An intelligent control method and system for an oil and gas well downhole visual manipulator and a medium

Abstract

The embodiment of the application provides a kind of oil and gas well downhole visual manipulator intelligent control method, system and medium, the method comprises: real-time acquisition downhole image in the process of manipulator movement, the downhole image is preprocessed, and enhanced image is obtained;The feature value of enhanced image is input into the preset gray scale calculation model, and the output image gray scale value is output;The gray scale value of enhanced image is differentially calculated with the preset gray scale threshold, and the gray scale difference is obtained;Determine whether the gray scale difference is greater than the preset gray scale difference threshold;If greater, it is determined that the falling object area is determined;If less than, then determine the current area as background area;According to the falling object parameter information, generate the grabbing parameter, control manipulator to move to the predetermined position and carry out grabbing;The application is by real-time image acquisition and accurate determination of falling object position information, and manipulator is adjusted according to falling object position information and moves position in real time, and according to the profile of the collected falling object, the intelligent grabbing of the manipulator is adjusted to the profile of the falling object and the grabbing state.

Description

Technical Field

[0001] This application relates to the field of robotic gripper control, and more specifically, to an intelligent control method, system, and medium for a downhole vision robotic gripper for oil and gas wells. Background Technology

[0002] A robotic arm is an automated device that mimics certain movements of the human hand and arm to grasp, move objects, or operate tools according to a fixed program. Its key feature is that it can be programmed to perform various pre-defined tasks, combining the advantages of both humans and machines in its construction and performance. The robotic arm was the earliest industrial robotic arm and also the earliest modern robotic arm. It can replace heavy human labor to achieve mechanization and automation of production, and can operate in hazardous environments to protect personal safety. The complex working environment of oil and gas wells necessitates the flexibility of robotic arms to collect and intelligently grasp falling objects. However, existing robotic arms cannot acquire images in real time and accurately determine the position of falling objects. They cannot adjust their movement position in real time based on the object's position information, and they cannot capture the object's outline to adjust their grasping posture accordingly. This results in poor grasping performance and a tendency for the grasped object to detach from the robotic arm again.

[0003] Effective technical solutions are urgently needed to address the above problems. Summary of the Invention

[0004] The purpose of this application is to provide an intelligent control method, system, and medium for a downhole vision robot in oil and gas wells. This method can acquire images in real time and accurately determine the location information of falling objects. The robot can adjust its movement position in real time according to the location information of the falling objects, and adjust its grasping state according to the contour of the falling objects to perform intelligent grasping.

[0005] This application also provides an intelligent control method for a downhole vision robot in oil and gas wells, including:

[0006] The robotic arm acquires downhole images in real time during its movement, and preprocesses these images to obtain enhanced images.

[0007] Extract the enhanced image feature values ​​and input the enhanced image feature values ​​into a preset grayscale calculation model to output the image grayscale values;

[0008] The grayscale difference is obtained by calculating the difference between the enhanced image grayscale value and the preset grayscale threshold.

[0009] Determine whether the grayscale difference is greater than the preset grayscale difference threshold;

[0010] If the value is greater than the specified value, the current area is extracted and identified as the area where the object fell.

[0011] If it is smaller than the specified value, then the current area is determined to be the background area;

[0012] Extract the outline of the falling object to obtain its parameter information;

[0013] Generate capture parameters based on the falling object parameter information.

[0014] The robotic arm is controlled to move to the predetermined position to perform the grasping action based on the grasping parameters.

[0015] Optionally, in the intelligent control method for a downhole vision robot in oil and gas wells according to embodiments of this application, downhole images are acquired in real time during the robot's movement, and the downhole images are preprocessed to obtain enhanced images; including:

[0016] Acquire downhole images and segment them into several sub-images;

[0017] The feature values ​​of several sub-images are extracted respectively, and the difference between the feature value of each sub-image and the preset feature threshold is calculated to obtain the feature difference;

[0018] If the feature difference is less than the first threshold, then several sub-images are filtered and their features are fused to obtain the processed image.

[0019] If the feature difference is greater than the first threshold and less than the second threshold, the feature values ​​of several sub-images are averaged to obtain the feature mean.

[0020] The feature mean is compared with the preset feature mean to obtain the feature deviation rate;

[0021] Determine whether the feature deviation rate is greater than the preset feature deviation threshold;

[0022] If the value is greater than the maximum and minimum feature values ​​of each sub-image, then the feature mean of several sub-images is calculated again.

[0023] If the value is less than 1, then the features of several sub-images are fused to obtain the processed image.

[0024] If the feature difference is greater than the second threshold, the corresponding sub-image is removed.

[0025] The first threshold is less than the second threshold.

[0026] Optionally, in the intelligent control method for downhole vision manipulators in oil and gas wells according to embodiments of this application, if the feature difference is greater than a second threshold, the corresponding sub-image is discarded, including:

[0027] If the feature value is greater than the second threshold, the corresponding sub-image is removed, and the number of removed sub-images is recorded as 1.

[0028] When the corresponding sub-image is removed again, the number of sub-images removed is recorded as 2, and so on;

[0029] Determine whether the number of sub-images to be removed exceeds a preset threshold.

[0030] If the value is greater than the specified value, delete the current downhole image and adjust the acquisition parameters to reacquire downhole images.

[0031] Optionally, in the intelligent control method for downhole vision manipulators in oil and gas wells according to embodiments of this application, extracting the outline of the falling object and obtaining the falling object parameter information further includes:

[0032] Obtain the outline of the falling object to obtain the outline dataset, and decompose the outline dataset into several components.

[0033] Solve for the wavelet coefficients of each component;

[0034] Determine the correlation between contour datasets acquired at different times and wavelet coefficients;

[0035] If the correlation is greater than a preset threshold, it is recorded as useful data and retained.

[0036] If the relevant data is less than the preset threshold, it is recorded as noise and discarded.

[0037] Useful data is fused together to calculate information about falling objects.

[0038] Optionally, in the intelligent control method for a downhole vision robot in oil and gas wells according to embodiments of this application, controlling the robot to move to a predetermined position for grasping based on grasping parameters includes:

[0039] Extract the outline of the falling object to obtain its parameter information. Then, input the falling object parameter information into the position prediction model to obtain the falling object's position information.

[0040] Obtain the position information of the robotic arm, input the position information of the robotic arm and the position information of the falling object into the path planning model, and generate the movement path of the robotic arm;

[0041] The robotic arm moves according to its movement path, and its real-time position information is collected.

[0042] The position deviation rate is obtained by comparing the real-time position information of the robotic arm with the preset position information.

[0043] Determine whether the position deviation rate is greater than the preset deviation rate threshold;

[0044] If the value is greater than the specified value, correction information is generated, and the robot's movement state parameters are adjusted based on the correction information.

[0045] Optionally, in the intelligent control method for a downhole vision robot in oil and gas wells according to embodiments of this application, controlling the robot to move to a predetermined position for grasping based on grasping parameters further includes:

[0046] Obtain the location information of the falling object and establish a three-dimensional coordinate system to obtain the location coordinate information of the falling object;

[0047] Collect robot arm posture information and calibrate the position of grasping joints;

[0048] A three-dimensional coordinate system is established based on the position of the grasping joint, and the position coordinate information of the grasping joint is obtained.

[0049] The coordinates of the captured joint points are transformed into three-dimensional coordinates to obtain the joint point coordinates corresponding to the position coordinates of the falling object.

[0050] The coordinates of the key points are compared with the coordinates of the falling object to obtain the coordinate deviation rate;

[0051] Determine whether the coordinate deviation rate is greater than the preset coordinate deviation threshold;

[0052] If the value is greater than the value, correction information is generated, and the robot's posture information is adjusted according to the correction information, and the position of the gripping joint is adjusted in real time.

[0053] Secondly, embodiments of this application provide an intelligent control system for a downhole vision robot for oil and gas wells. This system includes a memory and a processor. The memory contains a program for an intelligent control method for a downhole vision robot for oil and gas wells. When the program for the intelligent control method for a downhole vision robot for oil and gas wells is executed by the processor, it performs the following steps:

[0054] The robotic arm acquires downhole images in real time during its movement, and preprocesses these images to obtain enhanced images.

[0055] Extract the enhanced image feature values ​​and input the enhanced image feature values ​​into a preset grayscale calculation model to output the image grayscale values;

[0056] The grayscale difference is obtained by calculating the difference between the enhanced image grayscale value and the preset grayscale threshold.

[0057] Determine whether the grayscale difference is greater than the preset grayscale difference threshold;

[0058] If the value is greater than the specified value, the current area is extracted and identified as the area where the object fell.

[0059] If it is smaller than the specified value, then the current area is determined to be the background area;

[0060] Extract the outline of the falling object to obtain its parameter information;

[0061] Generate capture parameters based on the falling object parameter information.

[0062] The robotic arm is controlled to move to the predetermined position to perform the grasping action based on the grasping parameters.

[0063] Optionally, in the intelligent control system for the downhole vision robot of the oil and gas well in this application embodiment, downhole images are acquired in real time during the movement of the robot, and the downhole images are preprocessed to obtain enhanced images; including:

[0064] Acquire downhole images and segment them into several sub-images;

[0065] The feature values ​​of several sub-images are extracted respectively, and the difference between the feature value of each sub-image and the preset feature threshold is calculated to obtain the feature difference;

[0066] If the feature difference is less than the first threshold, then several sub-images are filtered and their features are fused to obtain the processed image.

[0067] If the feature difference is greater than the first threshold and less than the second threshold, the feature values ​​of several sub-images are averaged to obtain the feature mean.

[0068] The feature mean is compared with the preset feature mean to obtain the feature deviation rate;

[0069] Determine whether the feature deviation rate is greater than the preset feature deviation threshold;

[0070] If the value is greater than the maximum and minimum feature values ​​of each sub-image, then the feature mean of several sub-images is calculated again.

[0071] If the value is less than 1, then the features of several sub-images are fused to obtain the processed image.

[0072] If the feature difference is greater than the second threshold, the corresponding sub-image is removed.

[0073] The first threshold is less than the second threshold.

[0074] Optionally, in the intelligent control system for downhole vision manipulators in oil and gas wells according to embodiments of this application, if the feature difference is greater than a second threshold, the corresponding sub-image is discarded, including:

[0075] If the feature value is greater than the second threshold, the corresponding sub-image is removed, and the number of removed sub-images is recorded as 1.

[0076] When the corresponding sub-image is removed again, the number of sub-images removed is recorded as 2, and so on;

[0077] Determine whether the number of sub-images to be removed exceeds a preset threshold.

[0078] If the value is greater than the specified value, delete the current downhole image and adjust the acquisition parameters to reacquire downhole images.

[0079] Thirdly, embodiments of this application also provide a computer-readable storage medium, which includes a program for an intelligent control method of a downhole vision robot for oil and gas wells. When the program is executed by a processor, it implements the steps of the intelligent control method of the downhole vision robot for oil and gas wells as described above.

[0080] As can be seen from the above, the intelligent control method, system, and medium for a downhole vision robot in oil and gas wells provided in this application involves real-time acquisition of downhole images during the robot's movement, preprocessing the downhole images to obtain enhanced images, extracting feature values ​​from the enhanced images and inputting these feature values ​​into a preset grayscale calculation model to output image grayscale values, calculating the difference between the enhanced image grayscale values ​​and a preset grayscale threshold to obtain a grayscale difference, determining whether the grayscale difference is greater than a preset grayscale difference threshold, extracting the current area if it is greater, and identifying it as a falling object area, or identifying it as a background area if it is less than the threshold, extracting the falling object contour to obtain falling object parameter information, generating grasping parameters based on the falling object parameter information, and controlling the robot to move to a predetermined position for grasping based on the grasping parameters. This application achieves intelligent grasping by acquiring images in real time and accurately determining the falling object position information, allowing the robot to adjust its movement position in real time based on the falling object position information, and adjusting its grasping state according to the captured falling object contour.

[0081] Other features and advantages of this application will be set forth in the following description, and the advantages of this application will be apparent in part from the description, or may be learned by practicing the embodiments of this application. The objectives and other advantages of this application may be realized and obtained by means of the structures particularly pointed out in the written description, claims, and drawings. Attached Figure Description

[0082] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0083] Figure 1 A flowchart of the intelligent control method for a downhole vision robot in oil and gas wells provided in the embodiments of this application;

[0084] Figure 2 A flowchart of the downhole image preprocessing method for the intelligent control method of the downhole vision robot for oil and gas wells provided in the embodiments of this application;

[0085] Figure 3 A flowchart illustrating the method for calculating falling object parameter information in the intelligent control method for downhole vision manipulators in oil and gas wells provided in this application embodiment;

[0086] Figure 4 A flowchart of the method for obtaining manipulator movement state parameters in the intelligent control method for downhole vision manipulators in oil and gas wells provided in this application embodiment;

[0087] Figure 5 A flowchart illustrating the robotic gripping process of the intelligent control method for downhole vision robotic arms in oil and gas wells provided in this application embodiment;

[0088] Figure 6 A schematic diagram of the structure of the intelligent control system for the downhole vision robot arm of an oil and gas well provided in the embodiments of this application. Detailed Implementation

[0089] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of the embodiments. The components of the embodiments of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely represents selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0090] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, the terms "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0091] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating an intelligent control method for a downhole vision robot in oil and gas wells, as described in some embodiments of this application. This intelligent control method for a downhole vision robot in oil and gas wells is used in a terminal device and includes the following steps:

[0092] S101: The robot arm acquires downhole images in real time during its movement, and preprocesses the downhole images to obtain enhanced images;

[0093] S102, extract the enhanced image feature values ​​and input the enhanced image feature values ​​into the preset grayscale calculation model, and output the image grayscale value;

[0094] S103, calculate the difference between the enhanced image grayscale value and the preset grayscale threshold to obtain the grayscale difference;

[0095] S104, determine whether the grayscale difference is greater than the preset grayscale difference threshold; if it is greater, extract the current area and determine it as the falling object area; if it is less than, determine the current area as the background area.

[0096] S105, extract the outline of the falling object and obtain the falling object parameter information;

[0097] S106 generates grasping parameters based on the falling object parameter information, and controls the robotic arm to move to the predetermined position to grasp the object based on the grasping parameters.

[0098] It should be noted that judging the fallen object and background area in the downhole image based on the gray value and segmenting the fallen object and background area is more conducive to extracting the outline of the fallen object and locating its position.

[0099] The parameters of the falling object include its shape, size, weight, and material.

[0100] Please refer to Figure 2 , Figure 2 This is a flowchart illustrating a downhole image preprocessing method within an intelligent control method for an oil and gas well downhole vision robot, as described in some embodiments of this application. According to embodiments of the present invention, downhole images are acquired in real-time during the robot's movement, and preprocessed to obtain enhanced images; this includes:

[0101] S201, acquire downhole images and segment the downhole images into several sub-images;

[0102] S202, extract the feature values ​​of several sub-images respectively, and calculate the difference between the feature value of each sub-image and the preset feature threshold to obtain the feature difference;

[0103] S203, if the feature difference is less than the first threshold, then the several sub-images are filtered and the features of the several sub-images are fused to obtain the processed image; if the feature difference is greater than the first threshold and less than the second threshold, then the feature values ​​of the several sub-images are averaged to obtain the feature mean.

[0104] S204, compare the feature mean with the preset feature mean to obtain the feature deviation rate;

[0105] S205, determine whether the feature deviation rate is greater than the preset feature deviation threshold;

[0106] If the value is greater than the maximum and minimum feature values ​​of each sub-image, then the feature mean of several sub-images is calculated again.

[0107] If the value is less than 1, then the features of several sub-images are fused to obtain the processed image.

[0108] If the feature difference is greater than the second threshold, the corresponding sub-image is removed.

[0109] The first threshold is less than the second threshold.

[0110] According to an embodiment of the present invention, if the feature difference is greater than a second threshold, the corresponding sub-image is removed, including:

[0111] If the feature value is greater than the second threshold, the corresponding sub-image is removed, and the number of removed sub-images is recorded as 1.

[0112] When the corresponding sub-image is removed again, the number of sub-images removed is recorded as 2, and so on;

[0113] Determine whether the number of sub-images to be removed exceeds a preset threshold.

[0114] If the value is greater than the specified value, delete the current downhole image and adjust the acquisition parameters to reacquire downhole images.

[0115] It should be noted that the number of sub-images is 20-30, and the preset threshold is 15-20. If more than half of the sub-images are removed, it indicates that the collected downhole images are distorted, causing the downhole images to fail to accurately reflect the location and parameter information of obstacles, making it difficult to control and grasp the robot arm, resulting in a large deviation. In this case, it is necessary to re-collect downhole images.

[0116] Please refer to Figure 3 , Figure 3 This is a flowchart illustrating a method for calculating falling object parameter information in an intelligent control method for a downhole vision robot in oil and gas wells, as described in some embodiments of this application. According to embodiments of the present invention, the method further includes extracting the falling object contour to obtain falling object parameter information, and also includes:

[0117] S301, Obtain the outline of the falling object, obtain the outline dataset, and decompose the outline dataset into several components;

[0118] S302, solve for the wavelet coefficients of each component;

[0119] S303, Determine the correlation between contour datasets acquired at different times and wavelet coefficients;

[0120] S304, if the correlation is greater than the preset threshold, it is recorded as useful data and retained;

[0121] If the relevant data is less than the preset threshold, it is recorded as noise and discarded.

[0122] S305 integrates useful data to calculate information on falling objects.

[0123] Please refer to Figure 4 , Figure 4 This is a flowchart illustrating a method for acquiring robot movement state parameters in an intelligent control method for a downhole vision robot in oil and gas wells, as described in some embodiments of this application. According to an embodiment of the present invention, controlling the robot to move to a predetermined position for grasping based on grasping parameters includes:

[0124] S401, extract the outline of the falling object to obtain the falling object parameter information, and input the falling object parameter information into the position prediction model to obtain the falling object position information;

[0125] S402, Obtain the position information of the robotic arm, input the position information of the robotic arm and the position information of the falling object into the path planning model, and generate the movement path of the robotic arm;

[0126] S403, the robot arm moves according to the robot arm's movement path, and the robot arm's real-time position information is collected;

[0127] S404, compare the real-time position information of the robot with the preset position information to obtain the position deviation rate;

[0128] S405, determine whether the position deviation rate is greater than the preset deviation rate threshold; if it is, generate correction information and adjust the robot's movement state parameters according to the correction information.

[0129] Please refer to Figure 5 , Figure 5 This is a flowchart illustrating the robotic arm grasping process of an intelligent control method for a downhole vision robotic arm in oil and gas wells, as described in some embodiments of this application. According to embodiments of the present invention, controlling the robotic arm to move to a predetermined position for grasping based on grasping parameters further includes:

[0130] S501, acquire the location information of the falling object, establish three-dimensional coordinates, and obtain the location coordinate information of the falling object;

[0131] S502 collects robot arm posture information and calibrates the position of grasping joints;

[0132] S503, establish three-dimensional coordinates based on the position of the grasping joint, and obtain the position coordinate information of the grasping joint;

[0133] S504, perform three-dimensional coordinate transformation on the captured joint point position coordinate information to obtain the joint point coordinates corresponding to the position coordinate information of the falling object;

[0134] S505, compare the coordinates of the key points with the coordinates of the falling object to obtain the coordinate deviation rate;

[0135] S506, determine whether the coordinate deviation rate is greater than the preset coordinate deviation threshold; if it is, generate correction information, adjust the robot arm posture information according to the correction information, and adjust the gripping joint position in real time.

[0136] According to an embodiment of the present invention, it further includes:

[0137] The robot's posture information includes the robot's bending angle, rotation angle, and gripping force.

[0138] Please refer to Figure 6 , Figure 6 This is a schematic diagram of the structure of an intelligent control system for a downhole vision robot in oil and gas wells, as described in some embodiments of this application. Secondly, embodiments of this application provide an intelligent control system 6 for a downhole vision robot in oil and gas wells. This system includes a memory 61 and a processor 62. The memory includes a program for an intelligent control method for a downhole vision robot in oil and gas wells. When the program for the intelligent control method for a downhole vision robot in oil and gas wells is executed by the processor, it implements the following steps:

[0139] The robotic arm acquires downhole images in real time during its movement, and preprocesses these images to obtain enhanced images.

[0140] Extract the enhanced image feature values ​​and input the enhanced image feature values ​​into a preset grayscale calculation model to output the image grayscale values;

[0141] The grayscale difference is obtained by calculating the difference between the enhanced image grayscale value and the preset grayscale threshold.

[0142] Determine whether the grayscale difference is greater than the preset grayscale difference threshold;

[0143] If the value is greater than the specified value, the current area is extracted and identified as the area where the object fell.

[0144] If it is smaller than the specified value, then the current area is determined to be the background area;

[0145] Extract the outline of the falling object to obtain its parameter information;

[0146] Generate capture parameters based on the falling object parameter information.

[0147] The robotic arm is controlled to move to the predetermined position to perform the grasping action based on the grasping parameters.

[0148] It should be noted that judging the fallen object and background area in the downhole image based on the gray value and segmenting the fallen object and background area is more conducive to extracting the outline of the fallen object and locating its position.

[0149] The parameters of the falling object include its shape, size, weight, and material.

[0150] According to an embodiment of the present invention, downhole images are acquired in real time during the movement of the robotic arm, and the downhole images are preprocessed to obtain enhanced images; including:

[0151] Acquire downhole images and segment them into several sub-images;

[0152] The feature values ​​of several sub-images are extracted respectively, and the difference between the feature value of each sub-image and the preset feature threshold is calculated to obtain the feature difference;

[0153] If the feature difference is less than the first threshold, then several sub-images are filtered and their features are fused to obtain the processed image.

[0154] If the feature difference is greater than the first threshold and less than the second threshold, the feature values ​​of several sub-images are averaged to obtain the feature mean.

[0155] The feature mean is compared with the preset feature mean to obtain the feature deviation rate;

[0156] Determine whether the feature deviation rate is greater than the preset feature deviation threshold;

[0157] If the value is greater than the maximum and minimum feature values ​​of each sub-image, then the feature mean of several sub-images is calculated again.

[0158] If the value is less than 1, then the features of several sub-images are fused to obtain the processed image.

[0159] If the feature difference is greater than the second threshold, the corresponding sub-image is removed.

[0160] The first threshold is less than the second threshold.

[0161] According to an embodiment of the present invention, if the feature difference is greater than a second threshold, the corresponding sub-image is removed, including:

[0162] If the feature value is greater than the second threshold, the corresponding sub-image is removed, and the number of removed sub-images is recorded as 1.

[0163] When the corresponding sub-image is removed again, the number of sub-images removed is recorded as 2, and so on;

[0164] Determine whether the number of sub-images to be removed exceeds a preset threshold.

[0165] If the value is greater than the specified value, delete the current downhole image and adjust the acquisition parameters to reacquire downhole images.

[0166] It should be noted that the number of sub-images is 20-30, and the preset threshold is 15-20. If more than half of the sub-images are removed, it indicates that the collected downhole images are distorted, causing the downhole images to fail to accurately reflect the location and parameter information of obstacles, making it difficult to control and grasp the robot arm, resulting in a large deviation. In this case, it is necessary to re-collect downhole images.

[0167] According to an embodiment of the present invention, extracting the contour of the falling object to obtain falling object parameter information further includes:

[0168] Obtain the outline of the falling object to obtain the outline dataset, and decompose the outline dataset into several components.

[0169] Solve for the wavelet coefficients of each component;

[0170] Determine the correlation between contour datasets acquired at different times and wavelet coefficients;

[0171] If the correlation is greater than a preset threshold, it is recorded as useful data and retained.

[0172] If the relevant data is less than the preset threshold, it is recorded as noise and discarded.

[0173] Useful data is fused together to calculate information about falling objects.

[0174] According to an embodiment of the present invention, controlling a robotic arm to move to a predetermined position for grasping based on grasping parameters includes:

[0175] Extract the outline of the falling object to obtain its parameter information. Then, input the falling object parameter information into the position prediction model to obtain the falling object's position information.

[0176] Obtain the position information of the robotic arm, input the position information of the robotic arm and the position information of the falling object into the path planning model, and generate the movement path of the robotic arm;

[0177] The robotic arm moves according to its movement path, and its real-time position information is collected.

[0178] The position deviation rate is obtained by comparing the real-time position information of the robotic arm with the preset position information.

[0179] Determine whether the position deviation rate is greater than the preset deviation rate threshold;

[0180] If the value is greater than the specified value, correction information is generated, and the robot's movement state parameters are adjusted based on the correction information.

[0181] According to an embodiment of the present invention, controlling the robotic arm to move to a predetermined position for grasping based on grasping parameters further includes:

[0182] Obtain the location information of the falling object and establish a three-dimensional coordinate system to obtain the location coordinate information of the falling object;

[0183] Collect robot arm posture information and calibrate the position of grasping joints;

[0184] A three-dimensional coordinate system is established based on the position of the grasping joint, and the position coordinate information of the grasping joint is obtained.

[0185] The coordinates of the captured joint points are transformed into three-dimensional coordinates to obtain the joint point coordinates corresponding to the position coordinates of the falling object.

[0186] The coordinates of the key points are compared with the coordinates of the falling object to obtain the coordinate deviation rate;

[0187] Determine whether the coordinate deviation rate is greater than the preset coordinate deviation threshold;

[0188] If the value is greater than the value, correction information is generated, and the robot's posture information is adjusted according to the correction information, and the position of the gripping joint is adjusted in real time.

[0189] According to an embodiment of the present invention, it further includes:

[0190] The robot's posture information includes the robot's bending angle, rotation angle, and gripping force.

[0191] A third aspect of the present invention provides a computer-readable storage medium including a program for an intelligent control method of a downhole vision robot for oil and gas wells. When the program is executed by a processor, it implements the steps of the intelligent control method of the downhole vision robot for oil and gas wells as described above.

[0192] This invention discloses an intelligent control method, system, and medium for a downhole vision robot in oil and gas wells. The method involves real-time acquisition of downhole images during robot movement, preprocessing these images to obtain enhanced images, extracting feature values ​​from the enhanced images and inputting these feature values ​​into a preset grayscale calculation model to output image grayscale values, calculating the difference between the enhanced image grayscale values ​​and a preset grayscale threshold, determining whether the grayscale difference is greater than the preset grayscale difference threshold, identifying the current region as a falling object region if it is greater, and identifying the current region as a background region if it is less than the threshold, extracting the falling object contour to obtain falling object parameter information, generating grasping parameters based on the falling object parameter information, and controlling the robot to move to a predetermined position for grasping based on the grasping parameters. This application achieves intelligent grasping by acquiring images in real-time and accurately determining the falling object position information, allowing the robot to adjust its movement position in real-time based on the falling object position information, and adjusting its grasping state according to the captured falling object contour.

[0193] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods, such as: multiple units or components can be combined, or integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the various components shown or discussed can be through some interfaces, and the indirect coupling or communication connection between devices or units can be electrical, mechanical, or other forms.

[0194] The units described above as separate components may or may not be physically separate. The components shown as units may or may not be physical units. They may be located in one place or distributed across multiple network units. Some or all of the units may be selected to achieve the purpose of this embodiment according to actual needs.

[0195] In addition, in the various embodiments of the present invention, each functional unit can be integrated into one processing unit, or each unit can be a separate unit, or two or more units can be integrated into one unit; the integrated unit can be implemented in hardware or in the form of hardware plus software functional units.

[0196] Those skilled in the art will understand that all or part of the steps of the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable storage medium. When the program is executed, it performs the steps of the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0197] Alternatively, if the integrated units of this invention are implemented as software functional modules and sold or used as independent products, they can also be stored in a readable storage medium. Based on this understanding, the technical solutions of the embodiments of this invention, or the parts that contribute to the prior art, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as mobile storage devices, ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for intelligent control of a downhole vision robot for oil and gas wells, characterized in that, include: The robotic arm acquires downhole images in real time during its movement, and preprocesses these images to obtain enhanced images. Extract the enhanced image feature values ​​and input the enhanced image feature values ​​into a preset grayscale calculation model to output the image grayscale values; The grayscale difference is obtained by calculating the difference between the enhanced image grayscale value and the preset grayscale threshold. Determine whether the grayscale difference is greater than a preset grayscale difference threshold; If the value is greater than the specified value, the current area is extracted and identified as the area where the object fell. If it is smaller than the specified value, then the current area is determined to be the background area; Extract the outline of the falling object to obtain its parameter information; Generate capture parameters based on the falling object parameter information. The robotic arm is controlled to move to a predetermined position to perform the grasping action based on the grasping parameters. The robotic arm acquires downhole images in real time during its movement, and preprocesses these images to obtain enhanced images; including: Acquire downhole images and segment them into several sub-images; The feature values ​​of several sub-images are extracted respectively, and the difference between the feature value of each sub-image and the preset feature threshold is calculated to obtain the feature difference; If the feature difference is less than the first threshold, then several sub-images are filtered and their features are fused to obtain the processed image. If the feature difference is greater than the first threshold and less than the second threshold, the feature values ​​of several sub-images are averaged to obtain the feature mean. The feature mean is compared with the preset feature mean to obtain the feature deviation rate; Determine whether the feature deviation rate is greater than a preset feature deviation threshold; If the value is greater than the maximum and minimum feature values ​​of each sub-image, then the feature mean of several sub-images is calculated again. If the value is less than 1, then the features of several sub-images are fused to obtain the processed image. If the feature difference is greater than the second threshold, the corresponding sub-image is removed. The first threshold is less than the second threshold; The step of removing the corresponding sub-image if the feature difference is greater than the second threshold includes: If the feature value is greater than the second threshold, the corresponding sub-image is removed, and the number of removed sub-images is recorded as 1. When the corresponding sub-image is removed again, the number of sub-images removed is recorded as 2, and so on; Determine whether the number of sub-images to be removed is greater than a preset threshold. If the value is greater than the specified value, delete the current downhole image and adjust the acquisition parameters to reacquire downhole images.

2. The intelligent control method for a downhole vision robot in oil and gas wells according to claim 1, characterized in that, The step of extracting the contour of the falling object and obtaining the falling object parameter information also includes: Obtain the outline of the falling object to obtain the outline dataset, and decompose the outline dataset into several components. Solve for the wavelet coefficients of each component; Determine the correlation between contour datasets acquired at different times and wavelet coefficients; If the correlation is greater than a preset threshold, it is recorded as useful data and retained. If the relevant data is less than the preset threshold, it is recorded as noise and discarded. Useful data is fused together to calculate information about falling objects.

3. The intelligent control method for a downhole vision robot in oil and gas wells according to claim 2, characterized in that, The step of controlling the robotic arm to move to a predetermined position for grasping according to the grasping parameters includes: Extract the outline of the falling object to obtain its parameter information. Then, input the falling object parameter information into the position prediction model to obtain the falling object's position information. Obtain the position information of the robotic arm, input the position information of the robotic arm and the position information of the falling object into the path planning model, and generate the movement path of the robotic arm; The robotic arm moves according to its movement path, and its real-time position information is collected. The position deviation rate is obtained by comparing the real-time position information of the robotic arm with the preset position information. Determine whether the position deviation rate is greater than a preset deviation rate threshold; If the value is greater than the specified value, correction information is generated, and the robot's movement state parameters are adjusted based on the correction information.

4. The intelligent control method for a downhole vision robot in oil and gas wells according to claim 3, characterized in that, The step of controlling the robotic arm to move to a predetermined position for grasping based on grasping parameters also includes: Obtain the location information of the falling object and establish a three-dimensional coordinate system to obtain the location coordinate information of the falling object; Collect robot arm posture information and calibrate the position of grasping joints; A three-dimensional coordinate system is established based on the position of the grasping joint, and the position coordinate information of the grasping joint is obtained. The coordinates of the captured joint points are transformed into three-dimensional coordinates to obtain the joint point coordinates corresponding to the position coordinates of the falling object. The coordinates of the key points are compared with the coordinates of the falling object to obtain the coordinate deviation rate; Determine whether the coordinate deviation rate is greater than a preset coordinate deviation threshold; If the value is greater than the value, correction information is generated, and the robot's posture information is adjusted according to the correction information, and the position of the gripping joint is adjusted in real time.

5. An intelligent control system for a downhole vision robot in oil and gas wells, characterized in that, The system includes a memory and a processor. The memory contains a program for an intelligent control method of a downhole vision robot for oil and gas wells. When the program for the intelligent control method of the downhole vision robot for oil and gas wells is executed by the processor, it performs the following steps: The robotic arm acquires downhole images in real time during its movement, and preprocesses these images to obtain enhanced images. Extract the enhanced image feature values ​​and input the enhanced image feature values ​​into a preset grayscale calculation model to output the image grayscale values; The grayscale difference is obtained by calculating the difference between the enhanced image grayscale value and the preset grayscale threshold. Determine whether the grayscale difference is greater than a preset grayscale difference threshold; If the value is greater than the specified value, the current area is extracted and identified as the area where the object fell. If it is smaller than the specified value, then the current area is determined to be the background area; Extract the outline of the falling object to obtain its parameter information; Generate capture parameters based on the falling object parameter information. The robotic arm is controlled to move to a predetermined position to perform the grasping action based on the grasping parameters. The robotic arm acquires downhole images in real time during its movement, and preprocesses these images to obtain enhanced images; including: Acquire downhole images and segment them into several sub-images; The feature values ​​of several sub-images are extracted respectively, and the difference between the feature value of each sub-image and the preset feature threshold is calculated to obtain the feature difference; If the feature difference is less than the first threshold, then several sub-images are filtered and their features are fused to obtain the processed image. If the feature difference is greater than the first threshold and less than the second threshold, the feature values ​​of several sub-images are averaged to obtain the feature mean. The feature mean is compared with the preset feature mean to obtain the feature deviation rate; Determine whether the feature deviation rate is greater than a preset feature deviation threshold; If the value is greater than the maximum and minimum feature values ​​of each sub-image, then the feature mean of several sub-images is calculated again. If the value is less than 1, then the features of several sub-images are fused to obtain the processed image. If the feature difference is greater than the second threshold, the corresponding sub-image is removed. The first threshold is less than the second threshold; The step of removing the corresponding sub-image if the feature difference is greater than the second threshold includes: If the feature value is greater than the second threshold, the corresponding sub-image is removed, and the number of removed sub-images is recorded as 1. When the corresponding sub-image is removed again, the number of sub-images removed is recorded as 2, and so on; Determine whether the number of sub-images to be removed is greater than a preset threshold. If the value is greater than the specified value, delete the current downhole image and adjust the acquisition parameters to reacquire downhole images.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a program for intelligent control of a downhole vision robot for oil and gas wells. When the program is executed by a processor, it implements the steps of the intelligent control method for a downhole vision robot for oil and gas wells as described in any one of claims 1 to 4.

Images