A whole-page strain gauge marking path planning method

Abstract

The application discloses a kind of whole edition strain gauge mark path planning methods, comprising the following steps: whole edition strain gauge mark path planning is equivalent to LLTSP solution, motion starting point, end point and defect single piece coordinate in whole edition strain gauge correspond to the node of LLTSP, the access time between coordinates is the distance weight between the nodes of LLTSP, the edge formed by the left and right adjacent defect single piece coordinates, the edge formed by motion end point and starting point, i.e. the constraint edge of LLTSP;Then LLTSP is converted into classic TSP to solve, the solution result of TSP is converted into the solution result of ATSP, and the solution result of ATSP is converted into the solution result of LLTSP.The application realizes the path planning of strain gauge whole edition mark by algorithm input matrix conversion and algorithm output sequence conversion, solves LLTSP using existing high-performance TSP solver, improves the efficiency of whole edition strain gauge defect mark, is easy to operate and easy to realize.

Description

Technical Field

[0001] This invention relates to a path planning method, and more particularly to a path planning method for marking a whole sheet of strain gauges. Background Technology

[0002] The production of strain gauges involves uniformly integrating hundreds or even thousands of strain gauges onto a polyimide substrate of approximately 100mm x 100mm. Due to etching process limitations, the pass rate of strain gauges is only about 80%. Therefore, it is necessary to detect and mark defective strain gauges using a cutting tool. Marking the entire strain gauge plate is a critical step in the production process. Currently, defect detection and marking are done manually, which is inefficient and labor-intensive. To improve efficiency and reduce labor intensity, a strain gauge surface defect detection and marking device automates the entire plate defect detection process, followed by automatic marking. The main operation is as follows: the cutting tool in the device starts from a fixed starting point in the entire plate marking process, traverses each defective strain gauge (hereinafter referred to as a defective individual), and marks it. The marking operation includes three steps: cutting, retraction, and final cut, finally reaching the upper left corner of the entire plate of strain gauges (a fixed endpoint), which is the reset origin for the entire device to replace the entire plate of strain gauges.

[0003] The typical strain gauge full-plate marking path, i.e., the path along which the scriber visits each defective section during the full-plate marking process, can be directly solved using existing high-performance TSP solvers such as LKH and Concorde through modeling and appropriate transformation. The path for each section's visit is as follows: Figure 1 As shown, black blocks represent defective pieces, white blocks represent defect-free pieces, and black lines represent the path of the engraving tool to the defective pieces. However, in actual work, the steps of starting and ending the engraving tool marking are time-consuming, and marking each defective piece individually affects the efficiency of marking the entire sheet. To improve the efficiency of marking the entire sheet, a synchronous marking mechanism should be added to minimize the number of starting and ending cuts during engraving. That is, for a group of consecutively adjacent defective pieces, the tool should start from the leftmost defective piece, cut to the rightmost defective piece, and then start again. Through a single starting and ending cut, the synchronous marking of multiple adjacent defective pieces can be completed.

[0004] This synchronous marking introduces a constraint to the whole-page marking path planning problem: adjacent defect pieces must be visited consecutively from left to right. In other words, in the path planning of the TSP problem, some edges have constraints requiring passage in a fixed direction. For TSP problems with this constraint, the judgment conditions in the classic TSP algorithm's search process can be changed according to the constraint to achieve a solution. However, this method relies heavily on the data structure of the algorithm implementation, resulting in high coupling with the algorithm itself, making it difficult to update and upgrade high-performance TSP solvers in application. Furthermore, a high-performance TSP problem solver often integrates numerous computational techniques and optimization algorithms. Modifying the algorithm structure is a time-consuming, laborious, and challenging task, especially for some TSP solvers whose code is not open source, making it impossible to modify the internal algorithms. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for whole-plate strain gauge marking path planning, thereby achieving whole-plate marking path planning and improving whole-plate marking efficiency.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows:

[0007] A method for planning the marking path of a whole strain gauge includes the following steps:

[0008] Step 1: Description and introduction of problem constraints:

[0009] Planning the marking path for the entire strain gauge is equivalent to solving the LLTSP. The coordinates of the starting point, ending point, and defective individual piece in the entire strain gauge correspond to the nodes in the LLTSP. The access time between any two coordinates corresponds to the distance weight between the corresponding nodes in the LLTSP. The edge formed by the left and right adjacent defective individual pieces, the edge formed by the starting point and the ending point in the entire strain gauge correspond to the constraint edge in the LLTSP.

[0010] Step 2, Establishing the LLTSP problem:

[0011] A LLTSP problem is described by a point set V, a matrix D, and a set of constraint edges E1, which serve as the input for the problem transformation.

[0012] Step 3: Convert LLTSP to ATSP:

[0013] To eliminate the constraints of set E1, the point set V is transformed into the point set V1 of ATSP, and the matrix D and set E1 are transformed into the asymmetric weight matrix C of ATSP.

[0014] Step 4: Convert ATSP to TSP:

[0015] Transform point set V1 into point set V2 of TSP, and transform the asymmetric weight matrix C into a symmetric weight matrix with no negative elements of TSP.

[0016] Step 5, Solving the LLTSP problem:

[0017] First, weight the point set V2 of TSP and the asymmetric weight matrix. Input the TSP solver to obtain the TSP solution result, then convert the TSP solution result into the ATSP solution result, and then convert the ATSP solution result into the LLTSP solution result. The LLTSP solution result is the closed-loop path of the entire strain gauge marking. Disconnect the edge connecting the node at the end point of the entire strain gauge marking and the starting node. Starting from the starting point, follow the path to the end point of the marking, which is the access path of the entire strain gauge marking.

[0018] The specific method for step two in the above scheme is as follows:

[0019] Let the two endpoints of a constraint edge in an LLTSP be a and b. Then a directed edge is represented by two endpoints, that is, (a,b) represents a directed edge from endpoint a to endpoint b. The constraint is represented by the endpoint representation of the edge as: (a,b) must be visited.

[0020] Let V be the set of all nodes in the LLTSP problem, and let V = {1, 2, ..., f}, where f is the number of nodes, i.e., the number of defective pieces. For i, j ∈ V, construct the f-order matrix D of the LLTSP:

[0021]

[0022] Where, d i,j This represents the distance weight from node i to node j;

[0023] Establish a set E1 to store the corresponding constraint edges. For example, if (a,b) must be visited, then (a,b)∈E1. Finally, the LLTSP problem is represented by a point set V, a matrix D, and a constraint edge set E1.

[0024] The specific method for step three in the above scheme is as follows:

[0025] Create a set S1 to store the sub-paths formed by merging multiple adjacent edges in E1; create a set of deleted nodes. Used to store the intermediate nodes and right endpoints of the sub-paths in S1. The initial value is empty; the specific steps include the following:

[0026] ①Merge consecutive adjacent constraint edges in E1: For all i1,…,i k∈V, k≥2, where k represents the number of nodes taken from V, i1,…,i k Let (i, i) represent any k unique nodes in set V; if (i1, i2), ..., (i... k-1 i k )∈E1, and point i2,…,i k-1 join in Let (i1,i2),...,(i k-1 i k Remove (i1, i) from E1, and set (i1, i) to (i1, i) k Add E1, and add the subpath [i1,…,i] to E1. k Add S1, making Represents the distance from node i1 to node i k Distance weights, Indicates node i m To node i m+1 The distance weights, 1≤m≤k-1;

[0027] ②Merge constraint endpoints in E1: For all (a,b)∈E1, let d b,a =d a,b =0, merge points a and b into a, add point b to the set. Let d a,i =d b,i , i∈V and i≠b, i≠a;

[0028] ③Establishment of ATSP point set V1 and weight matrix C: Let V1 be The complement of V, i.e., V1, represents all sets that belong to V but not to V. Let V1 be a set of points, containing n points. Then V1 = {i1, ..., i...} n};

[0029] The weight matrix C is the distance weight matrix for all points in the point set V1, as shown in the following formula, for i k1 i k2 Let ∈V1,

[0030]

[0031] In the above scheme, the specific method for step four is as follows:

[0032] ① Let the temporary transformation matrix be set. Again For matrix The diagonal elements;

[0033] ② Let U be an n-dimensional matrix, where all elements in matrix U except the diagonal have the value INT_MAX / 2, where INT_MAX is the maximum integer value that the computer installed on the TSP solver can recognize;

[0034] ③ Order V2 = {i1,…,i n ,i1+f,…,i n +f};

[0035] ④ W is a matrix of the same order as C′. All elements in matrix W except for the diagonal elements are M, and the values ​​of M are as follows:

[0036] M = 2(c max -c min )+1

[0037] Among them, c max Let c be the maximum weight in C. min It represents the minimum weight in C excluding the diagonal elements.

[0038] In the above scheme, the specific method for step five is as follows:

[0039] ① Transformed distance matrix As input to the unconstrained TSP solver, the optimal sequence of TSP paths is obtained as L, as shown in the following equation;

[0040] i1→(i1+f)→i2→(i2+f)→…→i n →(i n +f)→i1

[0041] ②Remove all extended nodes from L to obtain sequence L2, as shown in the following formula;

[0042] i1→i2→…→i n →i1

[0043] ③ First, for all element sequences [i1, i2] in set E1, replace i1 in sequence L2 with i1→i2. Then, for all element sequences [i1,…,i2] in set S1... k ], transfer i1→…→i k Replace i1→i in sequence L2 k Thus, the solution for sequence L3 is an LLTSP;

[0044] ④ Disconnect the edge connecting the node at the end of the full-sheet strain gauge mark in L3 with the node at the start point. Start from the start point and follow the path to the end of the mark. This is the access path L4 for the full-sheet strain gauge mark.

[0045] Through the above technical solution, the present invention provides a method for path planning of full-plate strain gauge marking. First, the path planning problem of full-plate marking is described as a Line Flow Constraint Trajectory Problem (LLTSP) (LLTSP problem). Then, it is transformed into a classic TSP problem for solution. Specifically, the LLTSP can be solved using existing high-performance TSP solvers through transformation of the algorithm input matrix and algorithm output sequence, thereby realizing path planning for full-plate strain gauge marking. The method of the present invention has the following beneficial effects:

[0046] 1. Improved the efficiency of defect marking for the entire strain gauge sheet;

[0047] 2. High-performance unconstrained TSP solvers can be used to solve LLTSP problems;

[0048] 3. Only the input matrix of the algorithm needs to be changed, without modifying the structure of the existing algorithm. It is simple to operate and easy to implement. Attached Figure Description

[0049] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below.

[0050] Figure 1 This is a schematic diagram illustrating the path for accessing each defective strain gauge individually using existing strain gauges.

[0051] Figure 2 This is a schematic flowchart of a full-sheet strain gauge marking path planning method disclosed in this invention;

[0052] Figure 3 This is a schematic diagram showing the layout of the entire strain gauge and its defective individual gauges according to an embodiment of the present invention. Detailed Implementation

[0053] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention.

[0054] This invention provides a method for planning the marking path of a whole sheet of strain gauges, such as... Figure 2 As shown, the specific implementation is as follows:

[0055] The layout of the entire strain gauge and its defective individual gauges in this embodiment is as follows: Figure 3 As shown. During the entire strain gauge marking process, it is necessary to start from point 1 and reach point 9, passing through 7 defective individual piece coordinate points along the way, for a total of 9 points. The access times between any two points are given. The task is to plan the path so that, starting from the point, all defective individual piece coordinate points are visited without repetition, and then the destination is reached. Defective individual piece coordinate points 4, 5, and 6 are consecutively adjacent and need to be marked uniformly. Find the optimal access path for the defective individual pieces of the strain gauge.

[0056] Step 1: Description and introduction of problem constraints

[0057] The defective single-piece coordinate points 4, 5, and 6 are adjacent to each other on the left and right. The endpoint 9 and the starting point 1 are considered to be adjacent. According to the endpoint representation of the edge, it can be represented as (4,5), (5,6), (6,7), and (9,1), which must be visited.

[0058] Step 2, Establishing the LLTSP Problem

[0059] Given the corresponding LLTSP problem point set V = {1, 2, ..., 9}, construct the 9th order matrix D of LLTSP as shown in the following formula, and construct the set E1 = {(4, 5), (5, 6), (6, 7), (9, 1)}.

[0060]

[0061] Step 3: Convert LLTSP to ATSP

[0062] ① Merge consecutive adjacent constraint edges in E1: Since (4,5), (5,6), and (6,7) ∈ E1, add points 5 and 6. Remove (4,5), (5,6), and (6,7) from E1, add (4,7) to E1, add the subpath [4,5,6,7] to set S1, and let d 4,7 =d 7,4 =d 4,5 +d 5,6 +d 6,7 E1 = {(4,7), (9,1)};

[0063] ②Merge the constraint endpoints in E1: For (4,7)∈E1, let d 4,7 =d 7,4 Merge points 4 and 7 into 7, then add point 7. Let d 4,i =d 7,i (i∈V and i≠4, i≠7); for (9,1)∈E1, let d 9,1 =d 1,9 Merge points 9 and 1 to form 9, then add point 1. Let d 9,i =d 1,i (i∈V and i≠9, i≠1)

[0064] ③Establishment of ATSP point set V1 and weight matrix C: V = {1,2,3,4,5,6,7,8,9} Then we have V1 = {2,3,4,8,9}, n = 5, and the 5th order matrix C is shown in equation (7):

[0065]

[0066] Step 4: Convert ATSP to TSP

[0067] The calculation yields M = 2(c) max -c min )+1=2×(30-1)+1=69;

[0068] ① Order Again

[0069] ②Let U be a 5-dimensional square matrix, where all values ​​are INT_MAX / 2;

[0070] ③ Order V2={2,3,4,8,9,2+9,…,9+9};

[0071] ④ W is a 10-order matrix, and all values ​​in the matrix are 69.

[0072] Step 5, Solving the LLTSP problem

[0073] ① The transformed point set V2 and matrix As input to the unconstrained TSP solver, the optimal sequence of TSP paths is obtained as L:

[0074] 3→(3+9)→2→(2+9)→…→8→(8+f)→3;

[0075] ②Remove all extended nodes from L to obtain sequence L2: 3→2→4→9→8→3;

[0076] ③E1={(4,7), (9,1)}, replace 4 in sequence L2 with 4→7 and replace 9 in sequence L2 with 9→1 to get the sequence 3→2→4→7→9→1→8→3; S1={[4,5,6,7]}, replace 4→7 in sequence L2 with 4→5→6→7 to get the sequence L3: 3→2→4→5→6→7→9→1→8→3 is the solution of LLTSP.

[0077] ④ Break 9→1 in L3, starting from 1, the access path of the entire page mark is L4: 1→8→3→2→4→5→6→7→9.

[0078] 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 planning the marking path of a whole sheet of strain gauges, characterized in that, Includes the following steps: Step 1: Description and introduction of problem constraints: Planning the marking path for the entire strain gauge plate is equivalent to solving the LLTSP. The coordinates of the starting point, ending point, and defective individual piece in the entire strain gauge plate correspond to nodes in the LLTSP. The access time between any two coordinates corresponds to the distance weight between corresponding nodes in the LLTSP. The edge formed by the left and right adjacent defective individual pieces, the edge formed by the starting point and the ending point in the entire strain gauge plate correspond to the constraint edge in the LLTSP. The LLTSP is a line flow constraint TSP. Step 2, Establishing the LLTSP problem: via a set of points A matrix A set of constraint edges E1 is used to describe an LLTSP problem, which serves as the input for the problem transformation; Step 3: Convert LLTSP to ATSP: To eliminate the constraints on set E1, the point set... Convert to ATSP point set , matrix The set E1 is transformed into an asymmetric weight matrix of ATSP. ; Step 4: Convert ATSP to TSP: Point set Point set transformed into TSP The asymmetric weight matrix Transform into a symmetric weight matrix with no negative elements of TSP ; Step 5, Solving the LLTSP problem: First, set the points of the TSP. and asymmetric weight matrix Input the TSP solver to obtain the TSP solution result, then convert the TSP solution result into the ATSP solution result, and then convert the ATSP solution result into the LLTSP solution result. The LLTSP solution result is the closed-loop path of the entire strain gauge marking. Disconnect the edge connecting the node at the end point of the entire strain gauge marking and the starting node. Starting from the starting point, follow the path to the end point of the marking, which is the access path of the entire strain gauge marking.

2. The method for planning the marking path of a whole-plate strain gauge according to claim 1, characterized in that, The specific method for step two is as follows: Let the two endpoints of a constraint edge in an LLTSP be... and Then a directed edge is represented by two endpoints, i.e. Let represent a directed edge pointing from endpoint a to endpoint b; then the constraint represented by the endpoint representation of the edge is: Access is required; Let the set of all nodes in the LLTSP problem be . And number its nodes, i.e., defective pieces, that is, set them as follows: , The number of nodes, i.e., the number of defective individual pieces, for Establishing LLTSP 1-th order matrix : ； in, Represents a node To the node Distance weights; Establish a set E1 to store the corresponding constraint edges, such as Access is required, therefore... ∈E1, ultimately the LLTSP problem is solved using point sets. ,matrix The constraint edge set E1 is used for representation.

3. The method for planning the marking path of a whole-plate strain gauge according to claim 2, characterized in that, The specific method for step three is as follows: Create a set S1 to store the sub-paths formed by merging multiple adjacent edges in E1; create a set of deleted nodes. This is used to store the intermediate nodes and right endpoints of the sub-paths in S1. The initial value is empty; the specific steps include the following: ① Merge consecutive adjacent constraint edges in E1: For all , Indicates from The number of nodes taken from the middle, Represents a set Any k unique nodes in the array; if there are ∈E1, point join in ,Will Delete from E1, ,join in Subpath Add S1, making , Represents a node To the node Distance weights, Represents a node To the node Distance weights, ; ②Merge the constraint endpoints in E1: For all Let ∈E1, Merging point for , will point join in ,make and ; ③ATSP point set and weight matrix The establishment of: for exist The complement of the set, i.e. Indicates all belonging to But not belonging to Let the set of points be . The number of midpoints is Then there is ; weight matrix For point set The distance weight matrix for all points is shown in the following equation. make ; 。 4. The method for planning the marking path of a whole-plate strain gauge according to claim 3, characterized in that, The specific method for step four is as follows: ① Let the temporary transformation matrix be set. , and then , For matrix The diagonal elements; ② Order For one 3D matrix, matrix The values ​​of all elements except the diagonal are... , The maximum integer value that the computer installed on the TSP solver can recognize; ③ Order , ; ④ It is and Matrices of the same order, matrix All elements except the diagonal are M, and the values ​​of M are as follows: ； in, for The maximum weight in, for The smallest weight excluding the diagonal elements.

5. The method for planning the marking path of a whole-plate strain gauge according to claim 4, characterized in that, The specific method for step five is as follows: ① Transformed distance matrix As input to the unconstrained TSP solver, the optimal sequence of TSP paths is obtained as L, as shown in the following equation; ； ②Remove all extended nodes from L to obtain sequence L2, as shown in the following formula; ； ③ First, consider the sequence of all elements in set E1. ,Will Replacement sequence L2 Then, for all element sequences in set S1 ,Will Replacement sequence L2 Thus, the solution for sequence L3 is an LLTSP; ④ Disconnect the edge connecting the node at the end of the full-sheet strain gauge mark in L3 with the node at the start point. Start from the start point and follow the path to the end of the mark. This is the access path L4 for the full-sheet strain gauge mark.

Images