Automatic grinding decision planning system and method for intelligent grinding robot

Through image processing and genetic algorithms, the grinding area and parameters are determined, which solves the problem that existing intelligent grinding robots cannot dynamically adjust grinding parameters, and achieves a more efficient grinding effect.

CN118990134AActive Publication Date: 2024-11-22DONGFANG ELECTRIC CHENGDU INTELLIGENT TECH CO LTD +1

Patent Information

Application Number
CN202411060494.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-02
Publication Date
2024-11-22
Estimated Expiration
2044-08-02

AI Technical Summary

Technical Problem

Existing intelligent grinding robots cannot dynamically adjust grinding parameters according to the characteristics and defects of different workpiece surfaces, resulting in poor grinding effect.

Method used

Through image acquisition and processing, the grinding area on the surface of the workpiece and its roughness are determined, and the optimal grinding parameter combination is set using a genetic algorithm, and the grinding parameters are dynamically adjusted to achieve automatic grinding.

Benefits of technology

The grinding effect is improved, ensuring the parameters of each grinding area are optimized, and the fixed values ​​depend on by manual experience are avoided, and the dynamic adjustment ability of the grinding process is enhanced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118990134A_ABST
    Figure CN118990134A_ABST
Patent Text Reader

Abstract

The invention discloses an automatic polishing decision planning system and method for an intelligent polishing robot, and relates to the technical field of intelligent polishing robots, and the system comprises an image collection module which is used for shooting the surface of a workpiece to be polished, and obtaining an image of the surface of the workpiece; the image processing module is used for determining the height of each point on the surface of the workpiece in the image, determining a polishing area according to the height of each point on the surface of the workpiece, and determining the roughness degree of each polishing area; the polishing decision module is used for determining an optimal polishing parameter combination according to the roughness degree in the polishing area; and the execution module is used for automatically polishing each polishing area on the surface of the workpiece according to the optimal polishing parameter combination. Polishing parameters can be dynamically adjusted according to the roughness degree of the workpiece surface, and the polishing effect is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent polishing robots, and in particular relates to an automatic polishing decision-making planning system and a planning method for an intelligent polishing robot. Background Art

[0002] Grinding is a common surface treatment process that can improve the surface quality and geometric accuracy of workpieces, remove surface defects and oxide layers, and increase the service life and performance of workpieces. The grinding process usually requires manual operation, is time-consuming and labor-intensive, and has safety hazards and unstable quality.

[0003] A grinding robot is an automated device specially designed for surface grinding, polishing, deburring, etc. They are widely used in the manufacturing industry, especially in the surface treatment process of materials such as metal, plastic, and wood. Grinding robots usually have the following characteristics: High precision: they can accurately control the force and position of grinding to ensure consistent surface treatment quality for each workpiece; High efficiency: compared with manual grinding, robots can work continuously, significantly improving production efficiency.

[0004] Existing intelligent grinding robots usually use pre-set grinding strategies for grinding. Although they can make relatively accurate grinding decisions, since grinding parameters such as grinding speed, grinding force, angle, etc. are often fixed values ​​or ranges obtained based on experience or experiments, they cannot be dynamically adjusted according to different workpiece surface characteristics and defects, resulting in poor grinding effects.

[0005] For example, the Chinese patent document, publication number CN116533103A, publication date August 4, 2023, is titled Invention Patent for Grinding Actuator, Grinding Robot and Grinding Method. The grinding actuator in the prior art can be installed on a general robot for automated grinding. During the grinding process, the grinding actuator can adjust the grinding pressure by itself to achieve constant force grinding.

[0006] The above-mentioned prior art sets a preset grinding pressure, and always maintains the preset grinding pressure through the grinding actuator. However, when grinding different workpiece surfaces, the grinding pressures borne by the workpiece surfaces are different. Therefore, the prior art cannot be dynamically adjusted according to different workpiece surfaces, resulting in poor grinding effect. Summary of the invention

[0007] In order to solve the above-mentioned deficiencies of the prior art, the present invention provides an automatic polishing decision-making planning system and a planning method for an intelligent polishing robot, which can dynamically adjust the polishing parameters according to the roughness of the workpiece surface to improve the polishing effect.

[0008] The present invention is achieved through the following technical solutions:

[0009] The automatic polishing decision-making and planning method of an intelligent polishing robot is characterized by comprising the following steps:

[0010] S1. photographing the surface of the workpiece to be polished to obtain an image of the workpiece surface;

[0011] S2, processing the image in step S1, determining the height of each point on the workpiece surface in the image, determining the grinding area according to the height of each point on the workpiece surface, and determining the roughness of each grinding area, where the roughness is the average value of the heights of all points in the grinding area;

[0012] S3, setting grinding parameters and grinding parameter ranges according to the roughness in the grinding area, and determining the optimal grinding parameter combination through a genetic algorithm;

[0013] S4. According to the optimal grinding parameter combination, the intelligent grinding robot is controlled to automatically grind each grinding area on the surface of the workpiece.

[0014] Furthermore, step S2 specifically includes:

[0015] S2.1. Draw a circle with point i as the center and preset radius R, and calculate the height difference HDa=|Ha'-Hi|, where Ha' represents the height of the ath point in the circle, and Hi represents the height of point i. When the height difference is greater than or equal to the preset difference threshold, subtract the preset radius correction value ΔR from the preset radius R to obtain a new radius R1;

[0016] S2.2, repeating step S2.1 with the newly obtained radius R1 until there is no height difference greater than or equal to the preset difference threshold, and taking the corresponding circular area as a polishing area;

[0017] S2.3. Use the point that is not in the grinding area as the center, preset the radius R to make a circle, and repeat the above steps to determine all the grinding areas and the corresponding roughness.

[0018] Furthermore, in step S2.1, when there is no height difference greater than or equal to the preset difference threshold, the preset radius R is added to the preset radius correction value ΔR to obtain a new radius R2, and the height difference is calculated again until the height difference greater than or equal to the preset difference threshold appears for the first time, and the corresponding circle area at this time is used as a polishing area.

[0019] Furthermore, in step S2.3, a point that does not belong to the polishing area is taken as a pending point, and a circle is drawn with the pending point as the center and a preset radius R. If the circle does not intersect the determined polishing area, the polishing area is determined in the following manner: first, several gradient value ranges are set from low to high, and then the gradient values ​​of several pending points are calculated, and the gradient value of each pending point is placed in the corresponding gradient value range, and the area composed of the pending points in the same gradient value range is determined as the polishing area; the gradient value is calculated by the following formula, K represents the gradient value of the Kth pending point calculated, K=0, 1, 2...n.

[0020]

[0021] Furthermore, step S2 also includes obtaining the roughness of all polished areas and generating a judgment set C = (C1, C2, C3, ... Cn), and obtaining the maximum roughness value C according to the judgment set C. max and the minimum roughness C min , where Cn represents the roughness of the nth polishing area. max When Cfir is less than Cfir, there is no need to grind the workpiece. Cfir represents the preset first roughness threshold. min ≥Csec, an early warning message is sent to report an error, Csec=Cfir+ΔC, where ΔC represents the preset roughness correction value.

[0022] Furthermore, step S3 specifically includes:

[0023] S3.1. Set the grinding parameter range, which includes the grinding force range [Fmin, Fmax], the grinding speed range [Vmin, Vmax] and the grinding time range [Tmin, Tmax]. The grinding force is the force F during grinding, and the grinding speed is the moving speed V of the grinding tool on the workpiece surface, and the grinding time is the grinding time T in each grinding area;

[0024] S3.2. Obtain the roughness within the Ath polishing area, and use the corresponding negative value of the roughness as the fitness function, and determine the optimal polishing parameter combination (FA', VA', TA') within the Ath polishing area according to the genetic algorithm, where FA' represents the optimal polishing force within the Ath polishing area, VA' represents the optimal polishing speed within the Ath polishing area, and TA' represents the optimal polishing time within the Ath polishing area.

[0025] An intelligent polishing robot automatic polishing decision-making planning system that executes any of the above-mentioned intelligent polishing robot automatic polishing decision-making planning methods is characterized by comprising an image acquisition module, an image processing module, a polishing decision module and a polishing execution module,

[0026] The image acquisition module is used to photograph the surface of the workpiece to be polished and obtain an image of the workpiece surface;

[0027] The image processing module processes the image acquired by the image acquisition module, determines the grinding area according to the height of each point on the workpiece surface, and determines the roughness of each grinding area;

[0028] The grinding decision module sets the grinding parameters and grinding parameter range according to the roughness in the grinding area, and determines the optimal grinding parameter combination through the genetic algorithm;

[0029] The grinding execution module automatically grinds each grinding area on the workpiece surface according to the optimal grinding parameter combination.

[0030] The beneficial effects of the present invention are as follows:

[0031] Fix the workpiece on the plane of the workbench or fixture, and first collect the image of the workpiece surface, which is the basis for subsequent processing; it is worth noting that when collecting the height of each point on the workpiece surface, it is a single side, not all surfaces at the same time; then determine the height of each point on the workpiece surface, which is used as a reference standard for subsequent determination of the grinding area and quantitative analysis of the height of the roughness of the workpiece surface. The lower the roughness, the smoother the workpiece surface; then determine the grinding area to ensure that the height of each point in the grinding area is within a range and there will be no large fluctuations, which will lead to deviations in the subsequent roughness and the optimal grinding parameter set determined based on the roughness, affecting the grinding effect; finally, set the grinding parameter range to determine that the final optimal grinding parameter set is within the grinding parameter range and will not cause damage to the workpiece, and determine the optimal grinding parameter combination in each grinding area based on the genetic algorithm, and automatically grind each grinding area on the workpiece surface based on the optimal grinding parameter combination. The grinding parameters can be dynamically adjusted according to the roughness of the workpiece surface to improve the grinding effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 It is a flow chart of the present invention. DETAILED DESCRIPTION

[0033] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] Example 1

[0035] The automatic polishing decision-making planning method of the intelligent polishing robot includes the following steps:

[0036] S1. photographing the surface of the workpiece to be polished to obtain an image of the workpiece surface;

[0037] S2, processing the image in step S1, determining the height of each point on the workpiece surface in the image, determining the grinding area according to the height of each point on the workpiece surface, and determining the roughness of each grinding area, where the roughness is the average value of the heights of all points in the grinding area;

[0038] S3, setting grinding parameters and grinding parameter ranges according to the roughness in the grinding area, and determining the optimal grinding parameter combination through a genetic algorithm;

[0039] S4. According to the optimal grinding parameter combination, the intelligent grinding robot is controlled to automatically grind each grinding area on the surface of the workpiece.

[0040] During implementation, the workpiece is fixed on the plane of a workbench or a fixture, and the image of the workpiece surface is first collected, which is the basis for subsequent processing; it is worth noting that when collecting the height of each point on the workpiece surface, it is a single side, rather than all surfaces at the same time; then the height of each point on the workpiece surface is determined, which is used as a reference standard for subsequent determination of the grinding area and quantitative analysis of the roughness height of the workpiece surface. The lower the roughness, the smoother the workpiece surface; then the grinding area is determined to ensure that the height of each point in the grinding area is within a range and there will be no large fluctuations, which will lead to deviations in the subsequent roughness and the optimal grinding parameter set determined according to the roughness, affecting the grinding effect; finally, the grinding parameter range is set to determine that the final optimal grinding parameter set is within the grinding parameter range and will not cause damage to the workpiece, and the optimal grinding parameter combination in each grinding area is determined according to the genetic algorithm, and automatic grinding is achieved for each grinding area on the workpiece surface according to the optimal grinding parameter combination.

[0041] Example 2

[0042] This embodiment further elaborates and supplements the implementation mode of the present invention on the basis of Embodiment 1.

[0043] The S2 step specifically includes:

[0044] S2.1. Draw a circle with point i as the center and preset radius R, and calculate the height difference HDa=|Ha'-Hi|, where Ha' represents the height of the ath point in the circle, and Hi represents the height of point i. When the height difference is greater than or equal to the preset difference threshold, subtract the preset radius correction value ΔR from the preset radius R to obtain a new radius R1;

[0045] S2.2, repeating step S2.1 with the newly obtained radius R1 until there is no height difference greater than or equal to the preset difference threshold, and taking the corresponding circular area as a polishing area;

[0046] S2.3. Use the point that is not in the grinding area as the center, preset the radius R to make a circle, and repeat the above steps to determine all the grinding areas and the corresponding roughness.

[0047] In implementation, the difference between the height of each point in the circle and the height of the center of the circle is calculated, and when the absolute value of the difference is greater than or equal to the preset difference threshold, the radius of the circle is adjusted to ensure that the heights of all points in the same circle are within a certain range, thereby avoiding excessive height differences in a polishing area, resulting in deviations in the subsequent roughness and the optimal polishing parameter set determined based on the roughness, affecting the polishing effect.

[0048] As another implementation of this embodiment, in step S2.1, when there is no height difference greater than or equal to the preset difference threshold, the preset radius R is added to the preset radius correction value ΔR to obtain a new radius R2, and the height difference is calculated again until the height difference greater than or equal to the preset difference threshold appears for the first time, and the corresponding circle area at this time is used as a polishing area.

[0049] In actual situations, the height within the circle corresponding to the initially set radius range may be relatively uniform. By gradually increasing the radius, the maximized polishing area without obvious height differences can be determined, reducing the amount of data that needs to be processed later.

[0050] As another implementation of this embodiment, in step S2.3, a point that does not belong to the polishing area is taken as a pending point, and a circle is drawn with the pending point as the center and a preset radius R. If the circle does not intersect with the determined polishing area, the polishing area is determined in the following manner: first, several gradient value ranges are set from low to high, and then the gradient values ​​of several pending points are calculated, and the gradient value of each pending point is placed in the corresponding gradient value range, and the area composed of the pending points in the same gradient value range is determined as the polishing area; the gradient value is calculated by the following formula, K represents the gradient value of the Kth pending point calculated, K=0, 1, 2...n.

[0051]

[0052] The gradient is a vector whose components are the partial derivatives of the function in all directions. It can reflect the rate of change of the surface height of the workpiece. A larger gradient value indicates that the surface height changes more dramatically at that point, which means that the height at that point is larger. If the height itself is small, even if it suddenly drops to 0, there will not be a large gradient value. A smaller gradient value indicates that the height changes slowly.

[0053] As another implementation of this embodiment, step S2 also includes obtaining the roughness of all polishing areas and generating a judgment set C = (C1, C2, C3, ... Cn), and obtaining the maximum roughness value C according to the judgment set C. max and the minimum roughness C min , where Cn represents the roughness of the nth polishing area. max When Cfir is less than Cfir, there is no need to grind the workpiece. Cfir represents the preset first roughness threshold. min ≥Csec, an early warning message is sent to report an error, Csec=Cfir+ΔC, where ΔC represents the preset roughness correction value.

[0054] It is understandable that when C max When <Cfir, it means that the workpiece has met the standard and no grinding is required; when C min When ≥Csec, it means that the roughness of the workpiece is too high, which may be caused by the fact that the previous process did not meet the standards, so an early warning message is sent to remind the staff.

[0055] Example 3

[0056] This embodiment further elaborates and supplements the implementation mode of the present invention on the basis of Embodiment 1 or Embodiment 2.

[0057] The S3 step specifically includes:

[0058] S3.1. Set the grinding parameter range, which includes the grinding force range [Fmin, Fmax], the grinding speed range [Vmin, Vmax] and the grinding time range [Tmin, Tmax]. The grinding force is the force F during grinding, and the grinding speed is the moving speed V of the grinding tool on the workpiece surface, and the grinding time is the grinding time T in each grinding area;

[0059] S3.2. Obtain the roughness within the Ath polishing area, and use the corresponding negative value of the roughness as the fitness function, and determine the optimal polishing parameter combination (FA', VA', TA') within the Ath polishing area according to the genetic algorithm, where FA' represents the optimal polishing force within the Ath polishing area, VA' represents the optimal polishing speed within the Ath polishing area, and TA' represents the optimal polishing time within the Ath polishing area.

[0060] The fitness function represents the survival ability or superiority of an individual in the population. The lower the roughness, the smoother the workpiece surface and the better the grinding effect. Therefore, in order to maximize the fitness, the negative value of the roughness is used as the fitness function. The natural selection process is simulated by the genetic algorithm. The grinding parameter combination is continuously optimized from the initial random solution to gradually obtain the optimal grinding parameter combination.

[0061] Example 4

[0062] An intelligent polishing robot automatic polishing decision-making planning system that executes the intelligent polishing robot automatic polishing decision-making planning method described in any of the above embodiments includes an image acquisition module, an image processing module, a polishing decision module and a polishing execution module.

[0063] The image acquisition module is used to photograph the surface of the workpiece to be polished and obtain an image of the workpiece surface;

[0064] The image processing module processes the image acquired by the image acquisition module, determines the grinding area according to the height of each point on the workpiece surface, and determines the roughness of each grinding area;

[0065] The grinding decision module sets the grinding parameters and grinding parameter range according to the roughness of the grinding area, and determines the optimal grinding parameter combination through the genetic algorithm.

[0066] The grinding execution module automatically grinds each grinding area on the workpiece surface according to the optimal grinding parameter combination.

[0067] As an example, during implementation, the image acquisition module can be set above the workbench or the fixture, or the image acquisition module can be set on the grinding arm of the intelligent grinding robot, and each surface of the workpiece is collected separately. After all the surfaces are collected, they are transmitted to the image processing module for image processing, and each image is processed separately to obtain the height of each point on the workpiece surface to determine the grinding area, and determine the roughness of each grinding area, and obtain the roughness of all grinding areas. When the maximum roughness is less than the preset first roughness threshold, there is no need to grind the workpiece. If the minimum roughness is greater than or equal to the sum of the preset first roughness threshold and the preset roughness correction value, an early warning message is sent to report an error. If none of the above two situations occurs, the grinding decision module sets the grinding parameters and the grinding parameter range according to the roughness in the grinding area, and determines the optimal grinding parameter combination for each grinding area through a genetic algorithm. The grinding execution module controls the intelligent grinding robot to automatically grind each grinding area on the workpiece surface.

Claims

1. Automatic polishing decision-making and planning method of intelligent polishing robot, Features: The following steps are included: S1. photographing the surface of the workpiece to be polished to obtain an image of the workpiece surface; S2, processing the image in step S1, determining the height of each point on the workpiece surface in the image, determining the grinding area according to the height of each point on the workpiece surface, and determining the roughness of each grinding area, where the roughness is the average value of the heights of all points in the grinding area; S3, setting grinding parameters and grinding parameter ranges according to the roughness in the grinding area, and determining the optimal grinding parameter combination through a genetic algorithm; S4. According to the optimal grinding parameter combination, the intelligent grinding robot is controlled to automatically grind each grinding area on the surface of the workpiece.

2. The automatic polishing decision-making planning method of the intelligent polishing robot according to claim 1, Features: The S2 step specifically includes: S2.

1. Draw a circle with point i as the center and preset radius R, and calculate the height difference HDa=|Ha'-Hi|, where Ha' represents the height of the ath point in the circle, and Hi represents the height of point i. When the height difference is greater than or equal to the preset difference threshold, subtract the preset radius correction value ΔR from the preset radius R to obtain the new radius R. 1 ; S2.2, with the newly obtained radius R 1 Repeat step S2.1 until there is no height difference greater than or equal to the preset difference threshold, and the corresponding circular area at this time is regarded as a polishing area; S2.

3. Use the point that is not in the grinding area as the center, preset the radius R to make a circle, and repeat the above steps to determine all the grinding areas and the corresponding roughness.

3. The automatic polishing decision-making planning method of the intelligent polishing robot according to claim 2, Features: In step S2.1, when there is no height difference greater than or equal to the preset difference threshold, the preset radius R is added to the preset radius correction value ΔR to obtain a new radius R 2 , and calculate the height difference again until the height difference is greater than or equal to the preset difference threshold for the first time, and the corresponding circle area at this time is used as a polishing area.

4. The automatic polishing decision-making planning method of the intelligent polishing robot according to claim 2, Features: In step S2.3, a point that does not belong to the polishing area is taken as a pending point, and a circle is drawn with the pending point as the center and a preset radius R. If the circle does not intersect with the determined polishing area, the polishing area is determined in the following way: first, several gradient value ranges are set from low to high, and then the gradient values ​​of several pending points are calculated, and the gradient value of each pending point is placed in the corresponding gradient value range, and the area composed of the pending points in the same gradient value range is determined as the polishing area; the gradient value is calculated by the following formula, K represents the calculated gradient value of the Kth pending point, K = 0, 1, 2...n.

5. The automatic polishing decision-making planning method of the intelligent polishing robot according to claim 1, Features: In step S, the roughness of all polished areas is obtained and a judgment set C = (C1, C2, C3, ... Cn) is generated. The maximum roughness value C is obtained according to the judgment set C. max and the minimum roughness C min , where Cn represents the roughness of the nth polishing area. max When <Cfir, there is no need to grind the workpiece, and Cfir represents the preset first roughness threshold; When there is C min ≥Csec, an early warning message is sent to report an error, Csec=Cfir+ΔC, where ΔC represents the preset roughness correction value.

6. The automatic polishing decision-making planning method of the intelligent polishing robot according to claim 1, Features: The S3 step specifically includes: S3.

1. Set the grinding parameter range, which includes the grinding force range [Fmin, Fmax], the grinding speed range [Vmin, Vmax] and the grinding time range [Tmin, Tmax]. The grinding force is the force F during grinding, and the grinding speed is the moving speed V of the grinding tool on the workpiece surface, and the grinding time is the grinding time T in each grinding area; S3.

2. Obtain the roughness within the Ath polishing area, and use the corresponding negative value of the roughness as the fitness function, and determine the optimal polishing parameter combination (FA', VA', TA') within the Ath polishing area according to the genetic algorithm, where FA' represents the optimal polishing force within the Ath polishing area, VA' represents the optimal polishing speed within the Ath polishing area, and TA' represents the optimal polishing time within the Ath polishing area.

7. An intelligent polishing robot automatic polishing decision-making planning system that executes the intelligent polishing robot automatic polishing decision-making planning method according to any one of claims 1 to 6 above, Features: It includes image acquisition module, image processing module, polishing decision module and polishing execution module. The image acquisition module is used to photograph the surface of the workpiece to be polished and obtain an image of the workpiece surface; The image processing module processes the image acquired by the image acquisition module, determines the grinding area according to the height of each point on the workpiece surface, and determines the roughness of each grinding area; The grinding decision module sets the grinding parameters and grinding parameter range according to the roughness of the grinding area, and determines the optimal grinding parameter combination through the genetic algorithm. The grinding execution module automatically grinds each grinding area on the workpiece surface according to the optimal grinding parameter combination.

Citation Information

Patent Citations

  • Polishing actuator, polishing robot and polishing method

    CN116533103A

  • Visual positioning method and system of robot for polishing local discontinuous area

    CN114913229A

  • Grinding machining method and system for hardware tool

    CN116551475A

  • Robot polishing path automatic planning method based on three-dimensional point cloud data

    CN116841246A

  • Automatic workpiece grinding device and method

    CN117226655A

Cited By

  • Intelligent robot control system of grinding wheel deburring workstation

    CN120116074A

  • Autonomous detection method and system for grinding surface of precision workpiece

    CN121061669A

  • Strip steel polishing robot path optimization system, method, device and equipment

    CN121361089A

  • Strip steel polishing robot path optimization system, method, device and apparatus

    CN121361089B