A construction site exposed steel bar self-adaptive rust removal system and method based on an intelligent mechanical arm

Abstract

The present application belongs to the technical field of intelligent machinery, and specifically relates to a construction site exposed reinforcement self-adaptive rust removal system and method based on an intelligent mechanical arm. Through the combination of a multi-degree-of-freedom intelligent mechanical arm and a 3D laser and vision fusion device, high-precision reinforcement surface three-dimensional structure recognition and positioning are achieved, ensuring that the mechanical arm can accurately move to the specified position for rust removal operation. At the same time, the rust removal intensity parameters are dynamically adjusted according to the scanning results, so that each area can obtain the most suitable rust removal intensity, thereby improving the rust removal efficiency and quality. In addition, by recording the operation data in real time and analyzing the rust removal efficiency, the parameter settings of the next rust removal cycle are automatically optimized, forming a closed-loop feedback mechanism, and further improving the overall rust removal effect and equipment utilization. This method effectively solves the problems of low efficiency and inconsistent rust removal effect in the prior art, and provides an efficient, accurate and adaptive exposed reinforcement rust removal solution.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of intelligent machinery, and particularly relates to a construction site exposed steel bar self-adaptive rust removal system and method based on an intelligent mechanical arm. BACKGROUND

[0002] In modern construction, rust removal of exposed steel bars is a common and important maintenance step. Traditional rust removal methods usually rely on manual operation or semi-automatic equipment, which have problems such as low efficiency, high labor intensity, and inconsistent rust removal effect. Specifically, traditional methods often require workers to manually operate polishing tools for rust removal, which not only consumes time and effort, but also makes it difficult to ensure that each area is uniformly and effectively treated.

[0003] General solutions in the prior art:

[0004] Manual rust removal: workers use handheld polishers or other tools to remove rust from the surface of steel bars. This method highly depends on the experience and skill level of workers, and is prone to incomplete rust removal or excessive polishing.

[0005] Semi-automatic equipment: some construction sites use semi-automatic rust removal equipment, such as mechanical arms on fixed tracks combined with simple sensor systems. However, such equipment can only perform simple actions on preset paths and lack the ability to adapt to complex-shaped steel bar surfaces, and cannot dynamically adjust the rust removal force according to the actual rusting degree.

[0006] Current technical solutions cannot achieve efficient, accurate, and self-adaptive steel bar surface rust removal operations. Especially when facing complex-shaped steel bars with uneven rusting degrees, existing methods cannot provide consistent rust removal effects, and due to the lack of intelligent feedback mechanisms, it is difficult to adjust rust removal parameters in real time to optimize work efficiency. SUMMARY

[0007] The present application aims to provide a construction site exposed steel bar self-adaptive rust removal system and method based on an intelligent mechanical arm, which solves the problems of low efficiency and inconsistent rust removal effect in the prior art, and provides an efficient, accurate, and self-adaptive exposed steel bar rust removal solution.

[0008] To achieve the above-mentioned purpose, the present application proposes a construction site exposed steel bar self-adaptive rust removal method based on an intelligent mechanical arm, which includes the following steps:

[0009] A multi-degree-of-freedom intelligent mechanical arm is configured, and the intelligent mechanical arm is connected to a 3D laser and vision fusion device;

[0010] scanning a surface of the steel bar to be processed using the 3D laser and vision fusion device to obtain three-dimensional structure information of the surface of the steel bar to be processed, and determining position coordinates of a region of the surface of the steel bar to be processed according to the three-dimensional structure information;

[0011] According to the position coordinates, the intelligent mechanical arm is guided to move to a specified position for preliminary positioning, and after the preliminary positioning, the intelligent mechanical arm identifies the rust degree of the surface of the steel bar.

[0012] According to the rust degree, the rust removal force parameter of the intelligent mechanical arm is dynamically adjusted, the rust removal operation is performed by using the rust removal force parameter to control the intelligent mechanical arm, and operation data is recorded;

[0013] The operation data is analyzed to calculate the rust removal efficiency, and the parameter settings of the next rust removal cycle are automatically optimized according to the calculation result.

[0014] Preferably, the intelligent mechanical arm with multiple degrees of freedom is configured, and the intelligent mechanical arm is connected with the 3D laser and vision fusion device, comprising:

[0015] The intelligent mechanical arm is installed at a predetermined working position, and the 3D laser and vision fusion device is fixed on the end effector of the intelligent mechanical arm;

[0016] The space coordinate system of the 3D laser and vision fusion device relative to the intelligent mechanical arm is calibrated by using the coordinate transformation formulas X' = X*cos(A)-Y*sin(A) and Y' = X*sin(A)+Y*cos(A), wherein A is the rotation angle of the 3D laser and vision fusion device relative to the intelligent mechanical arm, X and Y are original coordinate values, and X' and Y' are calibrated coordinate values;

[0017] Based on the calibrated coordinate system, the intelligent mechanical arm is initialized to complete the connection of the intelligent mechanical arm and the 3D laser and vision fusion device.

[0018] Preferably, the 3D laser and vision fusion device is used to scan the surface of the steel bar to be processed to obtain three-dimensional structure information, comprising:

[0019] The intelligent mechanical arm is started and the 3D laser and vision fusion device is activated;

[0020] The intelligent mechanical arm is guided to move to above the steel bar to be processed according to a preset path, so that the 3D laser and vision fusion device is located at a scanning position;

[0021] The 3D laser and vision fusion device emits a laser beam and synchronously photographs an image of the surface of the steel bar, records a position (Xi, Yi) of each laser point in the image and a corresponding depth value Zi thereof to form a preliminary three-dimensional data set {(Xi, Yi, Zi)};

[0022] Based on the preliminary three-dimensional data set {(Xi, Yi, Zi)}, the distance D between each data point is calculated by using the spatial geometric relationship formula D = sqrt((X2-X1)^2+(Y2-Y1)^2+(Z2-Z1)^2) to construct the complete three-dimensional structure information of the reinforcing bar surface.

[0023] Preferably, according to the three-dimensional structure information, the position of the reinforcing bar surface treatment area is determined, including:

[0024] The preliminary three-dimensional data set {(Xi, Yi, Zi)} is analyzed to identify the height variation characteristics of different regions of the reinforcing bar surface. For each point (Xi, Yi, Zi), the height difference ΔH = |Zi-Zj| between it and the adjacent point is calculated, where i and j represent adjacent data points.

[0025] Based on the distribution of the height difference ΔH, a threshold T is set. When the height difference of a certain point exceeds the threshold T, it is marked as a potential treatment area. The set S of all potential treatment areas is determined by the formula S = {(Xi, Yi, Zi)|ΔH > T};

[0026] For each point (Xi, Yi, Zi) in the set S, the distance D between it and other marked points is calculated using the spatial geometric relationship formula D = sqrt((X2-X1)^2+(Y2-Y1)^2+(Z2-Z1)^2). If the distance between a certain point and other points is less than a certain value L, it is classified as the same treatment area.

[0027] According to all points (Xi, Yi, Zi) in the treatment area, the center coordinates (Xc, Yc, Zc) of each treatment area are calculated, where Xc = sum(Xi) / n, Yc = sum(Yi) / n, Zc = sum(Zi) / n, n is the number of points in the region. The center coordinates are taken as the position coordinates of the reinforcing bar surface treatment area.

[0028] Preferably, according to the position coordinates, the intelligent mechanical arm is guided to move to the specified position for preliminary positioning, including:

[0029] According to the center coordinates (Xc, Yc, Zc), the displacement vector ΔP = (ΔX, ΔY, ΔZ) of the intelligent mechanical arm from the current position to the target position is calculated, where ΔX = Xc-X0, ΔY = Yc-Y0, ΔZ = Zc-Z0, (X0, Y0, Z0) is the current position of the intelligent mechanical arm.

[0030] Calculate the angle adjustment value B of each joint using the displacement vector ΔP and the kinematics formula B = arctan2(ΔY, ΔX) + B_offset of the intelligent robot arm, where B_offset is the initial angle offset;

[0031] Send control instructions to the intelligent robot arm to move according to the calculated joint angle adjustment value B, and monitor the position (X', Y', Z') of the intelligent robot arm in real time and compare it with the target position (Xc, Yc, Zc) during the movement;

[0032] When the intelligent robot arm reaches the vicinity of the target position, fine-tune its position using the fine-tuning formula δX = Xc - X', δY = Yc - Y', δZ = Zc - Z' until the position error E = sqrt(δX^2 + δY^2 + δZ^2) of the intelligent robot arm is less than the set threshold ε, completing the preliminary positioning.

[0033] Preferably, after the preliminary positioning, the intelligent robot arm identifies the rust degree of the steel surface, including:

[0034] After the intelligent robot arm completes the preliminary positioning, start the 3D laser and visual fusion device to perform high-resolution scanning on the area to be processed, obtain image data and depth information in the area, and record the gray value G and depth value Z of each pixel point;

[0035] According to the gray value G and depth value Z, calculate the rust index RI = (G_max - G) / Z of each pixel point, where G_max is the preset maximum gray value for standardizing the gray difference;

[0036] Statistically analyze the rust index RI to generate a rust distribution map, including: for each pixel point (X, Y), the rust index RI(X, Y) is mapped to a two-dimensional matrix to form a rust distribution map M = {RI(X, Y)};

[0037] Based on the rust distribution map M, calculate the average rust index RI_avg = sum(RI(X, Y)) / N of the entire area to be processed, where N is the number of pixel points in the area; if RI_avg exceeds the preset threshold T_ri, mark the area as a severe rust area, otherwise mark it as a mild rust area.

[0038] Preferably, according to the rust degree, dynamically adjust the rust removal force parameter of the intelligent robot arm, including:

[0039] Determine the rust level of each pixel point (X, Y) according to the rust distribution map M and the average rust index RI_avg; define the rust index range of the mild rust area as 0 <= RI(X, Y) < T_ri, and the rust index range of the severe rust area as RI(X, Y) >= T_ri;

[0040] For each area to be processed, calculate the rust removal intensity coefficient F = (RI_avg - RI_min) / (RI_max - RI_min), where RI_min and RI_max are the preset minimum and maximum rust index thresholds respectively;

[0041] Based on the rust removal intensity coefficient F, adjust the working parameters of the rust removal tool of the intelligent robotic arm, including: for the severe rust area, use the formula P = P_base + F * ΔP to calculate the new rust removal power P, where P_base is the base power and ΔP is the power increment; for the mild rust area, use the formula P = P_base - F * ΔP to reduce the rust removal power;

[0042] Apply the adjusted rust removal power P to the intelligent robotic arm and monitor the rust removal effect in real time. Compare the change in the rust index after actual rust removal ΔRI = RI_before - RI_after. If ΔRI reaches the expected target, continue to operate with the current parameters; otherwise, readjust the rust removal intensity coefficient F and update the rust removal power P.

[0043] Preferably, use the rust removal intensity parameter to control the intelligent robotic arm to perform rust removal operations, and record operation data at the same time, including:

[0044] Input the adjusted rust removal power P into the intelligent robotic arm, start the rust removal tool, and the intelligent robotic arm adjusts the working state of the tool according to the rust removal power P;

[0045] During the rust removal process, collect the operation data of the intelligent robotic arm in real time, including the current rust removal power P, working time t, and position coordinates (X, Y, Z); for each collection point, record its original rust index RI_before and the rust index RI_after after rust removal;

[0046] Calculate the rust removal efficiency E = (RI_before - RI_after) / t of each processed area based on the operation data, where t is the processing time of this area, and calculate the average rust removal efficiency E_total of the entire working surface through the formula E_total = sum(E) / N, where N is the number of processed areas;

[0047] Store the operation data and calculation results in the database to form a historical record, including: for each processing cycle, generate a data packet:

[0048] {(P,t,X,Y,Z,RI_before,RI_after,E)} and save.

[0049] Preferably, the operation data is analyzed to calculate the rust removal efficiency, and the parameter settings of the next rust removal cycle are automatically optimized according to the calculation results, including:

[0050] Extract the operation data packet {(P,t,X,Y,Z,RI_before,RI_after,E)} of the last rust removal cycle from the database, and calculate the average rust removal efficiency E_avg = sum(E) / N for each processing area, where N is the number of processing areas;

[0051] Based on the average rust removal efficiency E_avg and the original rust index RI_before of each area, determine the optimization target, and define the optimization target to achieve a higher rust removal efficiency E_target = E_avg + ΔE in the next round of rust removal operation, where ΔE is the expected efficiency improvement value;

[0052] According to the optimization target, adjust the rust removal force parameter of the intelligent mechanical arm, and for each processing area, use the formula P_new = P*(E_target / E_avg) to calculate the new rust removal power P_new; if P_new exceeds the preset maximum power P_max, set P_new to P_max;

[0053] Apply the new rust removal power P_new to the intelligent mechanical arm, and update the parameter settings of the next rust removal cycle; at the same time, save the optimized parameter settings to the database to form a historical record, and continuously improve the rust removal operation by comparing the data of different cycles.

[0054] On the other hand, the present application proposes a construction site exposed steel bar self-adaptive rust removal system based on an intelligent mechanical arm, comprising:

[0055] The polishing equipment configuration module is configured to configure a multi-degree-of-freedom intelligent mechanical arm, and connect the intelligent mechanical arm with a 3D laser and visual fusion device;

[0056] The polishing coordinate acquisition module is configured to use the 3D laser and visual fusion device to scan the surface of the steel bar to be processed, acquire its three-dimensional structure information, and determine the position coordinates of the processing area on the surface of the steel bar according to the three-dimensional structure information;

[0057] The corrosion degree recognition module is configured to direct the intelligent mechanical arm to move to the specified position for preliminary positioning according to the position coordinates, and the intelligent mechanical arm recognizes the rust degree of the surface of the steel bar after preliminary positioning.

[0058] A dynamic rust removal module is used to dynamically adjust the rust removal force parameter of the intelligent mechanical arm according to the rust degree, and the rust removal force parameter is used to control the intelligent mechanical arm to perform the rust removal operation, and operation data is recorded at the same time.

[0059] A rust removal optimization module is used to analyze the operation data to calculate the rust removal efficiency, and automatically optimize the parameter setting of the next rust removal period according to the calculation result.

[0060] The technical effects and advantages of the present application are as follows:

[0061] The present application realizes high-precision steel surface three-dimensional structure identification and positioning through the combination of a multi-degree-of-freedom intelligent mechanical arm and a 3D laser and vision fusion device, ensures that the mechanical arm can accurately move to the specified position for rust removal operation, and dynamically adjusts the rust removal force parameter according to the scanning result, so that each area can obtain the most suitable rust removal force, thereby improving the rust removal efficiency and quality. BRIEF DESCRIPTION OF DRAWINGS

[0062] Figure 1 The flowchart of the construction site exposed steel bar adaptive rust removal method based on the intelligent mechanical arm of the present application is shown in the figure.

[0063] Figure 2 The block diagram of the construction site exposed steel bar adaptive rust removal system based on the intelligent mechanical arm of the present application is shown in the figure. DETAILED DESCRIPTION

[0064] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The specific embodiments described herein are only used to explain the present application, and are not used to limit the present application. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0065] The present application provides a construction site exposed steel bar adaptive rust removal system based on an intelligent mechanical arm, which comprises a rust degree detection module, a dynamic rust removal module, a rust removal optimization module and a rust removal execution module. Figure 1The illustrated construction site exposed steel bar self-adaptive rust removal method based on intelligent robot, by recording operation data in real time and analyzing rust removal efficiency, automatically optimizing parameter setting of next rust removal period, forming closed loop feedback mechanism, further improving overall rust removal effect and equipment utilization. This method effectively solves the problem of low efficiency and inconsistent rust removal effect in the prior art, and provides an efficient, accurate and self-adaptive exposed steel bar rust removal solution; specifically as follows:

[0066] In this embodiment, the construction site exposed steel bar self-adaptive rust removal method based on intelligent robot, comprising the following steps:

[0067] Step one: configure a multi-degree-of-freedom intelligent robot, and connect the intelligent robot with a 3D laser and vision fusion device; specifically including:

[0068] Install the intelligent robot at the predetermined working position, and fix the 3D laser and vision fusion device on the end effector of the intelligent robot; wherein the basic structure and principle of the 3D laser and vision fusion device are as follows:

[0069] 3D laser scanner:

[0070] Laser emitter: emits high-precision laser beams for scanning the surface of the target object.

[0071] Receiver: receives the reflected laser signals from the target object surface and converts them into electrical signals.

[0072] Optical system: including lenses and other optical elements, used for focusing and guiding laser beams and receiving reflected light.

[0073] Vision sensor (camera):

[0074] Image sensor: usually using CMOS or CCD sensor, used for capturing two-dimensional images of the target object.

[0075] Lens: used for focusing light and generating clear images.

[0076] Filter: used to filter light of specific wavelength, improve image quality.

[0077] Data processing unit:

[0078] Processor: responsible for processing data from laser scanner and vision sensor, performing coordinate transformation, point cloud generation, image stitching and other operations.

[0079] Memory: used for storing raw data and processed results.

[0080] Communication interface: exchanges data with external devices (such as intelligent robot control system).

[0081] Calibration tools:

[0082] Calibration board: used to calibrate the relative position and angular relationship between the 3D laser scanner and the vision sensor.

[0083] Software tools: used to perform coordinate transformation, error compensation, and other calibration operations.

[0084] 3D laser scanning principle:

[0085] Emit laser beams: the laser emitter emits one or more laser beams that illuminate the target object (such as the surface of steel bars).

[0086] Receive reflected signals: when the laser beams illuminate the object surface, part of the light is reflected back to the receiver. The receiver records the time difference or phase difference of the reflected light, and calculates the distance information of each point on the object surface.

[0087] Generate point cloud data: through multiple scans, obtain the distance information of multiple points, and record the position coordinates of these points to form a three-dimensional point cloud data set.

[0088] Visual imaging principle:

[0089] Take pictures: the vision sensor (camera) captures the two-dimensional image of the target object, generating image data containing color and brightness information.

[0090] Image processing: through image processing algorithms, feature points in the image are extracted, and the image is spliced and corrected to obtain a more comprehensive view.

[0091] Data fusion and processing:

[0092] Coordinate transformation: using pre-calibrated parameters, align and fuse the three-dimensional point cloud data obtained by laser scanning with the two-dimensional image data taken by the vision sensor.

[0093] Feature recognition: combining point cloud data and image data, identify the geometric shape and surface features of the target object, such as rusted areas, cracks, etc.

[0094] Real-time feedback: real-time transmission of processed data to the intelligent robot control system, guiding the robot to perform precise operations such as positioning and rust removal.

[0095] The spatial coordinate system of the 3D laser and visual fusion device relative to the intelligent mechanical arm is calibrated using the coordinate transformation formulas X' = X*cos(A)-Y*sin(A) and Y' = X*sin(A)+Y*cos(A), where A is the rotation angle of the 3D laser and visual fusion device relative to the intelligent mechanical arm, X and Y are the original coordinate values, and X' and Y' are the calibrated coordinate values. The formulas ensure that the data of the 3D laser and visual fusion device can be correctly mapped into the coordinate system of the intelligent mechanical arm, thereby achieving precise operation.

[0096] Suppose there is a point (X, Y) = (3, 4) and the rotation angle A of the 3D laser and visual fusion device relative to the intelligent mechanical arm is 45° (i.e. π / 4 radians). According to the formula, we have:

[0097] X' = X*cos(π / 4)-Y*sin(π / 4) = 3*cos(π / 4)-4*sin(π / 4) = 3*0.707-4*0.707 = -0.707;

[0098] Y' = X*sin(π / 4)+Y*cos(π / 4) = 3*sin(π / 4)+4*cos(π / 4) = 3*0.707+4*0.707 = 4.949.

[0099] After calibration, the coordinates of the point in the new coordinate system are (X', Y') = (-0.707, 4.949).

[0100] Through the coordinate transformation formula, the precise alignment of the coordinate system between the 3D laser and visual fusion device and the intelligent mechanical arm is achieved, ensuring the accuracy of subsequent scanning and operation.

[0101] Based on the calibrated coordinate system, the intelligent mechanical arm is initialized to complete the connection between the intelligent mechanical arm and the 3D laser and visual fusion device. The initialization instructions may include setting the initial angles of the joints of the mechanical arm, starting the sensor system, and confirming the smoothness of the data transmission channel, etc. For example, set the initial angles of the mechanical arm as (θ1, θ2, θ3,...) = (0, 0, 0,...), and ensure that the sensors of the 3D laser and visual fusion device are working normally, and start receiving and processing data. Technical effects:

[0102] The connection and initialization of the intelligent mechanical arm and the 3D laser and visual fusion device are completed, ensuring that the equipment can work together, providing a reliable hardware foundation for subsequent steel surface scanning and rust removal operations. Through initialization, the system can ensure that all components are in the best state, reducing the possibility of failure and improving overall work efficiency.

[0103] Step two: use the 3D laser and vision fusion device to scan the surface of the steel bar to be processed to obtain its three-dimensional structure information; specifically including:

[0104] Start the intelligent mechanical arm and activate the 3D laser and vision fusion device; send the start command through the control system of the intelligent mechanical arm to activate each component of the device, and prepare for the subsequent scanning operation.

[0105] Direct the intelligent mechanical arm to move to the top of the steel bar to be processed according to the preset path, so that the 3D laser and vision fusion device is located at the scanning position, ensuring the accuracy and integrity of the scanning data, and avoiding data loss or errors caused by improper position.

[0106] Use the 3D laser and vision fusion device to emit a laser beam and simultaneously take an image of the steel bar surface, record the position (Xi, Yi) of each laser point in the image and its corresponding depth value Zi, and form a preliminary three-dimensional data set {(Xi, Yi, Zi)}; assuming that there are three laser points in a scanning area, their position coordinates and depth values are:

[0107] Point 1: (X1, Y1, Z1) = (10, 20, 5);

[0108] Point 2: (X2, Y2, Z2) = (15, 25, 7);

[0109] Point 3: (X3, Y3, Z3) = (12, 22, 6);

[0110] The preliminary three-dimensional data set is {(10, 20, 5), (15, 25, 7), (12, 22, 6)}. By synchronously collecting the position and depth information of the laser points, a three-dimensional data set containing the geometric shape of the steel bar surface is formed.

[0111] Based on the preliminary three-dimensional data set {(Xi, Yi, Zi)}, the spatial geometric relationship formula D = sqrt((X2-X1)^2+(Y2-Y1)^2+(Z2-Z1)^2) is applied to calculate the distance D between each data point. Through this formula, the distance between each laser point and its adjacent point can be calculated, thereby constructing a complete three-dimensional structure model of the steel bar surface. The specific application method is:

[0112] Calculate the distance between point 1 and point 2:

[0113] D12 = sqrt((X2-X1)^2+(Y2-Y1)^2+(Z2-Z1)^2) = sqrt((15-10)^2+(25-20)^2+(7-5)^2) = sqrt(5^2+5^2+2^2) = sqrt(25+25+4) = 7.35;

[0114] Similarly, calculate the distance between point 1 and point 3:

[0115] D13 = sqrt((X3 - X1)^2 + (Y3 - Y1)^2 + (Z3 - Z1)^2) = sqrt((12 - 10)^2 + (22 - 20)^2 + (6 - 5)^2) = sqrt(2^2 + 2^2 + 1^2) = sqrt(4 + 4 + 1) = 3.00.

[0116] By calculating the distance between each data point, a complete three-dimensional structural information of the steel surface can be constructed. This not only helps to identify the specific shape and features of the steel surface, but also provides accurate positioning basis for subsequent corrosion degree identification and rust removal operations.

[0117] Step three: determine the location coordinates of the steel surface treatment area according to the three-dimensional structural information; specifically including:

[0118] Analyze the preliminary three-dimensional data set {(Xi, Yi, Zi)} and identify the height variation characteristics of different regions of the steel surface. For each point (Xi, Yi, Zi), calculate the height difference ΔH = |Zi - Zj| between adjacent points, where i and j represent adjacent data points. Regions with larger height differences indicate the presence of corrosion, cracks, or other problems that need to be addressed.

[0119] Suppose there are four laser points in a scanning area, with their location coordinates and depth values as follows:

[0120] Point 1: (X1, Y1, Z1) = (10, 20, 5);

[0121] Point 2: (X2, Y2, Z2) = (15, 25, 7);

[0122] Point 3: (X3, Y3, Z3) = (12, 22, 6);

[0123] Point 4: (X4, Y4, Z4) = (18, 28, 9);

[0124] Calculate the height difference between point 1 and point 2:

[0125] ΔH12 = |Z2 - Z1| = |7 - 5| = 2;

[0126] Similarly, calculate the height difference between other adjacent points:

[0127] ΔH13 = |Z3 - Z1| = |6 - 5| = 1;

[0128] ΔH24 = |Z4 - Z2| = |9 - 7| = 2;

[0129] By calculating the height difference, the height variation characteristics of the rebar surface can be identified, providing a basis for subsequent marking of potential areas requiring treatment.

[0130] Based on the distribution of the height difference ΔH, a threshold T is set. When the height difference at a certain point exceeds the threshold T, it is marked as a potential area requiring processing. The set S of all potential areas requiring processing is determined by the formula S = {(Xi, Yi, Zi) | ΔH > T}. Assuming the set threshold T = 1.5, then based on the previously calculated height difference:

[0131] ΔH12 = 2 (greater than T), marked as a potential area requiring processing;

[0132] ΔH13 = 1 (less than T), no marking;

[0133] ΔH24 = 2 (greater than T), marked as a potential area requiring processing;

[0134] Therefore, the set S = {(10,20,5),(15,25,7),(18,28,9)}. Technical effect: By setting thresholds and marking potential areas requiring processing, areas needing special attention can be effectively filtered out.

[0135] For each point (Xi, Yi, Zi) in the set S, the distance D between it and other marked points is calculated using the spatial geometric relation formula D = sqrt((X2-X1)^2 + (Y2-Y1)^2 + (Z2-Z1)^2). If the distance between a point and other points is less than a specific value L, it is classified into the same processing region. Assuming the specific value L = 5, the distances between points in set S are calculated as follows:

[0136] D12 = sqrt((X2-X1)^2+(Y2-Y1)^2+(Z2-Z1)^2) = sqrt((15-10)^2+(25-20)^2+(7-5)^2) = sqrt(5^2+5^2+2^2) = 7.35 (greater than L, not belonging to the same processing region);

[0137] D24 = sqrt((X4-X2)^2+(Y4-Y2)^2+(Z4-Z2)^2) = sqrt((18-15)^2+(28-25)^2+(9-7)^2) = sqrt(3^2+3^2+2^2) = 4.69 (less than L, belonging to the same processing region);

[0138] By calculating and classifying the distances between points, adjacent and similar regions can be grouped into one category, facilitating subsequent centralized processing and improving operational efficiency and accuracy.

[0139] According to all points (Xi, Yi, Zi) in the processing area, the center coordinates (Xc, Yc, Zc) of each processing area are calculated, where Xc = sum(Xi) / n, Yc = sum(Yi) / n, Zc = sum(Zi) / n, n is the number of points in the area, and the center coordinates are taken as the position coordinates of the steel bar surface processing area. These center coordinates represent the main position of the processing area, facilitating accurate positioning and operation of the intelligent robot arm.

[0140] Assuming that the processing area contains two points (10, 20, 5) and (18, 28, 9), then: Xc = (10 + 18) / 2 = 14; Yc = (20 + 28) / 2 = 24; Zc = (5 + 9) / 2 = 7; the center coordinates of the processing area are (14, 24, 7). By calculating the center coordinates of the processing area, an accurate target position can be provided for the intelligent robot arm, ensuring the efficiency and accuracy of the rust removal operation.

[0141] Step four: according to the position coordinates, instruct the intelligent robot arm to move to the designated position for preliminary positioning; specifically including:

[0142] According to the center coordinates (Xc, Yc, Zc), the displacement vector ΔP = (ΔX, ΔY, ΔZ) of the intelligent robot arm from the current position to the target position is calculated, where ΔX = Xc - X0, ΔY = Yc - Y0, ΔZ = Zc - Z0, (X0, Y0, Z0) is the current position of the intelligent robot arm; assuming that the intelligent robot arm is currently located at (X0, Y0, Z0) = (5, 10, 3) and the target position is (Xc, Yc, Zc) = (14, 24, 7), then:

[0143] ΔX = Xc - X0 = 14 - 5 = 9;

[0144] ΔY = Yc - Y0 = 24 - 10 = 14;

[0145] ΔZ = Zc - Z0 = 7 - 3 = 4;

[0146] Therefore, the displacement vector is ΔP = (9, 14, 4). Calculating the displacement vector can clearly indicate the distance and direction that the intelligent robot arm needs to move, providing basic data for subsequent path planning and motion control.

[0147] The angle adjustment value B of each joint is calculated using the displacement vector ΔP and the kinematics formula B = arctan2(ΔY, ΔX) + B_offset, where B_offset is the initial angle offset; assuming that the initial angle offset B_offset = 0, then:

[0148] B = arctan2(AY, AX) = arctan2(14, 9) = 1.004 radian (approximately equal to 57.5°).

[0149] By calculating the joint angle adjustment value, the intelligent robot arm can be guided to rotate and move accurately, ensuring that it can reach the target position along the shortest path, improving operation efficiency and accuracy.

[0150] Send control instructions to the intelligent robot arm to move according to the calculated joint angle adjustment value B, and monitor the position (X', Y', Z') of the intelligent robot arm in real time during the movement and compare it with the target position (Xc, Yc, Zc) to ensure that the robot arm moves along the planned path and adjusts the deviation in time. During the movement, assume that the actual position of the intelligent robot arm at a certain time is (X', Y', Z') = (10, 18, 5), compared with the target position (14, 24, 7), it is found that there is a deviation, the control system will adjust the motion trajectory of the robot arm according to the deviation.

[0151] When the intelligent robot arm reaches the vicinity of the target position, fine-tune its position using the fine-tuning formula δX = Xc - X', δY = Yc - Y', δZ = Zc - Z' until the position error E = sqrt(δX^2 + δY^2 + δZ^2) of the intelligent robot arm is less than the set threshold ε, completing the preliminary positioning.

[0152] Assume that the actual position of the intelligent robot arm when approaching the target position is (X', Y', Z') = (13.8, 23.9, 6.9) and the target position is (14, 24, 7), then:

[0153] δX = Xc - X' = 14 - 13.8 = 0.2;

[0154] δY = Yc - Y' = 24 - 23.9 = 0.1;

[0155] δZ = Zc - Z' = 7 - 6.9 = 0.1;

[0156] Calculate the position error:

[0157] E = sqrt(δX^2 + δY^2 + δZ^2) = sqrt(0.2^2 + 0.1^2 + 0.1^2) = sqrt(0.04 + 0.01 + 0.01) = 0.245;

[0158] If the threshold value ε is set to 0.3, E < ε, indicating that the mechanical arm has reached the target position. Through the fine adjustment formula and position error calculation, fine adjustment can be made when the mechanical arm approaches the target position, ensuring that its final position is highly consistent with the target position, reducing errors and improving positioning accuracy. This fine adjustment mechanism helps to improve the reliability and accuracy of the overall operation.

[0159] Step five: After the initial positioning of the intelligent mechanical arm, the rust degree of the steel bar surface is identified; specifically including:

[0160] After the initial positioning of the intelligent mechanical arm, the 3D laser and visual fusion device is started to perform high-resolution scanning on the area to be processed, obtaining image data and depth information in the area, recording the gray value G and depth value Z of each pixel point;

[0161] According to the gray value G and depth value Z, the rust index RI of each pixel point is calculated, RI = (G_max - G) / Z, where G_max is the preset maximum gray value for standardizing the gray difference; assuming that the gray value G of a certain pixel point (X, Y) is 150, the depth value Z is 5, and the preset maximum gray value G_max is 255, then: RI = (G_max - G) / Z = (255 - 150) / 5 = 105 / 5 = 21; By calculating the rust index, the rust degree of each pixel point can be quantified, providing a basis for subsequent rust distribution map generation.

[0162] Statistical analysis is performed on the rust index RI to generate a rust distribution map, including: for each pixel point (X, Y), the rust index RI(X, Y) is mapped to a two-dimensional matrix to form a rust distribution map M = {RI(X, Y)}; assuming that there is a 3x3 pixel area, the rust indexes are respectively:

[0163] RI(1,1) = 21, RI(1,2) = 18, RI(1,3) = 20;

[0164] RI(2,1) = 17, RI(2,2) = 22, RI(2,3) = 19;

[0165] RI(3,1) = 16, RI(3,2) = 23, RI(3,3) = 24;

[0166] Then the rust distribution map formed is:

[0167] M = [

[0168] [21, 18, 20],

[0169] [17, 22, 19],

[0170] [16, 23, 24] ];

[0172] The generated rust distribution map visually displays the rust condition of the entire treatment area, which helps to identify the local area with severe rust.

[0173] Based on the rust distribution map M, the average rust index RI_avg = sum(RI(X,Y)) / N of the entire treatment area is calculated, where N is the number of pixel points in the area; if RI_avg exceeds the preset threshold T_ri, the area is marked as a severe rust area, otherwise it is marked as a mild rust area. Assuming there are a total of 9 pixel points:

[0174] sum(RI(X,Y)) = 21 + 18 + 20 + 17 + 22 + 19 + 16 + 23 + 24 = 180;

[0175] N = 9;

[0176] RI_avg = 180 / 9 = 20;

[0177] Assuming the set threshold T_ri = 18, then RI_avg > T_ri, so the area is marked as a severe rust area. By calculating the average rust index and setting the threshold, the severity of rust in the entire area can be effectively determined.

[0178] Step six: dynamically adjusting the rust removal force parameter of the intelligent mechanical arm according to the rust degree; specifically including:

[0179] According to the rust distribution map M and the average rust index RI_avg, the rust grade of each pixel point (X,Y) is determined; the rust index range of the mild rust area is defined as 0 <= RI(X,Y) < T_ri, and the rust index range of the severe rust area is defined as RI(X,Y) >= T_ri;

[0180] Assuming T_ri = 20, for a pixel point (X,Y), its rust index RI(X,Y) = 18, because RI(X,Y) < T_ri, so the point is marked as a mild rust area. Another pixel point (X',Y'), its rust index RI(X',Y') = 25, because RI(X',Y') >= T_ri, so the point is marked as a severe rust area.

[0181] For each treatment area, the rust removal force coefficient F = (RI_avg - RI_min) / (RI_max - RI_min) of the area is calculated, where RI_min and RI_max are the preset minimum and maximum rust index thresholds respectively; assuming RI_avg = 20, RI_min = 10, RI_max = 30, then: F = (20-10) / (30-10) = 10 / 20 = 0.5.

[0182] By calculating the rust removal intensity coefficient, the working parameters of the rust removal tool can be quantitatively adjusted to adapt to different rust levels.

[0183] Based on the rust removal intensity coefficient F, the working parameters of the rust removal tool of the intelligent mechanical arm are adjusted, including: for heavy rust area, using the formula P=P_base+F*ΔP to calculate the new rust removal power P, where P_base is the base power and ΔP is the power increment; for light rust area, using the formula P=P_base-F*ΔP to reduce the rust removal power; assuming P_base=100W, ΔP=50W, F=0.5, then for heavy rust area: P=100+0.5*50=125W; for light rust area: P=100-0.5*50=75W;

[0184] Dynamic adjustment of rust removal power helps to optimize energy consumption and ensure the most appropriate rust removal strategy for different levels of rust.

[0185] Apply the adjusted rust removal power P to the intelligent mechanical arm and monitor the rust removal effect in real time, compare the change of rust index after actual rust removal ΔRI=RI_before-RI_after, if ΔRI reaches the expected target, keep the current parameters and continue to operate; otherwise, re-adjust the rust removal intensity coefficient F and update the rust removal power P.

[0186] Assuming RI_before=25, after rust removal RI_after=15, then: ΔRI=25-15=10; if the expected target is to reduce at least 8 units of rust index, then ΔRI=10 has reached the standard, and the current parameters can be kept to continue operation. Real-time monitoring and feedback mechanism can ensure the effectiveness and accuracy of the rust removal process, and timely adjust the strategy to cope with different rust conditions.

[0187] Step seven: use the rust removal intensity parameter to control the intelligent mechanical arm to perform rust removal operation, and record operation data; specifically including:

[0188] Input the adjusted rust removal power P into the intelligent mechanical arm to start the rust removal tool, and the intelligent mechanical arm adjusts the working state of the tool according to the rust removal power P; assuming the adjusted rust removal power P=125W, input this value into the intelligent mechanical arm control system, start the rust removal tool and set the working power to 125W. Dynamic adjustment of rust removal power can ensure the most appropriate rust removal strategy for different rust levels.

[0189] In the rust removal process, the operation data of the intelligent mechanical arm is collected in real time, including the current rust removal power P, the working time t and the position coordinates (X, Y, Z); for each collection point, the original rust index RI_before and the rust index RI_after after rust removal are recorded; within a certain processing period, the data of a certain collection point is assumed as follows:

[0190] The rust removal power P = 125 W;

[0191] The working time t = 30 seconds;

[0192] The position coordinates (X, Y, Z) = (14, 24, 7);

[0193] The original rust index RI_before = 25;

[0194] The rust index RI_after after rust removal = 15;

[0195] Real-time data collection provides detailed information about the rust removal process, which helps to evaluate the rust removal effect.

[0196] Based on the operation data, the rust removal efficiency E = (RI_before - RI_after) / t of each processing area is calculated, where t is the processing time of the area, and the average rust removal efficiency E_total of the entire work surface is calculated by the formula E_total = sum(E) / N, where N is the number of processing areas; assuming the processing time t = 30 seconds, then:

[0197] E = (25 - 15) / 30 = 10 / 30 = 0.333;

[0198] If there are three similar processing areas, then:

[0199] E_total = (0.333 + 0.35 + 0.30) / 3 = 0.328.

[0200] The operation data and calculation results are stored in the database to form a historical record, including: for each processing period, a data packet is generated:

[0201] {(P, t, X, Y, Z, RI_before, RI_after, E)}, and saved. By saving detailed operation data, not only can the specific situation of each rust removal operation be tracked, but also can provide basis for future optimization, improve the intelligent level of the system and the overall efficiency of the rust removal work.

[0202] Step eight: analyze the operation data to calculate the rust removal efficiency, and automatically optimize the parameter settings of the next rust removal period according to the calculation results; specifically including:

[0203] Extract the operation data packet {(P, t, X, Y, Z, RI_before, RI_after, E)} of the last rust removal cycle from the database, and calculate the average rust removal efficiency E_avg = sum(E) / N for each treatment area, where N is the number of treatment areas; assuming there are three treatment areas with rust removal efficiencies E1 = 0.333, E2 = 0.35, and E3 = 0.30, then:

[0204] E_avg = (0.333 + 0.35 + 0.30) / 3 = 0.328; calculating the average rust removal efficiency can quantitatively evaluate the performance of the entire rust removal cycle, helping to identify areas for improvement.

[0205] Based on the average rust removal efficiency E_avg and the original rust index RI_before of each area, determine the optimization target, define the optimization target to achieve higher rust removal efficiency E_target = E_avg + ΔE in the next round of rust removal operation, where ΔE is the desired efficiency improvement value; assuming E_avg = 0.328 and the desired efficiency improvement value ΔE = 0.05, then: E_target = 0.328 + 0.05 = 0.378; setting a clear optimization target helps guide the system to make targeted adjustments.

[0206] According to the optimization target, adjust the rust removal force parameter of the intelligent mechanical arm, for each treatment area, use the formula P_new = P * (E_target / E_avg) to calculate the new rust removal power P_new; if P_new exceeds the preset maximum power P_max, then set P_new to P_max; assuming the original rust removal power P = 125W for a certain area, then: P_new = 125 * (0.378 / 0.328) = 143.9W;

[0207] If the preset maximum power P_max = 150W, then P_new meets the requirements; otherwise, if P_new > P_max, then set P_new to P_max. Dynamically adjusting the rust removal power can be optimized flexibly for different rust levels, ensuring that each treatment area can obtain the most suitable rust removal force, while avoiding overloading of the equipment.

[0208] Apply the new rust removal power P_new to the intelligent mechanical arm and update the parameter settings for the next rust removal cycle; at the same time, save the optimized parameter settings to the database to form a historical record, and continuously improve the rust removal operation by comparing data from different cycles.

[0209] By saving the optimized parameter settings, not only can the specific conditions of each rust removal operation be tracked, but also can provide a basis for future optimization. This continuous improvement mechanism helps to continuously improve the performance and efficiency of the system, ensuring long-term stable and efficient rust removal operation. In addition, by comparing data of different periods, potential problems can be found and solved in time, further improving the overall work quality.

[0210] In another aspect, the present application proposes a construction site exposed steel bar self-adaptive rust removal system based on intelligent robot arm, as shown in the figure, comprising: Figure 2

[0211] A polishing device configuration module is configured to configure a multi-degree-of-freedom intelligent robot arm and connect the intelligent robot arm with a 3D laser and vision fusion device.

[0212] A polishing coordinate acquisition module is configured to use the 3D laser and vision fusion device to scan the surface of the steel bar to be processed and acquire its three-dimensional structure information, and determine the position coordinates of the steel bar surface treatment area according to the three-dimensional structure information.

[0213] A corrosion degree recognition module is configured to direct the intelligent robot arm to move to a specified position for preliminary positioning according to the position coordinates, and the intelligent robot arm recognizes the rust degree of the steel bar surface after preliminary positioning.

[0214] A dynamic rust removal module is configured to dynamically adjust the rust removal intensity parameters of the intelligent robot arm according to the rust degree, and use the rust removal intensity parameters to control the intelligent robot arm to perform rust removal operation, while recording operation data.

[0215] A rust removal optimization module is configured to analyze the operation data to calculate the rust removal efficiency, and automatically optimize the parameter settings of the next rust removal period according to the calculation results.

[0216] In addition, each of the above modules is also used to implement other steps of the above-mentioned construction site exposed steel bar self-adaptive rust removal method based on intelligent robot arm when executed, which will not be described here.

[0217] In summary, through the combination of multi-degree-of-freedom intelligent robot arm and 3D laser and vision fusion device, high-precision steel bar surface three-dimensional structure recognition and positioning are realized, ensuring that the robot arm can accurately move to the specified position for rust removal operation. At the same time, according to the scanning results, the rust removal intensity parameters are dynamically adjusted, so that each area can obtain the most suitable rust removal intensity, thereby improving the rust removal efficiency and quality.

[0218] ​In addition, by recording operation data in real time and analyzing rust removal efficiency, parameter settings of the next rust removal cycle are automatically optimized, a closed-loop feedback mechanism is formed, and overall rust removal effect and equipment utilization are further improved. This method effectively solves the problems of low efficiency and inconsistent rust removal effect in the prior art, and provides an efficient, accurate and adaptive exposed steel bar rust removal solution.

[0219] Finally, it should be noted that the above only describes the preferred embodiments of the present application and is not intended to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or equivalently replace some technical features thereof, as long as they are within the spirit and principles of the present application. Any modification, equivalent replacement, improvement, etc. made within the scope of the present application shall be included in the protection scope of the present application.

Claims

1. A construction site exposed steel bar self-adaptive rust removal method based on an intelligent robot arm, characterized in that, The method comprises the following steps: configuring a multi-degree-of-freedom intelligent robot arm and connecting the intelligent robot arm with a 3D laser and vision fusion device; scanning the surface of the steel bar to be processed using the 3D laser and vision fusion device to obtain three-dimensional structure information of the surface, and determining the position coordinates of the area of the steel bar surface to be processed according to the three-dimensional structure information; directing the intelligent robot arm to move to a specified position for preliminary positioning according to the position coordinates, and identifying the rust degree of the surface of the steel bar after the preliminary positioning of the intelligent robot arm; dynamically adjusting the rust removal force parameter of the intelligent robot arm according to the rust degree, and controlling the intelligent robot arm to perform rust removal operation using the rust removal force parameter, while recording operation data; analyzing the operation data to calculate the rust removal efficiency, and automatically optimizing the parameter setting of the next rust removal cycle according to the calculation result; wherein the rust degree of the surface of the steel bar is identified after the preliminary positioning of the intelligent robot arm, comprising: after the intelligent robot arm completes the preliminary positioning, starting the 3D laser and vision fusion device to perform high-resolution scanning on the area to be processed, obtaining image data and depth information in the area, and recording the gray value G and depth value Z of each pixel point; calculating the rust index RI=(G_max-G) / Z of each pixel point according to the gray value G and the depth value Z, wherein G_max is the maximum gray value preset for standardizing the gray difference; statistically analyzing the rust index RI to generate a rust distribution map, including: for each pixel point (X, Y), the rust index RI(X, Y) is mapped into a two-dimensional matrix to form a rust distribution map M={RI(X, Y)}; based on the rust distribution map M, calculating the average rust index RI_avg=sum(RI(X, Y)) / N of the entire area to be processed, wherein N is the number of pixel points in the area; if RI_avg exceeds a preset threshold T_ri, the area is marked as a heavy rust area, otherwise it is marked as a light rust area.

2. The construction site exposed reinforcement adaptive derusting method based on the intelligent mechanical arm according to claim 1, characterized in that, The configuration of a multi-degree-of-freedom intelligent robot arm and the connection of the intelligent robot arm with a 3D laser and vision fusion device comprise: installing the intelligent robot arm at a predetermined working position, and fixing the 3D laser and vision fusion device on the end effector of the intelligent robot arm; calibrating the spatial coordinate system of the 3D laser and vision fusion device relative to the intelligent robot arm using the coordinate transformation formulas X'=X*cos(A)-Y*sin(A) and Y'=X*sin(A)+Y*cos(A), wherein A is the rotation angle of the 3D laser and vision fusion device relative to the intelligent robot arm, X and Y are the original coordinate values, and X' and Y' are the calibrated coordinate values; based on the calibrated coordinate system, initializing the intelligent robot arm to complete the connection of the intelligent robot arm with the 3D laser and vision fusion device.

3. The method of claim 2, wherein the method further comprises: The scanning of the surface of the steel bar to be processed using the 3D laser and vision fusion device to obtain three-dimensional structure information comprises: starting the intelligent robot arm and activating the 3D laser and vision fusion device; Directing the intelligent mechanical arm to move to above the steel bar to be processed according to a preset path, so that the 3D laser and vision fusion device is located at a scanning position; Using the 3D laser and vision fusion device to emit a laser beam and synchronously capture an image of the surface of the steel bar, record the position (Xi, Yi) of each laser point in the image and its corresponding depth value Zi, and form a preliminary three-dimensional data set {(Xi, Yi, Zi)}; Based on the preliminary three-dimensional data set {(Xi, Yi, Zi)}, applying a spatial geometric relationship formula D=sqrt((X2-X1)^2+(Y2-Y1)^2+(Z2-Z1)^2) to calculate the distance D between each data point, so as to construct complete three-dimensional structure information of the surface of the steel bar.

4. The construction site exposed reinforcement adaptive derusting method based on the intelligent mechanical arm according to claim 3, characterized in that, According to the three-dimensional structure information, determining the position coordinates of the region of the surface of the steel bar to be processed, comprising: Analyzing the preliminary three-dimensional data set {(Xi, Yi, Zi)}, identifying the height variation characteristics of different regions of the surface of the steel bar, for each point (Xi, Yi, Zi), calculating the height difference ΔH=|Zi-Zj| between it and the adjacent point, wherein i and j represent adjacent data points; Based on the distribution of the height difference ΔH, setting a threshold value T, when the height difference of a certain point exceeds the threshold value T, marking it as a potential region to be processed, and determining the set S of all potential regions to be processed through the formula S={(Xi, Yi, Zi)|ΔH>T}; For each point (Xi, Yi, Zi) in the set S, using the spatial geometric relationship formula D=sqrt((X2-X1)^2+(Y2-Y1)^2+(Z2-Z1)^2) to calculate the distance D between it and other marked points, if the distance between a certain point and other points is less than a certain value L, then it is classified into the same processing region; According to all points (Xi, Yi, Zi) in the processing region, calculating the center coordinates (Xc, Yc, Zc) of each processing region, wherein Xc=sum(Xi) / n, Yc=sum(Yi) / n, Zc=sum(Zi) / n, n is the number of points in the region, and taking the center coordinates as the position coordinates of the region of the surface of the steel bar to be processed.

5. The construction site exposed reinforcement adaptive derusting method based on the intelligent mechanical arm according to claim 4, characterized in that, According to the position coordinates, directing the intelligent mechanical arm to move to a specified position for preliminary positioning, comprising: According to the center coordinates (Xc, Yc, Zc), calculating the displacement vector ΔP=(ΔX, ΔY, ΔZ) of the intelligent mechanical arm from the current position to the target position, wherein ΔX=Xc-X0, ΔY=Yc-Y0, ΔZ=Zc-Z0, (X0, Y0, Z0) is the current position of the intelligent mechanical arm; Using the displacement vector ΔP and the kinematics formula B=arctan2(ΔY, ΔX)+B_offset of the intelligent mechanical arm to calculate the angle adjustment value B of each joint, wherein B_offset is an initial angle offset; Sending a control instruction to the intelligent mechanical arm, so that it moves according to the calculated joint angle adjustment value B, and in the moving process, the position (X', Y', Z') of the intelligent mechanical arm is monitored in real time and compared with the target position (Xc, Yc, Zc). When the intelligent robot arm reaches the vicinity of the target position, its position is fine-tuned using the fine-tuning formula δX=Xc-X', δY=Yc-Y', δZ=Zc-Z', until the position error E=sqrt(δX^2+δY^2+δZ^2) of the intelligent robot arm is less than the set threshold ε, completing the preliminary positioning.

6. The construction site exposed reinforcement adaptive derusting method based on the intelligent mechanical arm according to claim 5, characterized in that, According to the rust degree, the rust removal force parameter of the intelligent robot arm is dynamically adjusted, including: According to the rust distribution map M and the average rust index RI_avg, the rust grade of each pixel point (X, Y) is determined; the rust index range of the mild rust area is defined as 0<=RI(X,Y)<T_ri, and the rust index range of the severe rust area is defined as RI(X,Y)>=T_ri; For each area to be processed, the rust removal force coefficient F=(RI_avg-RI_min) / (RI_max-RI_min) is calculated, where RI_min and RI_max are the preset minimum and maximum rust index thresholds, respectively; Based on the rust removal force coefficient F, the working parameters of the rust removal tool of the intelligent robot arm are adjusted, including: for the severe rust area, the new rust removal power P is calculated using the formula P=P_base+F*ΔP, where P_base is the base power and ΔP is the power increment; for the mild rust area, the rust removal power is reduced using the formula P=P_base-F*ΔP; The adjusted rust removal power P is applied to the intelligent robot arm and the rust removal effect is monitored in real time, and the rust index change ΔRI=RI_before-RI_after after actual rust removal is compared. If ΔRI reaches the expected target, the current parameters are kept and the operation continues; otherwise, the rust removal force coefficient F is re-adjusted and the rust removal power P is updated.

7. The construction site exposed reinforcement adaptive derusting method based on the intelligent mechanical arm according to claim 6, characterized in that, The rust removal force parameter is used to control the intelligent robot arm to perform the rust removal operation, and operation data is recorded, including: The adjusted rust removal power P is input into the intelligent robot arm, and the rust removal tool is started. The intelligent robot arm adjusts the working state of the tool according to the rust removal power P. During the rust removal process, the operation data of the intelligent robot arm is collected in real time, including the current rust removal power P, working time t and position coordinates (X, Y, Z); for each collection point, the original rust index RI_before and the rust index RI_after after rust removal are recorded; Based on the operation data, the rust removal efficiency E=(RI_before-RI_after) / t of each processing area is calculated, where t is the processing time of the area, and the average rust removal efficiency E_total of the entire work surface is calculated by the formula E_total=sum(E) / N, where N is the number of processing areas; The operation data and calculation results are stored in the database to form a historical record, including: for each processing period, a data packet is generated: {(P,t,X,Y,Z,RI_before,RI_after,E)}, and saved.

8. The construction site exposed reinforcement adaptive derusting method based on the intelligent mechanical arm according to claim 7, characterized in that, The operation data is analyzed to calculate the rust removal efficiency, and the parameter settings of the next rust removal period are automatically optimized according to the calculation results, including: extracting an operation data packet {(P, t, X, Y, Z, RI_before, RI_after, E)} of a last rust removal cycle from the database and calculating an average rust removal efficiency E_avg = sum(E) / N of each treatment area, where N is the number of treatment areas; determining an optimization target based on the average rust removal efficiency E_avg and the original rust index RI_before of each area, defining the optimization target to achieve a higher rust removal efficiency E_target = E_avg + AE in the next round of rust removal operation, where AE is the expected efficiency improvement value; adjusting the rust removal force parameter of the intelligent mechanical arm according to the optimization target, for each treatment area, using the formula P_new = P*(E_target / E_avg) to calculate a new rust removal power P_new; if P_new exceeds a preset maximum power P_max, P_new is set to P_max; applying the new rust removal power P_new to the intelligent mechanical arm and updating the parameter setting of the next rust removal cycle; at the same time, saving the optimized parameter setting to the database to form a historical record, and continuously improving the rust removal operation by comparing the data of different cycles.

9. A smart robot based exposed reinforcement adaptive rust removal system for construction sites for implementing the method as claimed in any one of claims 1 to 8, wherein, It comprises: a polishing device configuration module for configuring a multi-degree-of-freedom intelligent mechanical arm and connecting the intelligent mechanical arm with a 3D laser and vision fusion device; a polishing coordinate acquisition module for scanning the surface of the steel bar to be treated using the 3D laser and vision fusion device to obtain its three-dimensional structure information, and determining the position coordinates of the treatment area on the surface of the steel bar according to the three-dimensional structure information; a corrosion degree recognition module for directing the intelligent mechanical arm to move to the specified position for preliminary positioning according to the position coordinates, and recognizing the corrosion degree of the surface of the steel bar after the intelligent mechanical arm is preliminarily positioned; specifically comprising: after the intelligent mechanical arm completes the preliminary positioning, starting the 3D laser and vision fusion device to perform high-resolution scanning on the treatment area to obtain image data and depth information of the area, and recording the gray value G and depth value Z of each pixel point; calculating the rust index RI = (G_max - G) / Z of each pixel point according to the gray value G and the depth value Z, where G_max is a preset maximum gray value for standardizing the gray difference; statistically analyzing the rust index RI to generate a rust distribution map, including: for each pixel point (X, Y), the rust index RI(X, Y) is mapped into a two-dimensional matrix to form a rust distribution map M = {RI(X, Y)}; based on the rust distribution map M, calculating the average rust index RI_avg = sum(RI(X, Y)) / N of the entire treatment area, where N is the number of pixel points in the area; if RI_avg exceeds a preset threshold T_ri, the area is marked as a heavy rust area, otherwise it is marked as a light rust area; A dynamic rust removal module is configured to dynamically adjust a rust removal intensity parameter of the intelligent robot arm according to the rust degree, and use the rust removal intensity parameter to control the intelligent robot arm to perform a rust removal operation, and record operation data; A rust removal optimization module is configured to analyze the operation data to calculate a rust removal efficiency, and automatically optimize parameter settings for a next rust removal cycle according to a calculation result.

Images