Rust removal robot control method and system

Abstract

The invention provides a rust removal robot control method and system, and belongs to the technical field of robot control, the method comprises the following steps: monitoring a target area based on a visual sensor of a rust removal robot to obtain first corrosion data; based on the first corrosion data, determining a target corrosion state of the target area; and based on the target corrosion state of the target area, the rust removal robot is controlled. According to the rust removal robot control method and system, the first corrosion data of the target area is obtained through the visual sensor, the target corrosion state is determined, accurate judgment of the corrosion condition can be achieved, and the accuracy is improved. According to different target corrosion states, the rust removal robot is controlled in a targeted mode, the situation of excessive rust removal or insufficient rust removal is avoided, the rust removal efficiency is improved, the rust removal quality is guaranteed, and the rust removal robot is suitable for different application scenes and requirements.

Description

Technical Field

[0001] The present disclosure belongs to the technical field of robot control, and more specifically, to a rust removal robot control method and system. Background Art

[0002] In many fields such as industrial production and infrastructure construction, metal equipment or components are easily oxidized and corroded due to long-term exposure to the natural environment, resulting in rust. Rust not only affects the appearance of the equipment, but more seriously reduces the structural strength, service life and working performance of the equipment, which may lead to safety accidents and cause huge economic losses. Rust removal mainly relies on workers to hold rust removal tools, such as sandpaper, electric grinding wheels, etc., to grind the rusted parts. However, this method has many disadvantages. On the one hand, manual rust removal is inefficient, and a large amount of manpower and time costs are required for large areas of rust. On the other hand, the rust removal working environment is often harsh, with harmful factors such as dust and noise. Long-term work in this work will cause serious damage to the health of workers. With the continuous advancement of science and technology, robot technology has gradually been applied to the field of rust removal to solve the problems of manual rust removal. However, the existing rust removal robots have shortcomings in control and cannot accurately perform efficient and intelligent rust removal operations according to the rust status of different areas, resulting in poor rust removal effects and serious waste of resources. Summary of the invention

[0003] The purpose of the present disclosure is to provide a rust removal robot control method and system to solve the problem that the existing rust removal robots cannot accurately perform efficient rust removal operations according to the rust state.

[0004] A first aspect of the embodiments of the present disclosure provides a rust removal robot control method, comprising: Based on the visual sensor of the rust removal robot, the target area is monitored to obtain the first rust data; Processing the first corrosion data based on a first random forest model to obtain differences of multiple recognition results, and determining a target corrosion state of the target area according to a comparison result between the differences of the multiple recognition results and the first difference; The rust removal robot is controlled based on the target rust state of the target area.

[0005] A second aspect of the disclosed embodiment provides a rust removal robot control system, comprising: A detection module, used for monitoring a target area based on a visual sensor of the rust removal robot to obtain first rust data; a state determination module, configured to process the first corrosion data based on a first random forest model to obtain a plurality of difference degrees of recognition results, and determine a target corrosion state of the target area according to a comparison result between the difference degrees of the plurality of recognition results and the first difference degree; A control module is used to control the rust removal robot based on the target rust state of the target area.

[0006] The beneficial effects of the rust removal robot control method and system provided by the disclosed embodiments are as follows: the present invention obtains the first rust data of the target area through a visual sensor, and uses advanced image processing and machine learning technology to determine the target rust state, which can achieve accurate judgment of the rust situation and improve accuracy. According to different target rust states, the rust removal robot is controlled in a targeted manner to avoid excessive or insufficient rust removal, which not only improves the rust removal efficiency, but also ensures the rust removal quality, and can also adapt to different application scenarios and needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0008] Figure 1 A schematic flow chart of a rust removal robot control method provided by an embodiment of the present disclosure; Figure 2 A structural block diagram of a rust removal robot control system provided in one embodiment of the present disclosure; Figure 3 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. DETAILED DESCRIPTION

[0009] In the following description, specific details such as specific system structures and technologies are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present disclosure. However, it should be clear to those skilled in the art that the present disclosure may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid obstructing the description of the present disclosure with unnecessary details.

[0010] To make the purpose, technical solutions and advantages of the present disclosure clearer, the following will be combined with the attached Figure 1-3 The following description will be given by way of specific examples.

[0011] Please refer to Figure 1 , Figure 1A flow chart of a rust removal robot control method provided in an embodiment of the present disclosure, the method comprising: S101: Based on the visual sensor of the rust removal robot, the target area is monitored to obtain first rust data.

[0012] In this embodiment, the visual sensor uses a high-resolution camera, which can shoot the target area in all directions and at multiple angles, obtain image data containing information such as the surface color and texture of the target area, and obtain the first corrosion data through preliminary processing of the image data, such as graying and noise reduction. The first corrosion data is obtained by the visual sensor by monitoring the target area, including the corrosion area, corrosion degree, corrosion texture, etc. of the target area. The target area refers to the monitored target, such as the steel beam of a large steel bridge, or the metal shell of industrial equipment.

[0013] In addition, the visual sensor can be a fixed-position visual sensor that can monitor the rust status in a preset area (ie, a target area). The installation location includes not only the rust removal robot body but also other locations that are conducive to data collection.

[0014] S102: Processing the first corrosion data based on the first random forest model to obtain differences of multiple recognition results, and determining a target corrosion state of the target area according to a comparison result of the differences of the multiple recognition results with the first difference.

[0015] In this embodiment, the first random forest model is trained based on a large amount of corrosion sample data and has powerful pattern recognition and data analysis capabilities. The first corrosion data is processed in parallel by multiple decision trees, and each decision tree classifies and judges the first corrosion data according to its own algorithm rules, and then generates multiple recognition results. These recognition results are not completely consistent, and there are differences to a certain extent. The difference between these recognition results is calculated, and the difference is compared with the pre-set first difference. If the difference between multiple recognition results is less than the first difference, it indicates that the model's judgment on the corrosion state of the target area is more concentrated and stable, and it can be determined that the target area is in a relatively single and clear target corrosion state; on the contrary, if the difference is greater than the first difference, it means that there are large differences in the recognition results, and further analysis or other means are required to determine the target corrosion state. The first difference can be set based on experience or the requirements of the user.

[0016] S103: Controlling the rust removal robot based on the target rust state of the target area.

[0017] In this embodiment, different control strategies or control methods can be set according to different target corrosion states. For example, if the target corrosion state is light corrosion, and only a thin layer of rust is formed on the surface of the object, the laser power output by the robot is controlled to be low and the scanning speed is fast. Or the high-precision grinding device carried by the robot is controlled to operate at a lower speed to grind the rusted surface. At the same time, the matching dust collection equipment is controlled to suck away the rust generated by grinding. If the target corrosion state is moderate corrosion, the rust layer has a certain thickness, which poses a potential threat to the structure of the object. At this time, the laser power output by the robot is controlled to be medium and the scanning speed is fast. If the target corrosion state is severe corrosion, the rust layer is thick and may have penetrated deep into the object, seriously affecting the structural strength. At this time, the laser power output by the robot is controlled to be adjusted to the highest level, so that the laser continues to bombard the rusted area with high-frequency and high-energy pulses, and the scanning speed is slow.

[0018] It can be concluded from the above that the present invention obtains the first rust data of the target area through the visual sensor, and uses advanced image processing and machine learning technology to determine the target rust state, which can achieve accurate judgment of the rust situation and improve accuracy. According to different target rust states, the rust removal robot is controlled in a targeted manner to avoid excessive or insufficient rust removal, which not only improves the rust removal efficiency, but also ensures the rust removal quality, and is also adapted to different application scenarios and needs.

[0019] In one embodiment of the present disclosure, determining a target rust state of a target area based on a comparison result of a difference between a plurality of recognition results and a first difference includes: in response to a difference between a plurality of recognition results being greater than or equal to a first difference, monitoring the target area based on an ultrasonic sensor built into a rust removal robot to obtain second rust data; determining target rust data based on the first rust data and the second rust data; in response to a difference between a plurality of recognition results being less than the first difference, using the first rust data as target rust data; and determining a target rust state of the target area based on the target rust data.

[0020] Specifically, considering that the target area can be curved and have curvature, although some visual sensors have certain depth perception capabilities, it may not be possible to accurately measure the thickness of the rust layer by relying solely on visual sensors; some rust hidden in gaps, holes or depressions cannot be fully detected, resulting in an incomplete assessment of the rust state of the target area. For example, when inspecting mechanical parts with complex structures, the internal rust layer may not be accurately perceived by the visual sensor.

[0021] The ultrasonic sensor transmits ultrasonic waves to the target area and receives the reflected ultrasonic signals. According to the propagation characteristics and reflection conditions of ultrasonic waves in different media (such as rust layer and metal body), the second corrosion data is analyzed. The second corrosion data can reflect the thickness of the rust layer, the internal structure, and whether there are hidden rust areas, etc., which supplements and verifies the first corrosion data to better complete the monitoring of the target area.

[0022] It can be concluded from the above that the present disclosure processes the first corrosion data through the first random forest model, which can make full use of the ensemble learning advantages of the random forest algorithm, combine the output results of multiple decision trees, and effectively reduce the recognition errors that may be caused by a single decision tree. When the difference between multiple recognition results is large, the ultrasonic sensor is enabled for supplementary monitoring, so that the present embodiment can automatically respond to special monitoring environments, such as the internal structure of the target area, thereby improving the accuracy and robustness of monitoring.

[0023] In this embodiment, determining the target corrosion data based on the first corrosion data and the second corrosion data includes: performing feature fusion on the first corrosion data and the second corrosion data, and performing weighted calculation to obtain the target corrosion data.

[0024] Specifically, first corrosion data from a visual sensor and second corrosion data from an ultrasonic sensor are received, key features (such as corrosion area, corrosion depth, and rust spot distribution density) are extracted, and different weights are assigned to the first corrosion data and the second corrosion data.

[0025] In this embodiment, the first corrosion data and the second corrosion data are subjected to feature fusion and weighted calculation to obtain target corrosion data, including: determining a first weight adjustment step length based on the curvature of the target area; Alternatively, determining the first weight adjustment step length based on the standard deviation difference rate between the first corrosion data and the second corrosion data; Adjusting the first weight reference value based on the first weight adjustment step length to obtain a first weight; adjusting the second weight reference value based on the first weight adjustment step length to obtain a second weight; Perform weighted calculation based on the first weight, the second weight, the first corrosion data, and the second corrosion data to obtain target corrosion data; The first weight is a weight corresponding to the first corrosion data, the second weight is a weight corresponding to the second corrosion data, and the first weight reference value and the second weight reference value have different adjustment directions.

[0026] Specifically, the first weight adjustment step is determined based on the curvature of the target area: the curvature of the target area plays an important role in reflecting the complexity of the surface morphology of the area. The larger the curvature, the more drastic the surface change in the area. For the fusion of the first corrosion data and the second corrosion data, we can determine the first weight adjustment step based on a certain mapping relationship between the curvature and the weight adjustment step. For example, a correspondence table between the curvature range and the weight adjustment step can be set in advance. After the curvature value of the target area is obtained, the corresponding first weight adjustment step is determined by searching the correspondence table. Alternatively, the curvature value is used as input, and the first weight adjustment step is calculated by a specific algorithm.

[0027] The curvature of the target area is obtained by laser scanning, and if the curvature is high (C>C1), the weight step of the ultrasound is increased (because vision is easily distorted on a curved surface). The calculation formula for the first weight adjustment step is Δw=k×(C-C1), where k is the proportional coefficient, Δw is the first weight adjustment step, and C1 is the preset curvature threshold.

[0028] Alternatively, the first weight adjustment step is determined based on the standard deviation difference rate of the first rust data and the second rust data: the standard deviation reflects the degree of discreteness of the data. The standard deviation of the first rust data and the standard deviation of the second rust data are calculated, and then their standard deviation difference rate is calculated. The first weight adjustment step is determined according to the relationship between the standard deviation difference rate and the first weight adjustment step. For example, when the standard deviation difference rate is large, it means that the degree of discreteness of the first rust data and the second rust data is quite different. At this time, the first weight adjustment step can be appropriately increased to better balance the weight distribution of the two data in the fusion process; conversely, when the standard deviation difference rate is small, the first weight adjustment step can be reduced. Similarly, we can pre-set a corresponding table of the standard deviation difference rate range and the weight adjustment step, or adjust the adjustment step of the first weight according to the comparison result of the difference rate calculated by the formula and the standard deviation threshold.

[0029] The formula for calculating the difference rate is: R=

[0030] in, σ 1 is the standard deviation of the first corrosion data, σ 2 is the standard deviation of the second corrosion data; Δw=g( R ), g is a function expression determined according to actual conditions.

[0031] In this embodiment, the first weight reference value is adjusted based on the first weight adjustment step to obtain the first weight; the second weight reference value is adjusted based on the first weight adjustment step to obtain the second weight. It should be noted here that the adjustment directions of the first weight reference value and the second weight reference value are different. For example, assuming that the first weight reference value is w1, the second weight reference value is w2, and w1+w2=1. When the first weight adjustment step Δw is a positive number, the first weight w r1 =w1+Δw, the second weight w r2 =w2−Δw; when the first weight adjustment step Δw is negative, the first weight w r1 =w1+Δw, the second weight w r2 =w2−Δw. Through such adjustment, it is ensured that the sum of the first weight and the second weight is always 1, and it also reflects the reasonable adjustment of the weights of the two data according to different conditions.

[0032] Based on the first weight, the second weight, the first corrosion data and the second corrosion data, a weighted calculation is performed to obtain the target corrosion data. Assuming that the first corrosion data is D1 and the second corrosion data is D2, the target corrosion data D=w r1 ×D1+w r2 ×D2. By means of such a weighted calculation method, the first corrosion data and the second corrosion data are feature fused to obtain target corrosion data that can more accurately reflect the corrosion situation of the target area.

[0033] In this embodiment, determining the target corrosion state of the target area based on the target corrosion data includes: Extracting features of target corrosion data to obtain corrosion feature data; Calculate the target similarity between the corrosion feature data and each corrosion state in the corrosion state library; A target corrosion state is selected from the corrosion state library based on target similarity.

[0034] In this embodiment, feature extraction is performed on the target corrosion data to obtain features such as corrosion area, corrosion depth, and rust spot distribution density; and the target similarity between the corrosion feature data and each corrosion state in the corrosion state library is calculated. The corrosion state library is pre-constructed and stores a large number of known corrosion states and their corresponding standard corrosion features. The extracted corrosion feature data is compared with the standard features of the corrosion state in the corrosion state library. When calculating the similarity, a variety of suitable algorithms can be selected, such as the cosine similarity algorithm, which measures the degree of similarity by calculating the cosine value of the angle between two vectors (here, the corrosion feature data vector and the standard corrosion feature vector); or the Euclidean distance algorithm is used to calculate the straight-line distance between two feature vectors in multidimensional space to obtain the similarity.

[0035] In this embodiment, a target corrosion state is selected from the corrosion state library based on the target similarity. After obtaining the target similarity between the corrosion feature data and each corrosion state, the corrosion state with the highest similarity is found by comparing the values ​​of these similarities. The corrosion state with the highest similarity is the target corrosion state of the target area determined based on the target corrosion data. For example, if it is calculated that the similarity value between the corrosion feature data and the "moderate corrosion" state in the corrosion state library is the largest, then the target corrosion state of the target area is determined to be "moderate corrosion".

[0036] In this embodiment, a target corrosion state of a target area is determined based on target corrosion data, including: performing feature extraction on the target corrosion data to obtain texture complexity features, oxide layer thickness features, and porosity distribution features; matching the texture complexity features with the corrosion state map to determine a first corrosion state; performing a first correction on the first corrosion state based on the oxide layer thickness features to obtain a second corrosion state, and performing a second correction on the second corrosion state based on the porosity distribution features to obtain a target corrosion state of the target area.

[0037] Specifically, the rust state atlas is a pre-constructed reference atlas including the correspondence between different texture complexity features and rust states. The extracted texture complexity features are compared one by one with various situations in the atlas, and a matching algorithm (such as a pattern recognition algorithm) is used to find the rust state that best matches the current texture complexity features, and determine the first rust state. The oxide layer thickness is an important indicator for measuring the degree of corrosion, and different oxide layer thicknesses often correspond to different stages of corrosion development. The extracted oxide layer thickness feature data is compared with the theoretical oxide layer thickness range corresponding to the first rust state. If the actual oxide layer thickness is greater than the theoretical oxide layer thickness range, the first rust state is adjusted according to the deviation of the oxide layer thickness. If the actual oxide layer thickness is thicker than the theoretical range, the first rust state is adjusted to a higher level of rust state according to the deviation; conversely, if it is thinner, the rust state needs to be adjusted to a lower level, thereby obtaining the second rust state. The extracted porosity distribution feature data is compared and analyzed with the theoretical porosity distribution range corresponding to the second rust state. According to the difference in porosity distribution, the second corrosion state is adjusted again. If the porosity is too large or the uneven distribution of pores exceeds the theoretical range, it indicates that there are special circumstances in the corrosion state, and the second corrosion state needs to be further corrected to finally obtain the target corrosion state of the target area.

[0038] The extracted texture complexity features, oxide layer thickness features and porosity distribution features are quantified. Assume that F T is the quantized value of texture complexity feature, F O is the characteristic quantitative value of the oxide layer thickness, F PM is the quantitative value of porosity distribution characteristics; T is a mapping function that represents the matching relationship between the texture complexity feature and the rust state map. The rust states in the rust state map include: S1, S2, ..., S n , the first corrosion state S is obtained by matching the function first , the formula is: S first =M T (F T ), M T It is a classification function based on pattern recognition or machine learning. For example, a support vector machine (SVM) model can be trained, the texture complexity feature quantization value FT can be input, and the corresponding corrosion state category can be output.

[0039] Correct the first corrosion state according to the characteristic quantitative value F of the oxide layer thickness O For the first corrosion state S first Adjust to obtain the second corrosion state S second. The formula is: S second =C O (S first , F O ), where C O is the first correction function.

[0040] Assume T O1 and T O2 are two threshold values ​​of oxide layer thickness, and T O1 <T O2 , the first correction rule is as follows: F O <T O1 , then S second =S first −ΔS O , T O1 ≤F O <T O2 , then S second =S first F O ≥T O2 , then S second =S first +ΔS O Among them, ΔS O is the first calibration step.

[0041] Quantify the value F according to the porosity distribution characteristics P For the second corrosion state S second Correction is performed to obtain the final target corrosion state S target The formula is as follows: target =CP (S second , F P ), C P is the second correction function.

[0042] Set the threshold T of porosity distribution characteristics P1 and T P2 , and T P1 <T P2 , the second correction rule is as follows: F P <T P1 , then S target =S second −ΔS P T P1 ≤F P <T P2 , then S target =S second F P ≥T P2 , then S target =S second +ΔS P Among them, ΔS P is the second correction step length, which is a positive integer.

[0043] In this embodiment, based on the rust state of the target area, controlling the rust removal robot includes: Based on the target corrosion state of the target area, a target control strategy corresponding to the target corrosion state is selected from a target strategy library, where a plurality of control strategies are stored; The laser power output by the rust removal robot is controlled based on the target control strategy.

[0044] The control strategies in the target strategy library are preliminarily sorted from high to low according to the target similarity to obtain the first sorting list, and an initial dynamic evaluation matrix is ​​established. The rows of the matrix represent the control strategies in the strategy library, and the columns represent the strategy evaluation indicators of multiple different dimensions. Each control strategy in the first sorting list is evaluated according to the initial dynamic evaluation matrix, and the evaluation results of each dimension are filled into the dynamic evaluation matrix to obtain the target dynamic evaluation matrix. The comprehensive evaluation score of each control strategy in the first sorting list is calculated according to the target dynamic evaluation matrix, and the control strategy with the highest comprehensive evaluation score is selected as the target control strategy corresponding to the target corrosion state. The control strategy includes the basic laser power, the first scanning speed and the material correction coefficient table.

[0045] In this embodiment, the laser power output by the rust removal robot is controlled based on the target control strategy, including: determining the basic laser power based on the target rust state; adjusting the basic laser power based on the material type of the target area to obtain the laser power of the rust removal robot. Common material types include steel, aluminum alloy, copper alloy, etc. The basic laser power is preliminarily determined based on the target rust state, and is a basic laser power calculated or preset without considering the material. The basic laser power is the final laser power used to control the robot to actually perform the rust removal task. It is obtained after adjusting the basic laser power by comprehensively considering the material type of the target grasped object, which can ensure that the robot can safely and stably remove rust from the target area.

[0046] In this embodiment, the first scanning speed of the rust removal robot is determined based on the target control strategy; the first scanning speed is corrected based on the laser power output by the rust removal robot to obtain the scanning speed of the rust removal robot. If the laser power output by the rust removal robot is greater than the first preset threshold, the first scanning speed is corrected based on the difference between the first scanning speed of the rust removal robot and the first preset threshold to obtain the scanning speed of the rust removal robot; if the laser power output by the rust removal robot is less than or equal to the first preset threshold, the first scanning speed is corrected based on the mapping table to obtain the scanning speed of the rust removal robot; the mapping table is a mapping table of laser power and scanning speed.

[0047] When the laser power output by the rust removal robot is less than or equal to the first preset threshold, it indicates that the current laser energy is strong and may cause damage to the target area. At this time, the difference between the laser power and the first preset threshold is calculated, and the first scanning speed is corrected according to this difference. The larger the difference, the more the laser power exceeds the normal range, and the correction amplitude will also increase accordingly, usually by increasing the scanning speed, so as to reduce the potential damage to the material during the rust removal process. When the laser power is less than or equal to the first preset threshold, the first scanning speed is corrected with the help of the first mapping table. The first preset threshold is used to distinguish between conventional power adjustment and over-limit protection mode. The first mapping table constructs the relationship between the laser power and the corresponding scanning speed in advance. Through a large number of experimental tests, various rust samples are rusted under different laser power conditions, and the corresponding optimal scanning speed is recorded, so as to summarize the mapping law between the two. The current laser power finds the corresponding scanning speed correction value from the first mapping table, and then obtains the appropriate scanning speed of the rust removal robot, ensuring that the rust removal robot can complete the grasping task at the optimal speed under different laser powers. It can be concluded from the above that this embodiment ensures the accuracy and safety of the rust removal robot operation by dynamically adjusting the scanning speed to adapt to different laser powers.

[0048] Specifically, when P final >P th, the difference calculation formula is: ΔP=P final −P th , the speed correction formula is: v final =v1−Y×ΔP, where P final is the current laser power, P th is the first preset threshold, ΔP is the difference, Y is the power-speed attenuation coefficient, v final is the scanning speed, v1 is the current scanning speed. final ≤P th , select the speed correction coefficient α in the first mapping table according to the current laser power, and the speed correction formula is: final =v1×α.

[0049] Table 1 First mapping table

[0050] Where P is the laser power.

[0051] Corresponding to the rust removal robot control method of the above embodiment, Figure 2 This is a structural block diagram of a rust removal robot control system provided by an embodiment of the present disclosure. For ease of explanation, only the parts related to the embodiment of the present disclosure are shown. Figure 2 The rust removal robot control system 20 includes: a detection module 21, a state determination module 22 and a control module 23.

[0052] Among them, the detection module 21 is used to monitor the target area based on the visual sensor of the rust removal robot to obtain the first rust data; the state determination module 22 is used to process the first rust data based on the first random forest model to obtain the difference of multiple recognition results, and determine the target rust state of the target area according to the comparison result of the difference of multiple recognition results and the first difference; the control module 23 is used to control the rust removal robot based on the target rust state of the target area.

[0053] In one embodiment of the present disclosure, the state determination module 22 is specifically used to: in response to a difference between multiple recognition results being greater than or equal to a first difference, monitor the target area based on the ultrasonic sensor built into the rust removal robot to obtain second corrosion data; determine target corrosion data based on the first corrosion data and the second corrosion data; in response to a difference between multiple recognition results being less than the first difference, use the first corrosion data as target corrosion data; and determine a target corrosion state of the target area based on the target corrosion data.

[0054] In one embodiment of the present disclosure, the state determination module 22 is specifically configured to: The first corrosion data and the second corrosion data are subjected to feature fusion and weighted calculation to obtain target corrosion data.

[0055] In one embodiment of the present disclosure, the state determination module 22 is specifically configured to: determining a first weight adjustment step length based on the curvature of the target area; Alternatively, determining the first weight adjustment step length based on the standard deviation difference rate between the first corrosion data and the second corrosion data; Adjusting the first weight reference value based on the first weight adjustment step length to obtain a first weight; adjusting the second weight reference value based on the first weight adjustment step length to obtain a second weight; Perform weighted calculation based on the first weight, the second weight, the first corrosion data, and the second corrosion data to obtain target corrosion data; The first weight is a weight corresponding to the first corrosion data, the second weight is a weight corresponding to the second corrosion data, and the first weight reference value and the second weight reference value have different adjustment directions.

[0056] In one embodiment of the present disclosure, the state determination module 22 is specifically configured to: Feature extraction is performed on target corrosion data to obtain corrosion feature data; target similarity between the corrosion feature data and each corrosion state in the corrosion state library is calculated; and target corrosion state is selected from the corrosion state library based on the target similarity.

[0057] In one embodiment of the present disclosure, the control module 23 is specifically configured to: Based on the target rust state of the target area, a target control strategy corresponding to the target rust state is selected from a target strategy library, wherein a plurality of control strategies are stored in the target strategy library; based on the target control strategy, the laser power output by the rust removal robot is controlled.

[0058] In one embodiment of the present disclosure, the control module 23 is specifically configured to: Based on the target corrosion state, the basic laser power is determined; based on the material type of the target area, the basic laser power is adjusted to obtain the laser power of the rust removal robot.

[0059] In one embodiment of the present disclosure, the control module 23 is specifically configured to: The first scanning speed of the rust removal robot is determined based on the target control strategy; the first scanning speed is corrected based on the laser power output by the rust removal robot to obtain the scanning speed of the rust removal robot.

[0060] In one embodiment of the present disclosure, the control module 23 is specifically configured to: If the laser power output by the rust removal robot is greater than a first preset threshold, based on the difference between the first scanning speed of the rust removal robot and the first preset threshold, the first scanning speed is corrected to obtain the scanning speed of the rust removal robot; If the laser power output by the rust removal robot is less than or equal to the first preset threshold, the first scanning speed is corrected based on the mapping table to obtain the scanning speed of the rust removal robot; the mapping table is a mapping table of laser power and scanning speed.

[0061] See also Figure 3 , Figure 3 A schematic block diagram of an electronic device provided by an embodiment of the present disclosure. Figure 3 The electronic device 300 in the embodiment shown may include: one or more processors 301, one or more input devices 302, one or more output devices 303 and one or more memories 304. The processors 301, input devices 302, output devices 303 and memories 304 communicate with each other via a communication bus 305. The memory 304 is used to store computer programs, which include program instructions. The processor 301 is used to execute the program instructions stored in the memory 304. The processor 301 is configured to call the program instructions to execute the functions of the modules in the above-mentioned device embodiments, such as Figure 2 The functions of modules 21 to 23 are shown.

[0062] It should be understood that in the embodiment of the present disclosure, the processor 301 may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0063] The input device 302 may include a touch panel, a fingerprint collection sensor (for collecting the user's fingerprint information and fingerprint direction information), a microphone, etc., and the output device 303 may include a display (LCD, etc.), a speaker, etc.

[0064] The memory 304 may include a read-only memory and a random access memory, and provide instructions and data to the processor 301. A portion of the memory 304 may also include a non-volatile random access memory. For example, the memory 304 may also store information about the device type.

[0065] In a specific implementation, the processor 301, input device 302, and output device 303 described in the embodiments of the present disclosure can execute the implementation methods described in the first and second embodiments of the rust removal robot control method provided in the embodiments of the present disclosure, and can also execute the implementation methods of the electronic device described in the embodiments of the present disclosure, which will not be repeated here.

[0066] In another embodiment of the present disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by the processor, all or part of the processes in the above-mentioned embodiment method are implemented, and the computer program can also be completed by instructing the relevant hardware through the computer program. The computer program can be stored in a computer-readable storage medium, and when the computer program is executed by the processor, the steps of each of the above-mentioned method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0067] The computer-readable storage medium may be an internal storage unit of the electronic device of any of the aforementioned embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium may also be an external storage device of the electronic device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Furthermore, the computer-readable storage medium may also include both an internal storage unit of the electronic device and an external storage device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium may also be used to temporarily store data that has been output or is to be output.

[0068] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this disclosure.

[0069] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the electronic devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0070] In the several embodiments provided in the present application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, 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 mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces or units, or it can be an electrical, mechanical or other form of connection.

[0071] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiments of the present disclosure.

[0072] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0073] The above are only specific embodiments of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present disclosure, and these modifications or replacements should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.

Claims

1. A rust removal robot control method, characterized in that: include: Based on the visual sensor of the rust removal robot, the target area is monitored to obtain the first rust data; Processing the first corrosion data based on a first random forest model to obtain differences of multiple recognition results, and determining a target corrosion state of the target area according to a comparison result between the differences of the multiple recognition results and the first difference; The rust removal robot is controlled based on the target rust state of the target area.

2. The rust removal robot control method according to claim 1, characterized in that: Determining the target corrosion state of the target area according to the comparison result of the difference between the plurality of recognition results and the first difference comprises: In response to the difference between the multiple recognition results being greater than or equal to the first difference, monitoring the target area based on the ultrasonic sensor built into the rust removal robot to obtain second rust data; determining target corrosion data based on the first corrosion data and the second corrosion data; In response to the difference between the plurality of recognition results being less than a first difference, taking the first corrosion data as target corrosion data; A target corrosion state of the target area is determined based on the target corrosion data.

3. The rust removal robot control method according to claim 2, characterized in that: The determining target corrosion data based on the first corrosion data and the second corrosion data comprises: The first corrosion data and the second corrosion data are subjected to feature fusion and weighted calculation to obtain target corrosion data.

4. The rust removal robot control method according to claim 3, characterized in that: The feature fusion of the first corrosion data and the second corrosion data and weighted calculation to obtain the target corrosion data include: determining a first weight adjustment step length based on the curvature of the target area; Alternatively, determining the first weight adjustment step length based on the standard deviation difference rate between the first corrosion data and the second corrosion data; Adjusting a first weight reference value based on the first weight adjustment step length to obtain a first weight; adjusting a second weight reference value based on the first weight adjustment step length to obtain a second weight; Perform weighted calculation based on the first weight, the second weight, the first corrosion data, and the second corrosion data to obtain the target corrosion data; The first weight is a weight corresponding to the first corrosion data, the second weight is a weight corresponding to the second corrosion data, and the first weight reference value and the second weight reference value are adjusted in different directions.

5. The rust removal robot control method according to claim 4, characterized in that: The step of determining the target corrosion state of the target area based on the target corrosion data includes: Performing feature extraction on the target corrosion data to obtain corrosion feature data; Calculating the target similarity between the corrosion feature data and each corrosion state in the corrosion state library; The target corrosion state is selected from a corrosion state library based on the target similarity.

6. The rust removal robot control method according to claim 1, characterized in that: The controlling the rust removal robot based on the rust state of the target area comprises: Based on the target corrosion state, a target control strategy corresponding to the target corrosion state is selected from a target strategy library, wherein the target strategy library stores a plurality of control strategies; The laser power output by the rust removal robot is controlled based on the target control strategy.

7. The rust removal robot control method according to claim 6, characterized in that: The laser power output by the rust removal robot is controlled based on the target control strategy, including: Determining basic laser power based on the target corrosion state; The basic laser power is adjusted based on the material type of the target area to obtain the laser power of the rust removal robot.

8. The rust removal robot control method according to claim 7, characterized in that: Also includes: Determining a first scanning speed of the rust removal robot based on the target control strategy; The first scanning speed is corrected based on the laser power output by the rust removal robot to obtain the scanning speed of the rust removal robot.

9. The rust removal robot control method according to claim 8, characterized in that: Also includes: If the laser power output by the rust removal robot is greater than a first preset threshold, based on the difference between the first scanning speed of the rust removal robot and the first preset threshold, the first scanning speed is corrected to obtain the scanning speed of the rust removal robot; If the laser power output by the rust removal robot is less than or equal to a first preset threshold, the first scanning speed is corrected based on a mapping table to obtain the scanning speed of the rust removal robot; the mapping table is a mapping table of laser power and scanning speed.

10. A rust removal robot control system, characterized in that: include: A detection module, used for monitoring a target area based on a visual sensor of the rust removal robot to obtain first rust data; a state determination module, configured to process the first corrosion data based on a first random forest model to obtain a plurality of difference degrees of recognition results, and determine a target corrosion state of the target area according to a comparison result between the difference degrees of the plurality of recognition results and the first difference degree; A control module is used to control the rust removal robot based on the target rust state of the target area.

Images