Method for automatically planning laser cutting path of product with complex shape

Abstract

The invention discloses an automatic planning method for a laser cutting path of a product in a complex shape. The automatic planning method comprises the following steps that S1, the position and the shape of the product are confirmed through image recognition, and internal holes are recognized; s2, acquiring original cutting path data; s3, re-planning a cutting path based on a multidirectional projection matching iterative correction algorithm to form new cutting path data; s4, the new cutting path data is confirmed manually; compared with the prior art, the method has the advantages that through the design of combining the random offset and the asymmetric distribution of the random numbers, spiral optimization can be carried out in one direction after multiple planning failures, each position can be traversed in the maximum range, and the occurrence of a local optimal solution is avoided.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing, and in particular to a method for automatically planning a laser cutting path for a product with a complex shape. Background Art

[0002] Existing laser cutting equipment usually requires products to be placed at a fixed position and angle during the cutting process to ensure the accuracy of the cutting path. And there are the following problems:

[0003] 1. As the industry has high requirements for cutting accuracy, there are also strict requirements for the placement angle when cutting products. However, the existing cutting equipment is placed at a fixed position and angle, which cannot meet the requirements of cutting accuracy.

[0004] 2. Using mechanical equipment for precise placement requires high investment, while manual placement is less efficient and difficult to control in terms of accuracy;

[0005] 3. Existing equipment can only cut according to the preset path and cannot flexibly adjust the path according to different shapes, especially when the product has a complex shape or internal holes;

[0006] 4. For products with holes or complex shapes, existing technologies make it difficult to effectively plan cutting paths, resulting in material waste. Summary of the invention

[0007] The purpose of the present invention is to address the problems that the existing cutting technology has fixed angles and positions, and can only cut according to a preset path, cannot meet the cutting accuracy requirements, and cannot flexibly adjust the path, resulting in material waste. To address this shortcoming, a method for automatic planning of laser cutting paths for products with complex shapes is proposed.

[0008] In order to achieve the above object, the present invention adopts the following technical solution:

[0009] A method for automatically planning a laser cutting path for a product of complex shape comprises the following steps:

[0010] S1. Confirm the product position, shape and identify internal holes through image recognition;

[0011] S2, obtaining original cutting path data;

[0012] S3, re-planning the cutting path based on the multi-directional projection matching iterative correction algorithm to form new cutting path data;

[0013] S4, manually confirming the new cutting path data;

[0014] S5. Execute cutting.

[0015] As a further preferred embodiment of the present invention, S3 comprises the following steps:

[0016] S31, counting the number of cuts that need to be planned, and performing regional planning to form a plannable area;

[0017] S32, determining whether the plannable area is to be planned for a simple situation;

[0018] S33. If the simple planning of the current area fails, then complex situation planning is performed;

[0019] S34. After the planning is completed, the planning data is returned to manual confirmation.

[0020] As a further preferred embodiment of the present invention, S31 comprises the following steps:

[0021] Determine whether the number of plannable areas is equal to the number of areas that need to be cut;

[0022] If they are equal, each region is cut;

[0023] If the number of plannable areas is greater than the number of areas that need to be cut, the planning will be carried out in order from the largest to the smallest area of ​​the plannable areas;

[0024] If the current plannable area planning fails, then plan the next plannable area;

[0025] When all plannable areas have been planned, the planning data will be returned to manual confirmation;

[0026] If the number of plannable areas is less than the number of areas that need to be cut, then there are some plannable areas that require planning two cut areas;

[0027] Select the largest plannable area for planning and determine whether two cutting areas can be planned;

[0028] After the planning is successful, determine whether the current number of plannable areas is equal to the number that needs to be cut;

[0029] If they are equal, each region is cut;

[0030] If the planning fails, the planning and quantity judgment will be carried out in order from large to small areas until all planning is completed, and the planning data will be returned to manual confirmation.

[0031] As a further preferred embodiment of the present invention, the simple case planning includes:

[0032] Planning attempts were made from horizontal and vertical perspectives from the geometric center of the area;

[0033] If the planning fails, the center position of the product's circumscribed circle is planned;

[0034] If the planning is successful, the planning data will be returned to manual confirmation;

[0035] If planning fails, proceed to complex situation planning.

[0036] As a further preferred embodiment of the present invention, the complex situation planning includes:

[0037] Select the location of the last planning and calculate whether the planned area and the product overlap;

[0038] If the planning area and the product do not overlap, a point is randomly selected in the product as a new planning point for planning;

[0039] Select 8 directions as basic optimization directions;

[0040] If the planned area and product overlap, follow-up planning will be carried out;

[0041] A point is randomly selected as the initial point in the overlapped part, and optimization is performed in 8 directions to find the optimal iteration direction;

[0042] according to

[0043] a ni =0°+45°*n+ɑ i ,(ɑ∈[-1°,2°],n∈[0,7]),ɑ i =ɑ+ɑ i-1

[0044] L=10+θ i ,(θ∈[-5,10]),θ i =θ+θ i-1 ,

[0045] Among them, ɑ is a random angle, θ is a random step length; the final addressing iteration direction a is calculated n and step length L, i is the i-th iteration; form points in 8 directions, take the 8 points as the center of the circle, and combine the addressing iteration direction to obtain the planning area after iteration;

[0046] Calculate the overlap between the iterative planning area and the product from 8 directions;

[0047] Select the point with the highest overlap in the eight directions as the new initial point;

[0048] Restart the next iteration until the overlap reaches 100% and the planning is successful;

[0049] If the convergence values ​​of the points in the eight directions are all lower than the initial points, the iteration fails;

[0050] Reselect random angles and random step sizes, perform iterations and judgments, until the overlap is equal to 100% or the iteration point with the maximum overlap is obtained;

[0051] After the iteration is completed, the planning data is returned to manual confirmation;

[0052] When the product area is smaller than the planned area and is insufficient to plan the cutting area, after reaching the maximum number of iterations (10,000 times), the planning is considered to have failed and the planning failure result is returned for manual re-planning.

[0053] The method for automatically planning the laser cutting path of a product with a complex shape proposed by the present invention has the following beneficial effects compared with the prior art:

[0054] 1. The present invention uses a design combining random offset and asymmetric distribution of random numbers, which can search for optimal solution in a spiral along one direction after multiple planning failures, and can traverse each position to the maximum extent, thus avoiding the emergence of local optimal solutions.

[0055] 2. After an iteration fails, you can traverse the entire product and find the optimal planning area by jumping out and searching again. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of the automatic planning process of the laser cutting path involved in the present invention;

[0057] Figure 2 It is a schematic diagram of the multi-directional projection matching iterative correction process;

[0058] Figure 3 This is a simple planning diagram;

[0059] Figure 4 It is a schematic diagram of 8 basic addressing directions;

[0060] Figure 5 It is a one-way exploration curve to find the range;

[0061] Figure 6 It is the exploration curve search range in 8 basic directions;

[0062] Figure 7 It is a 500-times random exploration curve point graph in one direction;

[0063] Figure 8 It is a schematic diagram of planning paths for complex situations;

[0064] Fig. 9 It is another schematic diagram of planning a path for complex situations;

[0065] Fig.10 It is a schematic diagram of the planning results of a complex situation;

[0066] Fig.11 This is another schematic diagram of planning results for a complex situation;

[0067] Fig.12 It is a schematic diagram of the planning in the horizontal direction of the geometric center of a simple case;

[0068] Fig.13 It is a planning diagram of the vertical direction of the geometric center of a simple case;

[0069] Fig.14 It is a schematic diagram of the planning of the center of the circumscribed circle in a simple case. DETAILED DESCRIPTION

[0070] The present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0071] Combination Figure 1-Figure 14 The present invention relates to an automatic planning method for the laser cutting path of products with complex shapes. The method extracts product position and shape information through image recognition, combines dynamic optimization and adaptive area division technology, and realizes efficient cutting path planning for randomly placed products. The present invention supports dynamic obstacle avoidance, material redundancy retention and multi-objective optimization, and is suitable for cutting products with complex shapes or holes. The present invention significantly improves the intelligence level and flexibility of path planning, reduces production costs, and is suitable for the field of intelligent manufacturing.

[0072] S1. Confirm the product position and shape and identify internal holes through image recognition.

[0073] S2. Obtain original cutting path data.

[0074] S3. Replan the cutting path based on the multi-directional projection matching iterative correction algorithm to form new cutting path data; when replanning the cutting path, it is necessary to avoid the identified holes.

[0075] S31. Count the number of cuts that need to be planned, and perform regional planning to form a plannable area.

[0076] Determine whether the number of plannable areas is equal to the number of areas that need to be cut.

[0077] If they are equal, each region is cut.

[0078] If the number of plannable areas is greater than the number of areas that need to be cut, planning is performed in order from the largest to the smallest area.

[0079] If the current plannable area fails to be planned, the next plannable area is planned.

[0080] Planning failures include the following:

[0081] 1. Unable to find the product: The actual failure to find the product is likely due to the user forgetting to put the product, or the environment is too bright or too dark, forgetting to turn on the light, etc., which can be uniformly classified as user error;

[0082] 2. Insufficient available space for the product: For example, if a 10*10 product is set to be planned as a 15*15 rectangle, the planning will fail. Or if two 8*8 circles are planned on a 10*10 rectangular product, the planning will also fail;

[0083] 3. Special circumstances may also cause this: for example, an 8*8 rectangle is planned on a 10*10 product, but because there is a hole in the center, the user does not find it and avoids the hole during planning, which will result in a failure prompt.

[0084] When all plannable areas are planned, the planning data will be returned to manual confirmation. Manual confirmation will be made to confirm whether the path is planned correctly and whether there are any planning errors. During operation, it will be determined whether all personnel are wearing protective goggles.

[0085] If the number of plannable areas is less than the number of areas that need to be cut, there are some plannable areas that require planning two cutting areas.

[0086] Select the largest plannable area for planning and determine whether two cutting areas can be planned.

[0087] After the planning is successful, determine whether the current number of plannable areas is equal to the number that needs to be cut.

[0088] If they are equal, each region is cut.

[0089] If the planning fails, the planning and quantity judgment will be carried out in order from large to small areas until all planning is completed, and the planning data will be returned to manual confirmation.

[0090] S32. Determine whether the plannable area is to be planned for a simple situation.

[0091] Planning attempts were made from horizontal and vertical perspectives from the geometric center of the area;

[0092] If the planning fails, the center position of the product's circumscribed circle is planned;

[0093] If the planning is successful, the planning data will be returned to manual confirmation.

[0094] If planning fails, proceed to complex situation planning.

[0095] S33. If simple planning of the current area fails, complex situation planning is performed.

[0096] Select the location of the last planning and calculate whether the planned area overlaps with the product. Due to planning failure, the planned area may not completely fall into the product. Therefore, it is necessary to determine whether the planned area overlaps with the product.

[0097] If the planning area and the product do not overlap, a point is randomly selected in the product as a new planning point for planning.

[0098] Eight directions are selected as basic optimization directions.

[0099] If the planned area and product overlap, follow-up planning will be carried out.

[0100] A point is randomly selected in the overlapping part as the initial point, and optimization is performed in 8 directions to find the optimal iteration direction.

[0101] according to

[0102] a ni =0°+45°*n+ɑ i ,(ɑ∈[-1°,2°],n∈[0,7]),ɑ i =ɑ+ɑ i-1

[0103] L=10+θ i ,(θ∈[-5,10]),θ i =θ+θ i-1 ,

[0104] Among them, ɑ is a random angle, θ is a random step length; the final addressing iteration direction a is calculated n And step length L, i is the i-th iteration; form points in 8 directions, take the 8 points as the center of the circle, and combine the addressing iteration direction to obtain the planning area after the iteration.

[0105] The overlap between the iterative planning area and the product is calculated from 8 directions.

[0106] Select the point with the highest overlap in the 8 directions as the new initial point.

[0107] The next iteration is restarted until the overlap reaches 100% and the planning is successful.

[0108] If the convergence values ​​of the points in the eight directions are all lower than the initial points, the iteration fails.

[0109] Reselect the random angle and random step size, and perform iteration and judgment until the overlap is equal to 100% or the iteration point with the maximum overlap is obtained.

[0110] After the iteration is completed, the planning data will be returned to manual confirmation.

[0111] When the product area is smaller than the planned area and is insufficient to plan the cutting area, after reaching the maximum number of iterations (10,000 times), the planning is considered to have failed and the planning failure result is returned for manual re-planning.

[0112] S34. After the planning is completed, the planning data is returned to manual confirmation.

[0113] S4. Manually confirm the new cutting path data.

[0114] S5. Execute cutting.

[0115] As a further preferred embodiment of the present invention, S3 comprises the following steps:

[0116] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the above embodiments do not limit the present invention in any form, and any technical solution obtained by equivalent replacement or equivalent transformation falls within the protection scope of the present invention.

Claims

1. A method for automatically planning the laser cutting path of a product with a complex shape, characterized in that: The following steps are included: S1. Confirm the product position, shape and identify internal holes through image recognition; S2, obtaining original cutting path data; S3, re-planning the cutting path based on the multi-directional projection matching iterative correction algorithm to form new cutting path data; S4, manually confirming the new cutting path data; S5. Execute cutting.

2. The method for automatically planning the laser cutting path of a product with a complex shape according to claim 1, characterized in that: The S3 comprises the following steps: S31, counting the number of cuts that need to be planned, and performing regional planning to form a plannable area; S32, determining whether the plannable area is to be planned for a simple situation; S33. If the simple planning of the current area fails, then complex situation planning is performed; S34. After the planning is completed, the planning data is returned to manual confirmation.

3. The method for automatically planning the laser cutting path of a product with a complex shape according to claim 2, characterized in that: The S31 comprises the following steps: Determine whether the number of plannable areas is equal to the number of areas that need to be cut; If they are equal, each region is cut; If the number of plannable areas is greater than the number of areas that need to be cut, the planning will be carried out in order from the largest to the smallest area of ​​the plannable areas; If the current plannable area planning fails, then plan the next plannable area; When all plannable areas have been planned, the planning data will be returned to manual confirmation; If the number of plannable areas is less than the number of areas that need to be cut, then there are some plannable areas that require planning two cut areas; Select the largest plannable area for planning and determine whether two cutting areas can be planned; After the planning is successful, determine whether the current number of plannable areas is equal to the number that needs to be cut; If they are equal, each region is cut; If the planning fails, the planning and quantity judgment will be carried out in order from large to small areas until all planning is completed, and the planning data will be returned to manual confirmation.

4. The method for automatically planning the laser cutting path of a product with a complex shape according to claim 2, characterized in that: The simple scenario planning includes: Planning attempts were made from horizontal and vertical perspectives from the geometric center of the area; If the planning fails, the center position of the product's circumscribed circle is planned; If the planning is successful, the planning data will be returned to manual confirmation; If planning fails, proceed to complex situation planning.

5. The method for automatically planning the laser cutting path of a product with a complex shape according to claim 2, characterized in that: The complex situation planning includes: Select the location of the last planning and calculate whether the planned area and the product overlap; If the planning area and the product do not overlap, a point is randomly selected in the product as a new planning point for planning; Select 8 directions as basic optimization directions; If the planned area and product overlap, follow-up planning will be carried out; A point is randomly selected as the initial point in the overlapped part, and optimization is performed from 8 directions to find the optimal iteration direction; according to a ni =0°+45°*n+ɑ i ,(ɑ∈[-1°,2°],n∈[0,7]),ɑ i =ɑ+ɑ i-1 L=10+θ i ,(θ∈[-5,10]),θ i =θ+θ i-1 , Among them, ɑ is a random angle, θ is a random step length; the final addressing iteration direction a is calculated n and step length L, i is the i-th iteration; form points in 8 directions, take the 8 points as the center of the circle, and combine the addressing iteration direction to obtain the planning area after iteration; Calculate the overlap between the iterative planning area and the product from 8 directions; Select the point with the highest overlap in the eight directions as the new initial point; Restart the next iteration until the overlap reaches 100% and the planning is successful; If the convergence values ​​of the points in the eight directions are all lower than the initial points, the iteration fails; Reselect random angles and random step sizes, perform iterations and judgments, until the overlap is equal to 100% or the iteration point with the maximum overlap is obtained; After the iteration is completed, the planning data is returned to manual confirmation; When the product area is smaller than the planned area and is insufficient to plan the cutting area, the maximum number of iterations is reached and the planning is deemed to have failed. The planning failure result is returned for manual re-planning.

Images