Rapid point collecting method and system for temperature measuring robot at bottom of aluminum electrolysis cell

Abstract

The invention provides an aluminum electrolysis cell bottom temperature measurement robot rapid point collection method and system, and the method comprises the steps: fitting a patrol inspection straight line of each electrolysis cell based on the coordinates of a sample point collected by the electrolysis cell, and generating the point position serial number and navigation pose data of the patrol inspection point of the electrolysis cell through uniform interpolation; collecting and recording detection configuration parameters and a reference image of each inspection point of the first electrolytic cell, and assigning the detection configuration parameters of each inspection point in the first electrolytic cell to each inspection point of other electrolytic cells; and for each inspection point of the rest of the electrolytic cells, acquiring an actual measurement image based on the detection configuration parameters of the inspection points, correcting a holder angle parameter in the detection configuration parameters of the current inspection point based on the deviation between the actual measurement image and the reference image, and determining navigation pose data of each inspection point of all the electrolytic cells and the corrected detection configuration parameters. The method not only improves the overall operation efficiency, but also ensures the consistency and accuracy of data acquisition, and provides reliable technical support for automatic inspection.

Description

Technical Field

[0001] This disclosure relates to the field of intelligent robot inspection applications, specifically to a rapid sampling method and system for temperature measurement robots at the bottom of aluminum electrolysis cells. Background Technology

[0002] Aluminum electrolytic cells are the core equipment in aluminum electrolytic production. Real-time and accurate monitoring of the cell bottom temperature is crucial for preventing major safety accidents such as cell leakage and shutdown, and for ensuring the continuity and stability of production. However, the bottom of electrolytic cells typically presents extremely harsh environmental conditions, including high temperatures, strong magnetic fields, and abundant dust. Relying on manual handheld temperature measurement is not only labor-intensive and poses high safety risks, but also suffers from data lag and inconsistency, failing to meet the growing demand for precise monitoring and early warning in modern aluminum smelting processes.

[0003] With the development of intelligent inspection technology, using robots to replace manual labor in performing tank bottom temperature measurement has become a clear industry trend. However, the prerequisite for achieving automated temperature measurement by robots is to pre-set a series of precise inspection points for the robots. Each inspection point must include the robot's navigation and positioning information (i.e., "where to go") and the alignment parameters of sensors such as the gimbal and camera (i.e., "how to see").

[0004] Currently, the traditional method for collecting inspection point data is as follows: First, remotely control the robot to the target point and record the navigation coordinate information; then control the gimbal, adjust the gimbal angle, camera focus value and focal length until the gimbal camera can clearly identify the target object, and record the gimbal angle, camera focus value and focal length at this time; then continue to the next target point and repeat the above two steps.

[0005] However, a single aluminum electrolysis plant in China typically has dozens of electrolytic cells, each containing hundreds of inspection points, resulting in a total number of inspection points that can reach thousands or even tens of thousands. Taking a plant with 50 electrolytic cells and 200 inspection points per cell as an example, it would take approximately 50 hours to collect data on all inspection points using traditional methods, which is extremely time-consuming and labor-intensive.

[0006] In the prior art, patent document CN111136655B discloses a method for rapidly acquiring inspection points, which uses linear interpolation of the coordinates of two endpoints to generate coordinate information of intermediate points along the path in batches. While this method improves coordinate acquisition efficiency to some extent, the generated inspection point information only includes navigation coordinates and lacks the attitude and imaging parameters necessary for precise alignment of detection sensors such as gimbals and cameras. Furthermore, this method relies solely on two endpoints to determine a straight line, failing to consider the random errors inherent in robot positioning and navigation, resulting in insufficient accuracy in fitting the straight line and making it difficult to meet the high-precision temperature measurement requirements of electrolytic cells.

[0007] Therefore, there is an urgent need to propose a new rapid data collection scheme for inspection points to solve the problems of low efficiency, insufficient accuracy and incomplete data collection at the bottom of aluminum electrolysis cells. Summary of the Invention

[0008] This disclosure provides a method and system for rapid temperature collection by a robot at the bottom of an aluminum electrolysis cell, which not only improves overall operational efficiency but also ensures the consistency and accuracy of data collection, providing reliable technical support for automated inspection work.

[0009] In a first aspect, this disclosure provides a method for rapid temperature measurement of the bottom of an aluminum electrolysis cell using a robot, comprising the following steps:

[0010] S1. For each electrolytic cell, based on the coordinates of the sample points collected by the electrolytic cell, fit the inspection line of each electrolytic cell, and uniformly interpolate on the inspection line to generate the point number and navigation pose data of all inspection points of the electrolytic cell.

[0011] S2, collect and record the detection configuration parameters and corresponding reference images of each inspection point corresponding to the serial number of the first electrolytic cell. The detection configuration parameters include the detection target, gimbal angle parameters and camera parameters.

[0012] S3, traverse all inspection points of the remaining electrolytic cells, and assign the detection configuration parameters of the inspection points in the first electrolytic cell that have the same position number as the current inspection point to the current inspection point;

[0013] S4. For each inspection point of the remaining electrolytic cells, based on the detection configuration parameters of the inspection points, collect measured images, perform target identification and feature matching between the reference image of the first electrolytic cell and the measured image collected at the current inspection point, obtain the image deviation, and correct the gimbal angle parameter in the detection configuration parameters of the current inspection point according to the image deviation.

[0014] S5. Based on steps S1 to S4, determine the navigation pose data and corrected detection configuration parameters of all inspection points of all electrolytic cells.

[0015] In some embodiments, the navigation pose data includes the navigation coordinates and navigation heading angle of the inspection point.

[0016] In some embodiments, step S1 specifically includes:

[0017] S1.1 For each electrolytic cell, determine the number of inspection points k, and collect the start point, at least one intermediate point and the end point on the inspection path of the electrolytic cell as sample points; wherein, the start point and end point of different electrolytic cells are located at the same relative position on the bottom of the cell.

[0018] S1.2, Based on the coordinates of the sample points collected by the electrolytic cell, fit the inspection line of the electrolytic cell;

[0019] S1.3, correct the coordinates of the start point and the end point to the inspection line, and uniformly interpolate on the inspection line between the corrected start point and the end point to obtain k-2 interpolation points. The corrected start point, the end point and all interpolation points together constitute the k inspection points of the electrolytic cell, and determine the navigation coordinates of each inspection point.

[0020] S1.4, Determine the navigation heading angle of each inspection point based on the direction of the inspection line;

[0021] S1.5, according to the spatial order from the start point to the end point, sequentially assign increasing point numbers to the k inspection points, and generate the point numbers and navigation pose data of all inspection points of the electrolytic cell. The navigation pose data includes the navigation coordinates and navigation heading angle of the inspection points.

[0022] In some embodiments, step S2 specifically includes:

[0023] S2.1, Based on the navigation pose data of each inspection point of the first electrolytic cell, control the robot to automatically move to the inspection point corresponding to each point number;

[0024] S2.2 At each inspection point, for the detection target at the inspection point, manually control the robot's gimbal and camera to make the detection target clearly aligned in the image;

[0025] S2.3 Record the gimbal angle parameters and camera parameters corresponding to the time when the detection target is clearly aligned, and bind them to the detection target to form the detection configuration parameters of the inspection point corresponding to the point number. The gimbal angle parameters include gimbal pitch angle and gimbal yaw angle, and the camera parameters include camera focal length and camera focus value.

[0026] S2.4 Based on the detection configuration parameters, a clear image of the detection target is acquired and saved as the reference image of the inspection point corresponding to the point number.

[0027] In some embodiments, step S4 specifically includes:

[0028] S4.1 For each inspection point of the remaining electrolytic cells, based on the navigation pose data of each inspection point, control the robot to move automatically to the inspection point, and adjust the gimbal and camera according to the detection configuration parameters of the inspection point to capture and obtain the measured image.

[0029] S4.2, Based on the location number of the inspection point, obtain the reference image of the inspection point with the same location number in the first electrolytic cell;

[0030] S4.3, a deep learning-based visual recognition algorithm is used to detect and identify targets in the reference image and the test image respectively; based on cross-frame target registration technology, feature matching is performed on the detected targets in the test image and the reference image to obtain the pixel coordinate offset and pose deviation angle of the detected target in the test image relative to the detected target in the reference image;

[0031] S4.4 Based on the pixel coordinate offset, attitude deviation angle, and the field of view and resolution of the camera, the gimbal angle compensation amount is calculated to correct the gimbal angle parameter in the current inspection point detection configuration parameters.

[0032] In some embodiments, the navigation heading angle of each inspection point is the slope of the inspection line.

[0033] In some embodiments, for each electrolytic cell, the sample point includes a start point, an intermediate point, and an end point.

[0034] Secondly, this disclosure provides a rapid temperature sampling system for the bottom of an aluminum electrolysis cell using a robot, for running the aforementioned rapid temperature sampling method for the bottom of an aluminum electrolysis cell, including:

[0035] The navigation pose generation module is used to fit an inspection line for each electrolytic cell based on the coordinates of the sample points collected by the electrolytic cell, and to uniformly interpolate on the inspection line to generate the point number and navigation pose data of all inspection points of the electrolytic cell.

[0036] The reference parameter generation module is used to collect and record the detection configuration parameters and corresponding reference images of the inspection points corresponding to the serial numbers of each point in the first electrolytic cell. The detection configuration parameters include the detection target, the gimbal angle parameters, and the camera parameters.

[0037] The parameter assignment module is used to traverse all inspection points of the other electrolytic cells and assign the detection configuration parameters of the inspection points in the first electrolytic cell that have the same position number as the current inspection point to the current inspection point.

[0038] The parameter correction module is used to collect measured images for each inspection point of the other electrolytic cells based on the detection configuration parameters of the inspection points, perform target identification and feature matching between the reference image of the first electrolytic cell and the measured image collected at the current inspection point to obtain the image deviation, and correct the gimbal angle parameter in the detection configuration parameters of the current inspection point according to the image deviation.

[0039] The inspection parameter generation module is used to determine the navigation pose data and corrected detection configuration parameters of all inspection points in all electrolytic cells.

[0040] Thirdly, this disclosure provides an electronic device, including:

[0041] A memory is used to store computer programs; a processor is used to execute the programs stored in the memory to implement the steps of the method described.

[0042] Fourthly, this disclosure provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the method described.

[0043] The beneficial effects of this disclosure are that, compared with the prior art, this disclosure has the following advantages:

[0044] 1. Compared with the traditional point-by-point data collection method, the proposed solution innovatively achieves a significant improvement in data collection efficiency, with an improvement rate approaching n times the number of electrolytic cells (where n represents the number of electrolytic cells). This significantly reduces the data collection process, which originally took tens of hours, to a few hours, effectively improving the overall operational efficiency.

[0045] 2. The inspection point data generated by this disclosed solution not only includes precise navigation pose information but also detailed sensor parameters. By introducing a visual closed-loop correction mechanism, system errors can be automatically compensated, thereby ensuring that the robot can achieve precise alignment in each temperature measurement operation, guaranteeing the consistency and accuracy of data collection, and providing reliable technical support for automated inspection work. Attached Figure Description

[0046] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0047] Figure 1 This is a schematic diagram of a rapid temperature sampling method using a robot at the bottom of an aluminum electrolysis cell, provided in an embodiment of this disclosure.

[0048] Figure 2 This is a schematic diagram of a two-dimensional map of the bottom of an electrolytic cell provided in the disclosed embodiments;

[0049] Figure 3 This is a schematic diagram of another rapid temperature sampling method for the bottom of an aluminum electrolysis cell provided in this embodiment of the present disclosure;

[0050] Figure 4 This is a schematic diagram of another rapid temperature sampling method for the bottom of an aluminum electrolysis cell provided in this embodiment of the present disclosure;

[0051] Figure 5 This is a schematic diagram of another rapid temperature sampling method for the bottom of an aluminum electrolysis cell provided in this embodiment of the present disclosure;

[0052] Figure 6 This is a structural diagram of a rapid temperature sampling system for the bottom of an aluminum electrolysis cell provided in an embodiment of this disclosure.

[0053] The accompanying drawings have illustrated specific embodiments of this disclosure, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concepts of this disclosure to those skilled in the art through reference to particular embodiments. Detailed Implementation

[0054] The present disclosure will be further described below with reference to the accompanying drawings. The following embodiments are only used to illustrate the technical solutions of the present disclosure more clearly, and should not be used to limit the scope of protection of the present disclosure.

[0055] like Figure 1 As shown, the first embodiment of this disclosure provides a method for rapid temperature sampling using a robot at the bottom of an aluminum electrolysis cell, the method comprising the following steps:

[0056] S1. For each electrolytic cell, based on the coordinates of the sample points collected by the electrolytic cell, fit the inspection line of each electrolytic cell, and uniformly interpolate on the inspection line to generate the point number and navigation pose data of all inspection points of the electrolytic cell.

[0057] In this embodiment, the inspection point is a preset work point in the robot's inspection process.

[0058] In one alternative implementation, the navigation pose data includes the navigation coordinates and navigation heading angle of the inspection point.

[0059] like Figure 3 As shown, in one optional implementation, step S1 specifically includes the following sub-steps:

[0060] S1.1 For each electrolytic cell, determine the number of inspection points k, and collect the start point, at least one intermediate point and the end point on the inspection path of the electrolytic cell as sample points; wherein, the start point and end point of different electrolytic cells are located at the same relative position on the bottom of the cell.

[0061] S1.2, Based on the coordinates of the sample points collected by the electrolytic cell, fit the inspection line of the electrolytic cell;

[0062] S1.3, correct the coordinates of the start point and the end point to the inspection line, and uniformly interpolate on the inspection line between the corrected start point and the end point to obtain k-2 interpolation points. The corrected start point, the end point and all interpolation points together constitute the k inspection points of the electrolytic cell, and determine the navigation coordinates of each inspection point.

[0063] S1.4, Determine the navigation heading angle of each inspection point based on the direction of the inspection line;

[0064] S1.5, according to the spatial order from the start point to the end point, sequentially assign increasing point numbers to the k inspection points, and generate the point numbers and navigation pose data of all inspection points of the electrolytic cell. The navigation pose data includes the navigation coordinates and navigation heading angle of the inspection points.

[0065] Preferably, three sample points are collected, including a start point, an intermediate point, and an end point.

[0066] Preferably, to reduce random errors in the system, the least squares method is used to fit a straight line based on the navigation coordinates of the sample points.

[0067] Preferably, the coordinates of the start point and the end point are corrected to the inspection line using orthogonal projection.

[0068] Preferably, the navigation heading angle of each inspection point is the slope of the inspection line.

[0069] In this embodiment, since the robot inspects along the aforementioned straight line, the navigation heading angle... The slope 'a' of the above line satisfies:

[0070]

[0071] In actual operation, the bottom of the tank and the inspection route are first inspected manually to plan the location of inspection points and determine the number of inspection points, start point, middle point and end point of each tank. The remote-controlled robot collects 3 sample points in each tank. Based on the collected sample points, a fitted straight line is calculated and the navigation coordinates of all inspection points are generated by uniform interpolation. The navigation heading angle is determined based on the slope of the straight line. Finally, the navigation pose data of each inspection point is obtained. The navigation pose data is two-dimensional pose data, including the x-axis coordinate, y-axis coordinate and navigation heading angle yaw of the point, denoted as (x, y, yaw).

[0072] S2, collect and record the detection configuration parameters and corresponding reference images of each inspection point corresponding to the serial number of the first electrolytic cell. The detection configuration parameters include the detection target, gimbal angle parameters and camera parameters.

[0073] like Figure 4 As shown, in one optional implementation, step S2 specifically includes the following sub-steps:

[0074] S2.1, Based on the navigation pose data of each inspection point of the first electrolytic cell, control the robot to automatically move to the inspection point corresponding to each point number;

[0075] S2.2 At each inspection point, for the detection target at the inspection point, manually control the robot's gimbal and camera to make the detection target clearly aligned in the image;

[0076] S2.3 Record the gimbal angle parameters and camera parameters corresponding to the time when the detection target is clearly aligned, and bind them to the detection target to form the detection configuration parameters of the inspection point corresponding to the point number. The gimbal angle parameters include gimbal pitch angle and gimbal yaw angle, and the camera parameters include camera focal length and camera focus value.

[0077] S2.4 Based on the detection configuration parameters, a clear image of the detection target is acquired and saved as the reference image of the inspection point corresponding to the point number.

[0078] In this embodiment, the technical parameters for the detection configuration parameters of each inspection point include:

[0079] Inspection targets: Each inspection point corresponds to an inspection target at the bottom of the tank, such as steel bars, side walls, furnace bottom, etc.

[0080] Gimbal angle parameters: including gimbal pitch angle and gimbal yaw angle. The gimbal pitch angle is defined as the pitch angle that controls vertical rotation, and the gimbal yaw angle is defined as the yaw angle that controls horizontal rotation. These two angles together determine the gimbal's turning attitude, ensuring that the camera's optical axis is aligned with the target being detected.

[0081] Camera parameters include camera focal length and camera focus value. The former adjusts the field of view, while the latter adjusts the image sharpness. The two work together to ensure the sharpness of the detected target, providing high-quality image data for subsequent detection algorithms.

[0082] In actual operation, the system guides the robot to automatically move to the corresponding inspection point based on the navigation pose of the first electrolytic cell inspection point determined in step S1. After reaching the designated position, the operator manually controls the gimbal to rotate, precisely adjusting the pitch and yaw angles, while simultaneously adjusting the camera parameters, changing the focal length and focus value. This operation continues until the target object is clearly displayed in the monitoring screen and is centered. At this point, the system automatically records this set of parameters, specifically including the detected target, gimbal pitch angle, gimbal yaw angle, camera focal length, and camera focus value, and takes an image for storage. This image will serve as the reference image for that inspection point.

[0083] S3, traverse all inspection points of the remaining electrolytic cells, and assign the detection configuration parameters of the inspection points in the first electrolytic cell that have the same position number as the current inspection point to the current inspection point;

[0084] In this embodiment, for each inspection point of different electrolytic cells, a string format of "cell bottom number_point sequence number" is used, denoted as id, to uniquely distinguish different inspection points. For example, when the id is "48_136", it represents the 136th inspection point corresponding to cell number 48. This achieves orderly management of multiple cells and multiple points.

[0085] like Figure 2 As shown, an aluminum electrolysis plant typically has multiple electrolytic cells, each with a basically identical structure. The inspection route at the bottom of each cell is a horizontal straight line, and the inspection points are evenly distributed across the cells. Due to the consistency of the cell bottom structure, for cells with different numbers, the detection configuration parameters of the inspection points with the same position number are theoretically identical. For example, the detection configuration parameters of the inspection point with id n_m are theoretically identical to those of the inspection point with id 1_m.

[0086] By using the current inspection point ID, the detection configuration parameters of the inspection point corresponding to slot 1 collected in step S2 are obtained. This data is then directly assigned to the current inspection point, thus completing the automatic generation of detection configuration parameters for all inspection points.

[0087] S4. For each inspection point of the remaining electrolytic cells, based on the detection configuration parameters of the inspection points, collect measured images, perform target identification and feature matching between the reference image of the first electrolytic cell and the measured image collected at the current inspection point, obtain the image deviation, and correct the gimbal angle parameter in the detection configuration parameters of the current inspection point according to the image deviation.

[0088] Due to system errors such as robot positioning and navigation, and trench bottom construction, the detection configuration parameters at some inspection points deviate during actual inspections, affecting the temperature measurement results. Therefore, this embodiment proposes a closed-loop correction scheme of "reference image calibration—actual image acquisition—intelligent comparison and calibration" to automatically identify and compensate for the aforementioned deviations.

[0089] like Figure 5 As shown, in one optional implementation, step S4 specifically includes the following sub-steps:

[0090] S4.1 For each inspection point of the remaining electrolytic cells, based on the navigation pose data of each inspection point, control the robot to move automatically to the inspection point, and adjust the gimbal and camera according to the detection configuration parameters of the inspection point to capture and obtain the measured image.

[0091] S4.2, Based on the location number of the inspection point, obtain the reference image of the inspection point with the same location number in the first electrolytic cell;

[0092] S4.3, a deep learning-based visual recognition algorithm is used to detect and identify targets in the reference image and the test image respectively; based on cross-frame target registration technology, feature matching is performed on the detected targets in the test image and the reference image to obtain the pixel coordinate offset and pose deviation angle of the detected target in the test image relative to the detected target in the reference image;

[0093] S4.4 Based on the pixel coordinate offset, attitude deviation angle, and the field of view and resolution of the camera, the gimbal angle compensation amount is calculated to correct the gimbal angle parameter in the current inspection point detection configuration parameters.

[0094] As an example, and not a limitation, one specific implementation of the target detection and cross-frame target registration technology is as follows: First, YOLOv8 is used for target detection and recognition; then, the SuperPoint model is used to extract feature points of the detected targets in the reference image and the measured image, respectively; next, Hamming distance is used for coarse feature point matching; subsequently, the RANSAC algorithm is applied to optimize the coarse matching results, eliminating mismatched point pairs to obtain a set of accurately matched point pairs; finally, based on the set of accurately matched point pairs, the pixel coordinate offset of the detected target is calculated. and attitude deviation angle .

[0095] It should be noted that the target detection and recognition and cross-frame target registration techniques can also be implemented using other well-known techniques in the field. For specific implementation methods, please refer to standard literature or related tool libraries in the field of computer vision (such as OpenCV).

[0096] In one optional implementation, step S4.4 specifically includes the following sub-steps:

[0097] 1) Calculate the horizontal angle compensation amount The formula is as follows:

[0098]

[0099] 2) Calculate the vertical angle compensation amount The formula is as follows:

[0100]

[0101] Among them, the horizontal field of view of the camera is The horizontal resolution is U pixels; the vertical field of view is The vertical resolution is V pixels, and the horizontal offset of the pixel coordinates of the detected target is... The vertical offset of the pixel coordinates of the detected target is ;

[0102] 3) Calculate the gimbal pitch angle correction value by combining the inter-axis coupling compensation coefficient k. The formula is as follows:

[0103]

[0104] in, The initial test configuration parameters include the gimbal pitch angle; k is the two-axis coupling compensation proportional coefficient, ranging from 0.3 to 0.5, calibrated using a standard target experiment. To detect the attitude deviation angle of the target;

[0105] 4) Calculate the gimbal heading angle correction value The formula is as follows:

[0106]

[0107] in, The gimbal heading angle is used for initial detection configuration parameters.

[0108] Understandably, the horizontal field of view used in the calculation of converting pixel coordinate offset into angle compensation is... With vertical field of view These are not fixed values, but rather actual values ​​corresponding to the camera focal length in the current inspection point's detection configuration parameters. The system establishes a mapping table between focal length and field of view through pre-calibration for real-time lookup. The camera's horizontal resolution U and vertical resolution V are inherent properties of the camera's image sensor and are preset constants within the system.

[0109] The two-axis gimbal used in this embodiment has a mechanical coupling between its pitch and yaw axes. Specifically, adjusting the pitch angle will cause a slight additional change in the yaw angle. To address this, a coupling compensation coefficient k (which can be obtained through experimental calibration) is introduced into the pitch angle correction formula to eliminate the alignment error caused by this effect.

[0110] S5. Based on steps S1 to S4, determine the navigation pose data and corrected detection configuration parameters of all inspection points of all electrolytic cells.

[0111] This embodiment can generate a set of high-precision navigation and alignment parameters for all inspection points with the same serial number in all electrolytic cells. Based on this parameter set, the robot can automatically, accurately, and efficiently complete the image acquisition of all temperature measurement points at the bottom of the electrolytic cells, providing reliable data support for the accurate monitoring of the electrolytic cells.

[0112] like Figure 6 As shown, the second embodiment of this disclosure proposes a rapid sampling system 100 for aluminum electrolysis cell bottom temperature measurement robot, used to run the rapid sampling method for aluminum electrolysis cell bottom temperature measurement robot, including:

[0113] The navigation pose generation module 101 is used to fit an inspection line for each electrolytic cell based on the coordinates of the sample points collected by the electrolytic cell, and to uniformly interpolate on the inspection line to generate the point number and navigation pose data of all inspection points of the electrolytic cell.

[0114] The reference parameter generation module 102 is used to collect and record the detection configuration parameters and corresponding reference images of the inspection points corresponding to the serial numbers of each point in the first electrolytic cell. The detection configuration parameters include the detection target, the gimbal angle parameters and the camera parameters.

[0115] The parameter assignment module 103 is used to traverse all inspection points of the other electrolytic cells and assign the detection configuration parameters of the inspection points in the first electrolytic cell that have the same position number as the current inspection point to the current inspection point.

[0116] The parameter correction module 104 is used to collect measured images for each inspection point of the other electrolytic cells based on the detection configuration parameters of the inspection points, perform target identification and feature matching between the reference image of the first electrolytic cell and the measured image collected at the current inspection point to obtain the image deviation, and correct the gimbal angle parameter in the detection configuration parameters of the current inspection point according to the image deviation.

[0117] The inspection parameter generation module 105 is used to determine the navigation pose data and corrected detection configuration parameters of all inspection points of all electrolytic cells.

[0118] The third embodiment of this disclosure provides an electronic device, including: a memory for storing computer programs;

[0119] The processor is used to execute the program stored in the memory to implement the steps of the above embodiment of the method for rapid temperature sampling by a robot at the bottom of an aluminum electrolysis cell.

[0120] For details on the specific implementation of each step and related explanations, please refer to the aforementioned embodiment of a rapid sampling method for temperature measurement robot at the bottom of aluminum electrolysis cell, which will not be repeated here.

[0121] The memory of the electronic device mentioned in the fourth embodiment of this disclosure may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device.

[0122] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0123] This disclosure also proposes a computer-readable storage medium storing a computer program. When executed by a processor, this computer program implements the steps of the above-described method for rapid temperature sampling by a robot at the bottom of an aluminum electrolysis cell. For details on the specific implementation and explanations of each step, please refer to the foregoing method embodiments; further elaboration is not provided here.

[0124] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0125] It should be understood that the above embodiments are only used to illustrate the technical solutions of this disclosure, and not to limit them; although this disclosure has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this disclosure.

Claims

1. A rapid temperature sampling method for the bottom of an aluminum electrolysis cell using a robot, characterized in that, Includes the following steps: S1. For each electrolytic cell, based on the coordinates of the sample points collected by the electrolytic cell, fit the inspection line of each electrolytic cell, and uniformly interpolate on the inspection line to generate the point number and navigation pose data of all inspection points of the electrolytic cell. S2, collect and record the detection configuration parameters and corresponding reference images of each inspection point corresponding to the serial number of the first electrolytic cell. The detection configuration parameters include the detection target, gimbal angle parameters and camera parameters. S3, traverse all inspection points of the remaining electrolytic cells, and assign the detection configuration parameters of the inspection points in the first electrolytic cell that have the same position number as the current inspection point to the current inspection point; S4. For each inspection point of the remaining electrolytic cells, based on the detection configuration parameters of the inspection points, collect measured images, perform target identification and feature matching between the reference image of the first electrolytic cell and the measured image collected at the current inspection point, obtain the image deviation, and correct the gimbal angle parameter in the detection configuration parameters of the current inspection point according to the image deviation. S5. Based on steps S1 to S4, determine the navigation pose data and corrected detection configuration parameters of all inspection points of all electrolytic cells.

2. The rapid temperature sampling method for the bottom of an aluminum electrolysis cell using a robot according to claim 1, characterized in that, The navigation pose data includes the navigation coordinates and navigation heading angle of the inspection point.

3. The rapid temperature sampling method for the bottom of an aluminum electrolytic cell using a robot according to claim 1, characterized in that, Step S1 specifically includes: S1.1 For each electrolytic cell, determine the number of inspection points k, and collect the start point, at least one intermediate point and the end point on the inspection path of the electrolytic cell as sample points; wherein, the start point and end point of different electrolytic cells are located at the same relative position on the bottom of the cell. S1.2, Based on the coordinates of the sample points collected by the electrolytic cell, fit the inspection line of the electrolytic cell; S1.3, correct the coordinates of the start point and the end point to the inspection line, and uniformly interpolate on the inspection line between the corrected start point and the end point to obtain k-2 interpolation points. The corrected start point, the end point and all interpolation points together constitute the k inspection points of the electrolytic cell, and determine the navigation coordinates of each inspection point. S1.4, Determine the navigation heading angle of each inspection point based on the direction of the inspection line; S1.5, according to the spatial order from the start point to the end point, sequentially assign increasing point numbers to the k inspection points, and generate the point numbers and navigation pose data of all inspection points of the electrolytic cell. The navigation pose data includes the navigation coordinates and navigation heading angle of the inspection points.

4. The rapid temperature sampling method for the bottom of an aluminum electrolysis cell using a robot according to claim 1, characterized in that, Step S2 specifically includes: S2.1, Based on the navigation pose data of each inspection point of the first electrolytic cell, control the robot to automatically move to the inspection point corresponding to each point number; S2.2 At each inspection point, for the detection target at the inspection point, manually control the robot's gimbal and camera to make the detection target clearly aligned in the image; S2.3 Record the gimbal angle parameters and camera parameters corresponding to the time when the detection target is clearly aligned, and bind them to the detection target to form the detection configuration parameters of the inspection point corresponding to the point number. The gimbal angle parameters include gimbal pitch angle and gimbal yaw angle, and the camera parameters include camera focal length and camera focus value. S2.4 Based on the detection configuration parameters, a clear image of the detection target is acquired and saved as the reference image of the inspection point corresponding to the point number.

5. The rapid temperature sampling method for the bottom temperature measurement robot of the aluminum electrolysis cell according to claim 1, characterized in that, Step S4 specifically includes: S4.1 For each inspection point of the remaining electrolytic cells, based on the navigation pose data of each inspection point, control the robot to move automatically to the inspection point, and adjust the gimbal and camera according to the detection configuration parameters of the inspection point to capture and obtain the measured image. S4.2, Based on the location number of the inspection point, obtain the reference image of the inspection point with the same location number in the first electrolytic cell; S4.3, a deep learning-based visual recognition algorithm is used to detect and identify targets in the reference image and the test image respectively; based on cross-frame target registration technology, feature matching is performed on the detected targets in the test image and the reference image to obtain the pixel coordinate offset and pose deviation angle of the detected target in the test image relative to the detected target in the reference image; S4.4 Based on the pixel coordinate offset, attitude deviation angle, and the field of view and resolution of the camera, the gimbal angle compensation amount is calculated to correct the gimbal angle parameter in the current inspection point detection configuration parameters.

6. The rapid temperature sampling method for the bottom of an aluminum electrolytic cell using a robot according to claim 3, characterized in that, The navigation heading angle at each inspection point is the slope of the inspection line.

7. The rapid temperature sampling method for the bottom of an aluminum electrolytic cell using a robot according to claim 1, characterized in that, For each electrolytic cell, the sample point includes a start point, an intermediate point, and an end point.

8. A rapid temperature sampling system (100) for measuring the temperature at the bottom of an aluminum electrolytic cell using a robot, used to operate the rapid temperature sampling method for measuring the temperature at the bottom of an aluminum electrolytic cell as described in any one of claims 1-7, characterized in that... include: The navigation pose generation module (101) is used to fit the inspection line of each electrolytic cell based on the coordinates of the sample points collected by the electrolytic cell, and uniformly interpolate on the inspection line to generate the point number and navigation pose data of all inspection points of the electrolytic cell. The reference parameter generation module (102) is used to collect and record the detection configuration parameters and corresponding reference images of the inspection points corresponding to the serial numbers of each point in the first electrolytic cell. The detection configuration parameters include the detection target, the gimbal angle parameters and the camera parameters. The parameter assignment module (103) is used to traverse all inspection points of the other electrolytic cells and assign the detection configuration parameters of the inspection points in the first electrolytic cell that have the same position number as the current inspection point to the current inspection point. The parameter correction module (104) is used to collect measured images for each inspection point of the other electrolytic cells based on the detection configuration parameters of the inspection points, perform target identification and feature matching between the reference image of the first electrolytic cell and the measured image collected at the current inspection point, obtain the image deviation, and correct the gimbal angle parameter in the detection configuration parameters of the current inspection point according to the image deviation. The inspection parameter generation module (105) is used to determine the navigation pose data and corrected detection configuration parameters of all inspection points of all electrolytic cells.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor for executing a program stored in memory to implement the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1-7.

Images