A free-form surface scanning path planning method based on adaptive sampling

Abstract

The present invention discloses a free-form surface scanning path planning method based on adaptive sampling, including step S1, free-form surface adaptive sampling, adaptively discretizing the free-form surface into multiple guide lines, adaptive sampling on the multiple guide lines, and generating a set of sampling points; step S2, viewpoint calculation; the viewpoint position is calculated according to the position and direction of the sampling point and the parameters of the line laser scanner, and a coordinate system is established to determine the scanning direction and moving direction of the line laser scanner scanning probe; step S3, path planning, according to the position and posture of the obtained viewpoint, the execution path of the free-form surface is planned, the execution path is adjusted, spline smoothing is performed, and collision avoidance is performed to form a scanning path. The present invention can perform adaptive sampling according to the model characteristics of the free-form surface, obtain more sampling points at places where the surface shape changes greatly, and can make the corresponding viewpoints located in a plane, and can automatically generate the scanning path of the free-form surface.

Description

Technical Field

[0001] The present invention belongs to the field of three-dimensional scanning measurement, and in particular, relates to a free-form surface scanning path planning method based on adaptive sampling. Background Art

[0002] With the rapid development of the manufacturing industry, free-form surfaces have been widely used in the design of products such as aircraft, high-speed railways, and ships. Since the manufacturing accuracy of the surface directly affects the main performance of the equipment, the scanning and measurement of free-form surfaces has become particularly important. The traditional free-form surface scanning and measurement method usually uses an industrial robot to carry a scanning and measuring device to scan the free-form surface through teaching programming. However, this method requires manual control of the scanning robot to reach each teaching point and design the scanning path of the model based on experience. Not only is the scanning efficiency low, but the teaching path is difficult to simultaneously meet the requirements of the scanning and measurement robot for position and posture. This may result in some areas not being scanned and incomplete data acquisition, and may also cause some areas to be scanned repeatedly, resulting in data redundancy. Summary of the invention

[0003] The purpose of the present invention is to provide a simple, efficient, free-form surface scanning path planning method based on adaptive sampling, which can automatically generate a scanning path for a free-form surface through an isoparametric line-sampling point-viewpoint-path planning method.

[0004] In order to achieve the purpose of the invention, the technical solution provided by the present invention is:

[0005] A free-form surface scanning path planning method based on adaptive sampling comprises the following steps:

[0006] Step S1, free-form surface adaptive sampling;

[0007] Step S10, determining the sweep line and guide line directions of the free-form surface;

[0008] Step S11, adaptively discretizing the free-form surface into a plurality of isoparametric lines;

[0009] Step S12, performing adaptive sampling on the isoparametric line whose direction of the free-form surface guide line is located at the midpoint of the parameter domain to generate an isoparametric line sampling point set;

[0010] Step S13, obtaining the coordinates of the free surface sampling point and its normal vector on the parametric surface;

[0011] Step S2, viewpoint calculation;

[0012] The viewpoint position is calculated according to the position and direction of the sampling point and the parameters of the line laser scanner, and a coordinate system is established to determine the scanning direction and moving direction of the line laser scanner scanning probe;

[0013] The specific steps include:

[0014] Step S20, calculating the visual cone corresponding to each sampling point;

[0015] The point to be sampled is taken as the cone top, the normal vector of the point to be sampled on the parametric surface is taken as the cone axis, the scanning angle is taken as the cone angle, and each sampling point to be measured corresponds to a frustum.

[0016] Step S21, obtaining a viewpoint position area;

[0017] The intersection plane obtained in step S13 is intersected with the visual cone obtained in step S20 to obtain a series of polygons, which are the location area of ​​the viewpoint;

[0018] Step S22, performing Boolean intersection on the viewpoint position area obtained in step S21 to merge redundant viewpoints;

[0019] Step S23, determining whether the viewpoint is blocked;

[0020] Detect occlusions at each viewpoint and adjust the viewpoint position to avoid occlusions;

[0021] Step S3, path planning;

[0022] Plan the execution path of the free-form surface according to the position and posture of the obtained viewpoint, adjust the execution path, perform spline smoothing and avoid collision;

[0023] Specifically include:

[0024] Step S30, each viewpoint is connected once to form each path;

[0025] Step S31, each path segment is connected end to end to form a scanning path;

[0026] Step S32, performing spline curve-based smoothing processing on the path;

[0027] Step S33, adjusting the path to avoid collision;

[0028] The collision between the line laser scanner and the free-form surface at each viewpoint during the execution of the path is detected, and the collision is eliminated to form a scanning path.

[0029] A further preferred technical solution provided by the present invention is: the step S10 specifically includes:

[0030] Step S100, feature recognition is performed on the isoparametric lines of the free-form surface in the parameter domains in the u and v directions, and the isoparametric line equations y(x) are obtained. The isoparametric line arc lengths l of the free-form surface in the parameter domains in the u and v directions are calculated according to formula (1);

[0031]

[0032] In formula (1), a is the lower limit of the definite integral, b is the upper limit of the definite integral, dx is the differential, and l is the arc length of the isoparametric line.

[0033] Step S101, selecting the direction of the longer isoparametric line as the guide line direction of the free zone surface.

[0034] A further preferred technical solution provided by the present invention is: the step S11 specifically includes:

[0035] Step S110, obtaining the isoparametric line of the midpoint of the sweep line direction parameter domain, and sampling points on the isoparametric line in a manner of equal arc length;

[0036] Step S111: all the sampling points M in step S110 are i Save to queue T i ;

[0037] Step S112, setting the maximum spacing d according to the scanning probe resolution and field of view max , determine whether the distance between adjacent points is less than the field of view distance, if not, proceed to step S113; if yes, proceed to step S114;

[0038] Step S113, inserting sampling points between the curve segments of two adjacent sampling points in an isoparametric sampling manner, and returning to step S111;

[0039] Step S114, calculate the curvature K of each curve segment y(x) according to formula (2): i

[0040]

[0041] In formula (2), y' is the first-order derivative, y" is the second-order derivative, and det means the absolute value;

[0042] Step S115, setting the threshold to δ b , compare the degree of completion of the curve segment K i Is it less than the threshold value? b , if K i <δ b , then complete step S11 and proceed to the next step; if К i >δ b , then return to step S113, and add sampling points to the curve segment at the two sampling points in an isoparametric manner;

[0043] The above steps are repeated to adaptively sample the isoparametric lines according to the curvature. The parameter value of the sampling point is the parameter constant value of the isoparametric line to be obtained. Each isoparametric line is thus obtained, and finally the surface is adaptively discretized into a series of isoparametric lines.

[0044] Another preferred technical solution provided by the present invention is: in the step S12, specifically:

[0045] Step S120, sampling points on the isoparametric line where the direction of the free-form surface guide line is located in the midpoint of the parameter domain according to the equal arc length method;

[0046] Step S121: all sampling points M j Save to queue T j middle;

[0047] Step S122, calculate the curvature of each curve segment according to formula (2): j ;

[0048] Step S123, setting the threshold to δ b , compare the degree of completion of the curve segment K j Is it less than the threshold value? b , if K j <δ b , then complete step S12 and proceed to the next step S13; if К j >δ b , then return to step S124;

[0049] Step 124, adding sampling points to the curve segment at the two sampling points in an isoparametric manner, and returning to step S121.

[0050] A further preferred technical solution provided by the present invention is: in the step S13, specifically:

[0051] Step S130, calculating the coordinates of each point obtained in step S12;

[0052] Step S131, calculate the tangent vector of each point on the isoparametric line According to formula (3), the intersection plane is constructed at each point using the point method formula.

[0053] A(x-x0)+B(y-y0)+C(z-z0)=0 (3)

[0054] Step S132: The intersection plane obtained in step S131 is intersected with the isoparametric line obtained in step S11 to obtain a series of sampling points, and the tangent vector T of each sampling point in the two parameter directions of the parameter domain u and v of the surface is calculated. u , T v , according to formula (4):

[0055] n=T u ×T v (4)

[0056] The cross product of the tangent vectors in the two parameter directions at the surface sampling point is the normal vector at the sampling point, completing the adaptive sampling of the free-form surface.

[0057] Another further preferred technical solution provided by the present invention is: the step S22 specifically includes the following steps:

[0058] Step S220, pushing the polygons obtained in step S21 into the stack in sequence;

[0059] Step S221, determine whether the element in the stack is 1, if so, proceed to step S223; if not, proceed to step S222;

[0060] Step S222: pop the top two elements S1 and S2 from the stack and perform a Boolean intersection. The intersection result is counted as S i , proceed to step S224;

[0061] Step S223, store the center of polygon S as a viewpoint, and proceed to step

[0062] Step S224, determine S i Is it empty? If not, go to step S225; if yes, go to step S226;

[0063] Step S225, determine whether the field of view condition is met, if so, proceed to step S227; if not, proceed to step S226;

[0064] Step S226, store the center position of S1 as a viewpoint, push S2 to the top of the stack; return to step S221;

[0065] Step S227, S i Push to the top of the stack and return to step S221.

[0066] Compared with the prior art, the present invention has the following beneficial effects:

[0067] 1. The existing free-form surface scanning path planning method based on teaching programming not only has low scanning efficiency, but also cannot guarantee the scanning integrity of the free-form surface. The present invention proposes a simple and efficient path planning method, which can automatically generate the scanning path of the free-form surface through the planning method of isoparametric line-sampling point-viewpoint-path, which not only improves the measurement efficiency, but also ensures the scanning integrity of the free-form surface.

[0068] 2. Existing free-form surface scanning path planning methods based on offline programming all use isoparametric sampling to uniformly obtain sampling points on the surface of the part to be measured, which cannot reflect the local changes of the surface and affects the sampling accuracy. Moreover, if the centroid position of the visible cone is used as the scanning position, the scanning positions of the sampling points on the same plane will not be in the same plane, which will make the scanning path complex and difficult to execute. The present invention proposes a simple and efficient path planning method, which can perform adaptive sampling according to the model characteristics of the free-form surface, obtain more sampling points where the surface shape changes greatly, and make the corresponding viewpoints in one plane, and can formulate the optimal scanning strategy that meets the model characteristics of the free-form surface. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 It is the overall flow chart of the present invention;

[0070] Figure 2 Flowchart for adaptive sampling of free-form surfaces;

[0071] Figure 3 Schematic diagram of the sweep line and guide line of the free-form surface;

[0072] Figure 4 Flowchart for optimal viewpoint calculation

[0073] Figure 5 It is a schematic diagram of scanner related parameters;

[0074] Figure 6 Generate schematic diagrams for candidate viewpoints;

[0075] Figure 7 This is a schematic diagram of viewpoint connection;

[0076] Figure 8 It is a schematic diagram of the scanning path of the free-form surface;

[0077] Fig. 9 This is a simulation diagram of the free-form surface scanning path. DETAILED DESCRIPTION

[0078] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings.

[0079] Figure 1 It is the overall flow chart of the present invention, combined with Figure 1 With reference to other drawings, the present invention is described in detail as follows:

[0080] The present invention provides a free-form surface scanning path planning method based on adaptive sampling, comprising the following steps:

[0081] Step S1, free-form surface adaptive sampling;

[0082] Figure 2 The free surface adaptive sampling flow chart of step S1 is shown in FIG. Figure 2 As shown, the free-form surface adaptive sampling specifically includes:

[0083] Step S10, determining the sweep line and guide line directions of the free-form surface;

[0084] Step S100, feature recognition is performed on the isoparametric lines of the free-form surface in the parameter domains in the u and v directions, and the isoparametric line equations y(x) are obtained. The isoparametric line arc lengths l of the free-form surface in the parameter domains in the u and v directions are calculated according to formula (1);

[0085]

[0086] In formula (1), a is the lower limit of the definite integral, b is the upper limit of the definite integral, dx is the differential, and l is the arc length of the isoparametric line.

[0087] Step S101, select the direction of the longer isoparametric line as the guide line direction of the free zone surface (such as Figure 3 As shown in the figure, p is the arc length of the isoparametric line in the u direction of the free surface. d is the arc length of the isoparametric line in the v direction of the free surface, S is the free surface, and the isoparametric line with the longer l is selected. d The direction of is the guide line direction of the free zone surface, then l p The direction of the guide line is the sweep line direction of the free zone surface), that is, the isoparametric line set in this direction is used to replace the parametric surface. Selecting the appropriate guide line direction can greatly reduce the number of scanning probe posture adjustments, reduce the path segments that need to be planned, and improve scanning efficiency and accuracy.

[0088] Step S11, adaptively discretizing the free-form surface into multiple isoparametric lines, specifically including:

[0089] Step S110, the free-form surface has determined the guide line direction in step S101, and the isoparametric line of the value in the parameter domain of the sweep line direction can be obtained, and sampling points on the isoparametric line are performed in a manner of equal arc length;

[0090] Step S111: all the sampling points M in step S110 are i Save to queue T i ;

[0091] Step S112, setting the maximum spacing d according to the scanning probe resolution and field of view max , determine whether the distance between adjacent points is less than the field of view distance, if not, proceed to step S113; if yes, proceed to step S114;

[0092] Step S113, inserting sampling points between the curve segments of two adjacent sampling points in an isoparametric sampling manner, and returning to step S111;

[0093] Step S114, calculate the curvature K of each curve segment y(x) according to formula (2): i

[0094]

[0095] In formula (2), y' is the first-order derivative, y" is the second-order derivative, and det means the absolute value;

[0096] Step S115, setting the threshold to δ b , compare the degree of completion of the curve segment K i Is it less than the threshold value? b , if K i <δ b , then complete step S11 and proceed to the next step; if К i >δ b , then return to step S113, add sampling points in an isoparametric manner to the curve segment at the two sampling points

[0097] The above steps are repeated to adaptively sample the isoparametric lines according to the curvature. The parameter value of the sampling point is the parameter constant value of the isoparametric line to be obtained. Each isoparametric line is thus obtained, and finally the surface is adaptively discretized into a series of isoparametric lines.

[0098] Step S12, performing adaptive sampling on the isoparametric line whose direction of the free-form surface guide line is located at the midpoint of the parameter domain to generate an isoparametric line sampling point set;

[0099] Step S120, sampling points on the isoparametric line where the direction of the free-form surface guide line is located in the midpoint of the parameter domain according to the equal arc length method;

[0100] Step S121: all sampling points M j Save to queue T j middle;

[0101] Step S122, calculate the curvature of each curve segment according to formula (2): j ;

[0102] Step S123, setting the threshold to δ b , compare the degree of completion of the curve segment K j Is it less than the threshold value? b , if K j <δ b , then complete step S12 and proceed to the next step S13; if К j >δ b , then return to step S124;

[0103] Step 124, adding sampling points to the curve segment at the two sampling points in an isoparametric manner, and returning to step S121;

[0104] Step S13, obtaining the coordinates of the free surface sampling point and its normal vector on the parametric surface;

[0105] Step S130, calculating the coordinates of each point obtained in step S12;

[0106] Step S131, calculate the tangent vector of each point on the isoparametric line According to formula (3), the intersection plane is constructed at each point using the point method formula.

[0107] A(x-x0)+B(y-y0)+C(z-z0)=0 (3)

[0108] Step S132: The intersection plane obtained in step S131 is intersected with the isoparametric line obtained in step S11 to obtain a series of sampling points, and the tangent vector T of each sampling point in the two parameter directions of the parameter domain u and v of the surface is calculated. u , T v , according to formula (4):

[0109] n=T u ×T v (4)

[0110] The cross product of the tangent vectors in the two parameter directions at the surface sampling point is the normal vector at the sampling point, completing the adaptive sampling of the free-form surface.

[0111] Step S2, viewpoint calculation;

[0112] The viewpoint position is calculated according to the position and direction of the sampling point and the parameters of the line laser scanner, and a coordinate system is established to determine the scanning direction and moving direction of the line laser scanner scanning probe;

[0113] like Figure 5 As shown, the parameters of the line laser scanner include field of view (the measured point must be within the field of view to be acquired), scanning depth of field DOF (surface measurement point p i The scanning laser source should be offset by a certain distance), the scanning angle of view α (the size of the scanning angle of view constrains the upper limit of the direction angle β, and the condition that the direction angle β is less than 0.5α must be met to obtain the measured point).

[0114] The specific steps include:

[0115] Step S20, calculating the visual cone corresponding to each sampling point;

[0116] The point to be sampled is taken as the cone top, the normal vector of the point to be sampled on the parametric surface is taken as the cone axis, the scanning angle is taken as the cone angle, and each sampling point to be measured corresponds to a frustum.

[0117] Step S21, obtaining a viewpoint position area;

[0118] The intersection plane obtained in step S131 is intersected with the visual cone in step S20 to obtain a series of polygons, which are the location area of ​​the viewpoint;

[0119] Step S22, performing Boolean intersection on the viewpoint position area obtained in step S21 to merge redundant viewpoints;

[0120] The specific steps include:

[0121] Step S220, pushing the polygons obtained in step S21 into the stack in sequence;

[0122] Step S221, determine whether the element in the stack is 1, if so, proceed to step S223; if not, proceed to step S222;

[0123] Step S222: pop the top two elements S1 and S2 from the stack and perform a Boolean intersection. The intersection result is counted as S i , proceed to step S224;

[0124] Step S223, store the center of polygon S as a viewpoint, and proceed to step

[0125] Step S224, determine S i Is it empty? If not, go to step S225; if yes, go to step S226;

[0126] Step S225, determine whether the field of view condition is met, if so, proceed to step S227; if not, proceed to step S226;

[0127] Under the premise of meeting the field of view requirements, the center S of the figure obtained after the intersection is regarded as the scanning position, so that the scanning positions of the sampling points located in the same plane are also on the same plane. Taking the scanning probe position as the starting point and the virtual sampling points that can be detected by the scanning position as the end point, a series of vectors are obtained, which are normalized and summed, and the final vector obtained is the scanning direction.

[0128] Step S226, store the center position of S1 as a viewpoint, push S2 to the top of the stack; return to step S221;

[0129] Step S227, S i Push to the top of the stack and return to step S221;

[0130] Step S23, determining whether the viewpoint is blocked;

[0131] Occlusions are detected at each viewpoint, and the position of the viewpoint is adjusted to avoid occlusions.

[0132] like Figure 6 As shown, the sampling point is taken as the origin, and the line connecting the sampling point and the corresponding viewpoint is taken as the direction to make a ray, and it is detected whether the ray intersects with the free surface. If so, it is considered that occlusion occurs; if so, that is, there is occlusion. For the viewpoint with occlusion, the sampling point corresponding to the current viewpoint is taken as the origin, the scanning direction is taken as the axis, and it is rotated according to a predetermined angle β (generally 15°) to generate multiple candidate viewpoints, and then the occlusion of each viewpoint is detected one by one until a viewpoint without occlusion is found, and this candidate viewpoint replaces the original viewpoint.

[0133] Step S3, path planning;

[0134] Plan the execution path of the free-form surface according to the position and posture of the obtained viewpoint, adjust the execution path, perform spline smoothing and avoid collision;

[0135] Specifically include:

[0136] Step S30, each viewpoint is connected once to form each path;

[0137] like Figure 7 As shown, the first viewpoints on adjacent intersecting planes are sequentially connected to form a first path segment, and then the second viewpoints on adjacent intersecting planes are connected to form a second path segment, until all viewpoints are connected;

[0138] Step S31, each path segment is connected end to end to form a scanning path;

[0139] Connect each path end to end to get a zigzag font path;

[0140] Step S32, as Figure 8 As shown, the path is smoothed based on the spline curve, and the final scanning path is smoothed.

[0141] Step S33, adjusting the path to avoid collision;

[0142] Detecting the collision between the line laser scanner and the free-form surface at each viewpoint during the execution of the path, and eliminating the collision to form a scanning path;

[0143] Collision detection checks whether the scanning probe contacts the free-form surface during movement. The bounding box of the scanning probe is intersected with the free-form surface model to determine whether there is an intersection. Because the scanning probe moves continuously, the movement of the scanning probe can be regarded as the movement of the cross section of the probe bounding box with the moving direction as the normal vector. With this cross section as the initial surface, the scan is performed along the path, and the obtained three-dimensional shape is then Boolean-intersected with the free-form surface workpiece. If there is no intersection, it means that the scanning process will not collide with the free-form surface workpiece. After the collision is detected, the viewpoint position needs to be readjusted until no collision occurs during the scanning process.

[0144] According to the above steps, after the free-form surface is adaptively sampled, the sampling point coordinates are shown in Table 1, which also gives the viewpoint coordinates and scanning vector data obtained through corresponding calculations. The scanning path simulation diagram generated according to the data in Table 1 is shown in Fig. 9 Shown.

[0145] Table 1 Free-form surface scanning path data table

[0146]

[0147]

[0148] This embodiment can perform adaptive sampling according to the model features of the free-form surface through the planning method of isoparametric line-sampling point-viewpoint-path, obtain more sampling points where the surface shape changes greatly, and make the corresponding viewpoints located in a plane. The scanning path of the free-form surface can be automatically generated, which not only improves the measurement efficiency and formulates the optimal scanning strategy that meets the model features of the free-form surface, but also ensures the scanning integrity of the free-form surface.

[0149] The described embodiments are only a part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

Claims

1. A free-form surface scanning path planning method based on adaptive sampling, characterized in that: The following steps are involved: Step S1, free-form surface adaptive sampling; Step S10, determining the sweep line and guide line directions of the free-form surface; Step S11, adaptively discretizing the free-form surface into a plurality of isoparametric lines; Step S12, performing adaptive sampling on the isoparametric line whose direction of the free-form surface guide line is located at the midpoint of the parameter domain to generate an isoparametric line sampling point set; Step S13, obtaining the coordinates of the free surface sampling point and its normal vector on the parametric surface; Step S2, viewpoint calculation; The viewpoint position is calculated according to the position and direction of the sampling point and the parameters of the line laser scanner, and a coordinate system is established to determine the scanning direction and moving direction of the line laser scanner scanning probe; The specific steps include: Step S20, calculating the visual cone corresponding to each sampling point; The sampling point to be sampled is taken as the cone top, the normal vector of the sampling point on the parametric surface is taken as the cone axis, the scanning angle is taken as the cone angle, and each sampling point to be measured corresponds to a frustum. Step S21, obtaining a viewpoint position area; The intersection plane obtained in step S13 is intersected with the visual cone obtained in step S20 to obtain a series of polygons, which are the location area of ​​the viewpoint; Step S22, performing Boolean intersection on the viewpoint position area obtained in step S21 to merge redundant viewpoints; Step S23, determining whether the viewpoint is blocked; Detect occlusions at each viewpoint and adjust the viewpoint's position to avoid occlusions; Step S3, path planning; Plan the execution path of the free-form surface according to the position and posture of the obtained viewpoint, adjust the execution path, perform spline smoothing and avoid collision; Specifically include: Step S30, each viewpoint is connected once to form each path; Step S31, each path segment is connected end to end to form a scanning path; Step S32, performing spline curve-based smoothing processing on the path; Step S33, adjusting the path to avoid collision; The collision between the line laser scanner and the free-form surface at each viewpoint during the execution of the path is detected, and the collision is eliminated to form a scanning path.

2. A free-form surface scanning path planning method based on adaptive sampling according to claim 1, characterized in that: The step S10 specifically includes: Step S100, feature recognition is performed on the isoparametric lines of the free-form surface in the parameter domains in the u and v directions, and the isoparametric line equations y(x) are obtained. The isoparametric line arc lengths l of the free-form surface in the parameter domains in the u and v directions are calculated according to formula (1); In formula (1), a is the lower limit of the definite integral, b is the upper limit of the definite integral, dx is the differential, and l is the arc length of the isoparametric line. Step S101, selecting the direction of the longer isoparametric line as the guide line direction of the free zone surface.

3. A free-form surface scanning path planning method based on adaptive sampling according to claim 2, characterized in that: The step S11 specifically includes: Step S110, obtaining the isoparametric line of the midpoint of the sweep line direction parameter domain, and sampling points on the isoparametric line in a manner of equal arc length; Step S111: all the sampling points M in step S110 are i Save to queue T i ; Step S112, setting the maximum spacing d according to the scanning probe resolution and field of view max , determine whether the distance between adjacent points is less than the field of view distance, if not, proceed to step S113; if yes, proceed to step S114; Step S113, inserting sampling points between the curve segments of two adjacent sampling points in an isoparametric sampling manner, and returning to step S111; Step S114, calculate the curvature K of each curve segment y(x) according to formula (2): i In formula (2), y' is the first-order derivative, y" is the second-order derivative, and det means the absolute value; Step S115, setting the threshold to δ b , compare the degree of completion of the curve segment K i Is it less than the threshold value? b , if K i <δ b , then complete step S11 and proceed to the next step; if К i >δ b , then return to step S113, and add sampling points to the curve segment at the two sampling points in an isoparametric manner; The above steps are repeated to adaptively sample the isoparametric lines according to the curvature. The parameter value of the sampling point is the parameter constant value of the isoparametric line to be obtained. Each isoparametric line is thus obtained, and finally the surface is adaptively discretized into a series of isoparametric lines.

4. A free-form surface scanning path planning method based on adaptive sampling according to claim 3, characterized in that: The step S12 is specifically as follows: Step S120, sampling points on the isoparametric line where the direction of the free-form surface guide line is located in the midpoint of the parameter domain according to the equal arc length method; Step S121: all sampling points M j Save to queue T j middle; Step S122, calculate the curvature of each curve segment according to formula (2): j ; Step S123, setting the threshold to δ b , compare the degree of completion of the curve segment K j Is it less than the threshold value? b , if K j <δ b , then complete step S12 and proceed to the next step S13; if К j >δ b , then return to step S124; Step 124, adding sampling points to the curve segment at the two sampling points in an isoparametric manner, and returning to step S121.

5. A free-form surface scanning path planning method based on adaptive sampling according to claim 1, characterized in that: The step S13 is specifically as follows: Step S130, calculating the coordinates of each point obtained in step S12; Step S131, calculate the tangent vector of each point on the isoparametric line According to formula (3), the intersection plane is constructed at each point using the point method formula. A(x-x0)+B(y-y0)+C(z-z0)=0 (3) Step S132: The intersection plane obtained in step S131 is intersected with the isoparametric line obtained in step S11 to obtain a series of sampling points, and the tangent vector T of each sampling point in the two parameter directions of the parameter domain u and v of the surface is calculated. u , T v , according to formula (4): n=T u ×T v (4) The cross product of the tangent vectors in the two parameter directions at the surface sampling point is the normal vector at the sampling point, completing the adaptive sampling of the free-form surface.

6. A free-form surface scanning path planning method based on adaptive sampling according to claim 1, characterized in that: The step S22 specifically includes the following steps: Step S220, pushing the polygons obtained in step S21 into the stack in sequence; Step S221, determine whether the element in the stack is 1, if so, proceed to step S223; if not, proceed to step S222; Step S222: pop the top two elements S1 and S2 from the stack and perform a Boolean intersection. The intersection result is counted as S i , proceed to step S224; Step S223, store the center of polygon S as a viewpoint, and proceed to step Step S224, determine S i Is it empty? If not, go to step S225; if yes, go to step S226; Step S225, determine whether the field of view condition is met, if so, proceed to step S227; if not, proceed to step S226; Step S226, store the center position of S1 as a viewpoint, push S2 to the top of the stack; return to step S221; Step S227, S i Push to the top of the stack and return to step S221.

Images