Intelligent robot control system of grinding wheel deburring workstation

Abstract

The invention discloses an intelligent robot control system of a grinding wheel deburring workstation, and belongs to the technical field of grinding control. The method comprises the following steps: firstly, carrying out gradient operation on a surface image of a polished area in x and y directions, calculating texture roughness, and extracting surface glossiness; by analyzing the texture progress difference value and the glossiness change in the x direction and the y direction, the progress proportionality coefficient is extracted, and the polishing speed is adjusted in real time according to the progress proportionality coefficient. According to the multi-dimensional intelligent control method, the accuracy and adaptability of the polishing process are remarkably improved, and the problem that in a traditional polishing technology, parameters cannot be dynamically adjusted according to the machining progress is solved.

Description

Technical Field

[0001] The invention relates to the technical field of grinding control, and in particular to an intelligent robot control system of a grinding wheel deburring workstation. Background Art

[0002] In the field of industrial manufacturing, deburring with grinding wheels is a crucial process, and its quality directly affects the final performance and appearance of the product. Traditional grinding operations often rely on manual operations, which are not only inefficient, but also difficult to ensure the consistency and accuracy of the grinding effect. With the continuous expansion of industrial production scale and increasingly stringent requirements for product quality, this labor-intensive grinding method has gradually exposed many disadvantages, such as high labor intensity leading to worker fatigue, which in turn causes fluctuations in grinding quality, and when processing complex-shaped workpieces, it is difficult for humans to accurately control the grinding force and direction. In addition, the traditional grinding process lacks real-time, accurate monitoring and quantitative evaluation of the surface state of the grinding area, and cannot dynamically adjust the grinding parameters according to the progress of grinding. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides an intelligent robot control system for a grinding wheel deburring workstation, which solves the problem that the prior art cannot dynamically adjust the grinding parameters according to the progress of grinding.

[0004] In order to achieve the above-mentioned invention object, the technical solution adopted by the present invention is: an intelligent robot control system for a grinding wheel deburring workstation, comprising: a texture roughness calculation subsystem, a surface gloss extraction subsystem, a progress difference extraction subsystem, a progress proportional coefficient extraction subsystem and a multi-unit speed adjustment subsystem;

[0005] The texture coarseness calculation subsystem is used to perform gradient calculations on the surface image of the polishing area in the x and y directions respectively, obtain the x-direction gradient image and the y-direction gradient image, and calculate the x-direction texture coarseness and the y-direction texture coarseness;

[0006] The surface gloss extraction subsystem is used to extract the surface gloss from the surface image of the polishing area;

[0007] The progress difference extraction subsystem is used to extract the texture progress difference in the x direction, the texture progress difference in the y direction and the glossiness progress difference according to the texture coarseness in the x direction, the texture coarseness in the y direction and the surface glossiness;

[0008] The progress scale coefficient extraction subsystem is used to extract the x-direction texture progress scale coefficient, the y-direction texture progress scale coefficient and the glossiness progress scale coefficient according to the x-direction texture progress difference, the y-direction texture progress difference and the glossiness progress difference;

[0009] The multi-unit speed adjustment subsystem is used to adjust the polishing speed according to the texture progress proportional coefficients in the x-direction, y-direction, and glossiness progress proportional coefficients at multiple moments.

[0010] Furthermore, the texture roughness calculation subsystem includes: an x-direction gradient image construction unit, a y-direction gradient image construction unit, an x-direction texture roughness calculation unit, and a y-direction texture roughness calculation unit;

[0011] The x-direction gradient image construction unit is used to perform a gradient operation on the surface image of the polishing area in the x-direction using the Sobel operator to obtain the x-direction gradient amplitude, and replace the original pixel value of each pixel point with the x-direction gradient amplitude to obtain the x-direction gradient image;

[0012] The y-direction gradient image construction unit is used to perform a gradient operation on the surface image of the polishing area in the y-direction using the Sobel operator to obtain the y-direction gradient amplitude, and replace the original pixel value of each pixel point with the y-direction gradient amplitude to obtain the y-direction gradient image;

[0013] The x-direction texture roughness calculation unit is used to calculate the x-direction texture roughness according to the number of pixel points in the x-direction gradient image where the x-direction gradient amplitude is equal to 0, and the mean value of the x-direction gradient amplitudes;

[0014] The y-direction texture roughness calculation unit is used to calculate the y-direction texture roughness according to the number of pixel points in the y-direction gradient image where the y-direction gradient amplitude is equal to 0, and the mean value of the y-direction gradient amplitudes.

[0015] Furthermore, the formula for calculating the x-direction texture roughness is:

[0016] , where C x is the x-direction texture roughness, R is the number of pixel points on the surface image of the polishing area, N x is the number of pixel points in the x-direction gradient image where the x-direction gradient amplitude is equal to 0, and G x,avg is the mean value of each x-direction gradient amplitude in the x-direction gradient image;

[0017] The formula for calculating the y-direction texture roughness is:

[0018] , where C y is the y-direction texture roughness, N y is the number of pixel points in the y-direction gradient image where the y-direction gradient amplitude is equal to 0, and G y,avg is the mean value of each y-direction gradient amplitude in the y-direction gradient image.

[0019] Further, the surface glossiness extraction subsystem includes: a color space mapping unit and a surface glossiness calculation unit;

[0020] The color space mapping unit is used to convert the surface image of the polished area into the HSV color space and extract the brightness values in the HSV color space;

[0021] The surface glossiness calculation unit is used to calculate the surface glossiness according to each brightness value.

[0022] Further, the calculation formula for surface glossiness is:

[0023] , where g is the surface glossiness, L avg is the average value of each brightness value of the surface image of the polished area in the HSV color space, and σ L is the standard deviation of each brightness value.

[0024] Further, the progress difference extraction subsystem includes: an x-direction texture progress difference extraction unit, a y-direction texture progress difference extraction unit, and a glossiness progress difference extraction unit;

[0025] The x-direction texture progress difference extraction unit is used to subtract the x-direction texture coarseness at adjacent times, take the absolute value of the subtraction result as the x-direction texture progress value, and subtract the x-direction texture progress value from the x-direction texture progress target value to obtain the x-direction texture progress difference;

[0026] The y-direction texture progress difference extraction unit is used to subtract the y-direction texture coarseness at adjacent times, take the absolute value of the subtraction result as the y-direction texture progress value, and subtract the y-direction texture progress value from the y-direction texture progress target value to obtain the y-direction texture progress difference;

[0027] The glossiness progress difference extraction unit is used to subtract the surface glossiness at adjacent times, take the absolute value of the subtraction result as the glossiness progress value, and subtract the glossiness progress value from the glossiness progress target value to obtain the glossiness progress difference.

[0028] Further, the progress ratio coefficient extraction subsystem includes: an x-direction progress ratio coefficient extraction unit, a y-direction progress ratio coefficient extraction unit, and a glossiness progress ratio coefficient extraction unit;

[0029] The x-direction progress ratio coefficient extraction unit is used to assign a value of 1 to the x-direction progress ratio coefficient when the x-direction texture progress difference is greater than or equal to 0, and when the x-direction texture progress difference is less than 0, the x-direction progress ratio coefficient is equal to the ratio of the x-direction texture progress difference to the x-direction texture progress target value;

[0030] The y - direction progress ratio coefficient extraction unit is used to assign the y - direction texture progress ratio coefficient as 1 when the y - direction texture progress difference is greater than or equal to 0, and when the y - direction texture progress difference is less than 0, the y - direction texture progress ratio coefficient is equal to the ratio of the y - direction texture progress difference to the y - direction texture progress target value;

[0031] The gloss progress ratio coefficient extraction unit is used to assign the gloss progress ratio coefficient as 1 when the gloss progress difference is greater than or equal to 0, and when the y - direction texture progress difference is less than 0, the gloss progress ratio coefficient is equal to the ratio of the gloss progress difference to the gloss progress target value.

[0032] Furthermore, the multi - unit speed adjustment subsystem includes: a first speed adjustment subunit, a second speed adjustment subunit, a third speed adjustment subunit, an addition unit, an increment suppression unit, and a speed output unit;

[0033] The first speed adjustment subunit is used to calculate the first speed increment according to the x - direction texture progress ratio coefficients at multiple moments;

[0034] The second speed adjustment subunit is used to calculate the second speed increment according to the y - direction texture progress ratio coefficients at multiple moments;

[0035] The third speed adjustment subunit is used to calculate the third speed increment according to the gloss progress ratio coefficients at multiple moments;

[0036] The addition unit is used to add the first speed increment, the second speed increment, and the third speed increment to obtain the total increment;

[0037] The increment suppression unit is used to suppress the total increment according to the gap between the total increment and the target increment to obtain the suppressed increment;

[0038] The speed output unit is used to add the suppressed increment and the current grinding speed to obtain the grinding speed at the next moment.

[0039] Furthermore, the expression of the first speed adjustment subunit is:

[0040] , where, V 1,z,t+1 is the first speed increment at the (t + 1) - th moment, V t is the grinding speed at the t - th moment, γ x,t is the x - direction texture progress ratio coefficient at the t - th moment, γ x,t-τ is the x - direction texture progress ratio coefficient at the (t - τ) - th moment, T is the length of the historical moment, t is the number of the current moment, τ is the number of the historical moment, β 1 is the first adjustment factor;

[0041] The expression of the second speed adjustment subunit is:

[0042] , where V 2,z,t+1 is the second velocity increment at the (t + 1)-th moment, and γ y,t is the y-direction texture progress proportionality coefficient at the t-th moment, and γ y,t-τ is the y-direction texture progress proportionality coefficient at the (t - τ)-th moment, and β 2 is the second adjustment factor;

[0043] The expression of the third velocity adjustment subunit is:

[0044] , where V 3,z,t+1 is the third velocity increment at the (t + 1)-th moment, and γ g,t is the glossiness progress proportionality coefficient at the t-th moment, and γ g,t-τ is the glossiness progress proportionality coefficient at the (t - τ)-th moment, and β 3 is the third adjustment factor.

[0045] Furthermore, the expression of the increment suppression unit is:

[0046] , where is the suppression increment at the (t + 1)-th moment, V z,t+1 is the total increment at the (t + 1)-th moment, V z,tar is the target increment, is the suppression increment at the t-th moment, E t+1 is V z,t+1 minus V z,tar , and t is the number of the current moment.

[0047] The beneficial effects of the present invention are as follows:

[0048] 1. By performing gradient operations on the surface image of the grinding area in the x and y directions, the present invention obtains the texture roughness in the x and y directions, enabling the system to sensitively capture the subtle changes in the texture of the grinding area in different directions.

[0049] 2. The present invention directly extracts the glossiness from the surface image of the grinding area, allowing the system to have an intuitive and quantitative evaluation of the grinding effect. Based on this, the progress difference extraction subsystem further combines the texture roughness and glossiness data, calculates the texture progress differences and glossiness progress differences in the x and y directions, and then converts the texture progress differences and glossiness progress differences into texture progress proportionality coefficients and glossiness progress proportionality coefficients to reflect the real-time progress during the grinding process.

[0050] 3. Based on the texture progress proportionality coefficients in the x-direction, y-direction, and gloss progress proportionality coefficient at multiple moments, as well as the texture progress in different directions and the gloss change, the present invention can reflect the grinding state of the workpiece surface from different angles, adjust the grinding speed based on multi-dimensional data, improve the grinding accuracy, and achieve dynamic adjustment of grinding parameters according to the progress. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a system block diagram of an intelligent robot control system for a deburring workstation of a grinding wheel;

[0052] Figure 2 is a schematic structural diagram of a multi-unit speed regulation subsystem. DETAILED DESCRIPTION OF THE INVENTION

[0053] The following describes the specific embodiments of the present invention to facilitate those skilled in the art of the present technology to understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those of ordinary skill in the art of the present technology, as long as various changes are within the spirit and scope of the present invention defined and determined by the appended claims, these changes are obvious, and all inventions made using the inventive concept of the present invention are within the scope of protection.

[0054] As Figure 1 shown, an intelligent robot control system for a deburring workstation of a grinding wheel includes: a texture roughness calculation subsystem, a surface gloss extraction subsystem, a progress difference extraction subsystem, a progress proportionality coefficient extraction subsystem, and a multi-unit speed regulation subsystem;

[0055] The texture roughness calculation subsystem is used to perform gradient operations on the surface image of the grinding area in the x and y directions respectively to obtain the x-direction gradient image and the y-direction gradient image, and calculate the x-direction texture roughness and the y-direction texture roughness;

[0056] The surface gloss extraction subsystem is used to extract the surface gloss of the surface image of the grinding area;

[0057] The progress difference extraction subsystem is used to extract the x-direction texture progress difference, the y-direction texture progress difference, and the gloss progress difference according to the x-direction texture roughness, the y-direction texture roughness, and the surface gloss;

[0058] The progress proportionality coefficient extraction subsystem is used to extract the x-direction texture progress proportionality coefficient, the y-direction texture progress proportionality coefficient, and the gloss progress proportionality coefficient according to the x-direction texture progress difference, the y-direction texture progress difference, and the gloss progress difference;

[0059] The multi-unit speed adjustment subsystem is used to adjust the grinding speed according to the texture progress proportional coefficients in the x direction, the texture progress proportional coefficients in the y direction, and the gloss progress proportional coefficients at multiple moments.

[0060] In this embodiment, the texture roughness calculation subsystem includes: an x-direction gradient image construction unit, a y-direction gradient image construction unit, an x-direction texture roughness calculation unit, and a y-direction texture roughness calculation unit;

[0061] The x-direction gradient image construction unit is used to perform a gradient operation on the surface image of the grinding area in the x direction using the Sobel operator to obtain the x-direction gradient amplitude, and replace the original pixel value of each pixel point with the x-direction gradient amplitude to obtain the x-direction gradient image;

[0062] The y-direction gradient image construction unit is used to perform a gradient operation on the surface image of the grinding area in the y direction using the Sobel operator to obtain the y-direction gradient amplitude, and replace the original pixel value of each pixel point with the y-direction gradient amplitude to obtain the y-direction gradient image;

[0063] The x-direction texture roughness calculation unit is used to calculate the x-direction texture roughness according to the number of pixel points in the x-direction gradient image where the x-direction gradient amplitude is equal to 0, and the mean value of the x-direction gradient amplitude;

[0064] The y-direction texture roughness calculation unit is used to calculate the y-direction texture roughness according to the number of pixel points in the y-direction gradient image where the y-direction gradient amplitude is equal to 0, and the mean value of the y-direction gradient amplitude.

[0065] In this embodiment, the Sobel operator in the x direction: , the Sobel operator in the y direction: .

[0066] During the grinding process, the change of the workpiece surface texture reflects the effect and degree of grinding. The x-direction gradient image construction unit and the y-direction gradient image construction unit respectively perform gradient operations on the surface image of the grinding area in the x and y directions using the Sobel operator. The Sobel operator can highlight the change of pixel values in the image, that is, the edge information of the texture. By obtaining the gradient images in the x direction and the y direction, the details and change situations of the texture in different directions can be clearly captured.

[0067] In this embodiment, the formula for calculating the x-direction texture roughness is:

[0068] , where C x is the x-direction texture roughness, R is the number of pixel points on the surface image of the grinding area, N x is the number of pixel points in the x-direction gradient image where the x-direction gradient amplitude is equal to 0, Gx,avg is the mean value of the x-direction gradient magnitudes in the x-direction gradient image;

[0069] The formula for calculating the y-direction texture coarseness is:

[0070] , where C y is the y-direction texture coarseness, N y is the number of pixel points in the y-direction gradient image where the y-direction gradient magnitude is equal to 0, and G y,avg is the mean value of the y-direction gradient magnitudes in the y-direction gradient image.

[0071] In the present invention, the proportion of pixel points with non-zero gradient magnitudes in the total number of pixel points is calculated to measure the degree of texture change in that direction. A non-zero gradient magnitude means that there is a change in pixel values between pixel points, that is, there are texture undulations. The larger this proportion, the richer the texture change and the rougher the surface. Then, multiplying by the mean gradient magnitude G x,avg (x-direction) or G y,avg (y-direction) further combines the overall gradient magnitude change degree in that direction, so that the calculated texture coarseness can more comprehensively and accurately quantify and reflect the true state of the surface texture.

[0072] In this embodiment, the surface gloss extraction subsystem includes: a color space mapping unit and a surface gloss calculation unit;

[0073] The color space mapping unit is used to convert the surface image of the polished area into the HSV color space and extract the brightness values in the HSV color space;

[0074] The surface gloss calculation unit is used to calculate the surface gloss according to each brightness value.

[0075] In this embodiment, the formula for calculating the surface gloss is:

[0076] , where g is the surface gloss, L avg is the mean value of the brightness values of the surface image of the polished area in the HSV color space, and σ L is the standard deviation of each brightness value.

[0077] The color space mapping unit converts the image into the HSV color space. Compared with the common RGB color space, the brightness value (V channel) in the HSV space can more intuitively and accurately reflect the gloss information of the surface, excluding the interference of the color itself. The brightness mean represents the overall brightness degree of the polished area, and it can give a basic value about the average gloss situation of the surface. However, only relying on the mean value cannot judge the uniformity of the gloss, and the brightness dispersion degree reflected by the standard deviation just makes up for this point.

[0078] In this embodiment, the progress difference extraction subsystem includes: an x-direction texture progress difference extraction unit, a y-direction texture progress difference extraction unit, and a gloss progress difference extraction unit;

[0079] The x-direction texture progress difference extraction unit is used to subtract the x-direction texture roughness at adjacent times, take the absolute value of the subtraction result as the x-direction texture progress value, and subtract the x-direction texture progress value from the x-direction texture progress target value to obtain the x-direction texture progress difference;

[0080] The y-direction texture progress difference extraction unit is used to subtract the y-direction texture roughness at adjacent times, take the absolute value of the subtraction result as the y-direction texture progress value, and subtract the y-direction texture progress value from the y-direction texture progress target value to obtain the y-direction texture progress difference;

[0081] The gloss progress difference extraction unit is used to subtract the surface gloss at adjacent times, take the absolute value of the subtraction result as the gloss progress value, and subtract the gloss progress value from the gloss progress target value to obtain the gloss progress difference.

[0082] Specifically: The x-direction texture progress difference extraction unit is used to subtract the x-direction texture roughness at time t - 1 from the x-direction texture roughness at time t, take the absolute value of the subtraction result as the x-direction texture progress value, and subtract the x-direction texture progress value from the x-direction texture progress target value to obtain the x-direction texture progress difference;

[0083] The y-direction texture progress difference extraction unit is used to subtract the y-direction texture roughness at time t - 1 from the y-direction texture roughness at time t, take the absolute value of the subtraction result as the y-direction texture progress value, and subtract the y-direction texture progress value from the y-direction texture progress target value to obtain the y-direction texture progress difference;

[0084] The gloss progress difference extraction unit is used to subtract the surface gloss at time t - 1 from the surface gloss at time t, take the absolute value of the subtraction result as the gloss progress value, and subtract the gloss progress value from the gloss progress target value to obtain the gloss progress difference.

[0085] The present invention subtracts the x-direction texture roughness, y-direction texture roughness, and surface gloss at adjacent times to obtain the x-direction texture progress difference, y-direction texture progress difference, and gloss progress difference, reflecting the changes in texture and gloss.

[0086] In this embodiment, the progress ratio coefficient extraction subsystem includes: an x-direction progress ratio coefficient extraction unit, a y-direction progress ratio coefficient extraction unit, and a gloss progress ratio coefficient extraction unit;

[0087] The x-direction progress ratio coefficient extraction unit is used to assign a value of 1 to the x-direction texture progress ratio coefficient when the x-direction texture progress difference is greater than or equal to 0, and when the x-direction texture progress difference is less than 0, the x-direction texture progress ratio coefficient is equal to the ratio of the x-direction texture progress difference to the x-direction texture progress target value;

[0088] The y-direction progress ratio coefficient extraction unit is used to assign a value of 1 to the y-direction texture progress ratio coefficient when the y-direction texture progress difference is greater than or equal to 0, and when the y-direction texture progress difference is less than 0, the y-direction texture progress ratio coefficient is equal to the ratio of the y-direction texture progress difference to the y-direction texture progress target value;

[0089] The gloss progress ratio coefficient extraction unit is used to assign a value of 1 to the gloss progress ratio coefficient when the gloss progress difference is greater than or equal to 0, and when the y-direction texture progress difference is less than 0, the gloss progress ratio coefficient is equal to the ratio of the gloss progress difference to the gloss progress target value.

[0090] The present invention provides a quantitative evaluation standard for the polishing effect by comparing the x-direction and y-direction texture progress differences and the gloss progress difference with the target values. When the progress difference is greater than or equal to 0, it is assigned a value of 1, indicating that the polishing in this dimension reaches or exceeds the expectation; when the difference is less than 0, the ratio of the difference to the target value is used as the coefficient, which can intuitively reflect the degree to which the polishing in this dimension lags behind the target. For example, if the coefficient is 0.5, it means that only half of the target is completed.

[0091] In this embodiment, the x-direction texture progress target value, the y-direction texture progress target value, and the gloss progress target value are pre-stored x-direction texture progress values, y-direction texture progress values, and gloss progress values.

[0092] As Figure 2 shown, the multi-unit speed adjustment subsystem includes: a first speed adjustment subunit, a second speed adjustment subunit, a third speed adjustment subunit, an addition unit, an increment suppression unit, and a speed output unit;

[0093] The first speed adjustment subunit is used to calculate the first speed increment according to the x-direction texture progress ratio coefficients at multiple moments;

[0094] The second speed adjustment subunit is used to calculate the second speed increment according to the y-direction texture progress ratio coefficients at multiple moments;

[0095] The third speed adjustment subunit is used to calculate the third speed increment according to the gloss progress ratio coefficients at multiple moments;

[0096] The addition unit is used to add the first speed increment, the second speed increment, and the third speed increment to obtain the total increment;

[0097] The increment suppression unit is used to suppress the total increment according to the gap between the total increment and the target increment to obtain the suppressed increment;

[0098] The speed output unit is used to add the suppressed increment and the current grinding speed to obtain the grinding speed at the next moment.

[0099] The present invention calculates the speed increment based on the progress ratio coefficients at multiple moments from three dimensions: the texture in the x direction, the texture in the y direction, and the glossiness. The addition unit integrates the speed increments in the three dimensions to obtain the total increment, and the increment suppression unit then suppresses it according to the gap between the total increment and the target increment. Finally, the speed output unit determines the grinding speed at the next moment. The increment suppression unit can effectively avoid the sudden change of the grinding speed caused by the excessive fluctuation of the total increment. In actual grinding, external factors or changes in the grinding state may cause large fluctuations in the increment, and the suppression unit can control it within a reasonable range to ensure the smooth change of the grinding speed.

[0100] In this embodiment, the expression of the first speed adjustment subunit is:

[0101] , where V 1,z,t+1 is the first speed increment at the (t + 1)-th moment, V t is the grinding speed at the t-th moment, γ x,t is the progress ratio coefficient of the texture in the x direction at the t-th moment, γ x,t-τ is the progress ratio coefficient of the texture in the x direction at the (t - τ)-th moment, T is the length of the historical moment, t is the number of the current moment, τ is the number of the historical moment, and β 1 is the first adjustment factor.

[0102] The expression of the second speed adjustment subunit is:

[0103] , where V 2,z,t+1 is the second speed increment at the (t + 1)-th moment, γ y,t is the progress ratio coefficient of the texture in the y direction at the t-th moment, γ y,t-τ is the progress ratio coefficient of the texture in the y direction at the (t - τ)-th moment, and β 2 is the second adjustment factor.

[0104] The expression of the third speed adjustment subunit is:

[0105] , where V 3,z,t+1 is the third speed increment at the (t + 1)-th moment, γ g,t is the progress ratio coefficient of the glossiness at the t-th moment, γ g,t-τ is the progress ratio coefficient of the glossiness at the (t - τ)-th moment, and β 3 is the third adjustment factor.

[0106] The present invention comprehensively considers the progress ratio coefficients at historical moments and the current moment. Taking the first speed adjustment subunit as an example, the sum of the x-direction texture progress ratio coefficients at the historical T moments is calculated, reflecting the historical trend of x-direction texture polishing; represents the progress ratio coefficient at the current moment. This method neither ignores the information accumulated in the past polishing process nor can it respond in a timely manner to the current polishing state.

[0107] In this embodiment, the expression of the increment suppression unit is:

[0108] , where is the suppression increment at the (t + 1)-th moment, V z,t+1 is the total increment at the (t + 1)-th moment, V z,tar is the target increment, is the suppression increment at the t-th moment, E t+1 is V z,t+1 minus V z,tar , and t is the number of the current moment.

[0109] In this embodiment, the target increment is the set speed increment.

[0110] When the total increment V z,t+1 is greater than or equal to the target increment V z,tar , take the suppression increment at the t-th moment, and calculate the speed adjustment ratio according to the difference between V z,t+1 and V z,tar . When the total increment V z,t+1 is less than the target increment V z,tar , directly output the total increment V z,t+1 to prevent excessive speed fluctuations from causing uneven polishing.

[0111] In this embodiment, the first adjustment factor β 1 , the second adjustment factor β 2 , and the third adjustment factor β 3 can be specifically set through experiments or experience, and can also be calibrated through a genetic algorithm. Taking β 1 , β 2 , and β 3 as the population individuals of the GA genetic model, through crossover and mutation, the specific values of β 1 , β 2 , and β 3 are screened out.

[0112] In this embodiment, a group of β 1 , β 2 , and β 3Set the values, such as all set to 0.3, and then select a certain number of workpieces for grinding tests. According to the test results, such as whether the grinding quality meets the standard and the grinding efficiency, adjust the adjustment factors.

[0113] The present invention performs gradient operations on the surface image of the grinding area in the x and y directions to obtain the texture roughness in the x and y directions, enabling the system to sensitively capture the subtle changes in the texture of the grinding area in different directions.

[0114] The present invention directly extracts the glossiness from the surface image of the grinding area, allowing the system to have an intuitive and quantitative evaluation of the grinding effect. Based on this, the progress difference extraction subsystem further combines the texture roughness and glossiness data, calculates the texture progress differences and glossiness progress differences in the x and y directions, and then converts the texture progress differences and glossiness progress differences into texture progress proportionality coefficients and glossiness progress proportionality coefficients to reflect the real-time progress during the grinding process.

[0115] The present invention can reflect the grinding state of the workpiece surface from different angles based on the texture progress proportionality coefficients in the x direction, the texture progress proportionality coefficients in the y direction, and the glossiness progress proportionality coefficients at multiple moments, as well as the texture progress and glossiness changes in different directions. By adjusting the grinding speed based on multi-dimensional data, the grinding accuracy is improved, and the grinding parameters can be dynamically adjusted according to the progress.

[0116] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent robot control system for a grinding wheel deburring workstation, characterized in that: include: Texture coarseness calculation subsystem, surface gloss extraction subsystem, progress difference extraction subsystem, progress proportional coefficient extraction subsystem and multi-unit speed regulation subsystem; The texture coarseness calculation subsystem is used to perform gradient calculations on the surface image of the polishing area in the x and y directions respectively, obtain the x-direction gradient image and the y-direction gradient image, and calculate the x-direction texture coarseness and the y-direction texture coarseness; The surface gloss extraction subsystem is used to extract the surface gloss from the surface image of the polishing area; The progress difference extraction subsystem is used to extract the texture progress difference in the x direction, the texture progress difference in the y direction and the glossiness progress difference according to the texture coarseness in the x direction, the texture coarseness in the y direction and the surface glossiness; The progress scale coefficient extraction subsystem is used to extract the x-direction texture progress scale coefficient, the y-direction texture progress scale coefficient and the glossiness progress scale coefficient according to the x-direction texture progress difference, the y-direction texture progress difference and the glossiness progress difference; The multi-unit speed adjustment subsystem is used to adjust the polishing speed according to the x-direction texture progress proportional coefficient, the y-direction texture progress proportional coefficient and the glossiness progress proportional coefficient at multiple moments.

2. The intelligent robot control system of the grinding wheel deburring workstation according to claim 1 is characterized in that: The texture coarseness calculation subsystem includes: an x-direction gradient image construction unit, a y-direction gradient image construction unit, an x-direction texture coarseness calculation unit and a y-direction texture coarseness calculation unit; The x-direction gradient image construction unit is used to perform a gradient operation in the x-direction on the surface image of the polishing area using a Sobel operator to obtain an x-direction gradient amplitude, and replace the original pixel value of the same pixel point with each x-direction gradient amplitude to obtain an x-direction gradient image; The y-direction gradient image construction unit is used to perform a gradient operation in the y-direction on the surface image of the polishing area using a Sobel operator to obtain a y-direction gradient amplitude, and replace the original pixel value of the same pixel point with each y-direction gradient amplitude to obtain a y-direction gradient image; The x-direction texture coarseness calculation unit is used to calculate the x-direction texture coarseness according to the number of pixel points whose x-direction gradient amplitude is equal to 0 in the x-direction gradient image and the average value of the x-direction gradient amplitude; The y-direction texture coarseness calculation unit is used to calculate the y-direction texture coarseness according to the number of pixel points whose y-direction gradient amplitude is equal to 0 in the y-direction gradient image and the average of the y-direction gradient amplitudes.

3. The intelligent robot control system of the grinding wheel deburring workstation according to claim 1 is characterized in that: The formula for calculating the texture coarseness in the x direction is: , where C x is the texture coarseness in the x direction, R is the number of pixels on the surface image of the polished area, and N x is the number of pixels whose x-direction gradient magnitude is equal to 0 in the x-direction gradient image, G x,avg is the mean value of the gradient amplitudes in each x direction in the x direction gradient image; The formula for calculating the texture coarseness in the y direction is: , where C y is the texture coarseness in the y direction, N y is the number of pixels whose y-direction gradient magnitude is equal to 0 in the y-direction gradient image, G y,avg is the mean of the gradient amplitudes in the y direction in the y direction gradient image.

4. The intelligent robot control system of the grinding wheel deburring workstation according to claim 1 is characterized in that: The surface gloss extraction subsystem includes: a color space mapping unit and a surface gloss calculation unit; The color space mapping unit is used to convert the surface image of the polishing area into the HSV color space and extract the brightness value in the HSV color space; The surface gloss calculation unit is used to calculate the surface gloss according to each brightness value.

5. The intelligent robot control system of the grinding wheel deburring workstation according to claim 1 is characterized in that: The calculation formula for surface gloss is: , where g is the surface gloss, L avg is the mean value of each brightness value of the surface image of the polished area in the HSV color space, σ L is the standard deviation of each brightness value.

6. The intelligent robot control system of the grinding wheel deburring workstation according to claim 1, characterized in that: The progress difference extraction subsystem includes: an x-direction texture progress difference extraction unit, a y-direction texture progress difference extraction unit and a glossiness progress difference extraction unit; The x-direction texture progress difference extraction unit is used to subtract the x-direction texture coarseness at adjacent moments, take the absolute value of the subtraction result as the x-direction texture progress value, and subtract the x-direction texture progress value from the x-direction texture progress target value to obtain the x-direction texture progress difference; The y-direction texture progress difference extraction unit is used to subtract the y-direction texture coarseness at adjacent moments, take the absolute value of the subtraction result as the y-direction texture progress value, and subtract the y-direction texture progress value from the y-direction texture progress target value to obtain the y-direction texture progress difference; The glossiness progress difference extraction unit is used to subtract the surface glossiness at adjacent moments, take the absolute value of the subtraction result as the glossiness progress value, and subtract the glossiness progress value from the glossiness progress target value to obtain the glossiness progress difference.

7. The intelligent robot control system of the grinding wheel deburring workstation according to claim 1, characterized in that: The progress ratio coefficient extraction subsystem includes: an x-direction progress ratio coefficient extraction unit, a y-direction progress ratio coefficient extraction unit and a glossiness progress ratio coefficient extraction unit; The x-direction progress ratio coefficient extraction unit is used to assign a value of 1 to the x-direction texture progress ratio coefficient when the x-direction texture progress difference is greater than or equal to 0, and when the x-direction texture progress difference is less than 0, the x-direction texture progress ratio coefficient is equal to the ratio of the x-direction texture progress difference to the x-direction texture progress target value; The y-direction progress proportional coefficient extraction unit is used to assign a value of 1 to the y-direction texture progress proportional coefficient when the y-direction texture progress difference is greater than or equal to 0, and when the y-direction texture progress difference is less than 0, the y-direction texture progress proportional coefficient is equal to the ratio of the y-direction texture progress difference to the y-direction texture progress target value; The glossiness progress proportional coefficient extraction unit is used to assign a value of 1 to the glossiness progress proportional coefficient when the glossiness progress difference is greater than or equal to 0, and when the texture progress difference in the y direction is less than 0, the glossiness progress proportional coefficient is equal to the ratio of the glossiness progress difference to the glossiness progress target value.

8. The intelligent robot control system of the grinding wheel deburring workstation according to claim 1, characterized in that: The multi-unit speed regulating subsystem comprises: a first speed regulating subunit, a second speed regulating subunit, a third speed regulating subunit, an adding unit, an increment suppressing unit and a speed output unit; The first speed adjustment subunit is used to calculate the first speed increment according to the x-direction texture progress ratio coefficients at multiple moments; The second speed adjustment subunit is used to calculate the second speed increment according to the y-direction texture progress ratio coefficients at multiple moments; The third speed adjustment subunit is used to calculate the third speed increment according to the glossiness progress proportional coefficients at multiple moments; The adding unit is used for adding the first speed increment, the second speed increment and the third speed increment to obtain a total increment; The increment suppression unit is used to suppress the total increment according to the difference between the total increment and the target increment to obtain the suppressed increment; The speed output unit is used to add the suppression increment to the current grinding speed to obtain the grinding speed at the next moment.

9. The intelligent robot control system of the grinding wheel deburring workstation according to claim 8, characterized in that: The expression of the first speed adjustment subunit is: , where V 1,z,t+1 is the first velocity increment at time t+1, V t is the grinding speed at the tth moment, γ x,t is the x-direction texture progression coefficient at the tth moment, γ x,t-τ is the x-direction texture progress ratio coefficient at the t-τth moment, T is the length of the historical moment, t is the number of the current moment, τ is the number of the historical moment, and β1 is the first adjustment factor; The expression of the second speed adjustment subunit is: , where V 2,z,t+1 is the second velocity increment at time t+1, γ y,t is the y-direction texture progression coefficient at the tth moment, γ y,t-τ is the y-direction texture progress ratio coefficient at the t-τth moment, β2 is the second adjustment factor; The expression of the third speed adjustment subunit is: , where V 3,z,t+1 is the third velocity increment at time t+1, γ g,t is the glossiness progression coefficient at the tth moment, γ g,t-τ is the glossiness progression coefficient at the t-τth moment, and β3 is the third adjustment factor.

10. The intelligent robot control system of the grinding wheel deburring workstation according to claim 8, characterized in that: The expression of the incremental suppression unit is: ,in, is the suppression increment at the t+1th moment, V z,t+1 is the total increment at time t+1, V z,tar is the target increment, is the suppression increment at the tth moment, E t+1 V z,t+1 With V z,tar , t is the number of the current moment.

Images