A centering device and a centering method for a robot to grip a profile

Abstract

The present application relates to the field of profile grabbing and transferring, and particularly relates to a centering device and a centering method for profile grabbing by a mechanical hand, comprising a first mounting plate, the first mounting plate is fixedly connected with a mechanical arm mechanism, a set of positioning mechanisms are fixedly connected with the first mounting plate at both ends respectively, the positioning mechanisms comprise a fixing frame, a laser radar and a laser ranging device are fixedly connected with the fixing frame respectively, the laser radar forms a fan-shaped scanning area downward, the laser ranging device detects vertically downward, the laser radar and the laser ranging device are electrically connected with a controller of the mechanical arm.

Description

Technical Field

[0001] This invention relates to the field of profile gripping and transfer, and specifically to a centering device and method for a robotic arm to grip profiles. Background Technology

[0002] When stacked I-beams or profiles need to be moved, the gaps between the profiles are small. To grab the material from the stacked profiles, the gripper needs to be inserted into the gaps between the materials. A typical mechanical gripper has at least two gripping points. During the gripping process, the line connecting the two gripping points needs to be parallel to the length of the profile, and the two grippers at each gripping point should be symmetrically distributed on both sides of the profile. Therefore, how to adjust the posture of the robotic arm to achieve a suitable gripping position becomes a problem that needs to be solved. Summary of the Invention

[0003] To address the aforementioned problems, this invention provides a centering device and method for a robotic arm to grasp profiles.

[0004] The objective of this invention is achieved in the following manner: a centering device and method for a robotic arm to grasp profiles, comprising a first mounting plate, the first mounting plate being fixedly connected to a robotic arm mechanism, and a set of positioning mechanisms being fixedly connected to both ends of the first mounting plate, the positioning mechanism comprising a fixed frame, on which a laser radar and a laser ranging device are fixedly connected respectively, the laser radar forming a fan-shaped scanning area downwards, the laser ranging device detecting vertically downwards, and the laser radar and the laser ranging device being electrically connected to the controller of the robotic arm;

[0005] The centering method includes the following steps:

[0006] S1. Set the radar scanning range: Determine the numerical range of the cross-sectional area of ​​the profile to be scanned. Calculate the radar scanning range based on the maximum cross-sectional area and the installation height of the laser radar, and take n values ​​from it as valid values.

[0007] S2. Read and filter information: Read the raw scanning information of the LiDAR and store it in listOri; then perform mean filtering and Kalman filtering on it, and store the new data in OldList;

[0008] S3. Obtain the cross-sectional area information of the support column:

[0009] (1) Obtain the minimum value minValue and its index value minIndex of the distance from the scan to the profile;

[0010] (2) Set a threshold value based on the minimum value minValue, retain all data within the threshold value range, set values ​​outside the range to zero, and store the acquired data in a List;

[0011] (3) Obtain the effective value of the cross-sectional area: Based on the minimum value minValue and List, find multiple consecutive non-zero values ​​before and after the minimum value, and store the obtained array and index into new arrays newList and newIndex respectively;

[0012] (4) Smoothing of valid values: Non-zero values ​​in newList are processed, the average of the previous and next values ​​is taken, and the newly acquired data is stored in newData;

[0013] S4. Establish a Cartesian coordinate system that conforms to the right-hand rule and is consistent with the robot arm's coordinate system;

[0014] S5. Calculate the included angle and translation distance between the two radars: Based on the radar characteristics, find the index min90Index and value min90Value corresponding to 90°. Based on minValue and minIndex, calculate the included angle agnle and translation distance dis between the two radars.

[0015] S6. Obtain the distance measured by the laser rangefinder;

[0016] S7. Based on the above values ​​and the geometric relationship of the alignment device installation, calculate the absolute error values ​​xValue, yValue, zValue, xRotate, yRotate, and zRotate with the robotic arm, and send them to the controller of the robotic arm. Use the end-effector coordinate system to control the movement of the robotic arm to achieve the purpose of alignment and grasping.

[0017] Furthermore, in S3, the radar scanning characteristics are utilized, as the distance to the cross-sectional area of ​​the support column is relatively short, and abrupt changes occur at the edge. Using this characteristic, the minimum value minValue and its index minIndex are found in OldList.

[0018] Furthermore, in step S3, a threshold of 0.15 is set. Based on the minimum value minValue, values ​​outside the threshold range are zeroed out to obtain a new data List. The List is traversed, and multiple consecutive non-zero values ​​before and after the minimum value are taken until three consecutive zero values ​​appear. The effective cross-sectional areas newList and newIndex are obtained and smoothed to obtain newdData.

[0019] Furthermore, the filtered data data1 and data2 scanned by the lidars on both sides, as well as the effective cross-sectional area data newdData1 and newdData2, can be obtained respectively.

[0020] Furthermore, in S5, the included angle is calculated as follows: angle = math.fabs(min90Index-minIndex) * 0.333 * math.PI / 180; the translation distance is calculated as follows: dis = math.sqrt(min90Value*min90Value+minValue*minValue-2*min90Value*minValue*math.cos(angle)).

[0021] Furthermore, the required angle and translation distance for the two lidars are obtained respectively, and are defined as: dis1, mid90Index1, mid90Value1, midIndex1, midValue1;

[0022] dis2,mid90Index2,mid90Value2,midIndex2,midValue2.

[0023] Furthermore, zValue = -(min(data1,data2)), xValue = 0.

[0024] Furthermore, the installation distance between the two lidars is a constant L. The zRotate value and yValue value are calculated in four cases:

[0025] (1) The radar line must intersect the center line of the support column, and the Z rotation must be positive at this time;

[0026] zRotate=math.atan(dis1+dis2) / L*180 / math.PI;

[0027] yValue=L / 2*(dis1-dis2) / math.sqrt(L*L+math.pow(dis1+dis2,2));

[0028] (2) Radar 1 is biased at a 45-degree angle, and radar 2 is biased at a 135-degree angle. The radar connection line must not intersect the center line of the support. The rotation of z needs to be determined according to the magnitude of dis1 and dis2.

[0029] temp = math.atan(math.fabs(dis1-dis2) / L)

[0030] inter1 = midIndex1 - mid90Index1

[0031] inter2 = mid90Index2 - midIndex2

[0032] if inter1 < 0 or inter2 < 0:

[0033] angle=angle

[0034] elif inter1 <inter2:

[0035] angle=-angle

[0036] zRotate = temp * 180 / math.PI

[0037] yValue = L / 2*(dis1+dis2) / math.sqrt(L*L+math.pow(dis1+dis2,2));

[0038] (3) Radar 1 is deflected at 135 degrees, radar 2 is deflected at 135 degrees, the radar connection line must not intersect the center line of the support, and Z needs to be rotated clockwise.

[0039] temp= math.atan((dis1 + dis2) / L);

[0040] zRotate = - temp * 180 / math.PI;

[0041] yValue = L / 2 * (dis1 - dis2) / math.sqrt(L * L + math.pow(dis1+dis2,2));

[0042] (4) Radar 1 is deflected at 135 degrees, Radar 2 is deflected at 45 degrees, and the radar line connecting them must not intersect the centerline of the support pillar. Therefore:

[0043] temp= math.tan(math.fabs(dis1 - dis2) / L)

[0044] inter1 = mid90Index1 - midIndex1

[0045] inter2 = midIndex2 - mid90Index2

[0046] if inter1 < 0 or inter2 < 0:

[0047] pass

[0048] elif inter2 <inter1:

[0049] angle=-angle

[0050] zRotate = temp * 180 / math.PI

[0051] yValue = - L / 2 * (dis1 + dis2) / math.sqrt(L * L + math.pow(dis1 +dis2, 2)).

[0052] Furthermore, based on data from two laser rangefinders installed on the same side:

[0053] The measurement distances of the two laser rangefinders are L1 and L2, respectively; the distance between the emission points of the two rangefinder installation positions is S; then: X rotation angle xRotate=-math.atan((L1-L2) / S)*180 / math.PI;

[0054] Y rotation angle: yRotate = math.asin((mid90Value2 - mid90Value1) / L) * 180 / math.PI.

[0055] Compared to existing technologies, this invention, when gripping a profile, uses radars located on both sides of the centering device to scan downwards. Through radar data analysis, the relative coordinates between the radar position and the profile can be determined. By comparing two sets of radar data, the deviation between the gripping device and the profile can be identified, and the relevant data is transmitted to the robotic arm. The robotic arm then drives the gripping device to adjust its posture, ensuring that the line connecting the two grippers of the centering device is parallel to the long side of the profile. The laser rangefinder detects the vertical distance between the gripping device and the profile and sends the relevant data to the robotic arm. The robotic arm then moves the gripping device downwards, inserting the grippers into the gaps in the profile and merging the grippers to complete the profile gripping function. Attached Figure Description

[0056] Figure 1 This is one of the schematic diagrams of a robotic arm structure;

[0057] Figure 2 This is a cross-sectional view of the rotary lifting cylinder;

[0058] Figure 3 This is a schematic diagram of the guide groove on the piston rod;

[0059] Figure 4 This is the second schematic diagram of the robotic arm structure;

[0060] Figure 5 This is a schematic diagram of the deflection mechanism;

[0061] Figure 6 This is a schematic diagram of the translation mechanism;

[0062] Figure 7 This is a sectional view of the translation mechanism;

[0063] Figure 8 This is a schematic diagram of the centering device;

[0064] Figure 9 It is a radar scanning characteristic map;

[0065] Figure 10 This is a schematic diagram of establishing a coordinate system for the centering device;

[0066] Figure 11 This is a schematic diagram showing the relationship between the radar angle and the translation distance.

[0067] The components include: 1. Connecting frame; 11. First mounting plate; 111. Pad; 12. Slide rail; 2. Rotary lifting cylinder; 21. Cylinder body; 211. Guide pin; 212. Oil port; 22. Piston rod; 221. Guide groove; 23. Front oil plug; 24. Rear oil plug; 25. Gripper plate; 3. Second mounting frame; 31. Slider; 32. Pull-in cylinder; 4. Third mounting plate; 5. Fixing frame; 51. Deflection cylinder; 6. Hinge shaft; 7. Positioning mechanism; 71. Fixing frame; 72. LiDAR; 721. LiDAR scanning sector; 73. Laser rangefinder; 8. Connecting flange. Detailed Implementation

[0068] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0069] In this invention, unless otherwise explicitly specified and limited, the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and simplifying the description, and are not intended to indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention.

[0070] As attached Figure 1As shown, an I-beam gripping robot includes a posture adjustment mechanism and a clamping mechanism connected to the posture adjustment mechanism. The clamping mechanism includes a first mounting plate 11, with a pair of rotary lifting cylinders 2 at each of the front and rear ends of the first mounting plate 11. The top of the piston rod 22 of the rotary lifting cylinder 2 is fixedly connected to a gripper plate 25, so that when the piston rod 22 extends, the gripper plates 25 of each pair of rotary lifting cylinders 2 extend parallel downwards from both sides of the I-beam. When the piston rod 22 retracts, the gripper plates 25 of each pair of rotary lifting cylinders 2 rotate toward the direction of the I-beam, thereby limiting and clamping the top plate of the I-beam.

[0071] Compared with the prior art, the gripper of the present invention does not produce horizontal displacement, and the gripper plate 25 only rotates. In actual working conditions, only a gap of the width of one gripper plate is needed between the I-beams so that the gripper plate 25 can pass through both sides of the I-beam, thereby increasing the placement density and indirectly improving production efficiency.

[0072] The rotary lifting cylinder 2 refers to a cylinder that rotates relative to the cylinder body during the extension and retraction of the piston rod. Existing technologies such as CN202210790853.4 (a rotary cylinder) or CN201910861681.3 (a hydraulic cylinder with a rotatable piston rod) can be used, or the cylinder structure proposed in this application can be used.

[0073] As attached Figure 2-3 As shown, the rotary lifting cylinder 2 includes a cylinder body 21, a piston rod 22 is disposed inside the cylinder body 21, and a front oil plug 23 and a rear oil plug 24 are fixedly connected to the front and rear ends of the cylinder body 21, respectively. A guide pin 211 is fixedly connected to the side of the cylinder body 21, and the head end of the guide pin 211 passes through the cylinder body 21 and extends into the cylinder body 21. A spiral guide groove 221 is disposed on the side of the piston rod 22. The guide groove 221 cooperates with the head end of the guide pin 211, so that the piston rod 22 rotates relative to the cylinder body 21 when it extends and retracts. The spiral direction of the guide groove 221 on the piston rod 22 of each pair of rotary lifting cylinders 2 is opposite.

[0074] This structure can be directly improved from an existing hydraulic cylinder. The position of the oil port 212 and the arrangement of the front and rear oil plugs do not need to be changed. A through hole is provided on the side of the cylinder body 21, through which the guide pin 211 passes. A sealing ring should be provided between the through hole and the guide pin 211. A guide groove 221 is machined on the side of the piston rod 22. Preferably, the shape of the guide groove 221 is as follows: Figure 3 As shown, it includes a vertical section and a spiral section. The spiral section corresponds to the rotation (preferably 90°) of the gripper plate 25 from extension to retraction. When the vertical section corresponds to the gripper plate 25 retracting to the clamping position, it further retracts to clamp the top plate of the I-beam.

[0075] like Figure 4As shown, a pad 111 is provided at the bottom of the first mounting plate 11 where it is clamped to the gripper plate 25. The pad 111 should be a soft elastic pad. Preferably, a soft pad is also provided at the top of the gripper plate 25.

[0076] The attitude adjustment mechanism can use the attitude adjustment mechanism of the upper part of the pole-mounted robot arm proposed by the applicant on August 28, 2020 (CN202021847163.0). This application can improve only the gripper structure based on the existing structure, or it can use some of the improvements proposed in this application.

[0077] like Figure 6-7 As shown, the attitude adjustment mechanism includes a translation mechanism and a deflection mechanism. The translation mechanism includes a connecting frame 1, with a first mounting plate 11 fixedly connected to the bottom of the connecting frame 1. L-shaped slide rails 12 are fixedly connected to both sides of the upper part of the connecting frame 1. A second mounting frame 3 is provided on the upper part of the connecting frame 1. Inverted L-shaped sliders 31 are fixedly connected to both sides of the lower part of the second mounting frame 3. The L-shaped slide rails 12 and the inverted L-shaped sliders 31 are interlocked to make the connecting frame 1 and the second mounting frame 3 slide together. A first hinge is fixedly connected to the top of the connecting frame 1, which hinges to the output end of the pull-in cylinder 32. A second hinge is fixedly connected to the bottom of the second mounting frame 3, which hinges to the cylinder body of the pull-in cylinder 32. This allows the relative sliding between the connecting frame 1 and the second mounting frame 3 to be controlled by the extension and retraction of the pull-in cylinder 32.

[0078] like Figure 5 As shown, a third mounting plate 4 is provided above the second mounting bracket 3. A third hinge is fixedly connected to the top of the second mounting bracket 3, and a fourth hinge is fixedly connected to the bottom of the third mounting plate 4. The third hinge and the fourth hinge are hinged together by a hinge shaft 6. The deflection mechanism includes a fixed frame 5. The fixed frame 5 is fixedly connected to the side of the third mounting plate 4. A deflection cylinder 51 is hinged to the cylinder body on each side of the fixed frame 5. The output end of the deflection cylinder 51 is hinged to a fifth hinge. The fifth hinge is fixedly connected to the second mounting bracket 3.

[0079] In detail, the third and fourth hinges are two corresponding sets. The fixing frame 5 is set on one side of the axis of the hinge shaft 6, and the deflection cylinders 51 are respectively set on both sides of the axis of the hinge shaft 6 to cooperate in realizing the deflection in two directions between the second mounting frame 3 and the third mounting plate 4. The third mounting plate 4 is provided with mounting holes for installation with the robotic arm.

[0080] As shown in the figure, reinforcing ribs are provided between the second mounting bracket 3 and the third hinge, between the third mounting plate 4 and the fourth hinge, and between the fixing bracket 5 and the third mounting plate 4 as needed.

[0081] When gripping the support column, the robotic arm of this application, in conjunction with the robotic arm, can perform multiple actions to adjust the posture of the gripper, so that the gripper plate 25 is positioned precisely at the gap between the two support columns. Under the action of the robotic arm, it moves downward, causing the pad 111 to fit tightly against the support column. The four gripper plates 25 first rotate, causing them to swing to the inside of the support column. Simultaneously, the grippers, driven by the rotary lifting cylinder 2, perform a lifting action, thereby pressing the support column firmly onto the pad 111, completing the gripping action of the robotic arm.

[0082] A centering device for a robotic arm to grasp profiles, such as Figure 8 As shown, the system includes a first mounting plate 11, which is fixedly connected to a robotic arm mechanism. A set of positioning mechanisms 7 are fixedly connected to both ends of the first mounting plate 11. The positioning mechanism 7 includes a fixing frame 71, on which a laser radar 72 and a laser ranging device 73 are fixedly connected. The laser radar 72 forms a fan-shaped scanning area 721 downwards, and the laser ranging device 73 detects vertically downwards. The laser radar 72 and the laser ranging device 73 are electrically connected to the controller of the robotic arm.

[0083] Preferably, the centering device can be an independent structure, with a connecting flange 8 provided in the middle of the first mounting plate 11 for assembly with different robotic arms, but preferably the robotic arm described above.

[0084] The neutralization method includes the following steps:

[0085] S1. Set the radar scanning range: The cross-sectional area of ​​the support column is generally in the range of (240mm-300mm). Calculate the radar scanning range based on the maximum cross-sectional area and the radar installation height. As an example, the scanning angle is α=(arctan(150 / 234))*2, α=65.32°, the radar resolution is 0.33°, and the scanning frequency is 15Hz. Within this range, 197 values ​​can be taken as valid values.

[0086] S2. Read and filter information: Read the raw scanning information of the radar and store it in listOri; then perform mean filtering and Kalman filtering on it, and store the new data in OldList.

[0087] S3. Obtain the cross-sectional area information of the support column:

[0088] (1) Obtain the minimum value and its index: such as Figure 9 As shown, based on the characteristics of radar scanning, the distance to the cross-sectional area of ​​the support column is relatively short, and abrupt changes will occur at the edge. Using this characteristic, the minimum value minValue and its index minIndex can be found in OldList.

[0089] (2) Set a threshold based on the minimum value minValue, retain all data within the threshold range, set values ​​outside the range to zero, and store the acquired data in a List;

[0090] (3) Obtain the effective cross-sectional area: Set the threshold to 0.15, and zero out the values ​​outside the threshold range to obtain a new data List; traverse the List, take multiple consecutive non-zero values ​​before and after the minimum value until three consecutive zero values ​​appear, and obtain the effective cross-sectional area newList and newIndex;

[0091] (4) Smoothing of valid values: Non-zero values ​​in newList are processed, the average of the previous and next values ​​is taken, and the newly acquired data is stored in newData;

[0092] The effective values ​​are smoothed by zeroing them: based on the minimum value, find the three consecutive non-zero values ​​before and after it, and store the obtained arrays and indices into new arrays, newList and newIndex respectively; since there will be a few serialized 0 values ​​in the new array, they need to be non-zeroed. Traverse newList, and use the average of the previous and next values ​​at the zero value as the current new value, and store it in newdData. According to the above principle, the filtered data data1 and data2 scanned by the two lidars (72) and the effective cross-sectional area data newdData1 and newdData2 can be obtained respectively.

[0093] S4. Establish a Cartesian coordinate system that conforms to the right-hand rule and is consistent with the robot arm's coordinate system, such as... Figure 10 As shown.

[0094] S5. Calculate the relationship between the 90-degree extreme point and the minimum extreme point: Based on the radar characteristics, find the index min90Index and the extreme value min90Value corresponding to the 90-degree value, such as... Figure 11 As shown, the included angle can be calculated.

[0095] angle=math.fabs(min90Index-minIndex)*0.333*math.PI / 180;

[0096] According to the Law of Cosines:

[0097] math.sqrt(min90Value*min90Value+minValue*minValue-2*min90Value*minValue*math.cos(angle)).

[0098] S6. Based on the above principle, the required angle and translation distance between the two radars can be calculated respectively, and defined as follows:

[0099] dis1,mid90Index1,mid90Value1,midIndex1,midValue1;

[0100] dis2,mid90Index2,mid90Value2,midIndex2,midValue2;

[0101] According to the coordinate system, the Z-axis points upwards, so the Z-value must be negative.

[0102] zValue = -(min(data1,data2));

[0103] xValue=0.

[0104] S7. Given that the radar installation distance is a constant L; with counterclockwise rotation as positive and clockwise rotation as negative; calculate other values ​​separately: divide them into four cases:

[0105] (1) The radar line must intersect the center line of the support column, and the Z rotation must be positive at this time;

[0106] zRotate=math.atan(dis1+dis2) / L*180 / math.PI;

[0107] yValue=L / 2*(dis1-dis2) / math.sqrt(L*L+math.pow(dis1+dis2,2));

[0108] (2) Radar 1 is biased at a 45-degree angle, and radar 2 is biased at a 135-degree angle. The radar connection line must not intersect the center line of the support. The rotation of z needs to be determined according to the magnitude of dis1 and dis2.

[0109] temp = math.atan(math.fabs(dis1-dis2) / L)

[0110] inter1 = midIndex1 - mid90Index1

[0111] inter2 = mid90Index2 - midIndex2

[0112] if inter1 < 0 or inter2 < 0:

[0113] angle=angle

[0114] elif inter1 <inter2:

[0115] angle=-angle

[0116] zRotate = temp * 180 / math.PI

[0117] yValue = L / 2*(dis1+dis2) / math.sqrt(L*L+math.pow(dis1+dis2,2));

[0118] (3) Radar 1 is deflected at 135 degrees, radar 2 is deflected at 135 degrees, the radar connection line must not intersect the center line of the support, and Z needs to be rotated clockwise.

[0119] temp= math.atan((dis1 + dis2) / L);

[0120] zRotate = - temp * 180 / math.PI;

[0121] yValue = L / 2 * (dis1 - dis2) / math.sqrt(L * L + math.pow(dis1+dis2,2));

[0122] (4) Radar 1 is deflected at 135 degrees, Radar 2 is deflected at 45 degrees, and the radar line connecting them must not intersect the centerline of the support pillar. Therefore:

[0123] temp= math.tan(math.fabs(dis1 - dis2) / L)

[0124] inter1 = mid90Index1 - midIndex1

[0125] inter2 = midIndex2 - mid90Index2

[0126] if inter1 < 0 or inter2 < 0:

[0127] pass

[0128] elif inter2 <inter1:

[0129] angle=-angle

[0130] zRotate = temp * 180 / math.PI

[0131] yValue = - L / 2 * (dis1 + dis2) / math.sqrt(L * L + math.pow(dis1 +dis2, 2)).

[0132] S8. Calculate the X rotation angle: Calculate based on data from two laser rangefinders installed on the same side.

[0133] The measuring distance of rangefinder 1 is L1; the measuring distance of rangefinder 2 is L2; ​​the distance between the launching points of the two rangefinder installation positions is S; then:

[0134] xRotate=-math.atan((L1-L2) / S)*180 / math.PI.

[0135] S9. Calculate the Y rotation angle:

[0136] yRotate = math.asin((mid90Value2 - mid90Value1) / L) * 180 / math.PI.

[0137] S10. Send xValue, yValue, zValue, xRotate, yRotate, and zRotate to the robotic arm controller and use the end-effector coordinate system to control the movement of the robotic arm.

[0138] When a profile needs to be gripped, radars located on both sides of the centering device scan downwards in a fan-shaped area. When the fan-shaped area detects the profile, the size, dimensions, and gaps between profiles are determined through reflection analysis. The radars, mounted on the centering device, provide the relative coordinates between the radar position and the profile through radar data analysis. LiDAR is installed at both ends of the centering device. By comparing data from the two sets of radars, the deviation between the gripping device and the profile is identified, and the relevant data is transmitted to the robotic arm, which then adjusts the gripping device's posture. Real-time comparison of radar data during adjustment ensures that the line connecting the two grippers of the centering device is parallel to the long side of the profile. A set of grippers can be symmetrically distributed on both sides of the profile, positioning the gripping device directly above it. A laser rangefinder detects the vertical distance between the gripping device and the profile and sends the data to the robotic arm. The robotic arm then moves the gripping device downwards, inserting the grippers into the gaps in the profile and closing them to complete the gripping function.

[0139] The above description is only a preferred embodiment of the present invention. It should be noted that those skilled in the art can make several changes and improvements without departing from the overall concept of the present invention, and these should also be considered within the scope of protection of the present invention.

Claims

1. A centering method applied to a centering device for a profile gripped by a robot, said centering device comprising a first mounting plate (11) fixedly connected to a robot arm mechanism, characterized in that, The first mounting plate (11) is fixedly connected with a set of positioning mechanisms (7) at both ends, the positioning mechanisms (7) include a fixed frame (71), the fixed frame (71) is fixedly connected with a laser radar (72) and a laser ranging device (73) respectively, the laser radar (72) forms a fan-shaped scanning area downwardly, the laser ranging device (73) detects vertically downwardly, the laser radar (72) and the laser ranging device (73) are electrically connected with the controller of the mechanical arm; The centering method comprises the following steps: S1, setting the radar scanning range: determining the numerical range of the cross-sectional area of the profile to be grabbed, calculating the scanning range of the laser radar (72) according to the maximum cross-sectional area and the installation height of the laser radar (72); S2, reading information and filtering: reading the original scanning information of the laser radar (72) and storing it in listOri; then performing mean filtering and Kalman filtering on it, and storing the new data in OldList; S3, obtaining the cross-sectional area information of the profile: (1) obtaining the minimum value minValue of the distance scanned by the laser radar (72) to the profile and its index value minIndex; (2) setting a threshold value, taking the minimum value minValue as the basis, retaining all data within the floating range of the threshold value, setting the values outside the range to zero, and storing the obtained data in List; (3) obtaining the effective value of the cross-sectional area: taking the minimum value minValue and List as the basis, finding a plurality of non-zero value data before and after the minimum value respectively, and storing the obtained array and index in the new array newList and newIndex respectively; (4) smoothing the effective values: performing non-zero processing on the zero values in newList, taking the mean of the previous and next values, and storing the newly obtained data in newData; S4, establishing a Cartesian coordinate system in accordance with the right-hand rule and keeping consistent with the coordinate system of the mechanical arm; S5, calculating the included angle and translation distance of the two radars: according to the characteristics of the radar, finding the index min90Index and the value min90Value corresponding to 90°, and calculating the included angle agnle and the translation distance dis of the two radars according to minValue and minIndex; S6, obtaining the measurement distance of the laser ranging device; S7, according to the above values and the geometric relationship of the centering device installation, calculating the absolute error values xValue, yValue, zValue, xRotate, yRotate, zRotate of the mechanical arm, sending them to the controller of the mechanical arm, and controlling the movement of the mechanical arm by using the end coordinate system to achieve the purpose of centering and grabbing.

2. A centering method for a centering device applied to a profile to be gripped by a robot, according to claim 1, characterized in that, In S3, the scanning characteristics of the radar are used to scan the cross-sectional area of the profile, and a sudden change occurs at the edge. By using this characteristic, the minimum value minValue and its index minIndex are found in OldList.

3. A centering method for a centering device for mechanical hand gripping profiles according to claim 2, characterized in that, In S3, a threshold value of 0.15 is set, and all values outside the threshold value range are set to zero based on the minimum value minValue, obtaining a new array newList. Traverse the new list, take the minimum value before and after a plurality of non-zero values, until three zero values appear in succession, obtain the effective array newData, newIndex, and perform smoothing processing to obtain newData.

4. A centering method for a centering device applied to a profile to be gripped by a robot according to claim 3, characterized in that, Respectively, the filtered data scanned by the two laser radars (72) data1, data2, and the effective cross-sectional area data newData1, newData2 are obtained.

5. A centering method for a centering device applied to a profile to be gripped by a robot, according to claim 4, characterized in that, In S5, the included angle angle=math.fabs(min90Index-minIndex)*0.333*math.PI / 180; the translation distance dis=math.sqrt(min90Value*min90Value+minValue*minValue-2*min90Value*minValue*math.cos(angle)).

6. A centering method for a centering device applied to a profile to be gripped by a robot, according to claim 5, characterized in that, Respectively, the required included angle and translation distance of the two laser radars (72) are obtained, respectively defined as: dis1, mid90Index1, mid90Value1, midIndex1, midValue1; dis2, mid90Index2, mid90Value2, midIndex2, midValue2.

7. A centering method for a centering device applied to a profile to be gripped by a robot, according to claim 6, characterized in that, The zValue=-(min(data1, data2)), xValue=0.

8. A centering method for a centering device applied to a profile to be gripped by a robot, according to claim 6, characterized in that, The installation distance of the two laser radars is a constant L, and the zRotate value and yValue value are calculated, which is divided into four cases: (1) The radar connecting line intersects the profile center line, and the Z rotation is positive at this time; zRotate=math.atan(dis1+dis2) / L*180 / math.PI; yValue=L / 2*(dis1-dis2) / math.sqrt(L*L+math.pow(dis1+dis2,2)); (2) Radar 1 deviates to the direction of 45 degrees, radar 2 deviates to the direction of 135 degrees, and the radar connecting line does not intersect the profile center line, and the direction of z rotation needs to be determined according to the size of dis1 and dis2; temp = math.atan(math.fabs(dis1-dis2) / L) inter1=midIndex1-mid90Index1 inter2=mid90Index2-midIndex2 if inter1<0 or inter2<0: angle=angle elif inter1<inter2: angle=-angle zRotate = temp * 180 / math.PI yValue = L / 2*(dis1+dis2) / math.sqrt(L*L+math.pow(dis1+dis2,2)); (3) Radar 1 is inclined to 135-degree angle direction, radar 2 is inclined to 135-degree angle direction, radar connecting line does not cross profile center line, Z rotates clockwise; temp = math.atan((dis1 + dis2) / L); zRotate = - temp * 180 / math.PI; yValue = L / 2 * (dis1 - dis2) / math.sqrt(L * L + math.pow(dis1+dis2,2)); (4) Radar 1 is inclined to 135-degree angle direction, radar 2 is inclined to 45-degree angle direction, radar connecting line does not cross profile center line, then: temp = math.tan(math.fabs(dis1 - dis2) / L) inter1 = mid90Index1-midIndex1 inter2 = midIndex2-mid90Index2 if inter1<0 or inter2<0: pass elif inter2<inter1: angle = -angle zRotate = temp* 180 / math.PI yValue = - L / 2 * (dis1 + dis2) / math.sqrt(L * L + math.pow(dis1 +dis2, 2)).

9. The centering method of the centering device for the mechanical hand to grab the profile according to claim 6, characterized in that, The measuring distances of the two laser ranging devices (73) are L1 and L2 respectively; the distance between the emitting points of the installation positions of the two laser ranging devices (73) is S; then: xRotate = -math.atan((L1-L2) / S)*180 / math.PI; yRotate = math.asin((mid90Value2 - mid90Value1) / L) * 180 / math.PI.

Images