Robot welding seam grinding system and method based on contact sensing
Through a robot weld grinding system based on contact sensing, the EHA constant force floating end effector and end contact wheel are used to detect the weld morphology in real time and generate the grinding trajectory. Combined with the optimization of adaptive process parameters, the cost in the existing technology is solved. The problem of high and poor accuracy achieves efficient and low-cost weld grinding effect.
Patent Information
- Application Number
- CN202411304839.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-18
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-09-18
AI Technical Summary
The existing weld grinding technology relies on expensive visual sensors and complex imaging environments, resulting in high cost, poor accuracy and poor stability, making it impossible to achieve high-precision automated weld grinding.
Using a robot weld grinding system based on contact sensing, the EHA constant force floating end effector and end contact wheel are used to detect the surface morphology of the weld through force sensors and displacement sensors in real time to generate grinding trajectories, and optimize them in combination with adaptive process parameters. , realize high-precision weld grinding.
Low-cost, high-precision weld grinding without additional visual sensors is achieved, reducing grinding costs, improving grinding efficiency and surface flatness, and avoiding over-grinding or under-grinding problems in traditional methods.
Smart Images

Figure CN119077504A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of weld grinding, and in particular to a robot weld grinding system and method based on contact sensing. Background Art
[0002] Weld grinding is widely used in the shipbuilding, automobile, aviation, aerospace and other industries. At present, most weld grinding is still done manually. Due to the harsh grinding environment, unstable manual grinding quality, rising labor costs and other reasons, it is urgent to realize robot welding to replace manual work.
[0003] The traditional teaching method is feasible for simple welds, but for complex welds on spatial curved surfaces, manual teaching cannot be applied on a large scale due to its operational errors and long time consumption. In addition, the pure force control grinding method is difficult to grind flat due to the unevenness of the weld, and it is easy to have quality problems such as over-grinding or under-grinding. It often requires a position control mode, similar to the machining method of machine tools, to achieve weld grinding flatness, but it is difficult for machine tools to achieve flexible and intelligent processing similar to robots. The surface morphology of the weld is an unstructured environment. The pure position mode requires external visual sensors to accurately measure the surface position of the weld, and then slice and mill to reduce material. Therefore, force control and position control modes are required to perform position mode milling to smooth the unevenness, and then perform force control mode combined with the parameter library for grinding to achieve weld grinding.
[0004] At present, the existing robot automated weld grinding is basically based on laser vision guidance or 3D vision guidance. The weld area is photographed by a laser camera or a 3D camera to obtain the weld area point cloud, and the weld is segmented using point cloud related algorithms. After that, the grinding trajectory is planned to obtain the grinding trajectory. However, this method cannot achieve high-precision residual height grinding of welds due to the high price of laser cameras or 3D cameras, the complex environment of the grinding site resulting in poor imaging effects, hand-eye calibration, and the robot's own absolute positioning errors. It is precisely the existence of these problems that greatly limits the actual use of robot automated grinding in factories.
[0005] Therefore, technicians in this field are committed to developing a contact sensing-based robot weld grinding system and method, which does not require additional visual sensors and achieves high-precision weld grinding effects at low cost. The weld surface morphology is obtained by combining the end contact wheel with the electro-hydraulic actuator EHA constant force floating end effector, and then the weld grinding trajectory is generated using a trajectory planning algorithm. Summary of the invention
[0006] In view of the above-mentioned defects of the prior art, the technical problems to be solved by the present invention are high shooting cost of the weld area, complex grinding site resulting in poor imaging effect, poor stability, high grinding cost and poor precision, etc.
[0007] To achieve the above-mentioned purpose, the present invention provides a contact sensing-based robot weld grinding system, comprising an EHA constant-force floating end effector, an EHA constant-force floating end control cabinet, an industrial robot, an industrial robot control cabinet, an industrial computer and a workbench, wherein the EHA constant-force floating end effector is connected to the end of the industrial robot to grind the weld, the industrial computer sends instructions to the industrial robot control cabinet to control the movement of the industrial robot, a workpiece is placed on the workbench, and the EHA constant-force floating end effector grinds the weld of the workpiece under the control of the industrial robot; the EHA constant-force floating end effector is a constant-force floating actuator driven by EHA hydraulic pressure, the EHA constant-force floating end effector is equipped with a force sensor, a displacement sensor, a grinding tool and an end contact wheel, the grinding tool and the end contact wheel are interchangeably installed at the end of the EHA constant-force floating end effector, the end contact wheel is used when the morphology on both sides of the weld is initially measured, and is replaced by the grinding tool during the grinding process.
[0008] Furthermore, the EHA constant force floating end effector includes a spindle motor, an EHA hydraulic servo motor, a force sensor, a connecting rod, a displacement sensor, a rotating spindle, a grinding tool, an end contact wheel and a floating part.
[0009] Furthermore, a constant output force is first set, and the force sensor and the EHA hydraulic servo motor form a constant force closed-loop control part to ensure a constant output force. The connecting rod transmits the constant force drive to the floating part, and the displacement sensor collects the displacement data of the floating part in real time. The floating part is connected to the rotating spindle and is driven to rotate by the spindle motor; the EHA hydraulic servo motor is connected to the floating part through the connecting rod, and the position of the floating part is adjusted through closed-loop control under the feedback of the force sensor to achieve constant force output.
[0010] Furthermore, the contact sensing-based morphology detection method is achieved through the joint action of the force sensor, the displacement sensor and the end contact wheel; first, a constant output force is set by the EHA constant-force floating end control cabinet, and the constant output force is jointly guaranteed by the force sensor and the EHA hydraulic servo motor, and the end contact wheel rolls in contact with the surface of the weld workpiece; then the displacement sensor is used to collect the displacement information of the floating part, and then converted into surface morphology information of the weld; during the grinding process, the grinding tool replaces the end contact wheel to contact the surface of the grinding area to realize real-time monitoring of the weld removal height.
[0011] Furthermore, before the profile detection of the metal roller with a radius of 5 mm at the end contact wheel, it is necessary to set the initial position of the floating part to zero, turn off the enable of the EHA servo hydraulic motor, perform TCP calibration at zero position, and calibrate the TCP at the end of the end contact wheel.
[0012] The present invention also provides a contact sensing-based robot weld grinding method, using the contact sensing-based robot weld grinding system, the method comprising the following steps:
[0013] Step 1, calibrating the TCP of the industrial robot and the end contact wheel respectively;
[0014] Step 2: Measure the topography of the areas on both sides of the weld;
[0015] Step 3, calculating the midpoint of the topography measurement points;
[0016] Step 4: Generate a polishing track according to the target prediction;
[0017] Step 5: Perform grinding operation;
[0018] Step 6: Optimize grinding parameters in real time;
[0019] Step 7: Determine whether the grinding operation is completed. If not, jump to step 5. If completed, end the grinding.
[0020] Furthermore, the step 1 specifically calibrates the TCP of the industrial robot at the grinding position of the grinding tool, and calibrates the TCP of the end contact wheel at the end of the end contact wheel; the step 2 measures the morphology of both sides of the weld by a morphology detection method based on contact sensing, and obtains two sets of position points L1[P 11 ,P 12 ,P 13 ,P 14 ,…,P 1n ] and L2[P 21 ,P 22 ,P 23 ,P24 ,…,P 2n ], and then calculate the points composed of the midpoints of the measurement base surfaces on both sides through step 3. In order to ensure the accuracy of the center point trajectory, the two sets of position points L1 and L2 are interpolated by the cubic B-spline difference method in step 3 to solve the curve trajectory points after interpolation, and the midpoint P is solved by the interpolation trajectory curve. m This is the grinding track.
[0021] Furthermore, in addition to the polishing trajectory position points composed of the midpoints obtained in the polishing trajectory processing step 3 in step 4, the polishing normal direction is also included, and the two together constitute the polishing trajectory file.
[0022] Furthermore, during the grinding process, by calculating the change Δd of the displacement sensor relative to the initial zero point f , you can get the grinding height that needs to be removed at this point.
[0023] Furthermore, efficient grinding with adaptive optimization of process parameters is achieved by dividing the weld excess height into two parts for removal. After detecting the weld height, the first part removal height and the second part removal height are calculated by the target excess height and the weld height. The first part removal height is the uneven part of the weld, and the second part removal height is the difference between the weld height and the first part. For the first part removal height, grinding parameters are generated through the process parameter library, and the grinding process parameters are adjusted to remove the uneven part and make it flat. After each grinding, the excess height after grinding is monitored in real time, and the grinding process parameters are adjusted according to the comparison between the monitored removal height and the target removal height of the grinding. For the second part removal height, appropriate grinding parameters are generated through the process parameter library. If the second part removal height is within the range of the process parameter library, the second part removal height can be removed by one grinding. If the second part removal height exceeds the maximum removal amount that can be removed by the grinding tool at one time, the second part removal height is divided into multiple removals.
[0024] The specific calculation method is as follows:
[0025]
[0026] In the formula, h r2 Indicates the height removed each time during the second grinding, expressed as h r2 Query the process parameter library to obtain the grinding parameters, h2 represents the removal height of the second part, h gmax It indicates the maximum amount that can be removed by the grinding tool at one time; the allowable error value σ of the grinding height is set, and whether the grinding operation is completed is judged based on whether the monitored height is within the range of δ±σ. If not, the grinding operation is performed again according to the adjusted grinding process parameters until the grinding operation is completed.
[0027] In a preferred embodiment of the present invention, the end of the EHA constant force floating end effector connected to the grinding tool or the end contact wheel can measure the surface morphology. However, since the initial grinding trajectory is obtained based on the morphology on both sides of the weld, the initial weld morphology detection uses the end contact wheel for morphology detection due to the interference problem of working conditions. During the grinding process, the removal height in the grinding depth direction can also be monitored in real time through the TCP position of the grinding tool. During the grinding process, the displacement sensor is calculated by calculating the change Δd relative to the initial zero point. f , the grinding height that needs to be removed at this point can be obtained. The grinding tool is a shutter blade or a diamond grinding wheel.
[0028] Compared with the prior art, the present invention has the following beneficial technical effects:
[0029] 1. The present invention uses a contact measurement method based on constant force control. The surface topography is obtained by contact sensing. The actual contact position of the robot can be detected in real time. The grinding removal height can be monitored in real time during the grinding process, thereby achieving high-precision surface measurement of the weld surface. The function of the end contact wheel is to facilitate the detection of the weld surface topography. Due to the high surface roughness and large height difference of the weld, the traditional probe-type topography detection structure is prone to collision and fracture problems, while the contact wheel form can effectively roll on the surface of the weld workpiece.
[0030] 2. The present invention uses automatic weld grinding without manual teaching, and does not require laser sensors and 3D cameras. The weld grinding trajectory generation method based on contact sensing can effectively generate weld grinding point information and robot posture based on the surfaces on both sides of the weld, thereby realizing low-cost robot automatic grinding.
[0031] 3. Aiming at the problem that the traditional method cannot adjust the process parameters adaptively, the present invention realizes adaptive process parameter optimization. The process parameter adjustment and adaptive process parameter optimization according to the weld detection height can realize the smooth surface after the weld is polished. After optimizing the parameters, the weld surface is smooth after polishing, and the polishing efficiency is higher.
[0032] The concept, specific structure and technical effects of the present invention will be further described below in conjunction with the accompanying drawings to fully understand the purpose, characteristics and effects of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 is a schematic diagram of a contact sensing-based robot weld grinding system according to a preferred embodiment of the present invention;
[0034] Figure 2 It is a schematic diagram of the internal structure of an EHA constant force floating end effector of a preferred embodiment of the present invention;
[0035] Figure 3 This is a working principle diagram of an EHA constant force floating end effector of a preferred embodiment of the present invention;
[0036] Figure 4 It is a schematic diagram of a topography detection principle based on contact sensing of a preferred embodiment of the present invention;
[0037] Figure 5 It is a flow chart of a weld grinding method based on contact sensing according to a preferred embodiment of the present invention;
[0038] Figure 6 It is a schematic diagram of weld grinding profile measurement of a preferred embodiment of the present invention;
[0039] Figure 7 It is a schematic diagram of the principle of generating a grinding track and a grinding direction according to a preferred embodiment of the present invention;
[0040] Figure 8 This is a schematic diagram of the real-time monitoring principle of the removal height in the grinding depth direction of a preferred embodiment of the present invention;
[0041] Fig. 9 It is a flow chart of an efficient polishing method for adaptively optimizing process parameters according to a preferred embodiment of the present invention;
[0042] Fig.10 It is a schematic diagram of weld grinding partial removal of a highly efficient grinding method with adaptive optimization of process parameters in a preferred embodiment of the present invention;
[0043] Among them: 1-EHA constant force floating end effector; 2-industrial robot; 3-industrial robot control cabinet; 4-industrial computer; 5-workpiece; 6-EHA constant force floating end control cabinet; 7-end contact wheel; 8-grinding tool; 9-EHA hydraulic servo motor; 10-force sensor; 11-connecting rod; 12-spindle motor; 13-floating part; 14-displacement sensor; 15-rotating spindle; 16-weld; 17-substrate; 18-weld measurement area. DETAILED DESCRIPTION
[0044] The following describes several preferred embodiments of the present invention with reference to the drawings in the specification, so that the technical content is clearer and easier to understand. The present invention can be embodied in many different forms of embodiments, and the protection scope of the present invention is not limited to the embodiments mentioned in the text.
[0045] In the drawings, components with the same structure are indicated by the same numerical reference numerals, and components with similar structures or functions are indicated by similar numerical reference numerals. The size and thickness of each component shown in the drawings are arbitrarily shown, and the present invention does not limit the size and thickness of each component. In order to make the illustration clearer, the thickness of the components is appropriately exaggerated in some places in the drawings.
[0046] like Figure 1 As shown, it is a schematic diagram of a robot weld grinding system based on tactile sensing. It includes an EHA constant force floating end effector 1, an EHA constant force floating end control cabinet 6, an industrial robot 2, an industrial robot control cabinet 3, an industrial computer 4 and a workbench. The end contact wheel 7 and the grinding tool 8 can be replaced and installed at the end of the EHA constant force floating end effector 1. Different from the pneumatic end effector used in conventional robot grinding, the EHA constant force floating end effector 1 is a constant force floating actuator driven by EHA hydraulic pressure. Its internal structure is as follows Figure 2 The main structure includes a spindle motor 12, an EHA hydraulic servo motor 9, a force sensor 10, a connecting rod 11, a displacement sensor 14, a rotating spindle 15, an end contact wheel 7, a grinding tool 8 and a floating part 13. Its working principle is as follows Figure 3 As shown. First, a constant output force is set. The force sensor 10 and the EHA hydraulic servo motor 9 form a constant force closed-loop control part to ensure a constant output force. The constant force drive is transmitted to the floating part 13 by the connecting rod 11, and the displacement sensor 14 can collect the displacement data of the floating part 13 in real time. The floating part 13 is connected to the rotating spindle 15 and is driven to rotate by the spindle motor 12. The EHA hydraulic servo motor 9 is connected to the floating part 13 through the connecting rod 11, and the position of the floating part 13 is adjusted through closed-loop control under the feedback of the force sensor 10 to achieve constant force output.
[0047] The schematic diagram of the topography detection method based on contact sensing is shown in Figure 4 The end detection wheel 7 is a metal roller with a radius of 5 mm. The initial position of the floating part 13 is set as the zero position. The EHA servo hydraulic motor is turned off, and the TCP is calibrated at the zero position, and the TCP is calibrated at the tip of the CNC probe.
[0048] When performing shape measurement, a constant output force is set at the end to collect the output position of the displacement sensor 14, thereby realizing the shape detection function of the end. Figure 4 As shown in the figure, the actual contact position of the end contact wheel is P[X, Y, Z], while the calibrated TCP position is P tcp [X tcp , Y tcp , Z tcp , A tcp , B tcp , C tcp], at this time, the change of the displacement sensor 14 relative to the initial zero point is Δd f , the position change of the floating part in the world coordinate system is written as [Δx f Δy f Δz f ]Then the relationship is as follows:
[0049]
[0050] The coordinate form written in the robot world coordinate system is as follows:
[0051]
[0052] According to the relevant principles of robotics, the following relationship can be obtained: (For the convenience of representation, α is A tcp , β is B tcp , γ is C tcp )
[0053]
[0054] Solving the matrix, we can obtain the following relationship:
[0055]
[0056] At this point, the actual contact position P of the robot end TCP has been calculated. The surface topography can be measured by calculating the measurement points composed of the actual contact position.
[0057] A weld grinding method based on contact sensing, the grinding method mainly includes weld morphology measurement and trajectory planning. The grinding process is as follows Figure 5 As shown, the following steps are included:
[0058] Step 1, calibrating the TCP of the industrial robot and the end contact wheel respectively;
[0059] Step 2: Measure the topography of the areas on both sides of the weld;
[0060] Step 3, calculating the midpoint of the topography measurement points;
[0061] Step 4: Generate a polishing track according to the target prediction;
[0062] Step 5: Perform grinding operation;
[0063] Step 6: Optimize grinding parameters in real time;
[0064] Step 7: Determine whether the grinding operation is completed. If not, jump to step 5. If completed, end the grinding.
[0065] Weld grinding usually removes the weld to the same height as the base surfaces on both sides, so the topography of the base surfaces on both sides of the weld needs to be measured first. Figure 6 shown.
[0066] Use end contact wheelsets Figure 6 The topography measurement area shown in the figure is measured, the TCP of the industrial robot is calibrated at the grinding position of the grinding tool, and the TCP of the end contact wheel is calibrated at the end of the end contact wheel. The topography of both sides of the weld is measured by the topography measurement principle based on contact sensing described above. After measuring on both sides, two sets of position points L1[P 11 ,P 12 ,P 13 ,P 14 ,…,P 1n ] and L2[P 21 ,P 22 ,P 23 ,P 24 ,…,P 2n ]. The grinding trajectory is composed of the midpoints of the two measuring base surfaces. However, since the points generated by the measurement are not uniform, directly solving the midpoint will affect the accuracy of the center point trajectory. Here, the cubic B-spline interpolation method is used to interpolate the two sets of position points L1 and L2. The cubic B-spline interpolation method is as follows:
[0067] The curve for cubic B-spline interpolation is defined as follows:
[0068]
[0069] Where C(u) is the B-spline interpolation function, L is the point measured on the base surface on both sides, that is, the control node of the B-spline interpolation, and N i,3 (u) is the basis function. The node vector U is defined as follows:
[0070] U={0,0,0,0,u4,u5,…,u n ,1,1,1,1} (6)
[0071] The cubic B-spline interpolation curve can be calculated by given control points and node vectors. The interpolation points of the node vectors can be expressed as follows:
[0072] u i+3 =u i+2 +|Δm i-1 |,i=1,2,3,…,n-3 (7)
[0073]
[0074] Where m i (i=0,1,…,n-2), where basis function Ni,3 (u) can be defined as follows:
[0075]
[0076] The fourth column in the above formula means that when the numerator or denominator is 0, the fraction is 0.
[0077] Substituting equation (8) into equation (5), we can obtain the cubic B-spline interpolation formula and the key contact point m i The relationship between (i=0,1,…,n-2) is as follows:
[0078]
[0079] Formula (11) can be rewritten into the following matrix form:
[0080]
[0081] As shown in formula (11), solving the cubic B-spline interpolation equation includes n+1 position control points and n-1 equations. Among them, the boundary condition is included, that is, the first control point and the last control point are consistent with the interpolation point, which can be expressed as the following formula:
[0082]
[0083] The interpolated curve trajectory points can be solved by equations (11) and (12). By calculating the node vectors at equal intervals, the interpolated trajectory curves C1 and C2 of the base surface measurement points on both sides can be obtained. Then the midpoint P can be solved by the interpolated trajectory curve. m This is the grinding track, such as Figure 7 (2) is shown in the figure, and the calculation method is as follows:
[0084]
[0085] For weld grinding, in addition to obtaining the grinding trajectory position points, it is also necessary to calculate the normal direction of the grinding. The schematic diagram of the grinding direction is as follows: Figure 7 (3) is shown in the figure. The calculation method of the grinding normal direction is as follows:
[0086] To calculate the normal vector direction of grinding, you must first establish a plane for the grinding target. Figure 7 As shown in the figure, the point set C1(u), C2(u) can be obtained by performing cubic B-spline interpolation on the weld substrate positions detected on both sides of the weld. Then, the point C1(u i ), C2(u i ) can generate a position vector The calculation formula is as follows:
[0087]
[0088] for The calculation of can be obtained through the first-order derivative of the cubic B-spline curve, and its calculation formula is as follows:
[0089]
[0090] So The calculation of is as follows:
[0091]
[0092] For the grinding track C that needs to retain the residual height δ δ The generation method is as follows:
[0093]
[0094] After the grinding trajectory points and grinding normal vectors are generated, they are arranged in the data format of [X, Y, Z, A, B, C] to generate a grinding trajectory file. During the grinding process, the removal height in the grinding depth direction can also be monitored in real time through the TCP position of the grinding tool. The principle is as follows Figure 8 shown.
[0095] During the grinding process, the displacement sensor is calculated by changing its initial zero point Δd f , you can get the grinding height that needs to be removed at this point.
[0096] On this basis, a method for efficient grinding of welds with adaptive optimization of process parameters is proposed. The weld excess height is divided into two parts for removal to achieve efficient grinding with adaptive optimization of process parameters. Fig. 9 , the schematic diagram of weld grinding and partial removal is as follows Fig.10 shown.
[0097] After detecting the weld height, the first part removal height and the second part removal height are calculated by the target residual height and the weld height. The first part removal height is the uneven part of the weld, and the second part removal height is the difference between the weld height and the first part. For the first part removal height, the grinding parameters are generated through the process parameter library, and the grinding process parameters are adjusted to remove the uneven parts and make them flat. After each grinding, the residual height after grinding is monitored in real time, and the grinding process parameters are adjusted according to the comparison between the monitored removal height and the target removal height of this grinding. For the second part removal height, appropriate grinding parameters are generated through the process parameter library. If the second part removal height is within the range of the process parameter library, the second part removal height can be removed through one grinding. If the second part removal height exceeds the maximum removal amount that the grinding tool can remove at one time, the second part removal height is divided into multiple removals. The specific calculation method is as follows:
[0098]
[0099] In the formula, h r2 Indicates the height removed each time during the second grinding, expressed as h r2 Query the process parameter library to obtain the grinding parameters, h2 represents the removal height of the second part, h gmax It indicates the maximum amount of removal that a grinding tool can remove in one time;
[0100] Set the allowable error value σ of the grinding height, and judge whether the grinding operation is completed based on whether the monitored height is within the range of δ±σ. If not, perform the grinding operation again according to the adjusted grinding process parameters until the grinding operation is completed. The grinding target height is achieved by dividing the weld into two parts.
[0101] The weld grinding method based on contact force position sensing proposed in the present invention does not require additional visual sensors and achieves high-precision weld grinding effects at a low cost. The method of obtaining weld morphology and performing grinding trajectory planning based on contact force position sensing obtains the weld surface morphology through the electro-hydraulic actuator EHA constant force floating end effector combined with the end contact wheel, and then generates the weld grinding trajectory through the trajectory planning algorithm. The present invention also proposes a robot absolute positioning correction method based on tactile servo, which corrects the robot position through tactile servo to improve the robot's positioning accuracy and achieve high-precision weld grinding.
[0102] The preferred specific embodiments of the present invention are described in detail above. It should be understood that ordinary technicians in the field can make many modifications and changes based on the concept of the present invention without creative work. Therefore, all technical solutions that can be obtained by technicians in the technical field based on the concept of the present invention through logical analysis, reasoning or limited experiments on the basis of the prior art should be within the scope of protection determined by the claims.
Claims
1. A contact sensing-based robot weld grinding system, characterized in that: It includes an EHA constant force floating end effector, an EHA constant force floating end control cabinet, an industrial robot, an industrial robot control cabinet, an industrial computer and a workbench, wherein the EHA constant force floating end effector is connected to the end of the industrial robot to grind the weld, the industrial computer sends instructions to the industrial robot control cabinet to control the movement of the industrial robot, the workpiece is placed on the workbench, and the EHA constant force floating end effector grinds the weld of the workpiece under the control of the industrial robot; the EHA constant force floating end effector is a constant force floating actuator driven by EHA hydraulics, the EHA constant force floating end effector is equipped with a force sensor, a displacement sensor, a grinding tool and an end contact wheel, the grinding tool and the end contact wheel are interchangeably installed on the end of the EHA constant force floating end effector, the end contact wheel is used when the morphology on both sides of the weld is initially measured, and is replaced by the grinding tool during the grinding process.
2. The contact sensing-based robot weld grinding system according to claim 1, characterized in that: The EHA constant force floating end effector includes a spindle motor, an EHA hydraulic servo motor, a force sensor, a connecting rod, a displacement sensor, a rotating spindle, a grinding tool, an end contact wheel and a floating part.
3. The contact sensing-based robot weld grinding system according to claim 2, characterized in that: First, a constant output force is set. The force sensor and the EHA hydraulic servo motor form a constant force closed-loop control part to ensure a constant output force. The connecting rod transmits the constant force drive to the floating part. The displacement sensor collects the displacement data of the floating part in real time. The floating part is connected to the rotating spindle and is driven to rotate by the spindle motor. The EHA hydraulic servo motor is connected to the floating part through the connecting rod. The position of the floating part is adjusted through closed-loop control under the feedback of the force sensor to achieve constant force output.
4. The contact sensing-based robot weld grinding system according to claim 2, characterized in that: The contact sensing-based morphology detection method is realized through the joint action of the force sensor, the displacement sensor and the end contact wheel; first, a constant output force is set by the EHA constant-force floating end control cabinet, and the constant output force is jointly guaranteed by the force sensor and the EHA hydraulic servo motor, and the end contact wheel rolls in contact with the surface of the weld workpiece; then the displacement sensor is used to collect the displacement information of the floating part, and then converted into the surface morphology information of the weld; during the grinding process, the grinding tool replaces the end contact wheel to contact the surface of the grinding area to realize real-time monitoring of the weld removal height.
5. The contact sensing-based robot weld grinding system according to claim 4, characterized in that: The end contact wheel is a metal roller with a radius of 5 mm. Before performing topography detection, it is necessary to set the initial position of the floating part to zero, turn off the enable of the EHA servo hydraulic motor, perform TCP calibration at zero position, and calibrate TCP at the end of the end contact wheel.
6. A robot weld grinding method based on contact sensing, characterized in that: Using the contact sensing-based robot weld grinding system as described in claims 1 to 5, the method comprises the following steps: Step 1, calibrating the TCP of the industrial robot and the end contact wheel respectively; Step 2: Measure the topography of the areas on both sides of the weld; Step 3, calculating the midpoint of the topography measurement points; Step 4: Generate a polishing track according to the target prediction; Step 5: Perform grinding operation; Step 6: Optimize grinding parameters in real time; Step 7: Determine whether the grinding operation is completed. If not, jump to step 5. If completed, end the grinding.
7. The contact sensing-based robot weld grinding method according to claim 6, characterized in that: The step 1 specifically includes calibrating the TCP of the industrial robot at the grinding position of the grinding tool and calibrating the TCP of the end contact wheel at the end of the end contact wheel; the step 2 includes measuring the morphology of both sides of the weld by a morphology detection method based on contact sensing, and obtaining two sets of position points L1[P 11 ,P 12 ,P 13 ,P 14 ,…,P 1n ] and L2[P 21 ,P 22 ,P 23 ,P 24 ,…,P 2n ], and then calculate the points composed of the midpoints of the measurement base surfaces on both sides through step 3. In order to ensure the accuracy of the center point trajectory, the two sets of position points L1 and L2 are interpolated by the cubic B-spline difference method in step 3 to solve the curve trajectory points after interpolation, and the midpoint P is solved by the interpolation trajectory curve. m This is the grinding track.
8. The contact sensing-based robot weld grinding method according to claim 6, characterized in that: The polishing trajectory in step 4 includes not only the polishing trajectory position points composed of the midpoints obtained in step 3, but also the normal direction of polishing, and the two together constitute the polishing trajectory file.
9. The contact sensing-based robot weld grinding method according to claim 6, characterized in that: During the grinding process, the displacement sensor is changed relative to the initial zero point by calculating Δd f , you can get the grinding height that needs to be removed at this point.
10. The contact sensing-based robot weld grinding method according to claim 9, characterized in that: By dividing the weld excess height into two parts for removal, efficient grinding with adaptive optimization of process parameters is achieved. After detecting the weld height, the first part removal height and the second part removal height are calculated by the target excess height and the weld height. The first part removal height is the uneven part of the weld, and the second part removal height is the difference between the weld height and the first part. For the first part removal height, grinding parameters are generated through the process parameter library, and the grinding process parameters are adjusted to remove the uneven part and make it flat. After each grinding, the excess height after grinding is monitored in real time, and the grinding process parameters are adjusted according to the comparison between the monitored removal height and the target removal height of the grinding. For the second part removal height, appropriate grinding parameters are generated through the process parameter library. If the second part removal height is within the range of the process parameter library, the second part removal height can be removed by one grinding. If the second part removal height exceeds the maximum removal amount that can be removed by the grinding tool at one time, the second part removal height is divided into multiple removals. The specific calculation method is as follows: In the formula, h r2 Indicates the height removed each time during the second grinding, expressed as h r2 Query the process parameter library to obtain the grinding parameters, h2 represents the removal height of the second part, h gmax It indicates the maximum amount that can be removed by the grinding tool at one time; the allowable error value σ of the grinding height is set, and whether the grinding operation is completed is judged based on whether the monitored height is within the range of δ±σ. If not, the grinding operation is performed again according to the adjusted grinding process parameters until the grinding operation is completed.
Citation Information
Patent Citations
Blade air inlet and exhaust edge three-dimensional non-contact measuring device
CN105180834A
Robot-based vehicle body weld joint polishing allowance classification method and equipment
CN112633235A
Force-controlled floating milling and grinding integrated device and operation method thereof
CN113714820A
Data processing and control method for automatic steel pipe welding seam polishing system
CN113941919A
Robot self-adaptive curved surface tracking constant-force grinding and polishing method and system
CN114454060A
Cited By
White vehicle body welding seam polishing system and method
CN119609820A
