Weld appearance detection method based on liquid control floating polishing head

Abstract

This invention relates to a weld morphology inspection method based on a hydraulically controlled floating grinding head, comprising the following steps: S1, calibrating the initial coordinates and initial angle of the center point of the non-floating end; S2, using a robot to drive the hydraulically controlled floating grinding head to multiple inspection points sequentially, at each inspection point, the telescopic drive device drives the floating end to extend and retract until the center point of the end contacts the weld and the pressure on the end is a set constant force, recording the coordinates of the center point before extension and retraction, the extension and retraction distance of the center point, and the attitude angle of the end, and calculating the coordinates of the center point after extension and retraction, which are the coordinates of the feature points of the weld; S3, obtaining the weld morphology based on the coordinates of each feature point of the weld. This invention can quickly obtain the key feature points of the weld morphology, achieve efficient weld inspection, and facilitate subsequent automated grinding.

Description

Technical Field

[0001] This invention relates to the field of welding technology, and in particular to a method for detecting weld morphology based on a hydraulically controlled floating grinding head. Background Technology

[0002] Welding is one of the three major material processing methods and has important applications in industrial production. Grinding is a commonly used material processing method. Grinding metal welds can, on the one hand, change the weld morphology to facilitate assembly, and on the other hand, reduce welding stress, improve weld fatigue strength, and improve weld formation. Therefore, grinding is an important means of post-weld processing.

[0003] Currently, most weld grinding is done manually. However, the harsh working environment, with its high levels of dust and noise, can cause physical and psychological harm to workers. Furthermore, manual grinding is inefficient and produces inconsistent quality. Therefore, replacing manual grinding with automated grinding is of great significance. While CNC machine tools offer high machining precision, they lack flexibility, requiring the workpiece dimensions to match the machine tool's size and limiting their outdoor operation capabilities. Therefore, robotic grinding represents the future trend in automated grinding.

[0004] There are three types of robot programming: manual teaching, offline programming, and automatic programming. Traditional manual teaching is time-consuming, inefficient, and has poor accuracy, making it unsuitable for high-efficiency, high-quality automated production. Offline programming requires strict model assembly relationships, but in actual production, it is difficult to ensure that the model assembly relationships match the actual situation, thus limiting its feasibility in practical production. Automatic programming involves the robot autonomously obtaining the workpiece model in conjunction with sensors, autonomously planning its trajectory, and achieving fully automated operation, which is true intelligent manufacturing.

[0005] Currently, the mainstream method for obtaining workpiece models for grinding is to scan the workpiece using laser vision or 3D vision to acquire a three-dimensional model for subsequent trajectory planning and autonomous grinding. However, both laser vision and 3D vision sensors are expensive, with a maximum accuracy of only 0.1mm. Furthermore, when scanning the ground metal surface, significant reflections result in poor point cloud imaging quality or even no imaging at all, severely impacting practical application. Contact sensing has been used in welding. Its principle involves applying a voltage between the welding wire and the workpiece. As the welding wire slowly approaches the workpiece, the voltage is detected at the moment of contact, and the position difference is calculated based on the voltage feedback. However, the welding wire used in traditional contact sensing is prone to breakage, and the welding torch and robot themselves are also susceptible to damage due to their lack of flexibility. How to achieve high-efficiency, high-precision, and low-cost weld inspection is a pressing technical problem in this field. Summary of the Invention

[0006] Therefore, the technical problem to be solved by the present invention is to achieve high-efficiency, high-precision, and low-cost weld morphology detection.

[0007] To solve the above technical problems, the present invention provides a weld morphology detection method based on a hydraulically controlled floating grinding head. The hydraulically controlled floating grinding head includes an end for contacting the weld, a rotary drive device for driving the end to rotate, a telescopic drive device for driving the end to extend and retract, a gyroscope for detecting the attitude angle of the end, a pressure sensor for detecting the pressure on the end, and a displacement sensor for detecting the extension and retraction distance of the end.

[0008] The weld morphology inspection method includes the following steps:

[0009] S1. The initial coordinates and initial angle of the end center point, which is set to be non-floating, are calibrated, and then proceed to step S2;

[0010] S2. The robot drives the hydraulic floating grinding head to reach multiple detection points in sequence. At each detection point, the telescopic drive device drives the floating end to extend and retract until the center point of the end contacts the weld and the pressure on the end is a set constant force. The coordinates of the center point of the end before extension and retraction, the extension and retraction distance of the center point of the end, and the attitude angle of the end are recorded respectively. The coordinates of the center point of the end after extension and retraction are calculated, which are the coordinates of the feature points of the weld. Proceed to step S3.

[0011] S3. Obtain the morphology of the weld by using the coordinates of each characteristic point of the weld.

[0012] The length direction of the weld is defined as the X-axis direction, the width direction of the weld is defined as the Y-axis direction, the height direction of the weld is defined as the Z-axis direction, the angle between the end and the X-axis direction is defined as A, the angle between the end and the Y-axis direction is defined as B, and the angle between the end and the Z-axis direction is defined as C. The coordinates of each feature point of the weld are calculated using the following formula.

[0013] ; ; ;

[0014] in, , , These are the X-axis coordinates, Y-axis coordinates, and Z-axis coordinates of the end center point before expansion and contraction, respectively, when the end center point contacts the i-th feature point of the weld.

[0015] The extension distance of the end center point when the end center point is in contact with the i-th feature point of the weld and the pressure on the end is a set constant force;

[0016] , , These are the angles between the end and the X-axis, Y-axis, and Z-axis directions, respectively, when the center point of the end contacts the i-th feature point of the weld and the pressure on the end is a set constant force.

[0017] , , These are the X-axis coordinates, Y-axis coordinates, and Z-axis coordinates of the i-th feature point of the weld, respectively.

[0018] Furthermore, the set constant force is 3-7 N.

[0019] Furthermore, the plurality of detection points are respectively located in a plurality of detection areas arranged sequentially along the length direction of the weld, and the plurality of detection points in the same detection area are arranged sequentially along the width direction of the weld.

[0020] Step S2 includes the following sub-steps:

[0021] S21. Set one of the detection areas located at one end of the weld along the length direction as the current detection area, and proceed to step S22.

[0022] S22. The robot drives the hydraulic floating grinding head to move a set distance along the width of the weld and reach multiple detection points in the current detection area in sequence, then proceeds to step S23.

[0023] S23. Determine if there is another unreached inspection area. If so, the robot drives the hydraulic floating grinding head to move a set distance along the length of the weld to reach the next inspection area, and sets the next inspection area as the current inspection area. Return to step S22. If not, the inspection ends.

[0024] Furthermore, in step S23, the next detection area is adjacent to the current detection area, and the robot drives the hydraulic floating grinding head to move alternately along the width and length directions of the weld and arrive at multiple detection points in sequence.

[0025] Furthermore, step S22 includes the following sub-steps:

[0026] S221. The robot drives the hydraulic floating grinding head to move a set distance along the width of the weld and sequentially reach the M detection points in the current detection area, then proceed to step S222.

[0027] S222. Determine whether the Z-axis coordinates of the N detection points reached afterward are continuously decreasing. If yes, proceed to step S23; otherwise, proceed to step S223.

[0028] S223. Continue to move the hydraulic floating grinding head along the width direction of the weld by a set distance using the robot to reach the next detection point in the current detection area, and return to step S222.

[0029] Furthermore, M is set to 3-7, N is set to 2-4, and M is greater than N.

[0030] Furthermore, M is set to 5 and N is set to 3.

[0031] Furthermore, the coordinates of the end-point center point before stretching are stored in array L, the stretching distance of the end-point center point is stored in array D, and the attitude angle of the end is stored in array R.

[0032] Furthermore, a tip structure is installed on the end, and the tip of the tip structure serves as the center point of the end.

[0033] Furthermore, the rotary drive device is a servo motor, and the telescopic drive device is an electro-hydraulic servo actuator.

[0034] Compared with the prior art, the above-mentioned technical solution of the present invention has the following advantages: The weld morphology detection method based on hydraulic grinding head described in the present invention uses EHA hydraulic grinding head as a contact sensing device, which does not require additional vision sensors, can realize weld morphology detection, and is not affected by the reflection phenomenon of metal surface. The detection accuracy is 0.01mm, which is far higher than that of laser vision and 3D vision sensors; The present invention proposes a strategy method for weld morphology detection based on grinding force contact sensing, which can quickly obtain key feature points of weld morphology, realize efficient weld detection, and facilitate subsequent automated grinding; The proposed weld morphology detection method can be used not only for trajectory planning before grinding, but also for weld formation detection after grinding. Attached Figure Description

[0035] To make the content of this invention easier to understand, the invention will be further described in detail below with reference to specific embodiments and accompanying drawings.

[0036] Figure 1 This is a schematic diagram of the structure of the hydraulically controlled grinding head disclosed in this invention;

[0037] Figure 2 This is a schematic diagram of the weld structure disclosed in this invention;

[0038] Figure 3 This is a detection curve of the weld along the width direction in one embodiment of the present invention.

[0039] Explanation of reference numerals in the accompanying drawings: 1. Hydraulic floating grinding head; 11. End point; 111. End center point; 112. Tip structure; 12. Rotary drive device; 13. Telescopic drive device; 14. Gyroscope; 15. Pressure sensor; 16. Displacement sensor;

[0040] 2. Weld seam; 3. Inspection curve; 31. First feature point; 32. Second feature point; 33. Third feature point; 34. Fourth feature point; 35. Fifth feature point. Detailed Implementation

[0041] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments described are not intended to limit the present invention.

[0042] like Figure 1 The diagram shown is a structural schematic of the hydraulically controlled floating grinding head 1; Figure 2 The diagram shown is a schematic of the weld 2 being inspected.

[0043] The hydraulic floating grinding head 1 includes an end 11 for contacting the weld, a rotary drive device 12 for rotating the end 11, a telescopic drive device 13 for extending and retracting the end 11, a gyroscope 14 for detecting the attitude angle of the end 11, a pressure sensor 15 for detecting the pressure on the end 11, and a displacement sensor 16 for detecting the extension and retraction distance of the end 11.

[0044] The weld morphology inspection method includes the following steps:

[0045] S1. Calibrate the initial coordinates and initial angle of the non-floating end center point 111, and proceed to step S2.

[0046] S2. The robot (not shown in the figure) drives the hydraulic floating grinding head 1 to reach multiple detection points in sequence. At each detection point, the telescopic drive device 13 drives the floating end 11 to extend and retract until the end center point 111 contacts the weld 2 and the pressure on the end is a set constant force. Record the coordinates of the end center point 111 before extension and retraction, the extension and retraction distance of the end center point 111, and the attitude angle of the end 11. Calculate the coordinates of the end center point 111 after extension and retraction, which are the coordinates of the feature points of the weld 2. Proceed to step S3.

[0047] S3. Obtain the morphology of the weld seam based on the coordinates of each feature point of weld seam 2;

[0048] The length direction of weld 2 is defined as the X-axis direction, the width direction of weld 2 is defined as the Y-axis direction, the height direction of weld 2 is defined as the Z-axis direction, the angle between end 11 and the X-axis direction is defined as A, the angle between end 11 and the Y-axis direction is defined as B, and the angle between end 11 and the Z-axis direction is defined as C. The coordinates of each feature point of weld 2 are calculated using the following formula.

[0049] ;

[0050] ;

[0051] ;

[0052] in, , , These are the X-axis coordinates, Y-axis coordinates, and Z-axis coordinates of the end center point 111 before extension and contraction, respectively, when the end center point 111 contacts the i-th feature point of weld 2.

[0053] The extension distance of the end center point 111 when the end center point 111 contacts the i-th feature point of the weld 2 and the pressure on the end 11 is a set constant force;

[0054] , , These are the angles between the end center point 111 and the i-th feature point of weld 2, respectively, and the pressure exerted on the end 11 is a set constant force, and the angles between the end 11 and the X-axis, Y-axis, and Z-axis directions.

[0055] , , These are the X-axis coordinates, Y-axis coordinates, and Z-axis coordinates of the i-th feature point of weld 2, respectively.

[0056] In the above technical solution, the magnitude of the constant force at the end is first set. The contact force at the end is detected by the pressure sensor 15. The magnitude of the contact force is compared with the set constant force at the end. The floating distance is adjusted by the telescopic drive device 13 to achieve the constant force at the end. The floating distance can be detected by the displacement sensor 16, and the posture of the grinding head end can be detected by the gyroscope 14. When the end 11 contacts the workpiece surface, due to the set constant force at the end, the height of the workpiece surface can be represented by the floating distance of the floating head end. The aforementioned weld morphology detection method based on a hydraulically controlled grinding head uses an EHA hydraulically controlled grinding head as a contact sensing device. It eliminates the need for additional vision sensors, enabling weld morphology detection without being affected by surface reflection. The detection accuracy is 0.01 mm, significantly higher than laser vision and 3D vision sensors. This invention proposes a strategy for weld morphology detection using grinding force contact sensing, which can quickly acquire key feature points of the weld morphology, achieving efficient weld detection and facilitating subsequent automated grinding. The proposed weld morphology detection method can be used not only for trajectory planning before grinding but also for weld formation detection after grinding.

[0057] In some preferred embodiments, the constant force is set to 3-7 N. Specifically, the constant force can be set to 3 N, 4 N, 5 N, 6 N, or 7 N. The constant force is maintained throughout the entire process of weld morphology inspection.

[0058] In some preferred embodiments, multiple detection points are located in multiple detection areas arranged sequentially along the length of weld 2, and multiple detection points in the same detection area are arranged sequentially along the width of weld; step S2 includes the following sub-steps:

[0059] S21. Set one of the detection areas located at one end of the weld 2 along the length direction as the current detection area, and proceed to step S22.

[0060] S22. The robot drives the hydraulic floating grinding head 1 to move a set distance along the width direction of the weld 2 and arrive at multiple detection points in the current detection area in sequence, then proceed to step S23.

[0061] S23. Determine if there is another unreached inspection area. If yes, the robot drives the hydraulic floating grinding head 1 to move a set distance along the length of the weld to reach the next inspection area, sets the next inspection area as the current inspection area, and returns to step S22. If no, the inspection ends.

[0062] In the above technical solution, morphological inspection of the weld along the width direction of the weld is more conducive to the execution of the inspection.

[0063] In some preferred embodiments, in step S23, the next detection area is adjacent to the current detection area. The robot drives the hydraulically controlled floating grinding head 1 to move alternately along the width and length directions of the weld 2 and sequentially reach multiple detection points. The movement direction of the hydraulically controlled grinding head in two adjacent detection areas is opposite. For example, if it moves along the positive Y-axis in the current detection area, it moves along the negative Y-axis in the next detection area. If it moves along the negative Y-axis in the current detection area, it moves along the positive Y-axis in the next detection area.

[0064] The above technical solution allows for regular, full-coverage inspection of welds, enabling more accurate detection of weld shape.

[0065] In some preferred embodiments, step S22 includes the following sub-steps:

[0066] S221. The robot drives the hydraulic floating grinding head 1 to move a set distance along the width direction of the weld 2 and arrive at the M detection points in the current detection area in sequence, and then proceeds to step S222.

[0067] S222. Determine whether the Z-axis coordinates of the N detection points reached afterward are continuously decreasing. If yes, proceed to step S23; otherwise, proceed to step S223.

[0068] S223. Continue to move the hydraulic floating grinding head 1 along the width direction of the weld 2 by the robot by a set distance to reach the next detection point in the current detection area, and return to step S222.

[0069] Through the above technical solution, during the inspection process, adaptive inspection can be carried out according to the actual shape of the weld, which can more accurately detect the shape of the weld.

[0070] In some preferred embodiments, M is set to 3-7, N is set to 2-4, and M is greater than N. Specifically, M can be set to 3, 4, 5, 6, or 7, and N can be set to 2, 3, or 4.

[0071] In some preferred embodiments, M is set to 5 and N is set to 3. When M is set to 5, the curve connected by 5 feature points is closer to the actual curve, which will not increase the workload of detection and calculation. When N is set to 3, the curve connected by 3 feature points can reflect the trend of the weld.

[0072] In some preferred embodiments, the coordinates of the end-point center point 111 before extension / retraction are stored in array L, the extension / retraction distance of the end-point center point 111 is stored in array D, and the attitude angle of the end-point 11 is stored in array R. Grouping and storing these data facilitates later calculations.

[0073] In some preferred embodiments, a tip structure 112 is mounted on the end 11, and the tip of the tip structure 112 serves as the end center point 111. The position of the end center point requires high precision, and the tip position of the tip structure can meet this requirement. During morphology inspection, the tip structure is mounted on the end, and during grinding, the tip structure is removed from the end.

[0074] In some preferred embodiments, the rotary drive device 12 is a servo motor, and the telescopic drive device 13 is an electro-hydraulic servo actuator. The servo motor can control speed and has very accurate positioning. It can convert voltage signals into torque and speed to drive the controlled object. The rotor speed of the servo motor is controlled by the input signal and can respond quickly. In automatic control systems, it is used as an actuator and has characteristics such as a small electromechanical time constant and high linearity. It can convert the received electrical signals into angular displacement or angular velocity output on the motor shaft. The electro-hydraulic servo actuator is a hydraulic actuator that can convert hydraulic energy from a hydraulic source into mechanical energy. It can also be servo controlled as needed through a built-in displacement sensor or limit switch. It is used to execute commands from the main controller, controlling the speed, direction, displacement, and force of the load, while simultaneously feeding back signals to the main controller. It has a large output force, accurate operating position, and small size.

[0075] The following describes a specific implementation method for weld morphology inspection:

[0076] Sa, set the end 11 to be non-floating, and install the tip structure 112 on the end 11, then proceed to step Sb;

[0077] Sb, calibrate the initial coordinates and initial angle of the end center point 111, and after calibration, set the end to be floating, then proceed to step Sc;

[0078] Sc. Set one of the detection areas located at one end of weld 2 along its length as the current detection area, and proceed to step Sd;

[0079] Sd, the robot drives the hydraulic floating grinding head 1 to move a set distance along the width of the weld 2 and sequentially reach the 5 inspection points in the current inspection area, then proceeds to step Se;

[0080] Se, determine whether the Z-axis coordinates of the three detection points reached after the judgment are continuously decreasing. If yes, proceed to step Sg; otherwise, proceed to step Sf.

[0081] Sf, continue to move the hydraulic floating grinding head 1 along the width direction of the weld 2 by the robot by a set distance to reach the next inspection point in the current inspection area, and return to step Sf;

[0082] Sg: Determine if there is another unreached inspection area. If so, the robot drives the hydraulic floating grinding head 1 to move a set distance along the length of the weld to reach the next inspection area, set the next inspection area as the current inspection area, and return to step Sd. If not, the inspection ends.

[0083] In step Se, at each detection point, the telescopic drive device 13 drives the floating end 11 to extend and retract until the end center point 111 contacts the weld 2 and the pressure on the end is 5N. The coordinates of the end center point 111 before extension and retraction are recorded respectively. , , ), the extension distance of the end center point 111 And the attitude angle of the end point 11 ( , , ), calculate the scaling coordinates of the end center point 111 ( , , ), which are the coordinates of the feature points of weld 2.

[0084] like Figure 3 As shown, this is the detection curve 3 along the width direction of the weld. In step Se, the highest point of the weld along the width direction is denoted as the first feature point 31, and the lowest point is denoted as the second feature point 32 and the third feature point 33. The midpoint between the first feature point 31 and the second feature point 32 is denoted as the fourth feature point 34, and the midpoint between the first feature point 31 and the third feature point 33 is denoted as the fifth feature point 35. These five points are the five feature points of the weld in this detection direction. The coordinates of the five points in the i-th detection are denoted as (…). ), ( ), ( ), ( ), ( ).

[0085] In step Se, the curve of the weld is acquired and the positions of five feature points of the weld are recorded to determine... If the above formula is true, then stop detecting to the right; otherwise, continue to shift to the right by a distance of m.

[0086] In step Sg, the detection width is set to m, and the robot drives the hydraulic floating grinding head 1 to move a distance of m along the length of the weld to reach the next detection area.

[0087] Based on the coordinates of the key points of the weld, the three-dimensional point cloud information of the weld is obtained, and the various points are connected. The trajectory T1 is obtained from point ( ), which is the line connecting the highest points of the weld. This line is the first path for automated grinding. Connect each ( ) ) point and ( Points ( ) are used to obtain trajectories T4 and T5, which form the subsequent polishing paths. Connect each ( ) ) point and ( The trajectories T2 and T3 are obtained, and the width information of the weld can be obtained from T2 and T3.

[0088] The above steps obtain information on the highest point of the weld, the width of the weld, and multiple paths for automated weld grinding. This information can be used not only for automated weld grinding path planning but also for weld formation inspection, determining whether the weld height and width meet standards. It boasts high detection accuracy, is unaffected by metal surface reflection, has a 3D reconstruction error of less than 0.01mm, requires no additional vision sensors, and offers advantages such as low cost, high efficiency, and high precision.

[0089] Obviously, the above embodiments are merely illustrative examples for clear explanation and are not intended to limit the implementation. Those skilled in the art will recognize that other variations or modifications can be made based on the above description. It is neither necessary nor possible to exhaustively list all possible implementations here. However, obvious variations or modifications derived therefrom are still within the scope of protection of this invention.

Claims

1. A method for weld morphology inspection based on a hydraulically controlled floating grinding head, wherein the hydraulically controlled floating grinding head includes an end for contacting the weld, a rotary drive device for driving the end to rotate, a telescopic drive device for driving the end to extend and retract, a gyroscope for detecting the attitude angle of the end, a pressure sensor for detecting the pressure on the end, and a displacement sensor for detecting the telescopic distance of the end. Its features are, The weld morphology inspection method includes the following steps: S1. The initial coordinates and initial angle of the end center point, which is set to be non-floating, are calibrated, and then proceed to step S2; S2. The robot drives the hydraulic floating grinding head to reach multiple detection points in sequence. At each detection point, the telescopic drive device drives the floating end to extend and retract until the center point of the end contacts the weld and the pressure on the end is a set constant force. The coordinates of the center point of the end before extension and retraction, the extension and retraction distance of the center point of the end, and the attitude angle of the end are recorded respectively. The coordinates of the center point of the end after extension and retraction are calculated, which are the coordinates of the feature points of the weld. Proceed to step S3. S3. Obtain the morphology of the weld by using the coordinates of each characteristic point of the weld. The length direction of the weld is defined as the X-axis direction, the width direction of the weld is defined as the Y-axis direction, the height direction of the weld is defined as the Z-axis direction, the angle between the end and the X-axis direction is defined as A, the angle between the end and the Y-axis direction is defined as B, and the angle between the end and the Z-axis direction is defined as C. The coordinates of each feature point of the weld are calculated using the following formula. ； ； ； in, , , These are the X-axis coordinates, Y-axis coordinates, and Z-axis coordinates of the end center point before expansion and contraction, respectively, when the end center point contacts the i-th feature point of the weld. a distance of the end center point from the i-th feature point of the weld seam when the end is in contact with the i-th feature point and the end is subjected to a constant force set to a predetermined value; , , respectively, the angle between the end and the X-axis direction, the Y-axis direction and the Z-axis direction when the end is in contact with the i-th feature point of the weld and the pressure on the end is a set constant force; , , Xi, Yi, Zi are the X-axis coordinate, Y-axis coordinate and Z-axis coordinate of the i-th feature point of the weld, respectively; The plurality of detection points are located in a plurality of detection areas arranged sequentially along the length of the weld, and the plurality of detection points in the same detection area are arranged sequentially along the width of the weld. Step S2 includes the following sub-steps: S21. Set one of the detection areas located at one end of the weld along the length direction as the current detection area, and proceed to step S22. S22. The robot drives the hydraulic floating grinding head to move a set distance along the width of the weld and reach multiple detection points in the current detection area in sequence, then proceeds to step S23. S23. Determine if there is another unreached inspection area. If yes, the robot drives the hydraulic floating grinding head to move a set distance along the length of the weld to reach the next inspection area, and sets the next inspection area as the current inspection area. Return to step S22. If no, the inspection ends. Step S22 includes the following sub-steps: S221. The robot drives the hydraulic floating grinding head to move a set distance along the width of the weld and sequentially reach the M detection points in the current detection area, then proceed to step S222. S222. Determine whether the Z-axis coordinates of the N detection points reached afterward are continuously decreasing. If yes, proceed to step S23; otherwise, proceed to step S223. S223. Continue to move the hydraulic floating grinding head along the width direction of the weld by a set distance using the robot to reach the next detection point in the current detection area, and return to step S222.

2. The weld morphology detection method based on a hydraulically controlled floating grinding head according to claim 1, characterized in that, The set constant force is 3-7 N.

3. The method of weld profile detection based on hydraulically controlled floating polishing head according to claim 1, characterized in that, In step S23, the next detection area is adjacent to the current detection area. The robot drives the hydraulic floating grinding head to move alternately along the width and length of the weld and arrive at multiple detection points in sequence.

4. The method of weld profile detection based on hydraulically controlled floating polishing head according to claim 1, wherein, M is set to 3-7, N is set to 2-4, and M is greater than N.

5. The method of weld profile detection based on hydraulically controlled floating polishing head according to claim 4, characterized in that, M is set to 5, and N is set to 3.

6. The liquid controlled floating abrader head based weld profile inspection method as claimed in claim 1, wherein, The coordinates of the end-point center point before stretching are stored in array L, the stretching distance of the end-point center point is stored in array D, and the attitude angle of the end is stored in array R.

7. The weld morphology detection method based on a hydraulically controlled floating grinding head according to claim 1, characterized in that, A tip structure is installed on the end, and the tip of the tip structure serves as the center point of the end.

8. The liquid controlled floating abrader head based weld profile inspection method as claimed in claim 1, wherein, The rotary drive device is a servo motor, and the telescopic drive device is an electro-hydraulic servo actuator.

Images