Multi-target-station-based long groove workpiece geometric quantity rapid measurement method

Abstract

The invention discloses a long groove workpiece geometric quantity rapid measurement method based on multiple target stations, and the implementation process of the method comprises the steps: installing an industrial robot, the tail end of which is additionally provided with a structured light sensor, on a ground rail or hanger rail system, and achieving the long-distance movement along the axial direction of a workpiece; a plurality of target stations are arranged on the side of the track, and coordinate systems of the target stations are subjected to global coordinate system unification through a laser tracker; in the segmented parking measurement process of the robot, synchronously collecting the profile of the section of a workpiece and the feature point data of an adjacent target station, and converting the coordinates of the section of the workpiece into a global coordinate system by using a pre-established conversion relation between a target station coordinate system and the global coordinate system; through fusion of scanning data of multiple target stations, a global measurement model containing three-dimensional coordinates of complete workpiece feature points is finally constructed. According to the scheme, the technical problem of unification of multi-station coordinates in a large-scale measurement scene is effectively solved, and the detection efficiency of the geometric parameters of the large-scale long-groove workpiece is remarkably improved.

Description

Technical Field

[0001] This invention relates to the fields of visual measurement and surface deformation measurement technology, specifically to a multi-target station-based measurement method that combines technologies such as industrial robots, laser trackers, and structured light sensors. It is widely used for the rapid detection and measurement of geometric quantities and deformation of large-sized workpieces, especially those with long grooves. Background Technology

[0002] High-precision 3D measurement of large workpieces is a crucial aspect of mechanical manufacturing, quality control, and structural analysis. As manufacturing demands for precision and efficiency continue to rise, traditional measurement methods, such as contact measurement and single-sensor-based systems, are no longer sufficient to meet the complex shapes and high-precision requirements of large workpieces. Contact measurement methods typically involve physical contact, which presents challenges such as cumbersome operation, limited measurement accuracy, and the risk of damage to the workpiece surface. Meanwhile, non-contact measurement methods based on single sensors often face problems such as insufficient coverage of the measurement area, inconsistent measurement accuracy, and low efficiency in applications involving large workpieces.

[0003] In recent years, non-contact measurement technology has rapidly become the mainstream in the industry, among which structured light 3D scanning and laser trackers have shown unique advantages in high-precision 3D measurement of large workpieces. Structured light 3D scanning projects structured light onto the surface of an object and reconstructs the object's 3D shape through image processing algorithms, making it suitable for measuring a wide range of shapes. However, it involves a large amount of data processing and is relatively sensitive to the external environment. On the other hand, laser tracker systems, which use the principle of laser interferometry, have an irreplaceable role in the precision inspection of complex curved surfaces in fields such as aerospace manufacturing, thanks to their measurement accuracy on the order of 0.5 ppm. However, when operating as a single machine, there are physical constraints that limit the measurement range, and the complexity of multi-view data fusion needs to be addressed, especially in multi-station measurement scenarios.

[0004] In the measurement of long workpieces, solving the problems of unifying the coordinate system under different measurement postures, fusing data from multiple measurement areas, and improving measurement accuracy remains a technical challenge. Current technologies have not yet effectively combined multi-sensor, multi-target station positioning systems, and robot-assisted systems to achieve efficient, accurate, and rapid high-precision measurement of the geometric quantities of long workpieces.

[0005] To address the aforementioned problems, this invention proposes a high-precision and rapid measurement method for the geometry of long slotted workpieces based on multiple target stations. This method not only improves the measurement accuracy of long workpieces but also significantly increases measurement efficiency and solves problems such as heavy data processing burden and inconsistent measurement accuracy in traditional methods. Summary of the Invention

[0006] The main objective of this invention is to provide a high-precision and rapid measurement method for the geometric quantities of long groove workpieces using a multi-target station, which can effectively solve the problems of low accuracy, low measurement efficiency, and difficulty in controlling local errors in the rapid measurement of geometric quantities of long groove workpieces.

[0007] The technical solution provided by this invention is: a high-precision and rapid measurement method for the geometry of long workpieces based on a multi-target station, comprising the following steps:

[0008] S1: An industrial robot is mounted on a guide rail system. The end effector of the robot is equipped with a structured light sensor, forming a composite measuring device that can move over a wide range along the track, facilitating scanning along the long groove.

[0009] S2: Calculate the optimal layout scheme based on the workpiece size D, the robot's maximum arm span l, and the track length T: Where N represents the required number of target stations, N reference target stations are set at the edge of the track, with the spacing not exceeding the robot's maximum reach, ensuring that each target station is within the robot's measurable range. The long slot workpiece is also divided into N measurement sections, serving as local measurement areas under a single target station. Each target station consists of at least two sets of stepped surfaces, with the edges of each set of stepped surfaces perpendicular to the edges of the other sets. The installation height of the target stations is adjusted according to the workpiece dimensions.

[0010] S3: Calibration phase, mainly involving the calibration of all target stations and the transformation relationship between the structured light sensor coordinate system and the robot base coordinate system. Specific steps include:

[0011] S31: Using a laser tracker and a handheld probe, the characteristic planes of all target stations are calibrated. The probe of the laser tracker is used to continuously scan the stepped surfaces on all targets, with the coordinate system of the laser tracker as the global coordinate system O. L For the i-th target station G i The j-th feature plane F j The global coordinate system O is obtained. L The set of coordinate points below

[0012] S32: The robot stops at the first target station. The robotic arm drives the structured light sensor to scan the first target station at least three times with varying angles, obtaining multiple sets of robotic arm joint parameters and structured light sensor point cloud coordinates. Finally, hand-eye calibration is performed between the structured light sensor and the end flange of the robotic arm, and the sensor coordinate system O is calculated. S With the robot base coordinate system O M Conversion parameters

[0013] S4: The robot stops at the subsequent target station, and the robotic arm drives the structured light sensor to continuously scan and measure the target station and workpiece cross-section from multiple angles. The specific steps include:

[0014] S41: For the i-th target station G i The robotic arm drives the structured light sensor to scan and measure the target station at multiple changing angles across all feature planes, obtaining the sensor coordinate system O. S The set of coordinate points of the characteristic plane below After passing through the sensor coordinate system O in step S32 S With the robot base coordinate system O M Conversion parameters Target station G i The set of coordinate points of all feature planes in the robot base coordinate system This represents the spliced ​​target station G. i All feature planes;

[0015] S42: For the same target station G i The characteristic planes are the sets of coordinate points in the robot base coordinate system. and global coordinate system O L The set of coordinate points below After fitting the characteristic plane, the coordinate system O of the robot's base is calculated when the robot stops for measurement at the i-th target station. M With global coordinate system O L Conversion parameters between

[0016] S43: For the i-th measurement section C of the long slot workpiece i The robotic arm drives the structured light sensor to continuously scan and obtain the sensor coordinate system O. S Set of coordinate points of the measurement section below It also passes through the sensor coordinate system O in step S31. S With the robot base coordinate system O M Conversion parameters The set of coordinate points of the measurement section in the robot base coordinate system was obtained. This represents the measurement section C after splicing. i All feature points;

[0017] S44: For the i-th measurement section C of the long slot workpiece i Section C in the robot base coordinate system i set of coordinate points The robot base coordinate system O obtained after step S53 M With global coordinate system O L Conversion parameters Transform to global coordinate system O L lower coordinate point set

[0018] S5: For the (i+1)th target station and workpiece cross-section, continue performing operation S4 until all N measurement cross-sections are obtained in the laser tracker coordinate system O. L Down That is, the point cloud of the cross section of the complete long groove is obtained.

[0019] S6: The workpiece in the long slot is in the global coordinate system Statistical filtering is used to remove noise coordinates representing anomalies before measuring the diameter and edge deformation of the long groove. For the long groove diameter measurement, the axial direction is first fitted, and then multiple vertical cross-sections are generated along the axial direction. The point cloud data of each cross-section is extracted and fitted to obtain the long groove diameter. For the edge measurement, the edge point cloud is detected based on curvature or normal abrupt change. The edge point cloud is compared with the fitted ideal shape (such as an ellipse or circle), and the distance from each point to the model is calculated as the deformation deviation.

[0020] According to specific embodiments of the present invention, the present invention provides the following technical effects: Through multi-target station arrangement and laser tracker calibration, global consistency of data for long workpieces across multiple measurement areas is achieved. The precise coordinate system method avoids the inconsistency of coordinate systems caused by changes in measurement posture in traditional measurement methods, thereby significantly improving measurement accuracy. Secondly, hand-eye calibration enables precise coordinate transformation between the robot end effector and the sensor, ensuring the consistency and accuracy of measurement data. Point cloud stitching using the least squares fitting method ensures efficient merging of measurement data from various local areas of the workpiece, ultimately forming a precise global point cloud model. This method enables rapid and accurate measurement of three-dimensional data for long workpieces, and is particularly suitable for high-precision, rapid measurement of long, grooved workpieces, providing real-time quality control and dynamic adjustment support for industrial production. Attached Figure Description

[0021] Figure 1 This is a diagram showing the composition of the measuring device in this invention;

[0022] Figure 2 This is a flowchart of the measurement process in this invention;

[0023] Among them: 101-Guide rail system, 102-Industrial robot, 103-Robot flange, 104-Robot base, 200-Structured light sensor, 300-Long slot workpiece, 400-Target station, 500-Laser tracker. Detailed Implementation

[0024] To achieve the above objectives, the technical solution will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] To address the problems existing in current technologies, such as Figure 1 As shown, this invention provides a high-precision and rapid method for measuring the geometry of long groove workpieces using a multi-target station, comprising the following steps:

[0026] S1: An industrial robot 102 is mounted on the guide rail system 101. A structured light sensor 200 is fixed to the end 103 of the robot flange, forming a composite measuring device that can move over a wide range along the guide rail system 101, which is convenient for scanning the workpiece 300 along the long groove.

[0027] S2: Calculate the optimal layout scheme based on the workpiece size D, the robot's maximum arm span l, and the track length T: Where N represents the required number of target stations, N reference target stations 400 are set at the edge of the track, with the spacing not exceeding the robot's maximum reach, ensuring that each target station is within the robot's measurable range. The long slot workpiece is also divided into N measurement sections, serving as local measurement areas under a single target station. Each target station consists of at least two sets of stepped surfaces, with the edges of each set of stepped surfaces perpendicular to the edges of the other sets. The installation height of the target stations is adjusted according to the workpiece size.

[0028] S3: Calibration phase, mainly involving the calibration of all target stations and the transformation relationship between the structured light sensor coordinate system and the robot base 104 coordinate system. Specific steps include:

[0029] S31: Using a laser tracker 500, the characteristic planes of all target stations are calibrated. The probe of the laser tracker is used to continuously scan the stepped surfaces on all targets, with the coordinate system of the laser tracker as the global coordinate system O. L For the i-th target station G i The j-th feature plane F j The global coordinate system O is obtained. L The set of coordinate points below

[0030] S32: The robot stops at the first target station. The robotic arm drives the structured light sensor to scan the first target station at least three times with varying angles, obtaining multiple sets of robotic arm joint parameters and structured light sensor point cloud coordinates. Finally, hand-eye calibration is performed between the structured light sensor and the end flange of the robotic arm, and the sensor coordinate system O is calculated. S With the robot base coordinate system O M Conversion parameters

[0031] S4: The robot stops at the subsequent target station, and the robotic arm drives the structured light sensor to continuously scan and measure the target station and workpiece cross-section from multiple angles. The specific steps include:

[0032] S41: For the i-th target station G i The robotic arm drives the structured light sensor to scan and measure the target station at multiple changing angles across all feature planes, obtaining the sensor coordinate system O. S The set of coordinate points of the characteristic plane below After passing through the sensor coordinate system O in step S32 S With the robot base coordinate system O M Conversion parameters Target station G i The set of coordinate points of all feature planes in the robot base coordinate system This represents the spliced ​​target station G. i All feature planes;

[0033]

[0034] S42: For target station G i The characteristic plane, which lies in the robot base coordinate system O M The set of coordinate points is represented in the middle. And in the global coordinate system O L set of coordinate points in After fitting the characteristic plane, the coordinate system O of the robot's base is calculated when the robot stops for measurement at the i-th target station. M With global coordinate system O L Conversion parameters between

[0035] S43: For the i-th measurement section C of the long slot workpiece i The robotic arm drives the structured light sensor to continuously scan and obtain the sensor coordinate system O. S Set of coordinate points of the measurement section below It also passes through the sensor coordinate system O in step S31. S With the robot base coordinate system O M Conversion parameters The set of coordinate points of the measurement section in the robot base coordinate system was obtained. This represents the measurement section C after splicing. i All feature points;

[0036]

[0037] S44: For the i-th measurement section C of the long slot workpiece i Section C in the robot base coordinate system iset of coordinate points The robot base coordinate system O obtained after step S53 M With global coordinate system O L Conversion parameters Transform to global coordinate system O L lower coordinate point set

[0038]

[0039] S5: For the (i+1)th target station and workpiece cross-section, continue performing operation S4 until all N measurement cross-sections are obtained in the laser tracker coordinate system O. L Down That is, the point cloud of the cross section of the complete long groove is obtained.

[0040]

[0041] S6: The workpiece in the long slot is in the global coordinate system Statistical filtering is used to remove noise coordinates representing anomalies before measuring the diameter and edge deformation of the long groove. For the long groove diameter measurement, the axial direction is first fitted, and then multiple vertical cross-sections are generated along the axial direction. The point cloud data of each cross-section is extracted and fitted to obtain the long groove diameter. For the edge measurement, the edge point cloud is detected based on curvature or normal abrupt change. The edge point cloud is compared with the fitted ideal shape (such as an ellipse or circle), and the distance from each point to the model is calculated as the deformation deviation.

[0042] This method for rapid geometric measurement of long slotted workpieces based on multi-target stations employs a process of precise local measurement followed by global stitching, enabling better measurement of the global geometric quantities and deformations of long-sized objects. In the global calibration stage, a laser tracker scans and measures the distributed target stations, obtaining the coordinates of all target station feature planes in the global coordinate system. In the local section measurement stage, for the registration of the base coordinate system point cloud and the global coordinate system point cloud of the target station feature planes, Singular Value Decomposition (SVD) matrix decomposition technology is used to calculate the minimum eigenvector of the covariance matrix to determine the plane normal vector. Combined with constraint conditions, high-precision plane equation fitting is achieved, effectively handling noise and outliers, thus providing a more stable and reliable transformation relationship between the base coordinate system and the global coordinate system, realizing the conversion of measured coordinate values ​​of local sections to global coordinate values.

[0043] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

Claims

1. A method for rapid measurement of geometric quantities of long slotted workpieces based on multiple target stations, comprising an industrial robot, a guide rail system, a structured light sensor, target stations, and a laser tracker, characterized in that... The following technical steps are included: Step 1: Construct a mobile measurement system by mounting an industrial robot on a track-type mobile platform consisting of a ground rail and a hanging rail. The robot's end effector integrates a structured light sensor, and multiple reference target stations are deployed along the track extension direction. The spacing between the targets does not exceed the robot's maximum working radius, and the height is adaptively adjusted according to the workpiece shape. Step 2: Through hand-eye calibration, establish the transformation from the sensor coordinate system to the robot base coordinate system, and establish a three-dimensional spatial reference. Use a laser tracker to calibrate each target station and obtain the coordinates of the feature plane of each target station in the global coordinate system. Step 3: Segmented dynamic measurement. Control the industrial robot to stop in segments along the track. When stopping to measure at a single station, drive the robotic arm to perform multi-view scanning and simultaneously acquire the local cross-section of the long slot workpiece and the three-dimensional point cloud of the corresponding target station feature plane. Step 4: Point cloud data fusion processing. Based on the coordinates of the feature plane of a single target station in the global coordinate system and the coordinates of the feature plane of the target station in the robot base coordinate system, the least squares fitting based on feature plane matching is used to obtain the transformation matrix between the robot base coordinate system and the global coordinate system when the target station is parked. Finally, the measured local section coordinates in the robot base coordinate system are transformed to the global coordinate system. At the same time, the multi-station collaborative adjustment algorithm is used to adaptively calibrate the cumulative error of robot motion. For the local scanning measurement results of multiple target stations, the point cloud data of multiple target stations are globally stitched together by fitting the feature plane to realize the three-dimensional point cloud reconstruction of the long groove workpiece, and finally generate a full-size three-dimensional geometric model of the workpiece that meets the preset accuracy requirements.

2. The method according to claim 1, characterized in that: By combining hand-eye calibration and laser tracker calibration, coordinate system transformation between sensors, robots, and target stations can be achieved.

3. The method according to claim 1, characterized in that: The spacing between adjacent target stations is dynamically optimized based on the workpiece length and the robot's working radius to ensure coverage of measurement blind spots. The height of the target stations is adjusted according to the curvature of the workpiece surface and then kept fixed.

4. The method according to claim 1, characterized in that: The guide rail system can be in the form of ground guide rails or suspended guide rails.

5. The method according to claim 2, characterized in that: Hand-eye calibration selects any target station, and the robot drives the structured light sensor to continuously scan the step surface from multiple angles to obtain the point cloud in the sensor coordinate system for plane fitting. The least squares fitting method is used to calculate the transformation matrix to realize the transformation relationship from the sensor coordinate system to the robot base coordinate system.

6. The method according to claim 2, characterized in that: A single target station consists of multiple sets of stepped surfaces, and the edges of each set of stepped surfaces are perpendicular to the interfaces of other sets of platforms. The step width of the stepped surfaces allows for the free arrangement of the reflective spheres of the laser interferometer.

Images