A method and system for calibrating degrees of freedom of a multi-group line structured light probe

Abstract

This invention discloses a method and system for calibrating the degrees of freedom of multiple sets of line structured optical probes, belonging to the field of precision testing technology and instruments. The method includes: constructing a digital three-dimensional model of a calibration element, the surface of which contains at least three different types of preset geometric features; scanning the calibration element under different poses using at least two sets of line structured optical probes to be calibrated to obtain measured point clouds; based on the digital three-dimensional model and the measured point clouds, iteratively adjusting the degree-of-freedom parameters of each set of line structured optical probes through an optimization algorithm until the deviation between the geometric features extracted from the measured point clouds and the corresponding preset geometric features in the digital three-dimensional model meets preset conditions, thereby determining the optimal degree-of-freedom parameters of each set of line structured optical probes and the relative pose relationship between the probes. This method offers fast calibration efficiency, high accuracy, and eliminates the need for calibration of the probe's intrinsic parameters, greatly improving its practicality.

Description

Technical Field

[0001] This invention relates to the field of precision testing technology and instruments, and more specifically to a method and system for calibrating the degrees of freedom of a multi-set line structure optical probe. Background Technology

[0002] With the advancement of sensing, computer, and information technologies, digital measurement has become a crucial step in processes such as component dimensional inspection and defect detection, and reverse engineering of products. Line structured light, as a type of active laser scanning measurement, features high efficiency, high repeatability, and high precision, and is increasingly being used for the digital measurement of high-precision curved surface components such as gears and aerospace blades.

[0003] Line structured light, based on the principle of laser triangulation, obtains the digital coordinates of the projection points or lines on the measured part. In digital measurement systems, there are typically two methods for acquiring the complex shape of the measured part: the moving probe method and the moving workpiece method. The moving probe method relies on precisely moving the line structured light probe to a circumferential position on the part to acquire its complex contour. The moving workpiece method relies on the workpiece moving around the line structured light probe in some way to acquire its complex contour shape.

[0004] The mobile workpiece type patent CN202110508393.7 allows a fixed sensor to observe from different angles by moving a simple cylinder with known geometric dimensions. By utilizing the geometric constraints of the cylinder, the three-dimensional coordinates of a large number of points on the laser plane can be deduced, thereby performing optical plane calibration. It has the advantages of simple calibration targets, low cost, easy availability, high flexibility, and suitability for narrow spaces. However, it has the disadvantages of low automation and low calibration efficiency. The calibration process requires manual placement of the cylinder and taking pictures multiple times, making it difficult to integrate into a fully automated calibration process. It is not conducive to the rapid and repeated calibration of sensors on the production line. The patent CN201910122872.8 for a moving probe proposes a method that uses a standard cylinder as a calibration component to accurately determine the position and angle of a line structured light sensor in space by analyzing the elliptical contour formed by the intersection of the laser plane and the cylinder. This method has the advantages of high precision, high reliability, and easy availability of calibration components. However, it has the disadvantage of strong dependence on the initial position. The convergence of the algorithm depends on algebraic fitting to provide a sufficiently good initial value. If the initial position deviation of the sensor is large, it may lead to incomplete or poor-quality elliptical arc segments, which will affect the initial fitting effect and cause iteration failure. Summary of the Invention

[0005] To address the above shortcomings, this invention proposes a method and system for calibrating the degrees of freedom of multiple line structured light probes. This method is based on acquiring complete point cloud data of the gear surface using line structured light probes. This method does not require the calibration elements to be placed in specific positions. The initial solution is obtained by matching the constructed digital three-dimensional model with the measured point cloud model. The calibration is efficient and accurate, and there is no need to calibrate the probe's intrinsic parameters, which greatly improves the practicality of the method.

[0006] To achieve the above objectives, the present invention adopts the following technical solution:

[0007] In a first aspect, embodiments of the present invention provide a method for calibrating the degrees of freedom of multiple sets of line structured optical probes, comprising the following steps:

[0008] S1: Construct a digital three-dimensional model of the calibration element, wherein the surface of the calibration element contains at least three different types of preset geometric features;

[0009] S2: The calibration element is scanned at different poses by at least two sets of line structured light probes to be calibrated to obtain the measured point cloud of the calibration element;

[0010] S3: Based on the digital 3D model and the measured point cloud, the degree of freedom parameters of each group of line structured light probes are iteratively adjusted through optimization algorithms until the deviation between the geometric features extracted from the measured point cloud and the corresponding preset geometric features in the digital 3D model meets the preset conditions, thereby determining the optimal degree of freedom parameters of each group of line structured light probes and the relative pose relationship between the probes.

[0011] In one embodiment, the calibration element includes a circumferential portion and an axial portion. The circumferential portion has a planar surface, an outer cylindrical surface, and an arcuate groove. The axial portion has a through-hole for clamping. The planar surface, the outer cylindrical surface, and the arcuate groove are all formed by precision grinding and polishing.

[0012] In one embodiment, step S1, constructing the digital three-dimensional model of the calibration element specifically includes: generating a three-dimensional solid model of the calibration element based on its design dimensions and processing accuracy, and converting it into a digital point cloud model with a point cloud density higher than that of the line structured light probe sampling density.

[0013] In one embodiment, step S2 specifically includes:

[0014] S2.1: Fix the calibration element on the rotating platform;

[0015] S2.2: Drive the calibration element to rotate around the axis, and synchronously collect its two-dimensional contour data at different rotation angles through at least two sets of line structure optical probes;

[0016] S2.3: Based on the initial degree of freedom parameters and rotation angle information of the line structured optical probe, the collected two-dimensional contour data is converted to the coordinate system of the calibration element to form the measured three-dimensional point cloud of the calibration element.

[0017] In one embodiment, step S3 specifically includes:

[0018] S3.1: Extract feature point cloud data corresponding to the preset geometric features from the measured point cloud;

[0019] S3.2: Fit the feature point cloud data to obtain planar feature parameters, cylindrical surface feature parameters, and arc groove feature parameters, respectively;

[0020] S3.3: Calculate the deviation between the fitted feature parameters and the actual parameters of the corresponding preset geometric features in the digital 3D model;

[0021] S3.4: With the goal of minimizing the overall deviation, the parameters in the homogeneous transformation matrix used to describe the probe pose are adjusted using an optimization algorithm. The parameters include translational degree of freedom parameters and at least one rotational degree of freedom parameter for each probe in the three coordinate axes.

[0022] S3.5: Repeat steps S2 to S3.4 until the overall deviation meets the preset threshold, and output the optimal parameters of the degree of freedom of each group of line structured light probes.

[0023] In one embodiment, the optimization algorithm employs the least squares method, and the overall deviation is the sum of squares of various characteristic deviations, including the sum of squares of planar deviation S1, the sum of squares of outer cylindrical surface deviation S2, and the sum of squares of arc groove deviation S3.

[0024] In one embodiment, step S3 further includes:

[0025] S3.6: Steps for determining the phase angle between two sets of line structured optical probes:

[0026] Obtain the expression of the normal vector of the same plane feature on the calibration element measured by each probe in the coordinate system of the calibration element;

[0027] Calculate the angle between the two normal vectors, which is taken as the phase angle between the two sets of line structured light probes.

[0028] In a second aspect, embodiments of the present invention provide a calibration system for a degree-of-freedom calibration method for implementing any one of the first aspects of a multi-set line structured optical probe, comprising:

[0029] At least two sets of line structured optical probes;

[0030] A rotary platform for mounting calibration elements, wherein the surface of the calibration elements comprises at least three different types of preset geometric features;

[0031] The drive motor and angle measuring device are connected to the rotating platform;

[0032] The data processing unit is configured as follows:

[0033] Store the digital three-dimensional model of the calibration element;

[0034] The drive motor is controlled to rotate, and the data collected by the line structure optical probe and the angle information of the angle measuring device are received to generate the measured point cloud of the calibration element.

[0035] Based on the digital 3D model and the measured point cloud, the degree-of-freedom parameters of each group of line structured light probes are iteratively adjusted through an optimization algorithm until the deviation between the geometric features extracted from the measured point cloud and the corresponding preset geometric features in the digital 3D model meets the preset conditions, thereby determining the optimal degree-of-freedom parameters of each group of line structured light probes and the relative pose relationship between the probes.

[0036] As can be seen from the above technical solution, compared with the prior art, the present invention has the following technical advantages:

[0037] 1. This invention improves the integrity of the point cloud of the calibration element under multiple sets of line structured light fields by designing a calibration element that combines plane, cylinder and arc groove.

[0038] 2. This invention proposes to match the geometric features of a pre-constructed digital 3D model with the measured geometric features of the calibration element to obtain an initial solution, thereby accelerating the convergence speed of the iteration. This not only improves the calibration efficiency of a single probe, but also improves the calibration efficiency of a measurement system containing multiple sets of sensing probes.

[0039] 3. This invention utilizes the measured point cloud of the calibration element and the digital three-dimensional model to construct an objective function for optimizing multiple sets of line structured light probe degree-of-freedom parameters. This objective function can be iteratively solved for the accurate line structured light probe degree-of-freedom parameters through an optimization algorithm.

[0040] 4. This invention can calibrate nine degrees of freedom of two sets of linear structure optical probes (A and B) when the probe intrinsic parameters are unknown and the calibration elements do not need to be in a specific initial position, which greatly improves the practicality of measurement systems containing multiple sets of sensing probes. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0042] Figure 1 This is a flowchart of the degree-of-freedom calibration method for the multi-set line structured optical probe of the present invention.

[0043] Figure 2 This is a flowchart illustrating the specific implementation process of the calibration method of the present invention.

[0044] Figure 3 This is a schematic diagram of the system device for calibrating the degrees of freedom of multiple sets of line structure optical probes according to the present invention.

[0045] Figure 4 This is a schematic diagram of the surface features of calibration element 5.

[0046] Figure 5 This is a schematic diagram of the digital calibration model constructed using calibration element 5.

[0047] Figure 6 This is a schematic diagram showing the relative pose relationship between the probe coordinate system and the calibration element 5 coordinate system in a multi-set line structured light probe degree-of-freedom calibration system.

[0048] Figure 7 This is a schematic diagram of the present invention for collecting feature point clouds of planes and arc grooves using the A-group line structured light probe 1.

[0049] Figure 8 This is a schematic diagram of the present invention for collecting feature point clouds of an outer cylindrical surface using a B-group line structured light probe 2.

[0050] Figure 9 The present invention uses a B-group line structured optical probe 2 along... A schematic diagram of the acquisition of surface feature point cloud of the calibration element in the plane direction.

[0051] Figure 10 The feature point cloud model diagram is obtained by using calibration element 5.

[0052] The labels in the diagram are as follows: 1-A group line structured light probe, 2-B group line structured light probe, 3-rotary turntable, 4-circular grating, 5-calibration element, 7-A group probe X-axis, 8-A group probe Y-axis, 9-A group probe Z-axis, 10-B group probe X-axis, 11-B group probe Y-axis, 12-B group probe Z-axis, 13-motor. Detailed Implementation

[0053] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0054] This invention provides a method for calibrating the degrees of freedom of a multi-set line structured light probe. By constructing a digital three-dimensional point cloud model of the calibration element and comparing the measured geometric features acquired by the multi-set line structured light, and by combining iterative algorithms to optimize the probe's degree of freedom parameters, the calibration time for the multi-set line structured light is reduced, and the accuracy of digital measurement of the measured part under the measurement field of view is improved.

[0055] The core idea of ​​this line structured light probe degree-of-freedom calibration method is to use a digitally generated virtual calibration model as a benchmark, and on the basis of the initial degree-of-freedom model of each group of probes, reversely adjust the degree-of-freedom parameters of each group of line structured light probes so that the point cloud of the measured calibration element coincides with the virtual calibration model. The degree-of-freedom calibration is considered complete when the feature distance fully meets the optimization index of least squares fitting.

[0056] Reference Figure 1 As shown, the present invention provides a method for calibrating the degrees of freedom of a multi-set line structured optical probe, comprising the following steps S1~S3:

[0057] S1: Construct a digital 3D model of the calibration element, the surface of which contains at least three different types of preset geometric features. The calibration element includes a circumferential portion and an axial portion. The circumferential portion has a planar surface, an outer cylindrical surface, and an arc-shaped groove. The axial portion has a through-hole for clamping. The planar surface, outer cylindrical surface, and arc-shaped groove are all machined through precision grinding and polishing. Based on the design dimensions and machining accuracy of the calibration element, generate its 3D solid model and convert it into a digital point cloud model with a point cloud density higher than the sampling density of the line structured light probe.

[0058] S2: The calibration element is scanned at different poses using at least two sets of line structured light probes to be calibrated, to obtain the measured point cloud of the calibration element. This step specifically includes:

[0059] S2.1: Fix the calibration element on the rotating platform;

[0060] S2.2: Drive the calibration element to rotate around the axis, and synchronously collect its two-dimensional contour data at different rotation angles through at least two sets of line structure optical probes;

[0061] S2.3: Based on the initial degree of freedom parameters and rotation angle information of the line structured optical probe, the collected two-dimensional contour data is converted to the coordinate system of the calibration element to form the measured three-dimensional point cloud of the calibration element.

[0062] S3: Based on the digital 3D model and the measured point cloud, the degree of freedom parameters of each group of line structured light probes are iteratively adjusted through optimization algorithms until the deviation between the geometric features extracted from the measured point cloud and the corresponding preset geometric features in the digital 3D model meets the preset conditions, thereby determining the optimal degree of freedom parameters of each group of line structured light probes and the relative pose relationship between the probes.

[0063] S3 specifically includes:

[0064] S3.1: Extract feature point cloud data corresponding to the preset geometric features from the measured point cloud;

[0065] S3.2: Fit the feature point cloud data to obtain planar feature parameters, cylindrical surface feature parameters, and arc groove feature parameters, respectively;

[0066] S3.3: Calculate the deviation between the fitted feature parameters and the actual parameters of the corresponding preset geometric features in the digital 3D model;

[0067] S3.4: With the goal of minimizing the overall deviation, the parameters in the homogeneous transformation matrix used to describe the probe pose are adjusted using an optimization algorithm. The parameters include translational degree of freedom parameters and at least one rotational degree of freedom parameter for each probe in the three coordinate axes.

[0068] S3.5: Repeat steps S2 to S3.4 until the overall deviation meets the preset threshold, and output the optimal parameters of the degree of freedom of each group of line structured light probes.

[0069] S3.6: Steps for determining the phase angle between two sets of line structured optical probes: Obtain the expression of the normal vector of the same plane feature on the calibration element measured by each probe in the coordinate system of the calibration element; calculate the angle between the two normal vectors as the phase angle between the two sets of line structured optical probes.

[0070] In step S3.4, the optimization algorithm uses the least squares method; and the overall deviation is the sum of squares of various characteristic deviations, including the sum of squares of plane deviation S1, the sum of squares of outer cylindrical surface deviation S2 and the sum of squares of arc groove deviation S3.

[0071] The method of the present invention will be further described in detail below:

[0072] Taking a calibration system containing two sets of line structured optical probes as an example, the overall calibration process is described in reference to... Figure 2 As shown.

[0073] First, such as Figure 3 As shown, the main components of the calibration system are: Group A line structured light probe 1, Group B line structured light probe 2, rotary stage 3, circular grating 4, calibration element 5, and Group A probe. Axis 7, Group A probe Axis 8, Group A probe Axis 9, Group B probe Shaft 10, Group B probe Shaft 11, Group B probe Shaft 12, Motor 13. This system uses two sets of line structured light probes, A and B, for calibration. The optical planes of the two sets of line structured light probes are perpendicular to each other. Line structured light probe 1 of set A measures the axial direction data, and line structured light probe 2 of set B measures... Planar direction data, the phase angle between the two sets of line structured light probes, A and B. It forms a 90° angle.

[0074] Among them, such as Figure 4 As shown, calibration element 5 is a precision part used for calibrating multiple sets of linear structure optical degrees of freedom. Calibration element 5 consists of a circumferential part and an axial part. The circumferential part consists of three features: a plane, an outer cylindrical surface, and an arc groove. The plane, outer cylindrical surface, and arc groove are all precision ground and then polished. The two end faces of the axial part are formed by high-precision grinding and are designed to be aligned with... A coaxial through-hole is used for clamping. For example... Figure 3 As shown, the coordinate system in which calibration element 5 is located is σ. s The origin is O s The axes are: X s Y s Z s .

[0075] like Figure 5 As shown, the digital 3D model in step S2 is a digital twin 3D model constructed based on the geometric features of calibration element 5. By comparing it with the measured point cloud features of calibration element 5, it is used to verify the accuracy of the calibration of the degrees of freedom of multiple sets of line structure optical probes.

[0076] The calibration of the degrees of freedom of a multi-group line structured light probe is achieved by determining the degree of freedom parameters between a single-group line structured light probe and multiple groups of line structured light probes. This requires iterative optimization using the least squares algorithm based on three categories of measured point cloud feature parameters to determine the degree of freedom parameters that meet the accuracy requirements of the probe.

[0077] Three main categories of measured point cloud feature parameters:

[0078] The first category is parameters of the plane characteristic equation. ;

[0079] The second major category is the coordinate system of the outer cylindrical surface axis. Unit direction vector of the axis The radius of the outer cylindrical surface ;

[0080] The third category is the radius of the arc groove. Unit direction vector of the axis Center point of the arc groove axis ;

[0081] like Figure 2 As shown, the method for calibrating the degrees of freedom of multiple line structured optical probes specifically includes:

[0082] W1: Initialize the measurement platform; specifically including sub-steps W1.1~W1.3:

[0083] W1.1: Using the axis of rotary table 3 as a reference line, along the B group probe... With axis 11 in the 0° direction and the probes arranged counterclockwise, motor 13 is in the 0° direction, B group line structure light probe 2 is in the 90° direction, and A group line structure light probe 1 is in the 180° direction.

[0084] W1.2: Initialize motor 13 and circular grating 4;

[0085] W1.3: Both groups A and B of the linear structured optical probes have... axis, axis, Translational degrees of freedom parameters in three directions along the axis and a along The angular degree of freedom parameter for axis rotation; and the phase angle degree of freedom parameter between the two sets of line structured light probes A and B. ;

[0086] W2: Constructing a digital calibration model is based on the actual outer cylindrical surface, arc groove, and planar feature parameters of calibration element 5 after high-precision manufacturing. This model generates a digital twin 3D point cloud model, and the point cloud density of this model needs to be at least twice the measured point cloud density of the line structured light probe. Specifically, this includes:

[0087] W2.1: Construct a 3D model based on the processed calibration element 5 and export it as a .STL format mesh file;

[0088] W2.2: Import the .STL format mesh file into CloudCompare software for resampling; adjust the point cloud density to twice the parameters of the line structured light probe to complete the construction of the digital calibration model;

[0089] W3: Obtain initial degrees of freedom parameters; fix calibration element 5 on the measurement platform, and take the surface features of calibration element 5 within the measurement fields of A and B of the two sets of line structured light probes as the target. Adjust each of the A and B sets of line structured light probes to obtain the initial degrees of freedom parameters of position and angle of the multi-set line structured light probe system; specifically including:

[0090] W3.1: Reference Figure 3 As shown, let the coordinate system of calibration element 5 be... for The coordinate system of the structured optical probe 1 in group A for The coordinate system of the B-group structured optical probe 2 for .like Figure 6 As shown, coordinate system Relative to the coordinate system of calibration element 5 The distance parameters are respectively , for Distance in direction for Distance in direction for Distance in direction, along The angular parameter of the direction is , The coordinate system representing the structured light probe 1 of group A. Relative to the calibrator 5 coordinate system Yaw angle; coordinate system Relative to the calibrator 5 coordinate system The distance parameters are respectively , for Distance in direction for Distance in direction for Distance in direction, along The angular parameter of the direction is , The coordinate system representing the B-group structured light probe 2 Relative to the calibrator 5 coordinate system Yaw angle;

[0091] W3.2: The calibration element 5 is clamped using a three-jaw chuck. The three jaws of the three-jaw chuck mechanism cooperate with the inner hole of the calibration element 5 to fix it on the rotary table 3. According to the height and orientation of the calibration element 5, the position and angle relationship between the two sets of line structure optical probes A and B and the calibration element 5 are adjusted to ensure that both sets of line structure optical probes A and B can measure all surface features of the calibration element 5.

[0092] W4: Obtain the measured surface point cloud features of calibration element 5; drive the rotary table 3 with motor 13, and use two sets of line structured light probes (A and B) to acquire the circumferential two-dimensional contour point cloud of calibration element 5 at various rotation angles. Each rotation angle is recorded in real-time by a circular grating 4. Based on the initial degree of freedom parameters and rotation angle information of the line structured light probe system, the acquired two-dimensional contour point cloud can be transformed into homogeneous transformation matrices to derive the measured surface point cloud models of calibration element 5 from the two sets of line structured light probes (A and B). Figure 7 The diagram shown is a schematic of collecting feature point clouds of a plane and an arc groove using the A-group line structured light probe 1. Figure 8 This is a schematic diagram of point cloud acquisition of features on an outer cylindrical surface using the B-group line structured light probe 2. Figure 9 To use the B-group line structured light probe 2 along A schematic diagram of the acquisition of surface feature point cloud of the calibration element in the plane direction.

[0093] like Figure 10 The diagram shown illustrates the obtained surface point cloud features of calibration element 5 as measured. Step W4 specifically includes sub-steps W4.1 to W4.2:

[0094] W4.1: In step W4, obtaining the surface data of calibration element 5 means obtaining data from two rings on the surface of calibration element 5 to ensure the validity of the data;

[0095] W4.2: Place calibration element 5 on rotary table 3, rotate rotary table 3 720°, and use two sets of line structured light probes, A and B, to scan calibration element 5 to obtain two-dimensional linear data of the contour scan; data acquisition is performed while the angular velocity of rotary table 3 remains constant; based on the acquired initial degree of freedom parameters of the probe system, the two sets of line structured light probes, A and B, derive the measured three-dimensional point cloud model of calibration element 5 in the calibration element coordinate system through homogeneous transformation matrices; coordinate system Relative to coordinate system Homogeneous transformation matrix:

[0096] (1)

[0097] In formula (1) The physical meaning is the same as in W3.1. The physical meaning is the same, and similarly, in formula (2) In formulas (3) and (4) The physical meaning is the same as in W3.1. , The physical meanings are the same;

[0098] coordinate system Point coordinates and coordinate system The point coordinate transformation relationship can be expressed as:

[0099] (2)

[0100] In formula (2), x A y A , z A : Represents the three-dimensional coordinates of the point in the coordinate system of probe group A, and represents the measured point cloud data. x S y S , z S : Representing points (x) A y A , z A The three-dimensional coordinates in the calibration element coordinate system represent the transformed unified coordinates.

[0101] Similarly, coordinate system Relative to coordinate system Homogeneous transformation matrix:

[0102] (3)

[0103] coordinate system Point coordinates and coordinate system The coordinate transformation relationship of a point can be expressed as:

[0104] (4)

[0105] In formula (4), x B y B , z B : Represents the three-dimensional coordinates of the point in the B-group probe coordinate system, and represents the measured point cloud data. x S y S , z S : Representing points (x) B y B , z B The three-dimensional coordinates in the calibration element coordinate system represent the transformed unified coordinates.

[0106] W5: Extract the feature point cloud of calibration element 5 based on the initial solution to provide a basis for subsequent feature deviation calculation; iteratively optimize the homogeneous transformation matrix parameters to make the generated point cloud features consistent with the true value of the digital calibration model. If inconsistent, return to W3 to restart the calibration until it is consistent with the true value of the digital calibration model, completing the degree of freedom calibration of the two sets of line structured light probes A and B; specifically including sub-steps W5.1~W5.6:

[0107] W5.1: Based on planar features, outer cylindrical surface features, and arc groove features, the actual point cloud model collected by the A-group line structured light probe 1 is matched with the digital calibration model to obtain the initial solution of the A-group line structured light probe 1. Data points related to each feature of the calibration element 5 are selected from the region where the features are located. At the same time, the edge transition region is cut off to avoid selecting data points in the transition region. The selected features are planar features, outer cylindrical surface features, and arc groove features.

[0108] W5.2: Fitting the first major category of feature parameters, constructing a mathematical model for the feature plane of the measured point cloud of calibration element 5:

[0109] (5)

[0110] Using the mathematical model of formula (5) as the fitting model, the least squares method is used for fitting optimization. The meanings of x, y, and z are: representing the three-dimensional coordinate values ​​of each data point in the measured point cloud under the coordinate system of the calibration element. The direction coefficient of the plane normal vector. Represented as the directed distance from the origin of the coordinate system of calibration element 5 to the characteristic plane of the measured point cloud. Multiply by the magnitude of the normal vector;

[0111] (6)

[0112] The directed distance from the valid data points to the fitting plane Sum of squares Minimize the objective function:

[0113] (7)

[0114] (8)

[0115] In formula (7), x i y i , z i : These represent the coordinates of the i-th measured point cloud data in the coordinate system of the calibration element. In formula (8), n represents the total number of point cloud data points. The plane equation parameters are adjusted through iterative optimization, and the sum of squared deviations is calculated. until the sum of squared deviations of two consecutive iterations. Less than a preset threshold (e.g., 10) -4 mm), to obtain the optimal parameters This completes the local planar feature fitting optimization.

[0116] W5.3: Fitting and optimizing the second major category of feature parameters, constructing a mathematical model for the outer cylindrical surface:

[0117] (9)

[0118] The mathematical model of formula (9) is used as the fitting model, where (x,y,z) represents the three-dimensional coordinates of the measured point cloud in the coordinate system of the calibration element, but here it is specifically used to describe the position of the point on the outer cylindrical surface feature. Let be the coordinates of the axis of the outer cylindrical surface, and let the radius of the outer cylindrical surface be . Unit direction vector of the outer cylindrical surface axis The least squares method was used for fitting optimization, with the vertical deviation distance from the data point to the outer cylindrical surface as the criterion. sum of squares Minimize the objective function, vertical deviation distance This is the absolute value of the difference between the perpendicular distance from the data point to the axis center and the radius:

[0119] (10)

[0120] (11)

[0121] The parameters of the outer cylindrical surface equation are adjusted using a numerical iterative algorithm. In each iteration, the sum of squares of the vertical deviation distances from all data points to the outer cylindrical surface under the current parameters is calculated, until the sum of squares of the deviations from two consecutive iterations is obtained. If the coordinates are less than a preset threshold, the cylinder axis coordinates are obtained. outer cylindrical surface radius Unit direction vector of the axis The optimal parameters are used to complete the fitting and optimization of the local outer cylindrical surface features;

[0122] W5.4: Fitting and optimizing the third category of feature parameters to construct a mathematical model for the arc-shaped groove:

[0123] (12)

[0124] Using the mathematical model of formula (12) as the fitting model, the radius of the arc groove is The unit direction vector of the arc groove axis Center point of the arc groove axis The least squares method was used for fitting optimization, with the deviation distance from the arc-shaped groove point cloud to the fitted cylindrical surface as the metric. sum of squares Minimum:

[0125] (13)

[0126] (14)

[0127] The parameters of the arc-shaped groove mathematical model are adjusted using a numerical iterative algorithm. In each iteration, the sum of squares of the distance deviations from all data points to the fitted cylindrical surface under the current parameters is calculated, until the sum of squares of the deviations from two consecutive iterations is obtained. If the value is less than a preset threshold, the unit direction vector of the arc groove axis is obtained. center point and the radius parameter of the arc groove The optimal parameters are used to complete the fitting and optimization of the local feature arc groove.

[0128] W5.5: The sum of squares of the overall deviations between the fitted results of all features and the true values ​​of the digitally calibrated model. Global fitting optimization is performed with the goal of minimizing:

[0129] (15)

[0130] The least squares method is used to fit and optimize formula (15). If the sum of squared deviations of the whole is... If the value exceeds a preset threshold, a numerical optimization algorithm is used to iteratively adjust the homogeneous transformation matrix. axis, axis, Translational degrees of freedom parameters in the three axes , and along angular degree of freedom parameters of axis rotation The surface data of calibration element 5 is remeasured and a new feature point cloud model is generated. Then, the local fitting steps from W5.1 to W5.4 are repeated and the sum of squared deviations between all features and the true values ​​of the digital calibration model is recalculated until the sum of squared global deviations is reached after two consecutive iterations. Within the preset threshold; the degree of freedom parameters obtained at this time are the optimal degree of freedom parameters of the A-group line structured light probe 1 relative to the calibration element 5, thus completing the calibration of the degree of freedom parameters of the A-group line structured light probe 1;

[0131] W5.6: The calibration of the B group line structured light probe 2 is the same as that of the A group line structured light probe 1, and is performed according to steps W5.1 to W5.5.

[0132] W6: Utilizing the geometric constraints between the true normal vector of calibration element 5 and the measured fitted normal vector of the probe, and combining it with circular grating 4, construct the phase angle between the normal vector and the two sets of linear structured optical probes A and B. The direct mapping relationship is used to obtain the phase angle between the two sets of line structured light probes A and B through the normal vector. Specifically, this includes sub-steps W6.1 to W6.3:

[0133] W6.1: The normal vector obtained by measuring the planar geometric features using the A-group linear structured light probe 1 is... , The normal vectors of the planar geometric features in the coordinate system of calibration element 5 are obtained through the homogeneous transformation matrix. The transformation relationship is as follows:

[0134] (16)

[0135] W6.2: The normal vector obtained by measuring the planar geometric features using the B-group line structured light probe 2 is... , The normal vectors of the planar geometric features in the coordinate system of calibration element 5 are obtained through the homogeneous transformation matrix. The transformation relationship is as follows:

[0136] (17)

[0137] W6.3: The phase angle between the two sets of line structured optical probes, A and B. :

[0138] (18)

[0139] The calibration of the pose degree of freedom parameters of the two line structured light probes A and B is completed, as is the calibration of the phase angle degree of freedom parameter between the probes. This means that the calibration of the degree of freedom parameters of multiple sets of line structured light probes is completed.

[0140] W7: Outputs the optimal 1-DOF parameters of the A-group line structured light probe. And the 2-DOF parameters of the B-group line structured light probe And the phase angle between the two sets of line structured optical probes, A and B. .

[0141] Based on the same inventive concept, embodiments of the present invention also provide a degree-of-freedom calibration system for multiple sets of line structured optical probes, such as... Figure 3 As shown, it includes:

[0142] At least two sets of line structured optical probes;

[0143] A rotary platform for mounting calibration elements, wherein the surface of the calibration elements comprises at least three different types of preset geometric features;

[0144] The drive motor and angle measuring device are connected to the rotating platform;

[0145] The data processing unit is configured as follows:

[0146] 1) Store the digital three-dimensional model of the calibration element;

[0147] 2) Control the rotation of the drive motor, and receive the data collected by the line structure optical probe and the angle information of the angle measuring device to generate the measured point cloud of the calibration element;

[0148] 3) Based on the digital 3D model and the measured point cloud, the degree of freedom parameters of each group of line structured light probes are iteratively adjusted through optimization algorithms until the deviation between the geometric features extracted from the measured point cloud and the corresponding preset geometric features in the digital 3D model meets the preset conditions, thereby determining the optimal degree of freedom parameters of each group of line structured light probes and the relative pose relationship between the probes.

[0149] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0150] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for calibrating the degrees of freedom of a multi-set line structured optical probe, characterized in that, Includes the following steps: S1: Construct a digital three-dimensional model of the calibration element, wherein the surface of the calibration element contains at least three different types of preset geometric features; S2: The calibration element is scanned at different poses by at least two sets of line structured light probes to be calibrated to obtain the measured point cloud of the calibration element; S3: Based on the digital 3D model and the measured point cloud, the degree of freedom parameters of each group of line structured light probes are iteratively adjusted through optimization algorithms until the deviation between the geometric features extracted from the measured point cloud and the corresponding preset geometric features in the digital 3D model meets the preset conditions, thereby determining the optimal degree of freedom parameters of each group of line structured light probes and the relative pose relationship between the probes. Specifically, step S3 includes: S3.1: Extract feature point cloud data corresponding to the preset geometric features from the measured point cloud; S3.2: Fit the feature point cloud data to obtain planar feature parameters, cylindrical surface feature parameters, and arc groove feature parameters, respectively; S3.3: Calculate the deviation between the fitted feature parameters and the actual parameters of the corresponding preset geometric features in the digital 3D model; S3.4: With the goal of minimizing the overall deviation, the parameters in the homogeneous transformation matrix used to describe the probe pose are adjusted using an optimization algorithm. The parameters include translational degree of freedom parameters and at least one rotational degree of freedom parameter for each probe in the three coordinate axes. S3.5: Repeat steps S2 to S3.4 until the overall deviation meets the preset threshold, and output the optimal parameters of the degree of freedom of each group of line structure optical probes. S3.6: Steps for determining the phase angle between two sets of line structured light probes: Obtain the expression of the normal vector of the same plane feature on the calibration element measured by each probe in the coordinate system of the calibration element; Calculate the angle between the two normal vectors, which is taken as the phase angle between the two sets of line structured light probes.

2. The method according to claim 1, characterized in that, The calibration element includes a circumferential portion and an axial portion. The circumferential portion has a planar surface, an outer cylindrical surface, and an arc-shaped groove. The axial portion has a through-hole for clamping. The planar surface, outer cylindrical surface, and arc-shaped groove are all formed by precision grinding and polishing.

3. The method according to claim 1, characterized in that, In step S1, constructing the digital three-dimensional model of the calibration element specifically includes: generating a three-dimensional solid model of the calibration element based on its design dimensions and processing accuracy, and converting it into a digital point cloud model with a point cloud density higher than that of the line structured light probe sampling density.

4. The method according to claim 1, characterized in that, Step S2 specifically includes: S2.1: Fix the calibration element on the rotating platform; S2.2: Drive the calibration element to rotate around the axis, and synchronously collect its two-dimensional contour data at different rotation angles through at least two sets of line structure optical probes; S2.3: Based on the initial degree of freedom parameters and rotation angle information of the line structured optical probe, the collected two-dimensional contour data is converted to the coordinate system of the calibration element to form the measured three-dimensional point cloud of the calibration element.

5. The method according to claim 1, characterized in that, The optimization algorithm uses the least squares method, and the overall deviation is the sum of squares of the deviations of various features.

6. A calibration system for implementing the method according to any one of claims 1-5, characterized in that, include: At least two sets of line structured optical probes; A rotary platform for mounting calibration elements, wherein the surface of the calibration elements comprises at least three different types of preset geometric features; The drive motor and angle measuring device are connected to the rotating platform; The data processing unit is configured as follows: Store the digital three-dimensional model of the calibration element; The drive motor is controlled to rotate, and the data collected by the line structure optical probe and the angle information of the angle measuring device are received to generate the measured point cloud of the calibration element. Based on the digital 3D model and the measured point cloud, the degree-of-freedom parameters of each group of line structured light probes are iteratively adjusted through an optimization algorithm until the deviation between the geometric features extracted from the measured point cloud and the corresponding preset geometric features in the digital 3D model meets the preset conditions, thereby determining the optimal degree-of-freedom parameters of each group of line structured light probes and the relative pose relationship between the probes.

Images