A vision measurement device and method based on parallel robot

Abstract

The application provides a kind of vision measurement device and method based on parallel robot, belongs to robot vision measurement technical field, including parallel robot, industrial camera, stereoscopic target, probe, contact detection component and processing module;Industrial camera is used to collect the characteristic point image of stereoscopic target;Stereoscopic target is provided with multiple active light characteristic points that are distributed in non-coplanar, for providing spatial geometric constraint;Probe is used to contact the surface of measured part;Contact detection component is used to detect the contact state of probe and measured part surface;Processing module is electrically connected with industrial camera and contact detection component respectively.When detecting effective contact, industrial camera collects target image, processing module solves target space pose, and obtains probe ball center coordinates by combining the fixed geometric relationship between target and probe, and then obtains the feature size of part by geometric fitting.The application realizes the fusion of vision measurement and contact measurement, improves the measurement accuracy and automation degree.

Description

Technical Field

[0001] This invention belongs to the field of robot vision measurement technology, and more specifically, relates to a vision measurement device and method based on parallel robots. Background Technology

[0002] Visual measurement technology, with its advantages of fast response, high automation, and easy integration with robotic systems, has been gradually applied in workpiece positioning and spatial measurement. However, in actual measurement processes, when relying solely on visual images for pose determination, the measurement accuracy is easily affected by factors such as feature point center positioning errors, unreasonable target spatial layout, and image noise. Especially when the number of target feature points is small, their distribution is coplanar, or their geometric constraints are weak, the pose calculation stability is poor, thus affecting the accuracy of the probe's spatial position calculation.

[0003] Furthermore, in contact vision measurement, the positional information of the probe at the instant of contact with the surface of the workpiece is crucial for subsequent dimensional calculations. Existing methods, if unable to accurately detect the contact state and simultaneously trigger the camera to acquire images, are prone to discrepancies between the acquisition time and the actual contact time, thus introducing measurement errors. In addition, the insufficient structural precision and reset consistency of existing contact triggering devices also limit the reliability and repeatability of contact detection. Summary of the Invention

[0004] The purpose of this invention is to provide a vision measurement device and method based on parallel robots, which aims to solve the technical problems of poor stability of target pose calculation, asynchronous contact time and image acquisition time, and low reliability of trigger detection.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is: to provide a vision measurement device based on a parallel robot, comprising: Parallel robots; An industrial camera, positioned on one side of the measurement area, is used to acquire feature point images of the three-dimensional target; A three-dimensional target is provided with multiple actively emitting feature points, which are distributed non-coplanarly in space to provide spatial geometric constraints for the three-dimensional pose calculation of the three-dimensional target after the industrial camera acquires images. The probe is rigidly connected to the three-dimensional target and is used to contact the surface of the part being tested. A contact detection component is used to detect the contact state between the probe and the surface of the part being tested; and The processing module is electrically connected to both the industrial camera and the contact detection component. The industrial camera is used to acquire an image of the stereo target when the contact detection component detects effective contact, and transmit the image to the processing module. The processing module is used to calculate the spatial pose of the stereo target based on the image of the stereo target, and obtain the three-dimensional coordinates of the probe center in the industrial camera coordinate system based on the fixed geometric relationship between the stereo target and the probe. Then, the feature dimensions of the measured part are obtained by acquiring multiple measurement points and geometric fitting.

[0006] In one possible implementation, the contact detection component includes a reset spring and a conductive sheet; when the probe contacts the surface of the part being tested, the probe causes the stereo target to shift or change its orientation and compresses the reset spring, causing the conductive sheet to break the circuit and generate a trigger signal; the industrial camera acquires feature point images of the stereo target based on the trigger signal.

[0007] In one possible implementation, the 3D target is connected to the probe via a target rod. A first trigger ring is fixedly connected to the outer wall of the target rod, and the upper end of the first trigger ring is connected to the bottom end of the reset spring. A shell with open ends is fitted around the outer side of the target rod. The upper end of the shell is connected to a top cover, and the lower end is connected to a base. Both the top cover and the base are annular and have a gap between them and the target rod. The base is fixedly connected to the parallel robot. The reset spring is located inside the shell, and the top end of the reset spring is fixedly connected to the top cover. A second trigger ring is connected to the upper end of the base inside the shell. The conductive sheet is disposed between the first trigger ring and the second trigger ring.

[0008] In one possible implementation, when the probe moves upward, the target rod drives the first trigger ring to compress the reset spring upward, causing the first trigger ring to separate from the second trigger ring, and the conductive sheet to break the circuit; when the probe detaches from the surface of the part being tested, the reset spring drives the first trigger ring to reset, causing the first trigger ring to re-contact the second trigger ring, and the conductive sheet to conduct.

[0009] In one possible implementation, the first trigger ring is provided with meshing teeth along its circumference, the meshing teeth being used to engage with the second trigger ring, the meshing teeth being used to restrict the first trigger ring from rotating around its own axis.

[0010] In one possible implementation, the active light-emitting feature points on the stereo target are infrared light-emitting diodes, and the number of active light-emitting feature points is seven, which constitute a redundant feature point structure.

[0011] In one possible implementation, three of the active light-emitting feature points are arranged in a circular array near the center of the outer wall of the three-dimensional target. The outer wall of the three-dimensional target has multiple protrusions of different thicknesses near its edge. The other four active light-emitting feature points are located on the side of the multiple protrusions away from the three-dimensional target, so as to form a non-coplanar spatial distribution.

[0012] In one possible implementation, the processing module is used to solve for the rotation matrix and translation vector of the 3D target relative to the industrial camera coordinate system based on the center coordinates of the actively emitting feature point in the image coordinate system and the spatial coordinates of the actively emitting feature point in the 3D target coordinate system; and to calculate the current position of the probe sphere center in the industrial camera coordinate system based on the pre-calibrated position of the probe sphere center in the 3D target coordinate system and the pose relationship of the 3D target relative to the industrial camera coordinate system.

[0013] In one possible implementation, the probe is a ruby ​​ball probe, and during the contact measurement process, the coordinates of the center of the ruby ​​ball probe are used as the coordinates of the measurement point.

[0014] The beneficial effects of the visual measurement device based on parallel robots provided by this invention are as follows: Compared with the prior art, this invention enhances the spatial geometric constraints of the target by setting a three-dimensional target with non-coplanar redundant feature points, thereby improving the stability of pose calculation; by adopting a layout of multiple actively emitting feature points, it can improve the recognizability of feature point distribution in a limited space and reduce the impact of image measurement errors on pose calculation results; by setting a contact detection component, it can synchronously trigger the industrial camera to acquire images when the probe contacts the surface of the part being measured, thereby improving the consistency between the contact time and the image acquisition time.

[0015] By utilizing the rigid connection between the probe and the 3D target, the pose of the 3D target can be converted into the spatial position of the probe's center, thereby enabling accurate acquisition of the spatial coordinates of the contact point on the surface of the measured part and improving the measurement accuracy, stability, and automation level of the vision measurement system.

[0016] This invention also provides a vision measurement method based on parallel robots, comprising the following steps: Establish a three-dimensional target coordinate system and obtain the spatial coordinates of each actively emitting feature point on the three-dimensional target in the three-dimensional target coordinate system; The fixed geometric relationship between the probe and the three-dimensional target is calibrated to obtain the position of the probe's center of sphere in the coordinate system of the three-dimensional target. The parallel robot drives the end effector to move, so that the probe contacts the surface of the part being tested; When the probe makes effective contact with the surface of the part being tested, the contact detection component generates a trigger signal and triggers the industrial camera to acquire feature point images of the stereo target; The industrial camera transmits the acquired feature point images to the processing module, which processes the feature point images and extracts the center coordinates of each actively emitting feature point. The processing module calculates the spatial pose of the 3D target relative to the industrial camera coordinate system based on the center coordinates of each active emitting feature point and the spatial coordinates of each active emitting feature point in the coordinate system of the 3D target. The processing module calculates the spatial coordinates of the probe ball center in the industrial camera coordinate system based on the position of the probe ball center in the 3D target coordinate system and the spatial pose of the 3D target relative to the industrial camera coordinate system. Repeat the contact acquisition process to obtain the probe sphere center coordinates corresponding to multiple contact measurement points, and perform geometric fitting based on the multiple probe sphere center coordinates to obtain the feature dimensions of the measured part.

[0017] The beneficial effects of the visual measurement method based on parallel robots provided by this invention are the same as those of the visual measurement device based on parallel robots described above, and will not be repeated here. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 A schematic diagram of the overall structure of a vision measurement device based on a parallel robot provided in an embodiment of the present invention; Figure 2 for Figure 1 Another perspective illustration; Figure 3 for Figure 1 Another perspective illustration; Figure 4 A schematic diagram of the structure of a parallel robot based on a vision measurement device for a parallel robot, provided as an embodiment of the present invention; Figure 5 for Figure 4 A diagram showing the view from below; Figure 6 for Figure 4 A top-down view; Figure 7 for Figure 6 Another perspective illustration; Figure 8 A schematic diagram of the contact detection component, stereo target, and probe of a vision measurement device based on a parallel robot provided in an embodiment of the present invention; Figure 9 for Figure 8 A diagram showing the view from below; Figure 10 for Figure 8 A top-down view; Figure 11 for Figure 10 Another perspective illustration; Figure 12 for Figure 8 A schematic diagram of the structure after removing the outer shell; Figure 13 for Figure 12 Another perspective illustration; Figure 14 for Figure 12 Another perspective illustration; Figure 15 for Figure 12 A schematic diagram of the structure of the target rod, three-dimensional target, probe, return spring, first trigger ring, second trigger ring and meshing teeth; Figure 16 for Figure 15 Another perspective illustration; Figure 17 for Figure 15 Another perspective illustration; Figure 18 for Figure 15 A schematic diagram of the structure after the first and second trigger rings are separated. Figure 19 This is a schematic diagram of the structure of a stereo target for a vision measurement device based on a parallel robot, provided as an embodiment of the present invention.

[0020] Explanation of reference numerals in the attached figures: 1. Parallel robot; 11. Static platform; 12. Moving platform; 13. Active arm; 14. Slave arm; 2. Industrial cameras; 3. Three-dimensional target; 31. Active luminescent feature point; 32. Target rod; 33. Protrusion; 34. Right-angle support for the target; 4. Probe; 5. Contact detection assembly; 51. Return spring; 52. Conductive sheet; 53. First trigger ring; 54. Housing; 55. Top cover; 56. Base; 57. Second trigger ring; 58. Engaging teeth; 6. Frame; 61. Linkage. Detailed Implementation

[0021] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0022] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0023] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0024] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0025] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0026] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0027] In the description of the embodiments of this application, the technical terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing the embodiments of this application and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the embodiments of this application.

[0028] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0029] Currently, in existing visual measurement technologies, when relying solely on visual images for pose determination, the measurement accuracy is easily affected by factors such as feature point center positioning errors, target spatial layout, and image noise. This is especially true when the number of target feature points is small, their distribution is coplanar, or their geometric constraints are weak, resulting in poor pose calculation stability and impacting the accuracy of probe spatial position calculation. Furthermore, in contact measurement processes, existing methods struggle to accurately detect the probe's contact state and simultaneously trigger the camera to acquire images. Discrepancies between the acquisition time and the actual contact time introduce measurement errors. Additionally, the structural precision and reset consistency of existing triggering devices are insufficient, limiting the reliability of contact detection and measurement repeatability.

[0030] To address the aforementioned issues, this invention enhances the spatial geometric constraints of the target by employing a 3D target with non-coplanar redundant feature points and a layout of multiple actively emitting feature points, thereby improving the accuracy and stability of pose calculation. Simultaneously, by integrating a contact detection component, an industrial camera is synchronously triggered to acquire images when the probe contacts the workpiece surface. The rigid connection between the probe and the 3D target is then used to calculate the probe's spatial position, thereby improving the measurement accuracy, stability, and automation level of the vision measurement system.

[0031] Please refer to the following together. Figures 1 to 19 This application provides a detailed description of a vision measurement device and method based on a parallel robot, as provided in the embodiments of this application.

[0032] The aforementioned visual measurement device based on a parallel robot includes a parallel robot 1, an industrial camera 2, a stereo target 3, a probe 4, a contact detection component 5, and a processing module (not shown in the figure).

[0033] Industrial camera 2 is positioned on one side of the measurement area (parallel robot 1 is located within the field of view of industrial camera 2) to acquire feature point images of the stereo target 3. The field of view of industrial camera 2 should cover the motion space of the moving platform of parallel robot 1 to ensure that clear target images can be acquired at each measurement position. Parallel robot 1 is positioned inside frame 6, and a vertically arranged link 61 is connected to one side of frame 6. Industrial cameras 2 (two sets, spaced apart) are connected to this link 61.

[0034] The stereo target 3 is provided with multiple actively emitting feature points 31, which are non-coplanarly distributed in space to provide spatial geometric constraints for the three-dimensional pose calculation of the stereo target 3 after the industrial camera 2 acquires images. The probe 4 is rigidly connected to the stereo target 3 and is used to contact the surface of the part being measured. The contact detection component 5 is used to detect the contact state between the probe 4 and the surface of the part being measured; the processing module is electrically connected to both the industrial camera 2 and the contact detection component 5. The processing module is an industrial computer, embedded processor, or other electronic device with data processing capabilities, as is available in the prior art.

[0035] The industrial camera 2 is used to acquire an image of the stereo target 3 when the contact detection component 5 detects effective contact, and transmit the image to the processing module. The processing module is used to calculate the spatial pose of the stereo target 3 based on the image of the stereo target 3, and obtain the three-dimensional coordinates of the center of the probe 4 in the coordinate system of the industrial camera 2 based on the fixed geometric relationship between the stereo target 3 and the probe 4. Then, the feature dimensions of the measured part are obtained by acquiring multiple measurement points and geometric fitting.

[0036] Parallel robot 1 includes a static platform 11, a moving platform 12, an active arm 13, and a driven arm 14. The moving platform 12 is used to drive the contact detection component 5 to move in space. Parallel robot 1 adopts a parallel structure, which has the advantages of high rigidity, high precision, and fast dynamic response, and is suitable for high-precision visual measurement tasks.

[0037] This invention provides a vision measurement device based on a parallel robot 1. Compared with the prior art, this invention enhances the spatial geometric constraints of the target by setting a three-dimensional target 3 with non-coplanar redundant feature points, thereby improving the stability of pose calculation. By employing multiple actively emitting feature points 31, the identifiability of feature point distribution within a limited space can be improved, reducing the impact of image measurement errors on pose calculation results. By setting a contact detection component 5, the industrial camera 2 can be triggered to acquire images simultaneously when the probe 4 contacts the surface of the part being measured, improving the consistency between the contact time and the image acquisition time.

[0038] By utilizing the rigid connection between probe 4 and the stereo target 3, the pose of the stereo target 3 can be converted into the spatial position of the center of probe 4, thereby achieving accurate acquisition of the spatial coordinates of the contact point on the surface of the measured part and improving the measurement accuracy, stability and automation level of the vision measurement system.

[0039] In some embodiments, please refer to Figures 8 to 18 The contact detection component 5 includes a reset spring 51 and a conductive sheet 52. When the probe 4 comes into contact with the surface of the part being tested, the probe 4 causes the stereo target 3 to shift or change its orientation and compresses the reset spring 51, causing the conductive sheet 52 to break the circuit and generate a trigger signal. The industrial camera 2 acquires feature point images of the stereo target 3 based on the trigger signal.

[0040] The 3D target 3 is connected to the probe 4 via the target rod 32 (the 3D target 3 is fixedly connected to the side of the target right-angle support 34, which is set at a right angle; the target rod 32 is cylindrical, with its upper end fixedly connected to the lower end of the target right-angle support 34, and the axis of the target rod 32 is perpendicular to the axis of the 3D target 3). A first trigger ring 53 is fixedly connected to the outer wall of the target rod 32, and the upper end of the first trigger ring 53 is connected to the bottom end of the reset spring 51; two... The outer shell 54 has an open end. The upper end of the outer shell 54 is connected to the top cover 55 and the lower end is connected to the base 56. Both the top cover 55 and the base 56 are annular and have a gap between them and the target rod 32. The base 56 is fixedly connected to the parallel robot 1. The reset spring 51 is located inside the outer shell 54. The top end of the reset spring 51 is fixedly connected to the top cover 55. The second trigger ring 57 is connected to the upper end of the base 56 inside the outer shell 54. The conductive sheet 52 is located between the first trigger ring 53 and the second trigger ring 57. The lower end of the target rod 32 is fixedly connected to the probe 4, and the upper end of the target rod 32 is fixedly connected to the upper end of the three-dimensional target 3. The outer shell 54 is cylindrical with open ends, forming a ring around the outer circumference of the target rod 32. The height of the outer shell 54 is less than the height of the target rod 32, and it is located in the middle of the target rod 32. The base 56 is fixedly connected to the moving platform 12 by bolts. The target rod 32 does not contact the top cover 55 and the base 56 to ensure that the target rod 32 can move up and down relative to the outer shell 54. The return spring 51 is located inside the outer shell 54 and its upper end is fixedly connected to the bottom end of the top cover 55. The lower end of the return spring 51 is fixedly connected to the upper end of the first trigger ring 53. The conductive sheet 52 is electrically connected to the processing module for outputting electrical signals.

[0041] The working principle of the contact detection component 5 is as follows: When the probe 4 moves upward, it indicates that the probe 4 has made contact with the surface of the part being measured. The probe 4 is subjected to an upward contact force, which causes the probe 4 and the target rod 32 to drive the first trigger ring 53 to compress the reset spring 51 upward. That is, the stereo target 3 and the target rod 32 move upward relative to the outer shell 54, causing the first trigger ring 53 to separate from the second trigger ring 57. The conductive sheet 52 changes from a conductive state to an open circuit state, and the open circuit of the conductive sheet 52 generates a trigger signal. After receiving the trigger signal, the processing module controls or responds to the industrial camera 2 to acquire the feature point image of the stereo target 3.

[0042] When probe 4 detaches from the surface of the part being measured, the contact force disappears, and the reset spring 51 drives the first trigger ring 53 to reset, so that the first trigger ring 53 and the second trigger ring 57 re-contact, the conductive sheet 52 returns to the conductive state, and the stereo target 3 and probe 4 return to the initial state so as to carry out contact acquisition at the next measurement point.

[0043] In some embodiments, please refer to Figures 12 to 18 The first trigger ring 53 has meshing teeth 58 arranged along its circumference. The meshing teeth 58 are used to mesh with the second trigger ring 57 and to restrict the first trigger ring 53 from rotating around its own axis. Since the target rod 32 is fixedly connected to the first trigger ring 53, and the three-dimensional target 3 and probe 4 are fixedly connected to the target rod 32, the meshing teeth 58 can restrict the three-dimensional target 3 and probe 4 from rotating around their own axes, thereby improving the consistency and accuracy of the reset position of the contact detection component 5 and ensuring the repeatability of multiple measurement results.

[0044] In some embodiments, please refer to Figure 19 The active light-emitting feature point 31 on the stereo target 3 is preferably an infrared light-emitting diode (LED) with a power of 0.3W. A circular lens is provided in the light emission direction of the active light-emitting feature point 31. The circular lens is used to constrain the light emission contour of the active light-emitting feature point 31 so that it forms a circular light spot in the image captured by the industrial camera 2, which facilitates the center positioning in subsequent image processing.

[0045] Preferably, the number of active luminescent feature points 31 is seven, and the seven active luminescent feature points 31 constitute a redundant feature point structure to improve the accuracy and stability of the 3D target pose calculation.

[0046] In some embodiments, please refer to Figure 19The seven active light-emitting feature points 31 are arranged as follows: three of the active light-emitting feature points 31 are arranged in a circular array near the center of the outer wall of the three-dimensional target 3, forming an inner circular layout. Multiple protrusions 33 of varying thicknesses are provided near the edge of the outer wall of the three-dimensional target 3. The other four active light-emitting feature points 31 are located on the side of the multiple protrusions 33 away from the three-dimensional target 3. Due to the different thicknesses of the protrusions 33, the active light-emitting feature points 31 located on the protrusions 33 are at different heights from the active light-emitting feature point 31 located at the center, thus forming a non-coplanar spatial distribution.

[0047] Through the above layout, the seven actively emitting feature points 31 form a spatially non-coplanar distribution, providing sufficient spatial geometric constraints for three-dimensional pose calculation. Compared with the target layouts in existing technologies that are coplanar or have a small number of feature points, this invention can effectively reduce the impact of feature point center positioning errors and image noise on the pose calculation results, ensuring the accuracy of the spatial position calculation of probe 4.

[0048] Preferably, the inner circular layout and the active light-emitting feature points 31 on each of the outer protrusions 33 have different radius dimensions, so as to form spatial geometric constraints of different scales in the effective target space, enhance the recognizability of the geometric features of the three-dimensional target 3, and further reduce the influence of image measurement error on the pose calculation results.

[0049] In some embodiments, the processing module is used to solve for the rotation matrix and translation vector of the stereo target 3 relative to the industrial camera 2 coordinate system based on the center coordinates of the actively emitting feature point 31 in the image coordinate system and the spatial coordinates of the actively emitting feature point 31 in the stereo target 3 coordinate system. Then, based on the pre-calibrated position of the probe 4 center in the stereo target 3 coordinate system and the pose relationship of the stereo target 3 relative to the industrial camera 2 coordinate system, the current position of the probe 4 center in the industrial camera 2 coordinate system is calculated.

[0050] Specifically, the processing module first calibrates the industrial camera 2 to obtain its internal parameters (including focal length, principal point coordinates, distortion coefficients, etc.) for subsequent pose calculation of the stereo target 3. Then, it establishes a coordinate system for the stereo target 3 and calculates the spatial coordinates of each actively emitting feature point 31 within this system based on their known geometric relationships. Since the relative positions of the seven actively emitting feature points 31 on the stereo target 3 are fixed and their geometric dimensions are known, the coordinates of each feature point in the stereo target 3 coordinate system can be obtained through design drawings or pre-calibration.

[0051] During the measurement process, after the industrial camera 2 acquires the feature point image of the stereo target 3, the processing module processes the image and extracts the center coordinates of each actively emitting feature point 31 in the image coordinate system. Preferably, the processing module adopts a sub-pixel-level center positioning algorithm to obtain the sub-pixel-level center coordinates of each actively emitting feature point 31, so as to improve the pose calculation accuracy.

[0052] The processing module calculates the rotation matrix and translation vector of the stereo target 3 relative to the coordinate system of the industrial camera 2 based on the image coordinates of each active luminous feature point 31 and the spatial coordinates in the coordinate system of the stereo target 3 using the binocular stereo matching PnP (Perspective-n-Point) algorithm, that is, the spatial pose of the stereo target 3.

[0053] The probe 4 and the stereo target 3 are rigidly connected, and the position of the center of probe 4 in the coordinate system of stereo target 3 remains unchanged. The processing module calculates the current position of the center of probe 4 in the coordinate system of industrial camera 2 based on the pre-calibrated position of the center of probe 4 in the coordinate system of stereo target 3 and the pose relationship of stereo target 3 relative to the coordinate system of industrial camera 2. Specifically, the coordinates of the center of probe 4 in the coordinate system of stereo target 3 are transformed to the coordinate system of industrial camera 2 using a rotation matrix and a translation vector.

[0054] The processing module is also used to perform fitting calculations on the circles, holes, planes, contours, or other geometric features of the measured part based on the coordinates of the probe 4 sphere centers corresponding to multiple contact measurement points, thereby obtaining the feature dimensions of the measured part. For example, when it is necessary to measure the diameter of a circular hole, multiple contact points can be collected on the inner wall of the hole to obtain multiple sphere center coordinates, and then these coordinates can be fitted with a circle to obtain the diameter and center position of the circle. When it is necessary to measure the flatness of a plane, multiple contact points can be collected on the plane to obtain multiple sphere center coordinates, and then these coordinates can be fitted with a plane to obtain the normal vector and flatness error of the plane.

[0055] In some embodiments, probe 4 is a ruby ​​ball probe, and during the contact measurement process, the coordinates of the center of the ruby ​​ball probe are used as the coordinates of the measurement point.

[0056] Please see Figures 1-19 The present invention also provides a vision measurement method based on parallel robots, which uses the above-mentioned vision measurement device based on parallel robots and includes the following steps: S1: Establish the coordinate system of the three-dimensional target 3, and obtain the spatial coordinates of each active luminous feature point 31 on the three-dimensional target 3 in the coordinate system of the three-dimensional target 3; S2: The fixed geometric relationship between probe 4 and 3D target 3 is calibrated to obtain the position of the center of probe 4 in the coordinate system of 3D target 3; S3: Parallel robot 1 drives the end effector to move, so that probe 4 contacts the surface of the part being tested; S4: When the probe 4 makes effective contact with the surface of the part being tested, the contact detection component 5 generates a trigger signal and triggers the industrial camera 2 to acquire feature point images of the stereo target 3. S5: The industrial camera 2 transmits the acquired feature point image to the processing module. The processing module processes the feature point image and extracts the center coordinates of each active luminous feature point 31. S6: The processing module calculates the spatial pose of the stereo target 3 relative to the coordinate system of the industrial camera 2 based on the center coordinates of each active light-emitting feature point 31 and the spatial coordinates of each active light-emitting feature point 31 in the coordinate system of the stereo target 3. S7: The processing module calculates the spatial coordinates of the center of probe 4 in the coordinate system of industrial camera 2 based on the position of the center of probe 4 in the coordinate system of 3D target 3 and the spatial pose of 3D target 3 relative to the coordinate system of industrial camera 2. S8: Repeat the contact acquisition process to obtain the coordinates of the probe 4 sphere center corresponding to multiple contact measurement points, and perform geometric fitting based on the coordinates of the multiple probe 4 sphere centers to obtain the feature dimensions of the measured part.

[0057] In step S1 above, specifically, the three-dimensional coordinates of the seven active luminescent feature points 31 in the coordinate system of the three-dimensional target 3 are determined according to the design dimensions of the three-dimensional target 3 or through pre-calibration. Since the relative positional relationship between each active luminescent feature point 31 is fixed and known, the coordinate value of each feature point in the coordinate system of the three-dimensional target 3 can be accurately obtained.

[0058] In step S2 above, since probe 4 and the 3D target 3 are rigidly connected by target rod 32, the position of the center of probe 4 in the coordinate system of the 3D target 3 remains unchanged. Through pre-calibration, the precise position of the center of probe 4 in the coordinate system of the 3D target 3 can be obtained.

[0059] Optionally, the order of steps S1 and S2 can be interchanged, or they can be performed simultaneously.

[0060] In step S3 above, specifically, the processing module controls the moving platform of the parallel robot 1 to move along a preset path, so that the probe 4, which is fixedly connected to the moving platform, gradually approaches and contacts the surface of the part being measured. The parallel robot 1 has advantages such as high rigidity, high precision, and fast dynamic response, and can accurately control the contact position and contact force of the probe 4.

[0061] In step S4 above, when probe 4 contacts the surface of the part being tested, probe 4 is subjected to contact force. Target rod 32 drives the first trigger ring 53 to compress the return spring 51 upwards, causing the first trigger ring 53 to separate from the second trigger ring 57. The conductive sheet 52 is then disconnected, generating a trigger signal. After receiving the trigger signal, the processing module immediately controls or responds to the industrial camera 2 to acquire an image of the stereo target 3.

[0062] Since the generation of the trigger signal is synchronized with the contact of probe 4 with the surface of the part being measured, the consistency between the image acquisition time and the contact time is ensured, and measurement errors caused by acquisition delay are avoided.

[0063] In step S5 above, the processing module preprocesses the acquired image, including filtering and denoising, binarization, etc., and then extracts the center coordinates of each active luminescent feature point 31 in the image coordinate system. Preferably, the processing module uses a sub-pixel-level center localization algorithm to obtain the sub-pixel-level center coordinates of each active luminescent feature point 31, so as to improve the accuracy of subsequent pose calculation.

[0064] In step S6 above, the processing module uses the PnP algorithm to solve the rotation matrix and translation vector of the stereo target 3 relative to the coordinate system of the industrial camera 2, based on the correspondence between the two-dimensional image coordinates and the three-dimensional spatial coordinates.

[0065] Before the calculation, the processing module calibrated the industrial camera 2 and obtained the internal parameters of the industrial camera 2, which were used to convert the image coordinates into normalized coordinates in the camera coordinate system.

[0066] In step S7 above, specifically, the coordinates of the center of probe 4 in the coordinate system of the 3D target 3 are calculated with the rotation matrix and translation vector of the 3D target 3 to obtain the three-dimensional coordinates of the center of probe 4 in the coordinate system of the industrial camera 2. These coordinates represent the spatial position of the contact point when probe 4 contacts the surface of the part being measured.

[0067] In step S8 above, the parallel robot 1 drives the probe 4 to sequentially contact multiple measurement positions on the surface of the part being measured, repeating steps S3 to S7 to obtain the coordinates of the probe 4's center of gravity corresponding to multiple contact points. The processing module performs geometric fitting based on these coordinates, such as circle fitting, plane fitting, cylinder fitting, etc., to obtain characteristic dimensions of the part being measured, such as diameter, flatness, roundness, and position.

[0068] In one embodiment of the present invention, the conductive sheet 52 of the contact detection component 5 is normally in a conductive state (i.e., the first trigger ring 53 and the second trigger ring 57 are in contact with each other). The processing module continuously monitors the voltage level of the conductive sheet 52. When the probe 4 contacts the surface of the part being tested, the first trigger ring 53 and the second trigger ring 57 separate, the conductive sheet 52 is open-circuited, and the voltage level changes. After detecting this voltage level change, the processing module determines that it is a valid contact and immediately triggers the industrial camera 2 to acquire an image.

[0069] In another embodiment of the present invention, the trigger signal can also be directly connected to the hardware trigger port of the industrial camera 2. When the conductive sheet 52 is open-circuited, the industrial camera 2 is directly triggered to acquire images without going through the processing module, thereby further shortening the response time and improving the synchronization accuracy.

[0070] In one embodiment of the present invention, all seven actively emitting feature points 31 are infrared light-emitting diodes and are driven by a constant current to ensure that the luminous intensity of each feature point is stable and consistent. A circular lens is provided in the light emission direction of the infrared light-emitting diodes so that the feature point image captured by the industrial camera 2 is a uniform circular light spot, facilitating center positioning.

[0071] Preferably, the luminescence intensity of the actively emitting feature point 31 is adjustable to adapt to different ambient lighting conditions and measurement distances. The processing module can automatically adjust the luminescence intensity of each feature point according to image quality, ensuring that clear and distinguishable feature point images can be obtained under various working conditions.

[0072] In one embodiment of the present invention, the reset spring 51 is a cylindrical helical compression spring, and its stiffness coefficient is selected according to the total mass of the probe 4 and the three-dimensional target 3 and the required contact force range. Preferably, the stiffness coefficient of the reset spring 51 is 0.5 N / mm-2.0 N / mm, so as to ensure that sufficient displacement can be generated under a small contact force to break the circuit of the conductive sheet 52, while ensuring that the probe 4 can reliably reset after being removed from the surface being measured.

[0073] The meshing teeth 58 are rectangular or triangular teeth, with at least six teeth, evenly distributed along the circumference of the first trigger ring 53, and cooperate with structures (such as grooves) on the second trigger ring 57. The meshing teeth 58 cooperate with the corresponding meshing grooves on the second trigger ring 57, restricting the rotational freedom of the first trigger ring 53 about its own axis, while allowing the first trigger ring 53 to reciprocate along the axial direction.

[0074] The beneficial effects of this invention are as follows: 1. By setting a three-dimensional target 3 with non-coplanar redundant feature points (seven actively emitting infrared feature points), the spatial geometric constraints of the target are enhanced, and the stability and accuracy of pose calculation are improved.

[0075] 2. The contact detection component 5 enables synchronous triggering of contact state detection and image acquisition, effectively avoiding measurement errors caused by the inconsistency between the acquisition time and the actual contact time.

[0076] 3. Through the specific structural design of the contact detection component 5, high reliability and high consistency of contact detection are achieved.

[0077] 4. By leveraging the rigid connection between probe 4 and the stereo target 3, and combining pose calculation and coordinate transformation, the organic integration of visual measurement and contact measurement is achieved.

[0078] 5. By using a parallel robot 1 to drive probe 4 for multi-point contact data acquisition and combining it with geometric fitting, automated and high-precision measurement of part dimensions was achieved.

[0079] In summary, this invention effectively solves the technical problems in the prior art, such as poor stability of target pose calculation, asynchronous contact time and image acquisition time, and insufficient trigger accuracy and reset consistency. It significantly improves the measurement accuracy, stability and automation level of the vision measurement system, and has high practical value and broad application prospects.

[0080] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A vision measurement device based on a parallel robot, characterized in that, include: Parallel robots; An industrial camera, positioned on one side of the measurement area, is used to acquire feature point images of the three-dimensional target; A three-dimensional target is provided with multiple actively emitting feature points, which are distributed non-coplanarly in space to provide spatial geometric constraints for the three-dimensional pose calculation of the three-dimensional target after the industrial camera acquires images. The probe is rigidly connected to the three-dimensional target and is used to contact the surface of the part being tested. A contact detection component is used to detect the contact state between the probe and the surface of the part being tested; as well as The processing module is electrically connected to both the industrial camera and the contact detection component. The industrial camera is used to acquire an image of the stereo target when the contact detection component detects effective contact, and then transmits the image to the processing module. The processing module is used to calculate the spatial pose of the stereo target based on the image of the stereo target, and obtain the three-dimensional coordinates of the probe sphere center in the industrial camera coordinate system based on the fixed geometric relationship between the stereo target and the probe. Then, the feature dimensions of the measured part are obtained by acquiring multiple measurement points and geometric fitting.

2. The vision measurement device based on a parallel robot as described in claim 1, characterized in that, The contact detection component includes a reset spring and a conductive sheet; when the probe comes into contact with the surface of the part being tested, the probe causes the stereo target to shift or change its orientation and compresses the reset spring, causing the conductive sheet to break the circuit and generate a trigger signal; the industrial camera acquires feature point images of the stereo target based on the trigger signal.

3. The vision measurement device based on a parallel robot as described in claim 2, characterized in that, The 3D target is connected to the probe via a target rod. A first trigger ring is fixedly connected to the outer wall of the target rod, and the upper end of the first trigger ring is connected to the bottom end of the reset spring. An open-ended shell is fitted around the target rod. The upper end of the shell is connected to a top cover, and the lower end is connected to a base. Both the top cover and the base are annular and have a gap between them and the target rod. The base is fixedly connected to the parallel robot. The reset spring is located inside the shell, and the top end of the reset spring is fixedly connected to the top cover. A second trigger ring is connected to the upper end of the base inside the shell. The conductive sheet is located between the first trigger ring and the second trigger ring.

4. The vision measurement device based on a parallel robot as described in claim 3, characterized in that, When the probe moves upward, the target rod drives the first trigger ring to compress the reset spring upward, causing the first trigger ring to separate from the second trigger ring, and the conductive sheet to break the circuit; when the probe detaches from the surface of the part being tested, the reset spring drives the first trigger ring to reset, causing the first trigger ring to re-contact with the second trigger ring, and the conductive sheet to conduct.

5. The vision measurement device based on a parallel robot as described in claim 3, characterized in that, The first trigger ring has meshing teeth along its circumference, which are used to mesh with the second trigger ring and to restrict the first trigger ring from rotating around its own axis.

6. The vision measurement device based on a parallel robot as described in claim 1, characterized in that, The active light-emitting feature points on the three-dimensional target are infrared light-emitting diodes, and there are seven active light-emitting feature points, which constitute a redundant feature point structure.

7. The vision measurement device based on a parallel robot as described in claim 6, characterized in that, Three of the active light-emitting feature points are arranged in a circular array near the center of the outer wall of the three-dimensional target. The outer wall of the three-dimensional target has multiple protrusions of different thicknesses near its edge. The other four active light-emitting feature points are located on the side of the multiple protrusions away from the three-dimensional target, so as to form a non-coplanar spatial distribution.

8. The vision measurement device based on a parallel robot as described in claim 1, characterized in that, The processing module is used to solve the rotation matrix and translation vector of the 3D target relative to the industrial camera coordinate system based on the center coordinates of the active luminous feature points in the image coordinate system and the spatial coordinates of the active luminous feature points in the 3D target coordinate system; and to calculate the current position of the probe sphere center in the industrial camera coordinate system based on the pre-calibrated position of the probe sphere center in the 3D target coordinate system and the pose relationship of the 3D target relative to the industrial camera coordinate system.

9. The vision measurement device based on a parallel robot as described in claim 1, characterized in that, The probe is a ruby ​​ball probe, and during the contact measurement process, the coordinates of the center of the ruby ​​ball probe are used as the coordinates of the measurement point.

10. A visual measurement method based on parallel robots, employing the visual measurement device based on parallel robots as described in any one of claims 1-9, characterized in that, Includes the following steps: Establish a three-dimensional target coordinate system and obtain the spatial coordinates of each actively emitting feature point on the three-dimensional target in the three-dimensional target coordinate system; The fixed geometric relationship between the probe and the three-dimensional target is calibrated to obtain the position of the probe's center of sphere in the coordinate system of the three-dimensional target. The parallel robot drives the end effector to move, so that the probe contacts the surface of the part being tested; When the probe makes effective contact with the surface of the part being tested, the contact detection component generates a trigger signal and triggers the industrial camera to acquire feature point images of the stereo target; The industrial camera transmits the acquired feature point images to the processing module, which processes the feature point images and extracts the center coordinates of each actively emitting feature point. The processing module calculates the spatial pose of the 3D target relative to the industrial camera coordinate system based on the center coordinates of each active emitting feature point and the spatial coordinates of each active emitting feature point in the coordinate system of the 3D target. The processing module calculates the spatial coordinates of the probe ball center in the industrial camera coordinate system based on the position of the probe ball center in the 3D target coordinate system and the spatial pose of the 3D target relative to the industrial camera coordinate system. Repeat the contact acquisition process to obtain the probe sphere center coordinates corresponding to multiple contact measurement points, and perform geometric fitting based on the multiple probe sphere center coordinates to obtain the feature dimensions of the measured part.

Images