A feeding control system and control method based on a robotic arm

Abstract

The present invention relates to the field of mechanical arm technology, and discloses a feeding control system and control method based on a mechanical arm, the method comprising: adjusting an orientation image to obtain a deviation image of a to-be-processed area, acquiring pixel points of each deviation image, and determining a cluster center point of the deviation image, and determining a deviation series according to the pixel distance between each pixel point and the cluster center point; generating a deviation calibration for the pixel points in the deviation series, and determining a deviation value of a grasped material according to a first quantity and a second quantity; comparing historical deviation data with the deviation value, and judging whether to adjust the initial grasping force of the mechanical arm according to the comparison result, determining a primary influence value of the force according to the friction coefficient between the grasped material and the mechanical arm, and determining a target influence value of the force according to the primary influence value of the force and the material weight of the grasped material; and adjusting the initial grasping force according to an adjustment factor. The present invention ensures the stability of material feeding by adjusting the grasping force.

Description

Technical Field

[0001] The present invention relates to the technical field of robotic arms, and in particular to a feeding control system and a control method based on a robotic arm. Background Art

[0002] In the automatic production process of CNC machine tools, the robot arm of the CNC machine tool is responsible for grabbing and moving materials. However, the existing robot arms have certain limitations in the process of automatic feeding. When the robot arm grabs the material and moves it, the initial grabbing force between the robot arm and the material will be affected by the weight and material properties of the material itself, resulting in a certain deviation in the grabbing position. If the initial grabbing force is not adjusted, there will be a risk of slipping during the material handling process. At present, when controlling the robot arm, it cannot adjust the initial grabbing force of the robot arm according to the weight and material properties of the material after grabbing the material, resulting in insufficient stability and efficiency of the robot arm's automatic feeding.

[0003] Therefore, how to provide a feeding control system and control method based on a robotic arm is a technical problem that technical personnel in this field urgently need to solve. Summary of the invention

[0004] In view of this, the present invention proposes a feeding control system and control method based on a robotic arm, aiming to solve the problem that the initial grasping force of the robotic arm cannot be adjusted according to the material's own weight and the material properties, resulting in insufficient stability and efficiency of automatic feeding of the robotic arm.

[0005] In one aspect, the present invention proposes a feeding control method based on a robotic arm, comprising:

[0006] Divide the grabbed materials into a plurality of areas to be processed, obtain an orientation image of the area to be processed, adjust the orientation image to obtain a deviation image of the area to be processed, obtain pixel points of each deviation image, determine a cluster center point of the deviation image, and determine a deviation series according to a pixel distance between each pixel point and the cluster center point;

[0007] Generating deviation calibration for pixel points in the deviation series, the deviation calibration including a first deviation calibration and a second deviation calibration, counting a first quantity of the first deviation calibration, counting a second quantity of the second deviation calibration, and determining a deviation value of the grasped material according to the first quantity and the second quantity;

[0008] Obtain historical deviation data of grasped materials, compare the historical deviation data with the deviation value, and determine whether to adjust the initial grasping force of the robot arm according to the comparison result. When it is determined to adjust the initial grasping force, determine the primary influence value of the force according to the friction coefficient between the grasped material and the robot arm, and determine the target influence value of the force according to the primary influence value of the force and the material weight of the grasped material;

[0009] The adjustment factor of the initial grasping force is determined according to the force target influence value, the initial grasping force is adjusted according to the adjustment factor, and the robotic arm is controlled according to the adjusted initial grasping force to complete the feeding operation.

[0010] Furthermore, when the orientation image is adjusted to obtain the deviation image of the area to be processed, it includes:

[0011] The orientation image is denoised by wavelet transform, the denoised orientation image is brightness adjusted according to an adaptive histogram equalization algorithm to obtain a brightness image, pyramid decomposition is used to capture image features of the brightness image, the brightness image is registered based on the image features, the registered orientation images are merged to obtain the deviation image of the area to be processed.

[0012] Furthermore, when determining the deviation series according to the pixel distance between each pixel point and the cluster center point, it includes:

[0013] Presetting a first preset pixel distance and a second preset pixel distance, wherein the first preset pixel distance is greater than the second preset pixel distance;

[0014] The deviation sequence includes a prominent sequence, a common sequence and an indirect sequence;

[0015] Allocating pixel points whose pixel distance is greater than or equal to the first preset pixel distance to the protrusion number series;

[0016] Allocate pixel points whose pixel distance is smaller than the first preset pixel distance and larger than the second preset pixel distance to the common number series;

[0017] Pixel points whose pixel distance is less than or equal to the second preset pixel distance are allocated to the indirect number series.

[0018] Furthermore, when generating deviation calibration for the pixel points in the deviation series, it includes:

[0019] Obtaining an average pixel distance of all pixels in the prominent series and recording it as a prominent value, obtaining an average pixel distance of all pixels in the common series and recording it as a common value, and obtaining an average pixel distance of all pixels in the indirect series and recording it as an indirect value;

[0020] The pixel points in the protrusion series whose pixel distance is greater than the protrusion value generate the first deviation calibration, and the pixel points in the protrusion series whose pixel distance is less than or equal to the protrusion value generate the second deviation calibration;

[0021] The pixel points in the common number series whose pixel distance is greater than the common value generate the first deviation calibration, and the pixel points in the common number series whose pixel distance is less than or equal to the common value generate the second deviation calibration;

[0022] The first deviation calibration is generated for pixel points in the indirect number series whose pixel distance is greater than the indirect value, and the second deviation calibration is generated for pixel points in the indirect number series whose pixel distance is less than or equal to the indirect value.

[0023] Further, when determining the deviation value of the grabbed material according to the first quantity and the second quantity, it includes:

[0024] The deviation value is obtained by the following formula:

[0025] ;

[0026] in, Indicates the deviation value, represents the first quantity, represents the second quantity, and represents the weight coefficient, and .

[0027] Furthermore, when obtaining historical deviation data of the grasped material, comparing the historical deviation data with the deviation value, and judging whether to adjust the initial grasping force of the robot arm according to the comparison result, it includes:

[0028] Obtain the historical deviation qualified mean value of the same material weight and the same material attribute as the grabbed material;

[0029] When the deviation value is greater than the qualified mean value of the historical deviation, it is determined to adjust the initial grasping force of the robot arm;

[0030] When the deviation value is less than or equal to the qualified mean value of the historical deviation, it is determined that the initial grasping force of the robot arm is not adjusted, and the feeding operation is completed according to the initial grasping force.

[0031] Further, when the primary force impact value is determined according to the friction coefficient between the grasped material and the robot arm, and the target force impact value is determined according to the primary force impact value and the material weight of the grasped material, it includes:

[0032] Preset a first preset force primary impact value, a second preset force primary impact value, and a third preset force primary impact value, wherein the first preset force primary impact value is smaller than the second preset force primary impact value, and the second preset force primary impact value is smaller than the third preset force primary impact value;

[0033] Determine the friction coefficient μ of the robotic arm according to the material properties of the grasped material and the material properties of the robotic arm;

[0034] When μ is greater than 0.3, the first preset force primary influence value is determined as the force primary influence value;

[0035] When μ is less than or equal to 0.3 and greater than 0.15, the second preset force primary influence value is determined as the force primary influence value;

[0036] When μ is less than or equal to 0.15, the third preset force primary influence value is determined as the force primary influence value;

[0037] The force target impact value is derived from the following formula:

[0038] ;

[0039] in, Indicates the intensity target impact value, Indicates the primary impact value of the force, Indicates the material weight of the grabbed material.

[0040] Furthermore, when determining the adjustment factor of the initial grasping force according to the force target influence value, it includes:

[0041] Preset a first preset force target impact value, a second preset force target impact value, and a third preset force target impact value, wherein the first preset force target impact value is greater than the second preset force target impact value, and the second preset force target impact value is greater than the third preset force target impact value;

[0042] A first preset adjustment factor N1, a second preset adjustment factor N2 and a third preset adjustment factor N3 are preset, and 1.1<N3<N2<N1<2;

[0043] When the force target impact value is greater than or equal to the first preset force target impact value, using the first preset adjustment factor N1 as the adjustment factor of the initial grasping force;

[0044] When the force target impact value is less than the first preset force target impact value and greater than the second preset force target impact value, using the second preset adjustment factor N2 as the adjustment factor of the initial grasping force;

[0045] When the force target influence value is less than or equal to the third preset force target influence value, the third preset adjustment factor N3 is used as the adjustment factor of the initial grasping force.

[0046] Furthermore, when the initial grasping force is adjusted according to the adjustment factor, it includes:

[0047] The adjustment factor is proportional to the initial grasping force.

[0048] Compared with the prior art, the beneficial effects of the present invention are as follows: by dividing the grasped material into multiple areas to be processed, adjusting the orientation image and generating the deviation image, the characteristics of the grasped material can be accurately identified and captured, so as to conduct a detailed analysis of the deviation in the material grasping and moving process, and then obtain an accurate adjustment basis, and, based on the pixel distance between the pixel point and the cluster center point, the deviation series is determined and the deviation calibration is generated, which improves the accuracy of classifying the characteristics of the grasped material, avoids the judgment error caused by a single data amount, ensures the accuracy of the deviation value, and effectively avoids the error of repeated adjustment by comparing with the historical deviation data, improves the accuracy and reliability of adjusting the initial grasping force, comprehensively determines the force target influence value according to the friction coefficient and the material weight, and obtains the force adjustment factor, so as to adjust the initial grasping force, ensure that the adjusted initial grasping force can match the material weight and the material properties of the material, prevent the risk of material falling off, improve the adaptability of the robot arm to different grasping environments, and ensure the stability and efficiency of automatic feeding of the robot arm.

[0049] On the other hand, the present application also provides a feeding control system based on a robotic arm, which is used to apply the above-mentioned feeding control method based on a robotic arm, including:

[0050] A sequence determination module is configured to divide the grabbed material into a plurality of areas to be processed, obtain an orientation image of the area to be processed, adjust the orientation image to obtain a deviation image of the area to be processed, obtain pixel points of each deviation image, determine a cluster center point of the deviation image, and determine a deviation sequence according to a pixel distance between each pixel point and the cluster center point;

[0051] a deviation analysis module, configured to generate a deviation calibration for the pixel points in the deviation series, the deviation calibration including a first deviation calibration and a second deviation calibration, counting a first quantity of the first deviation calibration, counting a second quantity of the second deviation calibration, and determining a deviation value of the grasped material according to the first quantity and the second quantity;

[0052] a force processing module, configured to obtain historical deviation data of grasped materials, compare the historical deviation data with the deviation value, determine whether to adjust the initial grasping force of the robot arm according to the comparison result, and when it is determined to adjust the initial grasping force, determine the primary influence value of the force according to the friction coefficient between the grasped material and the robot arm, and determine the target influence value of the force according to the primary influence value of the force and the material weight of the grasped material;

[0053] The force adjustment module is configured to determine the adjustment factor of the initial grasping force according to the force target influence value, adjust the initial grasping force according to the adjustment factor, and control the robotic arm according to the adjusted initial grasping force to complete the loading operation.

[0054] It can be understood that the above-mentioned feeding control system and control method based on a robotic arm have the same beneficial effects, which will not be repeated here. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0056] Figure 1 A flow chart of a feeding control method based on a robotic arm provided in an embodiment of the present invention;

[0057] Figure 2 A functional block diagram of a feeding control system based on a robotic arm provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0058] Exemplary embodiments of the present disclosure will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the accompanying drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided in order to enable a more thorough understanding of the present disclosure and to fully convey the scope of the present disclosure to those skilled in the art. It should be noted that, in the absence of conflict, the embodiments of the present invention and the features described in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0059] See also Figure 1 As shown, in some embodiments of the present application, this embodiment provides a feeding control method based on a robotic arm, comprising:

[0060] S100: Divide the grabbed materials into a plurality of areas to be processed, obtain the orientation images of the areas to be processed, adjust the orientation images to obtain deviation images of the areas to be processed, obtain the pixel points of each deviation image, determine the cluster center point of the deviation image, and determine the deviation series according to the pixel distance between each pixel point and the cluster center point;

[0061] S200: generating deviation calibration for pixel points in the deviation series, the deviation calibration including first deviation calibration and second deviation calibration, counting a first quantity of the first deviation calibration, counting a second quantity of the second deviation calibration, and determining a deviation value of the grasped material according to the first quantity and the second quantity;

[0062] S300: Obtain historical deviation data of the grasped material, compare the historical deviation data with the deviation value, and determine whether to adjust the initial grasping force of the robot arm according to the comparison result. When it is determined to adjust the initial grasping force, determine the primary influence value of the force according to the friction coefficient between the grasped material and the robot arm, and determine the target influence value of the force according to the primary influence value of the force and the weight of the grasped material;

[0063] S400: determining an adjustment factor of the initial grasping force according to the force target influence value, adjusting the initial grasping force according to the adjustment factor, and controlling the robot arm according to the adjusted initial grasping force to complete the loading operation.

[0064] Specifically, before the robotic arm performs the loading operation, the grasped material is first divided into multiple areas to be processed, preferably 10 areas to be processed, which can be dynamically adjusted according to the surface area of ​​the material. The orientation image of the area to be processed is obtained by using image acquisition equipment such as a camera, so as to comprehensively evaluate the actual grasping state of the material and determine the position of the material in three-dimensional space. These images are processed to capture the characteristics of the grasped material in the orientation image, so as to obtain the deviation image of the area to be processed, and obtain the information of each pixel point to determine the pixel distance between it and the cluster center point. The cluster center point obtained by the clustering algorithm can effectively describe the overall deviation of the material. According to the pixel distance between each pixel point and the cluster center point, a deviation series is generated, which lays the foundation for the subsequent deviation value calculation. After obtaining the deviation series, the pixel points in the deviation series are used to generate deviation calibration, which is the first deviation calibration and the second deviation calibration, so as to further refine the deviation of the material. The first number of the first deviation calibration and the second number of the second deviation calibration are statistically analyzed. The first number and the second number reflect the distribution characteristics of the actual deviation of the material. The deviation value reflects the changes in the material when the material is grasped and moved, providing data support for the subsequent force adjustment. The historical deviation data of the grasped material and the deviation value are compared to make a decision on whether to adjust the initial grasping force. When a large deviation appears in the comparison result, the initial grasping force needs to be adjusted appropriately. When adjusting the initial grasping force, the primary influence value of the force is determined according to the friction coefficient between the grasped material and the robot arm. The primary influence value reflects the initial judgment on the adjustment of the initial grasping force, and the target influence value of the force is determined according to the primary influence value of the force and the material weight of the grasped material. The target influence value of the force measures the relationship between the material weight and the friction coefficient, avoiding the risk of adjustment errors caused by relying on a single data, accurately responding to the actual grasping requirements of different materials, and improving the accuracy and reliability of adjusting the initial grasping force. Based on the calculated force target impact value, the adjustment factor of the initial grasping force is determined. The adjustment factor dynamically adjusts the initial grasping force according to the material deviation, historical data and material properties. This ensures that the robot arm can adjust the initial grasping force in real time when grasping different materials for movement, so as to meet the grasping requirements of different materials and avoid the problem of materials slipping, thereby completing the precise grasping and handling of materials, and then improving the stability and efficiency of the feeding operation.

[0065] It can be understood that by adjusting the initial grasping force according to the friction coefficient and weight of different materials, the robotic arm can adapt to the grasping needs of different types of materials, improve the adaptability of the robotic arm in various production environments, and avoid the problem of material slipping due to force mismatch. It not only improves the grasping accuracy and stability, but also improves the efficiency of automatic loading of the robotic arm.

[0066] In some embodiments of the present application, when the orientation image is adjusted to obtain the deviation image of the area to be processed, it includes:

[0067] The orientation image is denoised by wavelet transform, and the brightness of the denoised orientation image is adjusted according to the adaptive histogram equalization algorithm to obtain a brightness image. The image features of the brightness image are captured by pyramid decomposition, and the brightness image is registered based on the image features. The registered orientation images are merged to obtain the deviation image of the area to be processed.

[0068] Specifically, wavelet transform can determine the low-frequency and high-frequency components of the azimuth image, thereby removing the high-frequency noise components to achieve the denoising effect, ensuring the main features of the azimuth image and reducing the interference of noise on subsequent processing. The adaptive histogram equalization (AHE) algorithm dynamically adjusts the brightness characteristics of different areas by performing histogram equalization on the local area of ​​the azimuth image, improving the contrast of the azimuth image, enhancing the visual effect of the brightness image, and facilitating the subsequent extraction and registration of image features. Pyramid decomposition constructs the pyramid structure of the image layer by layer by performing multiple downsampling and high-pass filtering on the brightness image. In each layer, the detail information of the brightness image will be gradually compressed, and the structural features (such as edges, corners, etc.) will be retained, ensuring that the image features are captured in different orientations, thereby improving the accuracy of subsequent image registration. The brightness images are registered based on the image features extracted by pyramid decomposition. The brightness images are aligned. The registration process often involves similarity metrics (such as mutual information, correlation coefficient, etc.) to ensure accurate alignment of the images. The registered orientation images are merged, including weighted mean, multi-band fusion, etc., and one of them can be adopted according to the actual situation of image processing to finally obtain the deviation image of the area to be processed.

[0069] It can be understood that by performing image processing such as denoising, brightness adjustment, feature extraction and image registration on the orientation image, the quality and visual effect of the deviation image are improved, providing a reliable basis for subsequent analysis.

[0070] In some embodiments of the present application, when determining the deviation series according to the pixel distance between each pixel point and the cluster center point, it includes:

[0071] A first preset pixel distance and a second preset pixel distance are preset, the first preset pixel distance is greater than the second preset pixel distance, the deviation series includes a prominent series, an ordinary series and an indirect series, pixel points whose pixel distance is greater than or equal to the first preset pixel distance are allocated to the prominent series, pixel points whose pixel distance is less than the first preset pixel distance and greater than the second preset pixel distance are allocated to the ordinary series, and pixel points whose pixel distance is less than or equal to the second preset pixel distance are allocated to the indirect series.

[0072] Specifically, by setting the first and second preset pixel distances and dividing the pixel points into prominent series, ordinary series and indirect series according to the pixel distances, the pixel points in the deviation image are effectively classified, thereby improving the accuracy and reliability of identifying and analyzing features in the deviation image.

[0073] In some embodiments of the present application, when generating deviation calibration for pixel points in the deviation series, the process includes:

[0074] Get the average pixel distance of all pixels in the prominent series and record it as the prominent value, get the average pixel distance of all pixels in the ordinary series and record it as the ordinary value, get the average pixel distance of all pixels in the indirect series and record it as the indirect value, generate a first deviation calibration for pixels in the prominent series whose pixel distance is greater than the prominent value, generate a second deviation calibration for pixels in the prominent series whose pixel distance is less than or equal to the prominent value, generate a first deviation calibration for pixels in the ordinary series whose pixel distance is greater than the ordinary value, generate a second deviation calibration for pixels in the ordinary series whose pixel distance is less than or equal to the ordinary value, generate a first deviation calibration for pixels in the indirect series whose pixel distance is greater than the indirect value, and generate a second deviation calibration for pixels in the indirect series whose pixel distance is less than or equal to the indirect value.

[0075] In some embodiments of the present application, when determining the deviation value of the grabbed material according to the first quantity and the second quantity, it includes:

[0076] The deviation value is given by the following formula:

[0077] ;

[0078] in, Indicates the deviation value, represents the first quantity, represents the second quantity, and represents the weight coefficient, and .

[0079] Specifically, by generating deviation calibration for the pixel points in each series through the average value of the pixel distance of each series, more accurate deviation information can be obtained in each series. Multi-level calibration improves the accuracy of deviation image processing and effectively captures subtle deviations in the deviation series. In addition, the deviation value is calculated based on the first quantity, the second quantity and the weight coefficient, so that the deviation value can be optimized according to the grasping requirements of different types of materials, thereby improving the adaptability and reliability of the robot arm to material grasping.

[0080] In some embodiments of the present application, when obtaining historical deviation data of grasped materials, comparing the historical deviation data with the deviation value, and judging whether to adjust the initial grasping force of the robot arm according to the comparison result, it includes:

[0081] Obtain the qualified mean historical deviation of the same material weight and material properties as the grasped material. When the deviation value is greater than the qualified mean historical deviation, determine to adjust the initial grasping force of the robot arm. When the deviation value is less than or equal to the qualified mean historical deviation, determine not to adjust the initial grasping force of the robot arm, and complete the loading operation according to the initial grasping force.

[0082] Specifically, the material property represents the physical material of the material itself, such as iron, aluminum, and plastic. By comparing the historical deviation data, the current operation can be judged based on the conditions under which the robot arm successfully grasped the material in the past, avoiding the risk of the robot arm causing the material to fall off due to inappropriate force, and improving the stability and accuracy of the grasping and feeding operations. The historical deviation qualified mean represents the mean deviation value of the historical qualified grasped materials. The historical deviation qualified mean can be set according to the grasping requirements. As a judgment indicator, its specific value can be modified during the actual grasping movement. When the deviation value is less than or equal to the historical deviation qualified mean, there is no need to adjust the initial grasping force. The original force can be maintained to complete the feeding operation, avoiding unnecessary adjustments and improving the efficiency of the operation. When the deviation value is greater than the historical deviation qualified mean, it means that the initial grasping force does not match the current grasped material, and there is a risk of falling off, which provides a judgment basis for subsequent adjustments.

[0083] In some embodiments of the present application, when determining the primary force impact value according to the friction coefficient between the grasped material and the robot arm, and determining the target force impact value according to the primary force impact value and the weight of the grasped material, it includes:

[0084] A first preset force primary impact value, a second preset force primary impact value, and a third preset force primary impact value are preset, the first preset force primary impact value is less than the second preset force primary impact value, the second preset force primary impact value is less than the third preset force primary impact value, the friction coefficient μ of the robot arm is determined according to the material properties of the grasped material and the material properties of the robot arm, when μ is greater than 0.3, the first preset force primary impact value is determined as the force primary impact value, when μ is less than or equal to 0.3 and greater than 0.15, the second preset force primary impact value is determined as the force primary impact value, when μ is less than or equal to 0.15, the third preset force primary impact value is determined as the force primary impact value, and the force target impact value is obtained by the following formula:

[0085] ;

[0086] in, Indicates the intensity target impact value, Indicates the primary impact value of the force, Indicates the material weight of the grabbed material.

[0087] Specifically, the primary influence value of force is determined by the friction coefficient between the grasped material and the robot arm, which effectively measures the relationship between the material properties of the material and the grasping of the robot arm. Different types of materials and robot arms will have different friction coefficients. Determining the primary influence value of force by the friction coefficient not only improves the adaptability of the robot arm in different operating environments, but also improves the stability and reliability of the grasping process. The larger the friction coefficient, the greater the friction between the material and the robot arm, and the smaller the determined primary influence value of force, indicating that the degree of influence on adjusting the initial grasping force is smaller. Conversely, when the friction coefficient is smaller, the friction between the material and the robot arm is smaller, and the larger the determined primary influence value of force, indicating that the degree of influence on adjusting the initial grasping force is greater. By dynamically selecting the primary impact value and further calculating it in combination with the material weight, the accuracy of the initial grasping force adjustment is improved, and the adjustment error caused by a single set threshold is avoided. The accuracy of material grasping is ensured, and the efficiency and accuracy of the overall workflow are improved. Whether it is heavier or lighter materials, the force target impact value can be obtained based on the friction coefficient and its own weight, thereby adjusting the initial grasping force to meet the specific needs of different material grasping, thereby improving the accuracy of the feeding operation.

[0088] In some embodiments of the present application, when determining the adjustment factor of the initial grasping force according to the force target impact value, it includes:

[0089] A first preset force target influence value, a second preset force target influence value, and a third preset force target influence value are pre-set, the first preset force target influence value is greater than the second preset force target influence value, the second preset force target influence value is greater than the third preset force target influence value, a first preset adjustment factor N1, a second preset adjustment factor N2, and a third preset adjustment factor N3 are pre-set, and 1.1<N3<N2<N1<2, when the force target influence value is greater than or equal to the first preset force target influence value, the first preset adjustment factor N1 is used as the adjustment factor of the initial grasping force, when the force target influence value is less than the first preset force target influence value and greater than the second preset force target influence value, the second preset adjustment factor N2 is used as the adjustment factor of the initial grasping force, and when the force target influence value is less than or equal to the third preset force target influence value, the third preset adjustment factor N3 is used as the adjustment factor of the initial grasping force.

[0090] In some embodiments of the present application, when the initial grasping force is adjusted according to the adjustment factor, it includes:

[0091] The adjustment factor is proportional to the initial gripping force.

[0092] Specifically, according to different force target influence values, the appropriate adjustment factor can be dynamically selected to ensure that the initial grasping force can be adjusted accordingly according to the material properties and weight of the material itself to meet the grasping requirements of different materials, especially when grasping materials with different physical properties, such as slippery, irregular materials, etc., to avoid the risk of the grasped material falling off due to too weak grasping force, improve the flexibility of the robot arm control, ensure the stability and reliability of the material grasping and feeding process, improve the automation degree and work efficiency of the robot arm, and enhance the adaptability of the robot arm in complex environments. When the initial grasping force is adjusted according to the adjustment factor, if the initial grasping force is F, the adjusted initial grasping force is determined to be F*Ni, Ni represents one of the first preset adjustment factor N1, the second preset adjustment factor N2 and the third preset adjustment factor N3. When the maximum grasping force is required, the first preset adjustment factor N1 is selected as the adjustment factor of the initial grasping force to adjust the initial grasping force, so as to realize the precise control of the initial grasping force by the robot arm during the feeding process, and improve the accuracy and reliability of automatic feeding.

[0093] In summary, the beneficial effects of the present invention are as follows: by dividing the grasped material into multiple areas to be processed, adjusting the orientation image and generating the deviation image, the characteristics of the grasped material can be accurately identified and captured, so as to conduct a detailed analysis of the deviation in the material grasping and moving process, and then obtain an accurate adjustment basis, and the deviation series is determined based on the pixel distance between the pixel point and the cluster center point and the deviation calibration is generated, which improves the accuracy of classifying the characteristics of the grasped material, avoids the judgment error caused by a single data amount, ensures the accuracy of the deviation value, and effectively avoids the error of repeated adjustment by comparing with the historical deviation data, improves the accuracy and reliability of adjusting the initial grasping force, comprehensively determines the force target influence value according to the friction coefficient and the material weight, and obtains the force adjustment factor, so as to adjust the initial grasping force, ensure that the adjusted initial grasping force can match the material weight and the material properties of the material, prevent the risk of material falling off, improve the adaptability of the robot arm to different grasping environments, and ensure the stability and efficiency of automatic feeding of the robot arm.

[0094] In another preferred embodiment based on the above embodiment, refer to Figure 2 As shown, this embodiment provides a feeding control system based on a robot arm, which is used to apply the above-mentioned feeding control method based on a robot arm, including:

[0095] A sequence determination module is configured to divide the grabbed materials into a plurality of areas to be processed, obtain an orientation image of the area to be processed, adjust the orientation image, obtain a deviation image of the area to be processed, obtain pixel points of each deviation image, determine a cluster center point of the deviation image, and determine a deviation sequence according to a pixel distance between each pixel point and the cluster center point;

[0096] The deviation analysis module is configured to generate a deviation calibration for the pixel points in the deviation series, the deviation calibration includes a first deviation calibration and a second deviation calibration, calculate a first quantity of the first deviation calibration, calculate a second quantity of the second deviation calibration, and determine the deviation value of the grasped material according to the first quantity and the second quantity;

[0097] A force processing module is configured to obtain historical deviation data of grasped materials, compare the historical deviation data with the deviation value, and determine whether to adjust the initial grasping force of the robot arm according to the comparison result. When it is determined to adjust the initial grasping force, a primary force impact value is determined according to the friction coefficient between the grasped material and the robot arm, and a target force impact value is determined according to the primary force impact value and the material weight of the grasped material;

[0098] The force adjustment module is configured to determine an adjustment factor of the initial grasping force according to the force target influence value, adjust the initial grasping force according to the adjustment factor, and control the robotic arm according to the adjusted initial grasping force to complete the loading operation.

[0099] Specifically, the sequence determination module first divides the grasped materials into multiple areas to be processed, then obtains the orientation image of each area, and adjusts these orientation images to determine the deviation image, extracts the pixel points of each deviation image, and calculates the distance between each pixel point and the cluster center point of the deviation image. The sequence determination module determines the cluster center point based on the K-means algorithm, hierarchical clustering algorithm, spectral clustering algorithm, etc., which can be set according to actual needs. The determination of pixel distance is lengthy and relatively mature, so it will not be described in detail here. The sequence determination module provides a basis for subsequent analysis by determining the deviation sequence. The deviation analysis module calibrates the deviation of the pixels in the deviation sequence and counts their respective numbers (the first number and the second number). According to the statistical results of the deviation calibration, the deviation analysis module calculates the deviation value of the grasped material to determine whether to adjust the initial grasping The force measurement module provides data support. The force processing module obtains historical deviation data and compares it with the real-time deviation value. If there is a large deviation in the deviation value, it means that there is a problem with the current grasping situation and the initial grasping force of the robot arm needs to be adjusted in time. At this time, the force processing module calculates the primary influence value of the force based on the friction coefficient between the grasped material and the robot arm, and determines the force target influence value based on this value and the material weight. The force target influence value combines the material properties and material weight of the material, measures the factors affecting the initial grasping force, and improves the accuracy and reliability of the initial grasping force adjustment. The force adjustment module determines the appropriate initial grasping force adjustment factor based on the force target influence value. The relationship between the preset force target influence value and the adjustment factor can ensure that the robot arm can adapt to the grasping requirements of different materials and ensure the completion of the loading operation.

[0100] It can be understood that through image analysis and calculation based on the deviation series, the system can accurately evaluate the grasping status of the grasped material, thereby dynamically adjusting the initial grasping force of the robotic arm, so that the robotic arm can be optimized and adjusted according to the material weight and friction coefficient of the material, thereby improving the adaptability of the robotic arm in different grasping environments and ensuring the stability and accuracy of the feeding process.

[0101] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems or computer program products. Therefore, the present application may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0102] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems) and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0103] These computer program instructions may also be stored in a computer-readable storage device that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable storage device produce a product including an instruction device that implements the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0104] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0105] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A feeding control method based on a robotic arm, characterized in that: include: Divide the grabbed materials into a plurality of areas to be processed, obtain an orientation image of the area to be processed, adjust the orientation image to obtain a deviation image of the area to be processed, obtain pixel points of each deviation image, determine a cluster center point of the deviation image, and determine a deviation series according to a pixel distance between each pixel point and the cluster center point; When determining the deviation series based on the pixel distance between each pixel and the cluster center point, it includes: Presetting a first preset pixel distance and a second preset pixel distance, wherein the first preset pixel distance is greater than the second preset pixel distance; The deviation sequence includes a prominent sequence, a common sequence and an indirect sequence; Allocating pixel points whose pixel distance is greater than or equal to the first preset pixel distance to the protrusion number series; Allocate pixel points whose pixel distance is smaller than the first preset pixel distance and larger than the second preset pixel distance to the common number series; Allocating pixel points whose pixel distance is less than or equal to the second preset pixel distance to the indirect number series; Generating deviation calibration for pixel points in the deviation series, the deviation calibration including a first deviation calibration and a second deviation calibration, counting a first quantity of the first deviation calibration, counting a second quantity of the second deviation calibration, and determining a deviation value of the grasped material according to the first quantity and the second quantity; When generating deviation calibration for the pixel points in the deviation sequence, it includes: Obtaining an average pixel distance of all pixels in the prominent series and recording it as a prominent value, obtaining an average pixel distance of all pixels in the common series and recording it as a common value, and obtaining an average pixel distance of all pixels in the indirect series and recording it as an indirect value; The pixel points in the protrusion series whose pixel distance is greater than the protrusion value generate the first deviation calibration, and the pixel points in the protrusion series whose pixel distance is less than or equal to the protrusion value generate the second deviation calibration; The pixel points in the common number series whose pixel distance is greater than the common value generate the first deviation calibration, and the pixel points in the common number series whose pixel distance is less than or equal to the common value generate the second deviation calibration; The pixel points in the indirect number series whose pixel distance is greater than the indirect value generate the first deviation calibration, and the pixel points in the indirect number series whose pixel distance is less than or equal to the indirect value generate the second deviation calibration; When determining the deviation value of the grasped material according to the first quantity and the second quantity, it includes: The deviation value is obtained by the following formula: ； in, Indicates the deviation value, represents the first quantity, represents the second quantity, and represents the weight coefficient, and ; Obtain historical deviation data of grasped materials, compare the historical deviation data with the deviation value, and determine whether to adjust the initial grasping force of the robot arm according to the comparison result. When it is determined to adjust the initial grasping force, determine the primary influence value of the force according to the friction coefficient between the grasped material and the robot arm, and determine the target influence value of the force according to the primary influence value of the force and the material weight of the grasped material; The force target impact value is derived from the following formula: ； in, Indicates the intensity target impact value, Indicates the primary impact value of the force, Indicates the material weight of the grabbed material; The adjustment factor of the initial grasping force is determined according to the force target influence value, the initial grasping force is adjusted according to the adjustment factor, and the robotic arm is controlled according to the adjusted initial grasping force to complete the feeding operation.

2. The feeding control method based on a robotic arm according to claim 1 is characterized in that: When the orientation image is adjusted to obtain the deviation image of the area to be processed, it includes: The orientation image is denoised by wavelet transform, the denoised orientation image is brightness adjusted according to an adaptive histogram equalization algorithm to obtain a brightness image, pyramid decomposition is used to capture image features of the brightness image, the brightness image is registered based on the image features, the registered orientation images are merged to obtain the deviation image of the area to be processed.

3. The feeding control method based on a robotic arm according to claim 2 is characterized in that: When obtaining historical deviation data of the grasped material, comparing the historical deviation data with the deviation value, and judging whether to adjust the initial grasping force of the robot arm according to the comparison result, it includes: Obtain the historical deviation qualified mean value of the same material weight and the same material attribute as the grabbed material; When the deviation value is greater than the qualified mean value of the historical deviation, it is determined to adjust the initial grasping force of the robot arm; When the deviation value is less than or equal to the qualified mean value of the historical deviation, it is determined that the initial grasping force of the robot arm is not adjusted, and the feeding operation is completed according to the initial grasping force.

4. The feeding control method based on a robotic arm according to claim 3 is characterized in that: When the primary force impact value is determined according to the friction coefficient between the grasped material and the robot arm, and the target force impact value is determined according to the primary force impact value and the material weight of the grasped material, it includes: Preset a first preset force primary impact value, a second preset force primary impact value, and a third preset force primary impact value, wherein the first preset force primary impact value is smaller than the second preset force primary impact value, and the second preset force primary impact value is smaller than the third preset force primary impact value; Determine the friction coefficient μ of the robotic arm according to the material properties of the grasped material and the material properties of the robotic arm; When μ is greater than 0.3, the first preset force primary influence value is determined as the force primary influence value; When μ is less than or equal to 0.3 and greater than 0.15, the second preset force primary influence value is determined as the force primary influence value; When μ is less than or equal to 0.15, the third preset force primary influence value is determined as the force primary influence value.

5. The feeding control method based on a robot arm according to claim 4 is characterized in that: When determining the adjustment factor of the initial grasping force according to the force target influence value, it includes: Preset a first preset force target impact value, a second preset force target impact value, and a third preset force target impact value, wherein the first preset force target impact value is greater than the second preset force target impact value, and the second preset force target impact value is greater than the third preset force target impact value; A first preset adjustment factor N1, a second preset adjustment factor N2 and a third preset adjustment factor N3 are preset, and 1.1<N3<N2<N1<2; When the force target impact value is greater than or equal to the first preset force target impact value, using the first preset adjustment factor N1 as the adjustment factor of the initial grasping force; When the force target impact value is less than the first preset force target impact value and greater than the second preset force target impact value, using the second preset adjustment factor N2 as the adjustment factor of the initial grasping force; When the force target influence value is less than or equal to the third preset force target influence value, the third preset adjustment factor N3 is used as the adjustment factor of the initial grasping force.

6. The feeding control method based on a robot arm according to claim 5 is characterized in that: When the initial grasping force is adjusted according to the adjustment factor, it includes: The adjustment factor is proportional to the initial grasping force.

7. A feeding control system based on a robot arm, applied to the feeding control method based on a robot arm as claimed in any one of claims 1 to 6, characterized in that: include: A sequence determination module is configured to divide the grabbed material into a plurality of areas to be processed, obtain an orientation image of the area to be processed, adjust the orientation image to obtain a deviation image of the area to be processed, obtain pixel points of each deviation image, determine a cluster center point of the deviation image, and determine a deviation sequence according to a pixel distance between each pixel point and the cluster center point; a deviation analysis module, configured to generate a deviation calibration for the pixel points in the deviation series, the deviation calibration including a first deviation calibration and a second deviation calibration, counting a first quantity of the first deviation calibration, counting a second quantity of the second deviation calibration, and determining a deviation value of the grasped material according to the first quantity and the second quantity; a force processing module, configured to obtain historical deviation data of grasped materials, compare the historical deviation data with the deviation value, determine whether to adjust the initial grasping force of the robot arm according to the comparison result, and when it is determined to adjust the initial grasping force, determine the primary influence value of the force according to the friction coefficient between the grasped material and the robot arm, and determine the target influence value of the force according to the primary influence value of the force and the material weight of the grasped material; The force adjustment module is configured to determine the adjustment factor of the initial grasping force according to the force target influence value, adjust the initial grasping force according to the adjustment factor, and control the robotic arm according to the adjusted initial grasping force to complete the loading operation.

Images