x , F y , F z , M x , M y , M z) is converted into electrical signals and transmitted to the control computer through a data acquisition card (DAQ). The data acquisition card is used to convert the analog signals output by the sensor into digital signals for processing and analysis.

[0083] The collected signal contains noise and needs to be filtered. For example, a low-pass filter can be used to remove high-frequency noise and retain the valid low-frequency signal. In addition, the signal can be amplified or normalized to better analyze and use the data.

[0084] The collected pressure parameters are fed back to the control module in real time for the three-force closed-loop optimization process. These parameters can reflect the contact force and torque between the workpiece and the locating pin, providing a basis for adjusting the action of the robot arm and the locating pin.

[0085] For example, suppose a six-dimensional force sensor is installed at the end of the robot arm to measure the force and torque when the workpiece contacts the positioning pin. At the same time, a small force sensor is also installed on each positioning pin to measure the reaction force when the positioning pin contacts the workpiece. When the robot arm clamps the target workpiece into the force guide area and pre-engages with the positioning pin, the pressure parameters collected by the force sensor are as follows: F x =10N (force in x direction), F y =5N (force in the y direction), F z =20N (force in z direction), M x =0.5Nm (torque around x-axis), M y =0.3Nm (torque around the y-axis), M z = 0.2Nm (torque around the z-axis). The collected signal is filtered through a low-pass filter (cut-off frequency is 10Hz) to remove high-frequency noise. Then, the signal is amplified 10 times to observe the slight changes more clearly. The collected pressure parameters are transmitted to the control computer in real time for three-force closed-loop optimization. For example, if F z If the pressure in a certain direction exceeds a preset threshold (such as 25N), the system will automatically adjust the downward pressure of the robot arm to avoid excessive squeezing of the workpiece.

[0086] The auxiliary assembly module 50 is used to perform three-force closed-loop optimization based on the fitting pressure parameters, output three-force dynamic control parameters, assist the linkage displacement of the upper pressing mechanism, the lower push structure and K electric positioning pins, and perform fixture-assisted assembly of the target workpiece.

[0087] Specifically, the three-force closed-loop optimization refers to dynamically adjusting the actions of the upper pressure mechanism, the lower top structure and the electric positioning pin through a closed-loop control algorithm according to the real-time collected pressure parameters to ensure that the force and torque are within a preset range. Receive the real-time pressure parameters from the pressure parameter acquisition module 40, which reflect the contact force and torque between the workpiece and the positioning pin. Use a closed-loop control algorithm (such as a PID control algorithm) to optimize the actions of the upper pressure mechanism, the lower top structure and the electric positioning pin in real time. The closed-loop control algorithm dynamically adjusts the actions of each actuator according to the real-time collected pressure parameters to ensure that the force and torque are within a preset range. According to the results of the optimization algorithm, dynamic control parameters are generated, including the downward pressure of the upper pressure mechanism, the thrust of the lower top structure, the telescopic length of the positioning pin, etc. These parameters will be updated in real time to adapt to dynamic changes in the assembly process. The dynamic control parameters are converted into specific control signals and sent to the drive controllers of the upper pressure mechanism, the lower top structure and the electric positioning pin. For example, a pulse signal is generated to control the rotation angle of the servo motor to adjust the telescopic length of the positioning pin. For example, if F is detected z If the pressure in the F direction is too great, the system will automatically reduce the downward pressure of the upper pressure mechanism; x or F y If the pressure in the upper and lower directions is unbalanced, the system will adjust the thrust direction and size of the lower top structure. The upper pressure mechanism, the lower top structure and the electric positioning pin are controlled by a controller (such as a PLC or motion controller). Ensure that the actions of these actuators are coordinated and the target workpiece is assembled accurately. Use industrial bus communication technology (such as EtherCAT or Profibus) to achieve real-time data transmission and synchronous control between the actuators. Ensure that the actions of each mechanism can be precisely coordinated during the assembly process.

[0088] In the embodiment of the present application, after the robotic arm grasps the target workpiece in the coarse positioning area, it constructs a collision-free trajectory through posture deviation analysis, transfers it to the visual working area, triggers micro-deviation analysis, outputs the trajectory and positioning pin compensation action, and drives the robotic arm to the force guidance area according to the corrected trajectory. At the same time, the positioning pin compensation action is used to pre-extend the electric positioning pin. After the robotic arm and the positioning pin are pre-engaged, the engagement pressure parameters are collected, and the three-force closed-loop optimization is performed based on the parameters. The dynamic control parameters are output to assist the upper pressure mechanism, the lower top structure and the electric positioning pin in linkage displacement, and complete the technical means such as jig-assisted assembly, thereby achieving the technical effect of improving the assembly accuracy.

[0089] In one possible implementation, the auxiliary assembly module 50 includes: a three-force balance condition calling module, which is used to locally call the three-force balance condition; a three-force deviation solving module, which is used to solve the three-force deviation of the mating pressure parameters according to the three-force balance condition, and output the initial three-force control parameters, wherein the initial three-force control parameters include the upper pressure, the lower support force and the positioning pin contact force; a PID parameter adjustment module, which is used to perform PID parameter adjustment of the initial three-force control parameters according to the real-time mating pressure and the three-force deviation of the three-force balance condition during the process of using the initial three-force control parameters to drive the linkage displacement of the upper pressure mechanism, the lower support structure and the K electric positioning pins.

[0090] Specifically, the preset three-force balance conditions are called from the local database. These conditions include the balance relationship between the upper pressure, the lower support force and the contact force of the positioning pin, ensuring that the forces on the target workpiece in all directions reach a balanced state during the assembly process. The three-force balance condition can be expressed as a mathematical model, for example: 上压 +F 下顶 + =0, where F 上压 is the pressure applied by the upper pressure mechanism, F 下顶 is the supporting force exerted by the lower roof structure, is the contact force of the ith locating pin.

[0091] The real-time pressure data collected by the embedded pressure sensor is used in combination with the three-force balance condition to calculate the current three-force deviation. The initial three-force control parameters, including the upper pressure, the lower support force, and the positioning pin contact force, are solved by an optimization algorithm (such as the least squares method). In the process of using the initial three-force control parameters to drive the linkage displacement of the upper pressure mechanism, the lower support structure, and the K electric positioning pins, the PID controller is used to dynamically adjust the initial three-force control parameters. The PID controller adjusts the upper pressure, the lower support force, and the positioning pin contact force according to the deviation between the real-time embedded pressure and the three-force balance condition.

[0092] In the above, refer to Figure 1 The intelligent control upper pressing and lower lifting positioning pin auxiliary device according to the embodiment of the present invention is described in detail. Figure 2 The following describes an intelligently controlled upward-pressing and downward-pushing positioning pin auxiliary method according to an embodiment of the present invention.

[0093] The intelligently controlled upward pressing and downward pushing positioning pin auxiliary method according to the embodiment of the present invention is used to solve the technical problem of low assembly accuracy existing in the existing workpiece fixture auxiliary assembly, thereby achieving the technical effect of improving assembly accuracy.

[0094] The intelligently controlled upward-pressing and downward-pushing positioning pin auxiliary method comprises: after the robot arm grasps the target workpiece in the rough positioning area, a collision-free trajectory is constructed through posture deviation analysis; after the collision-free trajectory is used to drive the robot arm to transfer the target workpiece into the visual working area, micro-deviation analysis is triggered, and trajectory compensation action and positioning pin compensation action are output; in the process of driving the robot arm to clamp the target workpiece to the force-sensing guide area with the collision-free trajectory corrected by the trajectory compensation action, the positioning pin compensation action is simultaneously used to pre-extend K electric positioning pins on the fixture base; after the robot arm clamps the target workpiece into the force-sensing guide area and pre-engages with the K electric positioning pins, the force sensor is activated to collect the engagement pressure parameters; three-force closed-loop optimization is performed according to the engagement pressure parameters, and three-force dynamic control parameters are output to assist the linkage displacement of the upward-pressing mechanism, the downward-pushing structure and the K electric positioning pins, so as to perform fixture-assisted assembly of the target workpiece.

[0095] Among them, after the robotic arm grasps the target workpiece in the coarse positioning area, constructing a collision-free trajectory through posture deviation analysis can further include: triggering a line laser scanner to collect macro-position point cloud data according to the grasping behavior of the robotic arm in the coarse positioning area, wherein the macro-position point cloud data includes H groups of non-coplanar feature points of the workpiece and positioning pin point clouds; performing coarse positioning analysis based on the macro-position point cloud data, and outputting a 6-degree-of-freedom posture deviation; after constructing the collision-free trajectory according to the 6-degree-of-freedom posture deviation, starting the robotic arm to clamp the workpiece for displacement.

[0096] Among them, performing coarse positioning analysis based on the macro-position point cloud data and outputting a 6-DOF posture deviation can further include: locally calling a workpiece CAD model according to the workpiece ID of the target workpiece; aligning the macro-position point cloud with the workpiece CAD model through an ICP algorithm, and calculating the 6-DOF posture deviation of the target workpiece relative to the fixture, wherein the 6-DOF posture deviation is composed of a translation component and a rotation component.

[0097] Among them, after the target workpiece is transferred into the visual work area by the collision-free trajectory driving robot arm, micro-deviation analysis is triggered, and trajectory compensation action and locating pin compensation action are output, which can further include: after the target workpiece enters the visual work area, the main view camera and the auxiliary view camera are synchronously activated to collect the workpiece image and the locating pin image; edge detection and feature matching are performed on the workpiece image and the locating pin image respectively, and the K locating hole spatial positions of the K target locating holes on the target workpiece and the W locating pin spatial positions of the W electric locating pins on the fixture are identified, wherein W is a positive integer greater than K; according to the K locating hole spatial positions and the W locating pin spatial positions, the adjacent locating pins of the K target locating holes are matched to obtain K hole-pin matching groups; after extracting the K electric locating pins from the K hole-pin matching groups, according to the K locating hole spatial positions and the K locating pin spatial positions of the K electric locating pins, 6-degree-of-freedom micro-deviations are solved and output; according to the 6-degree-of-freedom micro-deviations, associated compensation correction is performed, and the trajectory compensation action and locating pin compensation action are output.

[0098] Among them, performing associated compensation correction according to the 6-DOF micro-deviation and outputting the trajectory compensation action and the locating pin compensation action can further include: decomposing the 6-DOF micro-deviation to obtain translation deviation and rotation deviation; calculating the interlocking end correction amount of the robot arm according to the translation deviation; calculating the clamping posture correction amount of the robot arm according to the rotation deviation, wherein the interlocking end correction amount and the clamping posture correction amount constitute the trajectory compensation action; after correcting the 6-DOF micro-deviation according to the trajectory compensation action, performing locating pin compensation calculation to obtain the locating pin compensation action.

[0099] Wherein, after correcting the 6-DOF micro-deviation according to the trajectory compensation action, performing positioning pin compensation calculation to obtain the positioning pin compensation action may further include: performing micro-deviation correction fitting of the 6-DOF micro-deviation according to the trajectory compensation action, and outputting a 6-DOF residual deviation; extracting the plane deviation from the 6-DOF residual deviation to calculate the positioning pin extension; extracting the angle deviation from the 6-DOF residual deviation to calculate the deflection pin rotation angle, wherein the positioning pin extension and the deflection pin rotation angle constitute the positioning pin compensation action.

[0100] Among them, three-force closed-loop optimization is performed according to the interlocking pressure parameters, and three-force dynamic control parameters are output to assist the linkage displacement of the upper pressure mechanism, the lower top structure and the K electric locating pins, so as to perform fixture-assisted assembly of the target workpiece, which can further include: locally calling the three-force balance condition; solving the three-force deviation of the interlocking pressure parameters according to the three-force balance condition, and outputting the initial three-force control parameters, wherein the initial three-force control parameters include the upper pressure, the lower top support force and the locating pin contact force; in the process of using the initial three-force control parameters to drive the linkage displacement of the upper pressure mechanism, the lower top structure and the K electric locating pins, the PID adjustment of the initial three-force control parameters is performed according to the real-time interlocking pressure and the three-force deviation of the three-force balance condition.

[0101] Among them, after constructing the collision-free trajectory according to the 6-DOF posture deviation, starting the robot arm to clamp the workpiece displacement can further include: presetting a grid resolution, and taking the grid resolution as a constraint, constructing a three-dimensional grid map of obstacles according to the 6-DOF posture deviation; taking collision-free as a search condition, after searching and obtaining K alternative trajectories in the three-dimensional grid map of obstacles, simulating the clamping behavior of the K alternative trajectories, and outputting K joint acceleration sequences and K clamping end velocity sequences; performing a clamping stability fusion evaluation based on the K joint acceleration sequences and the K clamping end velocity sequences, and screening and locating the collision-free trajectory according to the evaluation results.

[0102] Among them, performing a clamping stability fusion evaluation based on the K joint acceleration sequences and the K clamping end velocity sequences, and screening and locating the collision-free trajectory based on the evaluation results, can further include: matching the clamping end velocity constraint based on the quality characteristics of the target workpiece; retrieving and extracting the joint acceleration limit based on the robot ID; traversing the first joint acceleration sequence using the joint acceleration limit to obtain Q acceleration deviation rates of Q acceleration deviation nodes; weightedly fusing the Q acceleration deviation rates based on adjacent time differences of the Q acceleration deviation nodes, and outputting a first deviation feature; and so on, evaluating the first clamping end velocity sequence using the clamping end velocity, and outputting a second deviation feature; weightedly fusing the first deviation feature and the second deviation feature using a preset stability weight, and outputting a first stability coefficient; and so on, performing a clamping stability fusion evaluation on the K alternative trajectories, and after outputting K stability coefficients, arranging the K alternative trajectories in ascending order based on the K stability coefficients to screen the collision-free trajectory.

[0103] The intelligently controlled push-up and push-down positioning pin auxiliary device provided in the embodiment of the present invention can execute the intelligently controlled push-up and push-down positioning pin auxiliary method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0104] Although the present application makes various references to certain modules in the device according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0105] The above specific implementation manner does not constitute a limitation to the protection scope of the present application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present application should be included in the protection scope of the present application. In some cases, the actions or steps recorded in the present application can be performed in an order different from that in the embodiment and can still achieve the desired results. In addition, the process depicted in the accompanying drawings does not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. Intelligently controlled upward pressing and downward pushing positioning pin auxiliary device, characterized in that: include: The trajectory construction module is used to construct a collision-free trajectory through posture deviation analysis after the robot arm grabs the target workpiece in the rough positioning area; A deviation analysis module, which is used to trigger micro-deviation analysis and output trajectory compensation actions and positioning pin compensation actions after the target workpiece is transferred into the visual work area by the robot arm driven by the collision-free trajectory; The positioning pin extension module is used to pre-extend K electric positioning pins on the fixture base by using the positioning pin compensation action while driving the robot arm to clamp the target workpiece to the force guide area with a collision-free trajectory corrected by the trajectory compensation action; The pressure parameter acquisition module is used to activate the force sensor to collect the fitting pressure parameters after the robot arm clamps the target workpiece into the force guide area and pre-fits with K electric positioning pins; The auxiliary assembly module is used to perform three-force closed-loop optimization according to the fitting pressure parameters, output three-force dynamic control parameters, assist the linkage displacement of the upper pressing mechanism, the lower top structure and K electric positioning pins, and perform fixture-assisted assembly of the target workpiece.

2. The intelligently controlled upward pressing and downward pushing positioning pin auxiliary device according to claim 1, characterized in that: The trajectory construction module includes: A macro position point cloud data acquisition module is used to trigger a line laser scanner to acquire macro position point cloud data according to the grasping behavior of the mechanical arm in the rough positioning area, wherein the macro position point cloud data includes H groups of workpiece non-coplanar feature points and positioning pin point clouds; A coarse positioning analysis module, used for performing coarse positioning analysis based on the macro position point cloud data and outputting a 6-DOF posture deviation; The collision-free trajectory construction module is used to start the displacement of the workpiece clamped by the robot arm after constructing the collision-free trajectory according to the 6-DOF posture deviation.

3. The intelligently controlled upward pressing and downward pushing positioning pin auxiliary device according to claim 2, characterized in that: The coarse positioning analysis module comprises: A workpiece CAD model calling module, used for locally calling a workpiece CAD model according to the workpiece ID of the target workpiece; The 6-DOF posture deviation calculation module is used to align the macro position point cloud with the workpiece CAD model through the ICP algorithm, and calculate the 6-DOF posture deviation of the target workpiece relative to the fixture, wherein the 6-DOF posture deviation is composed of a translation component and a rotation component.

4. The intelligently controlled upward pressing and downward pushing positioning pin auxiliary device according to claim 1, characterized in that: The deviation analysis module includes: An image acquisition module, used to synchronously activate the main view camera and the auxiliary view camera after the target workpiece enters the visual working area, and to acquire the workpiece image and the positioning pin image; A feature matching module is used to perform edge detection and feature matching on the workpiece image and the positioning pin image respectively, and identify the spatial positions of K positioning holes of the K target positioning holes on the target workpiece and the spatial positions of W positioning pins of the W electric positioning pins on the fixture, wherein W is a positive integer greater than K; A proximity locating pin matching module is used to perform proximity locating pin matching of the K target locating holes according to the spatial positions of the K locating holes and the spatial positions of the W locating pins, so as to obtain K hole-pin matching groups; A 6-DOF micro-deviation solving module is used to solve and output the 6-DOF micro-deviation according to the spatial positions of the K locating holes and the spatial positions of the K locating pins of the K electric locating pins after extracting the K electric locating pins from the K hole-pin matching groups; The associated compensation correction module is used to perform associated compensation correction based on the 6-DOF micro-deviation and output the trajectory compensation action and the positioning pin compensation action.

5. The intelligently controlled upward pressing and downward pushing positioning pin auxiliary device according to claim 4, characterized in that: The association compensation correction module includes: A 6-DOF micro-deviation decomposition module is used to decompose the 6-DOF micro-deviation to obtain a translation deviation and a rotation deviation; A fitting end correction amount calculation module, used for calculating the fitting end correction amount of the robot arm according to the translation deviation; A clamping posture correction amount calculation module, used for calculating the clamping posture correction amount of the robot arm according to the rotation deviation, wherein the fitting end correction amount and the clamping posture correction amount constitute the trajectory compensation action; The positioning pin compensation calculation module is used to correct the 6-DOF micro-deviation according to the trajectory compensation action, and then perform positioning pin compensation calculation to obtain the positioning pin compensation action.

6. The intelligently controlled upward pressing and downward pushing positioning pin auxiliary device according to claim 5, characterized in that: The positioning pin compensation calculation module includes: A 6-DOF residual deviation output module is used to perform micro-deviation correction fitting of the 6-DOF micro-deviation according to the trajectory compensation action, and output the 6-DOF residual deviation; A positioning pin extension calculation module is used to extract the plane deviation from the 6-DOF residual deviation to calculate the positioning pin extension; The deflection pin angle calculation module is used to extract the angle deviation from the 6-DOF residual deviation to calculate the deflection pin angle, wherein the positioning pin extension amount and the deflection pin angle constitute the positioning pin compensation action.

7. The intelligently controlled upward pressing and downward pushing positioning pin auxiliary device according to claim 1, characterized in that: The auxiliary assembly module comprises: Three-force balance condition calling module, used to call the three-force balance condition locally; A three-force deviation solving module is used to solve the three-force deviation of the fitting pressure parameter according to the three-force balance condition, and output the initial three-force control parameters, wherein the initial three-force control parameters include the upper pressing pressure, the lower supporting force and the positioning pin contact force; A PID parameter adjustment module is used to perform PID parameter adjustment of the initial three-force control parameters according to the real-time fitting pressure and the three-force deviation of the three-force balance condition when the initial three-force control parameters are used to drive the linkage displacement of the upper pressure mechanism, the lower push structure and the K electric positioning pins.

8. The intelligently controlled upward pressing and downward pushing positioning pin auxiliary device according to claim 2, characterized in that: The collision-free trajectory building module includes: An obstacle three-dimensional grid map construction module is used to preset a grid resolution and use the grid resolution as a constraint to construct an obstacle three-dimensional grid map according to the six-degree-of-freedom posture deviation; A clamping behavior simulation module is used to use non-collision as a search condition, after obtaining K candidate trajectories through searching the three-dimensional grid map of the obstacle, simulate the clamping behavior of the K candidate trajectories, and output K joint acceleration sequences and K clamping end velocity sequences; The clamping stability fusion evaluation module is used to perform clamping stability fusion evaluation based on the K joint acceleration sequences and the K clamping end velocity sequences, and screen and locate the collision-free trajectory based on the evaluation results.

9. The intelligently controlled upward pressing and downward pushing positioning pin auxiliary device according to claim 8, characterized in that: The clamping stability fusion evaluation module includes: A clamping end speed constraint matching module, used for matching the clamping end speed constraint according to the quality characteristics of the target workpiece; A joint acceleration limit extraction module, used to retrieve and extract the joint acceleration limit according to the robot arm ID; An acceleration deviation rate acquisition module, used to traverse the first joint acceleration sequence using the joint acceleration limit to obtain Q acceleration deviation rates of Q acceleration deviation nodes; A weighted fusion module, used for weighted fusion of the Q acceleration deviation rates according to adjacent time differences of the Q acceleration deviation nodes, and outputting a first deviation feature; A second deviation feature output module is used to evaluate the first clamping end speed sequence by analogy using the clamping end speed and output a second deviation feature; A first stability coefficient output module, configured to weight and fuse the first deviation feature and the second deviation feature using a preset stability weight value to output a first stability coefficient; The collision-free trajectory screening module is used to perform a clamping stability fusion evaluation of the K candidate trajectories by analogy, and after outputting K stability coefficients, arrange the K candidate trajectories in ascending order according to the K stability coefficients to screen the collision-free trajectory.

10. Intelligently controlled push-up and push-down positioning pin auxiliary method, characterized in that: The method is implemented by the intelligently controlled upward pressing and downward pushing positioning pin auxiliary device according to any one of claims 1 to 9, and the device comprises: After the robot grasps the target workpiece in the rough positioning area, it constructs a collision-free trajectory through posture deviation analysis; After the target workpiece is transferred into the visual work area by driving the robot arm with the collision-free trajectory, micro-deviation analysis is triggered, and trajectory compensation actions and positioning pin compensation actions are output; In the process of driving the robot arm to clamp the target workpiece to the force sense guide area by using the collision-free trajectory corrected by the trajectory compensation action, the positioning pin compensation action is simultaneously used to pre-extend K electric positioning pins on the fixture base; After the robot arm clamps the target workpiece and enters the force sense guide area for pre-engagement with the K electric positioning pins, the force sensor is activated to collect engagement pressure parameters; The three-force closed-loop optimization is performed according to the fitting pressure parameters, and the three-force dynamic control parameters are output to assist the linkage displacement of the upper pressing mechanism, the lower top structure and the K electric positioning pins to perform the fixture-assisted assembly of the target workpiece.

Images