A method for detecting the tangent line of a curved surface

Abstract

The present invention relates to the field of numerical control machine tool detection technology, and discloses a method for detecting the tangent line of a curved surface. A probe on a numerical control machine tool is used to detect a target curve, and the tangent line position of the target curve perpendicular to the detection direction in a specified plane is obtained. It includes parameter determination: the target curve is segmented according to the segmentation method, the radius of the detected curve segment is estimated, and detection parameters are determined according to the segmented curve; detection point acquisition: according to the detection parameters, the detected curve segments are respectively detected to obtain multiple groups of detection points; tangent point selection: the detection points are projected onto the detection normal direction, and the projection maximum value point #imgabs0# is calculated; tangent line calculation: according to the tangent point #imgabs1# and the detection normal direction, the detected tangent line position is determined. The present invention can realize the detection of the tangent line of a curved surface on an ordinary machine tool, can adapt to probes with any precision, can eliminate random errors, improve the detection accuracy, improve the processing efficiency and product quality, and reduce the processing difficulty and cost.

Description

Technical Field

[0001] The present invention relates to the technical field of grinding processes for numerically controlled machine tools, and particularly to a method for detecting the tangent line of a curved surface. Background Art

[0002] In modern manufacturing, the detection technology of numerically controlled machine tools plays an important role in improving machining accuracy and efficiency. Through precise detection, key elements such as the position, size, and shape of the workpiece can be accurately grasped before machining, providing a strong guarantee for subsequent high-quality machining operations. At the same time, real-time detection during machining can also help reduce debugging time and adjust machining parameters in a timely manner according to the detection feedback, further improving the overall machining efficiency.

[0003] However, in the existing technology, most ordinary numerically controlled machine tools cannot directly use a probe as a measuring tool to detect the tangent line of a curved surface. In some machining situations where the tangent line at the machining location of the workpiece needs to be known, the workpiece needs to be removed and replaced on other detection equipment for detection. This off-line measurement method not only makes the machining process more cumbersome and reduces machining efficiency, but more seriously, each re-clamping will introduce new positioning errors and other errors in subsequent machining, resulting in a gradual decrease in machining accuracy. Even if a high-precision coordinate measuring machine is currently used for detection, its measurement results are difficult to directly feedback to the machining system to form a closed-loop control.

[0004] In addition, in most cases, the existing detection technology can only detect one end face and is difficult to detect the tangent line of a complex curved surface. Even if some current high-end measurement systems have certain curved surface measurement capabilities, they have strict requirements for the use environment and require special probes and complex compensation algorithms. More importantly, in the actual detection process, due to the influence of various uncontrollable factors, the probe often generates random errors, and the existing detection technology cannot effectively eliminate these random errors automatically. This leads to the accuracy of the detection results being affected, and further may make the machined workpiece difficult to meet the expected accuracy standards, seriously restricting the improvement of machining quality. Especially in some machining scenarios with extremely high precision requirements for parts, the impact of such errors is more prominent. Summary of the Invention

[0005] The present invention aims to provide a method for detecting the tangent line of a curved surface to solve the problem that existing ordinary numerically controlled machine tools cannot directly use a probe as a measuring tool to detect the tangent line of a curved surface, resulting in a cumbersome and low-precision existing curve tangent line detection process, and prone to errors affecting machining accuracy.

[0006] To achieve the above object, the present invention adopts the following technical solution. A method for detecting the tangent line of a curved surface uses a probe on a numerically controlled machine tool to detect a target curve and obtain the tangent line position of the target curve perpendicular to the detection direction in a specified plane. The method includes the following steps.

[0007] Parameter determination: segment the target curve according to the division method, estimate the radius R of the detected curve segment according to the segmented curve segments, and determine the detection parameters according to the segmented curve;

[0008] Detection point acquisition: According to the determined detection parameters, the detected curve segments are detected according to the corresponding detection methods to obtain multiple groups of detection points ;

[0009] Tangent point selection: Project it onto the detection normal n and calculate the maximum projection point As a cut point;

[0010] Tangent calculation: Based on the obtained tangent point and the detection normal n, determine the tangent position of the detection.

[0011] The principles and advantages of this solution are:

[0012] This solution can be used on ordinary machine tools to directly detect the curve where the target surface intersects the specified plane in the specified plane and along the specified direction through the probe built into the machine tool to obtain the tangent position of the curve perpendicular to the detection direction, thereby solving the problem that ordinary CNC machine tools cannot use probes as measurement tools to detect surface tangents. By dividing the complex curve into several segments, each segment is probed, and the detection normal n is known for each segment of the curve, a tangent perpendicular to the detection normal n is calculated in the specified plane. Then, by using the probe to detect the target curve, find the The tangent point is detected in the detection range one by one to obtain the contact point position information. The contact point position information is multiplied by the detection normal n, and the maximum value is selected, which is the tangent point information. Calculate the tangent information.

[0013] This solution breaks through the limitations of existing technologies that can only detect one end face and cannot detect the tangent of the surface, and can detect the tangent of the surface on ordinary machine tools. It is no longer limited to specific equipment such as three-coordinate measuring machines, and reduces the accuracy requirements of the probe, greatly expanding its application range in surface processing. It can achieve more accurate processing path planning, thereby improving processing efficiency and product quality, and effectively reducing the processing difficulty and cost caused by detection limitations. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 The figure is a flow chart of a curved surface tangent line detection method of the present invention.

[0015] Figure 2It is a schematic diagram of detection parameters for detecting curve segments of a surface tangent detection method of the present invention.

[0016] Figure 3 It is a structural schematic diagram of detecting deviation angle of a curved surface tangent detection method of the present invention.

[0017] Figure 4 It is a schematic diagram of the direction of the detection normal of a detection curve in two different forms of a surface tangent detection method of the present invention.

[0018] Figure 5 It is a schematic diagram of the detection step length of a surface tangent detection method of the present invention.

[0019] Figure 6 It is a schematic diagram of the detection process of Example 1 in a curved surface tangent detection method of the present invention. DETAILED DESCRIPTION

[0020] The following is further described in detail through specific implementation methods:

[0021] Example

[0022] A surface tangent detection method in this embodiment cleverly uses the principle of geometric projection to transform complex tangent detection into a mathematical problem of finding projection extreme points, which not only ensures accuracy but also realizes automated detection, thereby realizing a method in which the probe of an ordinary CNC machine tool can be directly used as a measuring tool to detect the surface tangent.

[0023] In the prior art, since the probe of an ordinary CNC machine tool is a spherical probe, the contact point measured is the coordinate of the probe's center of the sphere, not the actual contact point. When the curvature of the surface changes dramatically (such as a sharp edge or a steep slope), the offset between the contact point and the center of the sphere will increase sharply, which leads to measurement distortion, just like using a ping-pong ball to touch the tip of a needle, the position of the center of the sphere cannot reflect the true coordinates of the tip of the needle at all. At the same time, coupled with the fact that ordinary probes cannot adjust the angle adaptively, it is also difficult to compensate for the measurement distortion problem. In addition, the servo system of traditional machine tools is mainly designed for cutting. Its rigid structure and inertial mass make it difficult to achieve the micron-level high-frequency dynamic adjustment required by the probe, making its dynamic response insufficient and unable to handle multiple solutions or singular points. Therefore, in the existing detection operations, it is believed that a special detection instrument with higher accuracy, shorter measurement cycle, and better decision-making algorithm is needed to complete it, such as using a three-coordinate machine equipped with a special probe that can be actively deflected, and a built-in iterative detection protocol (such as ISO10360), which can automatically optimize the detection path and compensate for anisotropic errors. There is an inherent thinking that traditional machine tools cannot be directly used as measurement tools.

[0024] In the face of this problem, the existing technologies have also been attempting to adopt methods such as integrating a microscopic vision probe on the machine tool to assist in positioning and constructing a hybrid detection system; or establishing a probe-surface contact dynamics model to predict errors in real time and achieve digital twin compensation; or directly reading the contact point position through photon signals. However, all of the above methods require additional auxiliary equipment to further improve the performance of the existing machine tool, which undoubtedly increases the manufacturing cost and difficulty of the machine tool, and it is also difficult to ensure that the measurement accuracy and quality of dedicated equipment can be achieved.

[0025] However, all of the above methods integrate new technologies into the existing machine tool system to ensure the accuracy and feasibility of detection. In contrast, this solution breaks out of the cognitive shackles of traditional machine tools and breaks the original detection thinking that complex curved surface tangents cannot be directly detected using traditional machine tool probes. Without adding additional technologies or equipment, a new tangent detection method is creatively proposed, which converts complex tangent detection into the calculation of projection extreme points, greatly reducing the dependence on algorithms, ensuring the detection accuracy, realizing the automatic detection method using a probe on a traditional machine tool, reducing the complexity of the detection operation process, reducing the errors caused by repeated clamping, thereby improving the accuracy, and directly feeding the measurement results back into the processing system to form a closed-loop control.

[0026] In this embodiment, the curved surface tangent detection method is as shown in the appendix Figure 1 and obtains the intersection curve of the specified plane and the target curved surface by using a contact probe on a numerically controlled machine tool to detect the target curved surface according to fixed steps, and obtains the tangent position perpendicular to the detection direction on the intersection curve, including the following steps:

[0027] S1, segment the target curve according to the segmentation method, estimate the radius of the curve segment to be detected according to the segmented curve segments, and determine the detection parameters according to the segmented curves.

[0028] In this embodiment, the target curve is the intersection curve of the specified plane and the target curved surface. The segmentation method is to project the complex target curve onto the detection normal n, and segment the detection curve according to the shape of the projected detection curve, dividing it into K curve segments with at most one radian, K≥1; that is, each curve segment contains at most one convex segment or concave segment, so that there is only one highest point or lowest point, that is, an extreme point, on each projected curve segment to ensure the uniqueness of the tangent of each curve segment. After segmentation, estimate the radius of the segmented curve segments to obtain the local radius R of the curve to be detected.

[0029] For each segmented curve segment, determine the detection parameters of the curve segment respectively. In this embodiment, the detection parameters include the detection contact surface, the detection direction m, the detection normal n, the tangent direction (i.e., the tangential direction) , the radius r of the probe ball head, and the detection step size S.

[0030] Combined with the attached Figure 2 Shown is a certain section of the detection curve segment. Among them, the detection direction m is the direction in which the probe is perpendicular to the target surface, generally the direction of the vertical movement of the probe. The detection normal n is parallel to the detection direction m. According to the known detection normal n, the tangential can be calculated, where the tangential is a tangential direction perpendicular to the detection normal n in the specified plane . In this embodiment, the geometric algorithm tangential is obtained by rotating the detection normal n by +90° or -90° in the specified plane.

[0031] The detection contact surface is the surface where the probe contacts the target surface for detection. In this embodiment, before detection, it is necessary to determine the size of the detection deviation angle θ. When the detection deviation angle θ ≤ 45°, the detection method in this solution is applicable. As shown in the attached Figure 3 figure, in this embodiment, the probe has a spherical probe head. During the detection process, the spherical probe head contacts the target surface. The detection deviation angle θ is the angle between the normal f of the probe contact surface (i.e., the normal of the spherical head contact surface) and the detection normal n. When the normal f of the contact surface is parallel to the detection normal n, it means that θ is 0 at this time, and the deviation is the smallest, and the position of the probe at the contact point can be accurately obtained; when the normal f of the contact surface is not parallel to the detection normal n, the detection result will have a deviation. In this embodiment, by calculating the size of the detection deviation angle θ to determine the size of the deviation. In the formula, r is the radius of the probe ball head. If the detection deviation angle θ is too large, it means that the detected position error is large and it is difficult to meet the detection accuracy requirements. By this method, the problem that the positioning of the traditional machine tool probe will bring errors can be avoided, so as to ensure that the measured contact point is closer to the true contact point coordinates, rather than just the probe ball center position, improve the detection accuracy, and ensure the applicability of the traditional probe.

[0032] After meeting the requirements of the detection deviation angle θ, according to the concave and convex direction of the curve segment to be detected, the detection normal n, the radius r of the probe ball head, and the detection step S are determined.

[0033] In this embodiment, although not limited to the diameter size of the probe ball head, according to the shape of the detection curve, a suitable probe ball head can be selected for detection to improve the detection effectiveness and accuracy. In this embodiment, the detection curve includes the following two shapes:

[0034] Shape 1. The curve segment to be detected is a convex segment.

[0035] Shape 2. The curve segment to be detected is a concave segment.

[0036] When the curve to be detected is a convex segment of Shape 1, at this time, as shown in the attached Figure 4As shown in (a), set the detection normal direction \(n\) to be opposite to the detection direction \(m\). At this time, a probe ball with a relatively large probe ball radius \(r\) can be selected at the convex part of the detection curve. Calculate the detection step size \(S\) based on the determined probe ball radius \(r\) and the local radius \(R\) of the detected curve, as shown in the appendix Figure 5 As shown. In this embodiment, the calculation method of the detection step size \(S\) is as follows:

[0037] ; (Equation 1)

[0038] In the formula, is the position deviation along the detection normal direction. During the actual operation process, the position deviation is determined according to actual requirements and directly input as a known value. At the same time, the set detection deviation is achieved by adjusting the step size.

[0039] When the detected curve is a concave line segment of Shape 2, at this time, as shown in the appendix Figure 4 As shown in (b), set the detection normal direction \(n\) to be the same as the detection direction \(m\). At this time, select a probe ball with a ball head radius smaller than the local radius \(R\) of the curved surface at the concave part of the detection curve. Calculate the detection step size \(S\) based on the determined probe ball radius \(r\) and the local radius \(R\) of the detected curve. In this embodiment, the effective range of the detection step size \(S\) is the range near the extreme value of the curve segment, and its calculation method is as follows:

[0040] ; (Equation 2)

[0041] According to Formulas 1 and 2, it can be known that when \(r\) and \(R\) are determined, the smaller the step size \(S\), the smaller the deviation \(e\). As can be seen from the appendix Figure 5 As shown, the above formula can be satisfied only near the estimated local radius of the detected curve, and thus the step size \(S\) is determined. The value of \(S\) will affect the detection deviation and detection speed. When \(S\) is too large, it may skip the tangent point that needs to be detected. When \(S\) is too small, it will cause an increase in the detection duration, and the probability of generating random errors locally will increase. According to Equation 1, when detecting a convex curve, the local radius \(R\) remains fixed. When increasing the probe head \(r\) of the probe, the deviation can be reduced, and at the same time, the detection step size \(S\) can be increased to reduce the detection duration and improve the detection efficiency. According to Equation 2, when detecting a concave curve, the probe head \(r\) of the probe needs to be smaller than the estimated local radius \(R\) to avoid being unable to detect the target tangent point during detection. At this time, the step size \(S\) cannot be too large.

[0042] S2. According to the determined detection parameters, detect the detected curve segments respectively according to the corresponding detection methods to obtain multiple groups of detection points.

[0043] Before detection, first set the detection range according to the intersection curve of the target surface, that is, the curve range of the curve segment to be detected. In this embodiment, the detection range needs to include the farthest point on the curve segment, and there cannot be two or more extreme values in the detection normal direction n. When specifically determining the range, by projecting the curve segment onto the detection normal direction n, the detection range is delimited as a range area including the point with the farthest distance.

[0044] According to the delimited detection range, perform detection within this detection range according to the detection method. In this embodiment, the detection method is a scanning segmented combined detection. During detection, move the center of the probe tip of the probe into the specified plane and place it at the initial position of the detection range, that is, the end point of the tangential direction n t can be set as the initial detection position, and perform a scanning detection on the intersection curve of the target surface to be detected, that is, the curve segment to be detected, according to the determined detection step S. The detection direction is parallel to the detection normal direction n. That is, after detecting a point on the curve, move the probe along the tangential direction for a distance S, and then perform detection along the detection direction m. Repeat this process until the entire detection range to be detected is covered. Then, detect the next curve segment to be detected.

[0045] During the detection process, record the position information of each detected position, that is, the position information of the contact point between the probe and the curve, and project the recorded position information onto the detection normal direction n, and organize it into coordinate points, denoted as detection points After detecting each curve segment respectively, the coordinate point information of multiple groups of detection points is obtained.

[0046] In this embodiment, it also includes excluding the random error of detection to ensure the accuracy of measurement data. Among them, the random error is excluded by repeatedly detecting the position information at the tangent point multiple times to obtain multiple detection points, and performing an averaging process on the detection point data measured at this point. If in the detection normal direction, a situation where the projection is farther than the previous detection occurs, then detect this point three times repeatedly and take the intermediate value as the detection value at this position, and exclude the random error during tangent detection through this method.

[0047] S3. Calculate the maximum projection value in the detection normal direction according to the obtained detection points; determine the tangent position of the detection according to the maximum projection value point and the detection normal direction.

[0048] In this embodiment, through segmented combined detection, multiple curves with extreme values can be detected in the detection normal direction. According to the obtained detection points, calculate the maximum value point in the projection direction, and recognize this point as the point on the tangent line perpendicular to the detection direction of the target curve. Specifically, perform a dot product on the obtained detection points with the detection normal direction n and then compare them. Take the maximum value as the maximum value point in the projection direction, denoted as ·n max , denoted as the maximum point , which is the tangent point to be found.

[0049] Then, according to the found tangent point P m and the tangential direction n t calculate and combine the required tangent information, denoted as . Thus, according to the set detection normal direction, the tangent information at the concave or convex part of the curve can be obtained respectively, ensuring the accuracy and uniqueness of the tangent information.

[0050] Specific implementation process description

[0051] Example 1

[0052] In this example, taking a five-axis CNC machine tool as an example, a spherical probe of the five-axis CNC machine tool is used to detect the tangent of a roller fillet, aiming to solve the problem that it is impossible to detect the tangent at any angle of a micro fillet on the CNC machine tool. The implementation steps are as shown in the appendix Figure 6 .

[0053] Preparation: Install a probe with a spherical probe on the five-axis CNC machine tool. This probe can capture the machine tool coordinate information of the detection point when it contacts and triggers along the movement direction.

[0054] It is known that the radius r of the ball head of this probe is 1.5 mm, the roller fillet is a convex curve segment, the estimated radius R of the roller fillet to be detected is 0.1 - 0.2 mm, and the required detection deviation e is less than 0.001 mm.

[0055] Then, the detection step size S can be calculated according to Equation 1, and the maximum step size S obtained is 0.062 mm.

[0056] Set parameters: Set the initial detection position near the roller fillet, and set the detection scanning range to include the roller fillet. As shown in Figures 1 and 2 of the appendix Figure 6 , set the detection direction m according to the roller fillet, then the detection normal direction n is opposite to m, and calculate the fillet tangential direction to be detected according to the detection normal direction n , that is, the detection tangential direction, and set the detection scanning direction to move along the tangential direction .

[0057] During detection, as shown in Figures 3 to 6 of the appendix Figure 6 , use the detection ball head to gradually detect from the initial position according to the set detection step size, and record the contact points on the detection normal direction n . Multiply these points by the detection normal direction n and then compare them to obtain the maximum value, that is ·n max , denoted as . Through Pm and tangent information can be obtained, that is . To achieve the spherical probe of a five-axis CNC machine tool to detect the tangent of a roller fillet.

[0058] From the attached Figure 6 it can be directly seen that the probe ball head of the five-axis CNC machine tool finds the points on this tangent, and further determines a tangent of the roller fillet.

[0059] Embodiment 2

[0060] In this embodiment, different from Embodiment 1, taking a three-axis CNC machine tool as an example, the spherical probe of the three-axis CNC machine tool is used to detect the tangent of a specific tangential direction of the intersection line between a concave surface of a workpiece and a specified section, so as to solve the problem that the three-axis CNC machine tool cannot accurately detect the tangent of a specific direction of the intersection line between a surface and a specified section.

[0061] Preparation work: Install a probe with a spherical probe on the three-axis CNC machine tool. This probe can capture the machine tool coordinate information of the detection point when it contacts and triggers along the movement direction.

[0062] It is known that the radius r of the ball head of this probe is 2 mm, the workpiece is a concave surface, the estimated radius R of the surface to be detected is 50 mm, and the detection deviation e is required to be lower than 0.001 mm.

[0063] Then the detection step size S can be calculated according to Equation 2, and the maximum step size S can be obtained as 4.794 mm.

[0064] Set parameters: Set the initial detection position near the surface to be detected, set the scanning detection range to include the position to be detected, set the detection direction m, the detection normal direction is n, n is in the same direction as m, and calculate the tangential direction of the surface to be detected according to the detection normal direction n , set the detection scanning direction along the tangent Move. The remaining steps are the same as those in Embodiment 1, so as to achieve the spherical probe of the three-axis CNC machine tool to detect the tangent of a specific tangential direction of the intersection line between a concave surface of a workpiece and a specified section.

[0065] Embodiment 3

[0066] In this embodiment, different from Embodiment 1, taking the detection of the tangent of a curve with multiple concave and convex segments on a CNC machine tool as an example, it solves the technical bottleneck that the CNC machine tool cannot achieve high-precision real-time detection of the tangent of a complex continuous curve with multiple concave and convex segments.

[0067] It is known that the radius r of the ball head of this probe is 2 mm, the minimum local radius R of the curve to be detected at the concave part is 50 mm, the maximum local radius R at the convex part is 30 mm, and the detection deviation e is required to be lower than 0.001 mm.

[0068] Then, according to Equation 1, the detection step length S1 of the convex line segment is calculated, and the maximum step length S1 can be obtained as 1.012 mm.

[0069] According to Equation 2, the detection step length S2 of the concave line segment is calculated, and the maximum step length S2 can be obtained as 4.794 mm.

[0070] Set parameters: The detection parameters are set for each curve segment separately. For the convex line segment, the detection normal direction n is opposite to the detection direction m, and the corresponding tangential direction of the fillet to be detected is calculated. ; For the concave line segment, set the detection normal direction n to be the same as the detection direction m, and calculate the corresponding tangential direction of the fillet to be detected. .

[0071] During detection, move along the set tangential direction for detection scanning. The remaining steps are the same as those in Embodiment 1 to achieve high-precision real-time detection of the tangent of a multi-convexity and concavity complex continuous curve.

[0072] In this embodiment, by processing the complex curve in segments and using a segmented combination for detection, it can adapt to probes of any precision and has the ability to eliminate random errors, which greatly improves the accuracy and reliability of the detection results. The present invention breaks through the limitations in the prior art that only one end face can be detected and the tangent of a curved surface cannot be detected, and can achieve the detection of the tangent of a curved surface on an ordinary machine tool, no longer limited to a specific device such as a coordinate measuring machine, and greatly reduces the requirement for the accuracy of the probe, greatly expanding its application range in the field of curved surface processing. For fields such as mold manufacturing and impeller processing that involve a large amount of complex curved surface processing, this means that the processing path can be planned more accurately, the processing efficiency and product quality can be improved, and the processing difficulty and cost caused by detection limitations can be reduced.

[0073] The above are only embodiments of the present invention. Specific technical solutions and / or common knowledge such as characteristics known in the solutions are not described in detail here. It should be noted that for those skilled in the art, without departing from the technical solution of the present invention, several deformations and improvements can still be made, and these should also be regarded as the protection scope of the present invention, and these will not affect the implementation effect of the present invention and the practicality of the patent. The protection scope required by this application should be subject to the content of its claims, and the specific implementation manners described in the specification can be used to explain the content of the claims.

Claims

1. A method for detecting a tangent line of a curved surface, characterized in that, The probe on the numerically controlled machine tool is used to detect the target curve, and the tangent position of the target curve perpendicular to the detection direction in the specified plane is obtained; It includes the following steps, Parameter determination: The target curve is segmented according to the segmentation method, the radius R of the curve segment to be detected is estimated according to the segmented curve segments, and the detection parameters are determined according to the segmented curves; Detection point acquisition: According to the determined detection parameters, the detected curve segments are detected respectively according to the corresponding detection methods to obtain multiple groups of detection points ; The detection parameters include the detection contact surface, the detection direction m, the detection normal direction n, the tangent direction , the probe ball head radius r and the detection step size S; The detection method is a scanning type segmented combined detection, that is, after detecting a point on the curve, the probe is moved a distance S along the tangent direction, and then detected along the detection direction m, and so on until the detection range covers the entire range to be detected; Tangent point selection: Project the detection point onto the detection normal vector n, and calculate the point with the maximum projection value as the tangent point; Tangent calculation: Based on the obtained tangent point and the detected normal direction n, determine the position of the detected tangent.

2. The method for detecting a tangent line of a curved surface according to claim 1, wherein: The target curve is the intersection curve of the specified plane and the target surface; the segmentation method is to project the target curve onto the detection normal n, and segment it according to the shape of the projected detection curve, and divide it into K curve segments with at most one radian, where K≥1.

3. A method for detecting a tangent line of a curved surface according to claim 1, characterized in that: The detection direction m is parallel to the detection normal direction n; the tangent direction is a tangential direction perpendicular to the detection normal direction n in a specified plane .

4. A method for detecting a tangent line of a curved surface according to claim 1, characterized in that: It also includes determining the size of the detection deviation angle θ before detection, such that θ ≤ 45°; the detection deviation angle θ is the angle between the normal f of the probe contact surface and the detection normal n, where ; in the formula, r is the radius of the probe ball head, is the position deviation along the detection normal.

5. A method for detecting a tangent line of a curved surface according to claim 1, characterized in that: The detection step S is calculated according to the shape of the curve segment to be detected, When the curve segment to be detected is a convex segment, the detection step S is calculated as ； When the curve segment to be detected is a concave segment, the detection step S is calculated as ； In the formula, is the position deviation along the detection normal direction.

6. A method for detecting a tangent line of a curved surface according to claim 3, characterized in that: The tangent position information is represented as .

7. A method for detecting a tangent line of a curved surface according to claim 3, characterized in that: The tangent direction is that the detection normal n rotates +90° or -90° within the specified plane.

8. A method for detecting a tangent line of a curved surface according to claim 1, characterized in that: It also includes excluding the random error during detection. The exclusion method of the random error is to repeatedly detect the position information at the tangent point of the tangent line multiple times, and average the measured detection point data.

9. A method for detecting a tangent line of a curved surface according to claim 1, characterized in that: It also includes setting the detection range according to the curve segment to be detected before detection; the detection range includes the farthest points on the curve segment, and there cannot be two or more extreme values in the detection normal n for the detection range.

Images