Automatic tightening control system based on binocular vision recognition and positioning

Abstract

The present invention discloses an automatic tightening control system based on binocular vision recognition and positioning, comprising a main control device, a camera adjustment device, an automatic assembly mechanism, an electric tightening device, an in-place detection sensor, a distance detection sensor, and a binocular vision intelligent camera group. The main control device detects the start signal output by the vehicle body in-place detection sensor, thereby controlling and collecting the distance data of the fixed object output by the vehicle body distance detection sensor, and then controls the camera adjustment device based on this data to accurately collect the visual coordinate data obtained by the binocular vision intelligent camera group. After the visual data is processed, the position and posture data of the fixed object are generated, so that the main control device can automatically adjust the workbench position and posture of the automatic assembly mechanism and start the electric fixation to implement the fixing operation. The hardware architecture of the present invention is simple, debugging is convenient, and maintainability is strong, which can realize the flexible production of platform-based product vehicle production lines.

Description

Technical Field

[0001] The present invention relates to the field of automobile manufacturing, and in particular to an automatic tightening control system based on binocular vision recognition and positioning. Background Art

[0002] With the continuous rise in labor costs and the rapid development of intelligent detection technology and automated control technology, it has become possible to further reduce labor costs, improve productivity, improve product quality, reduce labor intensity, and improve working conditions. In addition, with fierce market competition and a substantial reduction in equipment manufacturing costs, most companies are increasingly favoring the use of advanced automated production technology.

[0003] Traditional new energy vehicle assembly plants generally use a production process that combines AGVs with manual power tool tightening. The AGVs transfer battery packs from a loading point at the lineside to the bottom of the chassis assembly conveyor line, operating synchronously with the assembly conveyor line. Operators, standing on accompanying AGVs, access the underbody to perform power tightening operations. This process has the advantages of a simple system, short on-site commissioning cycles, and simple debugging. However, this method also has many disadvantages:

[0004] (1) Generally, multiple people are required to perform tightening operations, resulting in high labor costs;

[0005] (2) Manual tightening operation one by one has low production efficiency;

[0006] (3) The tightening process is difficult to control and the tightening quality cannot be guaranteed;

[0007] (4) Operating on the vehicle body and the bottom of the battery pack poses a safety hazard. Summary of the Invention

[0008] In view of the above, the present invention aims to provide an automatic tightening control system based on binocular vision recognition and positioning, so as to reduce labor costs and production safety risks, and improve production efficiency and the stability of process quality.

[0009] The technical solution adopted in the present invention is as follows:

[0010] The present invention provides an automatic tightening control system based on binocular vision recognition and positioning, which includes:

[0011] The detection device includes a position detection sensor for the tightening object, a distance detection sensor, and two sets of smart cameras, and the two sets of smart cameras are arranged diagonally in the rectangular tooling area in a horizontal plane;

[0012] An automatic assembly mechanism, comprising a camera shift servo drive, a lifting platform servo motor, an X-direction servo motor for moving the tightened object, a Y-direction servo motor for rotating the tightened object around the Z-axis, and a multi-axis electric tightening device;

[0013] A main control device controls the execution of signal acquisition, data acquisition, data preprocessing, automatic visual positioning, automatic adjustment and automatic tightening; the main control device determines that the sling carrying the tightening object has arrived above the automatic assembly mechanism carrying the assembly part based on the signal output by the vehicle body arrival detection sensor, and starts a working cycle; after entering the working cycle, the main control device collects the tightening object distance data output by the distance detection sensor, and after comparing the data with the calibrated photographing distance sample data, generates the displacement correction value required by the camera shift servo driver, and the main control device controls the camera shift servo driver to move to the corresponding photographing position based on the correction value; the main control device filters the visual coordinate data collected by each of the two sets of the smart cameras, and generates visual coordinate position data, and processes the visual coordinate position data into world coordinate data; based on the two world coordinate data, the position and posture data of the tightening object are obtained;

[0014] The multi-axis electric tightening device is started to perform the tightening operation, and the working cycle ends.

[0015] In at least one possible implementation, the control system further comprises: a human-computer interaction unit connected to the main control device, the main control device displaying the tightening object distance, visual coordinate data, and tightening object position and posture data through the human-computer interaction unit;

[0016] The main control device determines whether the visual data is normal based on the range of the visual coordinate data and displays it on the human-computer interaction unit; if it is abnormal, the main control device alarms and stops running.

[0017] In at least one possible implementation, the main control device is further used to correct the position and posture of the automatic assembly mechanism before starting the multi-axis electric tightening device: the main control device calculates the position and posture correction value of the automatic assembly mechanism based on the position and posture data of the tightening object, the position and posture data of the pre-calibrated tightening sample and the camera displacement correction value.

[0018] In at least one possible implementation manner, obtaining the camera displacement correction value includes: selecting image features of the tightening sample by setting visual parameters of the two sets of smart cameras, and saving the image features in the smart cameras.

[0019] In at least one possible implementation, the method for calibrating the position and posture of the tightening sample includes: the main control device collects two sets of visual coordinate data of the smart cameras to generate the position and posture data of the tightening sample.

[0020] In at least one possible implementation manner, the main control device obtains photographing distance data of the tightening sample through a distance detection sensor.

[0021] In at least one possible implementation, the method for determining the world coordinate data includes: the main control device uses the center point of the field of view of one set of smart cameras as a reference point of the world coordinate system, measures the physical distance from the center of the field of view of another set of smart cameras to the reference point, and measures the angle of the line connecting the two field of view centers relative to a preset center line of the automatic tightening control system, thereby obtaining the position data of the field of view centers of the two sets of smart cameras in the world coordinate system.

[0022] In at least one possible implementation, determining a range conversion parameter for converting a coordinate position to world coordinate data includes: adjusting the tightening object to move to a set displacement along the X and Y directions at a set photographing distance, generating a change value of the visual coordinate position data, and using the ratio of the change value to the set displacement as the range conversion parameter.

[0023] In at least one possible implementation method, the data processing method of the main control device also includes performing coordinate data conversion based on the world coordinate value of the center of the field of view and the range conversion parameter: based on the center of the connection between the two world coordinate data, calculating the position data of the connection center and the angle data of the connection between the two in the world coordinate system, thereby representing the position and posture data of the tightening object.

[0024] In at least one possible implementation, the main control device is also used to control and correct the position and posture: control and correct the workbench of the automatic assembly mechanism to the required height, control and correct the pallet of the workbench to move the required distance along the X and Y directions, and control and correct the pallet to rotate the required angle along the Z-axis center of the workbench.

[0025] The main design concept of the present invention is that it includes a main control device, a detection part, and an execution part, wherein the execution part includes: a camera adjustment device, an automatic assembly mechanism, and an electric tightening device; the detection part includes: a body in position detection sensor, a body distance detection sensor, and a binocular vision intelligent camera group. The detection part and the execution part are both electrically connected to the main control device; the main control device detects the start signal output by the body in position detection sensor, controls and collects the distance data of the fixed object output by the body distance detection sensor according to the start signal, and then controls the camera adjustment device based on the distance data to accurately collect the visual coordinate data obtained by the binocular vision intelligent camera group. After the visual data is processed, the position and posture data of the fixed object are generated, so that the main control device can automatically adjust the workbench position and posture of the automatic assembly mechanism accordingly, and start the electric fixation to implement the tightening and fixing operation. The hardware architecture of the present invention is simple, easy to debug, and highly maintainable, so as to realize the flexible production of platform-based product vehicle production lines. In addition, while reasonably controlling the overall composition, the present invention ensures higher accuracy and reliability of identification, positioning, and automatic adjustment methods (the error rate is no more than one in ten thousand after actual verification), and is applicable to a wide range of workpieces to be assembled, and can be promoted in the processing and manufacturing links within the industry.

[0026] It should be emphasized that the present invention is based on the chassis assembly process of whole vehicle manufacturing, and fixes workpieces such as battery packs on the lifting floating workbench to the vehicle body with uncertain position and posture by screws. It includes a camera automatic adjustment device, an intelligent camera that directly provides feature position data, vehicle body position and posture calculation, and control of the floating workbench. It focuses on system engineering applicability rather than being limited to the machine vision principle level. Specifically, the technology of the present invention is biased towards the engineering application of intelligent cameras, which are machine vision products. After calibration, the intelligent camera directly provides the feature position data of the camera coordinate system, and then undergoes further spatial coordinate conversion to generate the world coordinate system position and posture data of the vehicle body. Then the control system receives the vehicle body position and posture data, and controls the floating workbench action to make the workpiece on the floating workbench consistent with the position and posture of the vehicle body before starting the fixing operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be further described below with reference to the accompanying drawings, in which:

[0028] Figure 1 A schematic diagram of the architecture of an automatic tightening control system based on binocular vision recognition and positioning provided by an embodiment of the present invention;

[0029] Figure 2 A schematic flow chart of an automatic tightening method based on binocular vision recognition and positioning provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0030] The following describes embodiments of the present invention in detail. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended only to explain the present invention and are not to be construed as limiting the present invention.

[0031] The present invention proposes an embodiment of an automatic tightening control system based on binocular vision recognition and positioning. Specifically, Figure 1 shown, including:

[0032] The detection device includes a position detection sensor for the tightening object, a distance detection sensor, and two sets of smart cameras, which are arranged diagonally in the horizontal plane in the rectangular tooling area and preferably have night vision capabilities;

[0033] An automatic assembly mechanism, comprising a camera shift servo drive, a lifting platform servo motor, an X-direction servo motor for moving the tightened object, a Y-direction servo motor for rotating the tightened object around the Z-axis, and a multi-axis electric tightening device;

[0034] A main control device controls the execution of signal acquisition, data acquisition, data preprocessing, automatic visual positioning, automatic adjustment and automatic tightening; the main control device determines that the sling carrying the tightening object has arrived above the automatic assembly mechanism carrying the assembly part based on the signal output by the vehicle body arrival detection sensor, and starts a working cycle; after entering the working cycle, the main control device collects the tightening object distance data output by the distance detection sensor, and after comparing the data with the calibrated photographing distance sample data, generates the displacement correction value required by the camera shift servo driver, and the main control device controls the camera shift servo driver to move to the corresponding photographing position based on the correction value; the main control device filters the visual coordinate data collected by the two sets of the intelligent cameras (binocular vision) respectively, and generates visual coordinate position data, and processes the visual coordinate position data into world coordinate data; according to the two world coordinate data, the position and posture data of the tightening object are obtained;

[0035] The multi-axis electric tightening device is started to perform the tightening operation, and the working cycle ends.

[0036] Furthermore, the control system further comprises: a human-computer interaction unit connected to the main control device, the main control device displays the distance of the tightening object, visual coordinate data, and the position and posture data of the tightening object through the human-computer interaction unit;

[0037] The main control device determines whether the visual data is normal based on the range of the visual coordinate data and displays it on the human-computer interaction unit; if it is abnormal, the main control device alarms and stops running.

[0038] Furthermore, the main control device is also used to correct the position and posture of the automatic assembly mechanism before starting the multi-axis electric tightening device: the main control device calculates the position and posture correction value of the automatic assembly mechanism based on the position and posture data of the tightening object, the position and posture data of the pre-calibrated tightening sample and the camera displacement correction value.

[0039] Based on this, the acquisition of the camera displacement correction value includes: selecting the image features of the tightening sample by setting the visual parameters of the two sets of smart cameras, and saving them in the smart cameras.

[0040] Based on this, the calibration method for the position and posture of the tightening sample includes: the main control device collects the visual coordinate data of the two sets of smart cameras to generate the position and posture data of the tightening sample.

[0041] Based on this, the main control device obtains the photographing distance data of the tightening sample through the distance detection sensor.

[0042] Furthermore, the method for determining the world coordinate data includes: the main control device uses the center point of the field of view of one set of smart cameras as a reference point of the world coordinate system, measures the physical distance from the center of the field of view of another set of smart cameras to the reference point, and measures the angle of the line connecting the two field of view centers relative to the preset center line of the automatic tightening control system, thereby obtaining the position data of the field of view centers of the two sets of smart cameras in the world coordinate system.

[0043] Furthermore, determining the range conversion parameter for converting the coordinate position to the world coordinate data includes: at a set shooting distance, adjusting the tightening object to move to a set displacement along the X and Y directions, generating a change value of the visual coordinate position data, and using the ratio of the change value to the set displacement as the range conversion parameter.

[0044] Therefore, the data processing method of the main control device also includes converting coordinate data according to the world coordinate value of the center of the field of view and the range conversion parameter: based on the center of the connection between the two world coordinate data, the position data of the connection center and the angle data of the connection between the two in the world coordinate system are calculated to represent the position and posture data of the tightening object.

[0045] Finally, it can be added that the main control device is also used to control and correct the position and posture: control and correct the workbench of the automatic assembly mechanism to the required height, control and correct the pallet of the workbench to move the required distance in the X and Y directions, and control and correct the pallet to rotate the required angle along the Z-axis center of the workbench.

[0046] The main control device detects the arrival of a sling carrying an object to be tightened above the automatic assembly mechanism via the vehicle body arrival detection sensor. Based on the data output by the vehicle body distance detection sensor, the main control device controls the camera adjustment device to reach a calibrated photographing distance. Once the camera position is adjusted, the main control device begins collecting data from the binocular vision intelligent camera group and processes the data to generate position and attitude correction values. Based on the correction values, the main control device controls the workbench carrying the assembly parts of the automatic assembly mechanism to adjust to the appropriate position and attitude. Once the correction is complete, the electric tightening process is triggered. Automatic tightening is completed, and the next work cycle begins. This invention reduces labor costs, reduces production safety risks, improves production efficiency, and enhances the stability of process quality.

[0047] Based on the above-mentioned embodiments, the following provides specific implementation examples for reference:

[0048] The system includes a main control device, a detection part, and an execution part; the execution part includes: a camera adjustment device, an automatic assembly mechanism, and an electric tightening device; the detection part of the system includes: a body position detection sensor, a body distance detection sensor, and a binocular vision intelligent camera group; the detection part and the execution part are both electrically connected to the main control device; the control system detects the start signal output by the body position detection sensor, controls according to the start signal, collects the tightening object distance data output by the body distance detection sensor, and controls the camera adjustment device according to the detected distance; the control system collects the visual coordinate data of the binocular vision intelligent camera group, generates the tightening object position and posture data after data processing; the control system automatically adjusts the workbench position and posture of the automatic assembly mechanism, and then starts the electric tightening device;

[0049] It should be noted that the tightening object in this example is the new energy vehicle body, and the assembly part is the battery pack. The battery pack needs to be fixed to the bottom of the new energy vehicle body through a workbench that can be raised and lowered, and the position and posture can be adjusted, and the electric tightening tool on it.

[0050] The top-down structure of the entire assembly station is as follows: overhead conveyor line, slings on the conveyor line, automatic assembly mechanism for battery packs, and roller conveyor device for transporting battery packs.

[0051] Specifically, the hoist carries the new energy vehicle body through tires, and its position and posture are uncertain within a certain range of the horizontal plane; different vehicle models and different tire pressures make the height of the vehicle body chassis battery pack assembly surface relative to the ground uncertain; the position and posture of the vehicle body are uncertain, but cannot be adjusted.

[0052] Specifically, the working table of the automatic battery pack assembly mechanism carries the battery pack, the position and posture of the battery pack relative to the working table are determined, and the position and posture of the working table can be adjusted.

[0053] Specifically, the battery pack automatic assembly mechanism includes: a vertical laser ranging sensor, a camera adjustment mechanism, an intelligent camera, an electric lifting mechanism, a workbench and an electric tightening tool.

[0054] Specifically, the workbench of the automatic battery pack assembly mechanism is a floating workbench. The workbench can be moved and adjusted in the X and Y directions under the servo control of the PLC control system, and can be rotated and adjusted along the vertical center line of the workbench to align all the locking bolts on the battery pack with the bolt holes of the vehicle body. After adjustment and alignment, the PLC control system starts the electric tightening task.

[0055] Specifically, an electric tightening tool is fixed to the bottom of the workbench of the battery pack automatic assembling mechanism.

[0056] Furthermore, the main control device displays the distance to the tightening object, the visual coordinate data, the position and posture data of the tightening object, the position and posture deviation value data, the speed and torque data of each tightening shaft, and the tightening qualification data through the human-computer interaction unit.

[0057] Specifically, the control system determines whether the visual data is normal based on whether the visual coordinate data is within a set range. The status is displayed on the human-computer interaction interface. If it is abnormal, the control system alarms, stops running, and recovers through manual intervention.

[0058] Furthermore, the control system determines that the sling carrying the tightening object has reached the automatic assembly mechanism carrying the assembly part according to the signal output by the vehicle body arrival detection sensor, and starts a working cycle;

[0059] Specifically, after entering the working cycle, the control system collects the tightening object distance data output by the vehicle body distance detection sensor, compares the data with the calibrated photographing distance sample data, and generates the displacement correction value required by the camera adjustment device. The control system controls the camera adjustment device to move to a suitable photographing position according to the correction value;

[0060] Specifically, after the control system controls the camera to the appropriate photographing position, it collects a set of visual coordinate data from each camera of the binocular vision intelligent camera group, generates visual coordinate position data after filtering the data; and processes the visual coordinate position data to generate world coordinate data of the feature;

[0061] Specifically, the control system generates the position and posture data of the tightening object according to the world coordinate data of the two features through data processing;

[0062] Specifically, the method includes obtaining the position and posture correction values ​​of the automatic assembly mechanism;

[0063] The acquisition of the position and posture correction values ​​of the automatic assembly mechanism includes: the control system calculates and processes the position and posture correction values ​​of the automatic assembly mechanism based on the tightening object position and posture data, the calibrated sample tightening object position and posture data, and the camera displacement correction value, and performs position and posture control correction.

[0064] Specifically, after the control system completes the correction, the electric tightening device is started to operate, the tightening operation is completed, and a working cycle ends;

[0065] Furthermore, before the automatic tightening system is put into use, a series of parameter calibrations are required. This includes feature calibration of the binocular vision intelligent camera set, which includes selecting features of a sample tightening object using the visual parameter setting software of the binocular vision intelligent camera set and saving them to the intelligent cameras of the binocular vision intelligent camera set.

[0066] Specifically, the world coordinate system is established and calibrated, including: manually selecting the center of the field of view of camera 1 of the binocular vision intelligent camera group as a special reference point in the world coordinate system. To minimize negative numbers in the control system calculation process, the coordinate value of this special reference point is manually set to (5000, 5000) in millimeters. The physical distance from the center of the field of view of camera 2 of the binocular vision intelligent camera group to the special reference point is measured, such as 1500 mm, and the angle between the two field of view centers and the centerline of the automatic tightening system is measured, such as 30°. The world coordinate value (a, b) of the field of view center of camera 2 can be obtained through simple mathematical calculation, where a = 5000 + 1500 * cos 30° mm, and b = 5000 + 1500 * sin 30° mm.

[0067] Specifically, there is a calibration of the range conversion parameters, including: determining the parameters of the algorithm for converting the visual coordinate position data to the world coordinate data of the feature, and the parameter generation method includes: at a set shooting distance, manually adjusting the tightening object feature to move to a set displacement in the X and Y directions, and the visual coordinate position data also changes accordingly, thereby generating a change value of the visual coordinate position data. The ratio of the change value to the set displacement is the range conversion parameter. Because the smart camera selected in the embodiment of the present invention does not cause image distortion or negligible distortion when taking pictures in the automatic tightening area, the range conversion parameter can be selected in either the X or Y direction. If the feature moves 20mm in the X direction and the visual coordinate changes 200 units in the X direction, the range conversion ratio is 10 units / mm.

[0068] Specifically, the calibration of the photographing distance data of the bottom body of a new energy vehicle includes: the main control device obtains the photographing distance sample data through the vehicle body distance detection sensor.

[0069] Specifically, the calibration of the position and posture data of the tightening object of the sample vehicle body includes: the main control device collects the visual coordinate data of the binocular vision intelligent camera group, and generates the position and posture data of the sample tightening object after data processing.

[0070] Specifically, during calibration and production application, the data processing process is as follows: based on the world coordinate value of the center of the field of view and the range conversion parameters, the control system realizes the conversion of the visual coordinate position data (u, v) to the world coordinate data (x, y) of the feature.

[0071] The coordinates of the center of the field of view of the smart camera are assumed to be (u0, v0).

[0072] Specifically, the visual coordinate position data (u1, v1) output by camera 1 and the world coordinates of the feature are (x1, y1). The basic conversion formula is as follows:

[0073] x1=5000+(u1-u0) / 10

[0074] y1=5000+(v1-v0) / 10

[0075] Specifically, the visual coordinate position data (u2, v2) output by the second camera is the world coordinate of the feature (x2, y2). The basic conversion formula is as follows:

[0076] x2=5000+(u2-u0) / 10

[0077] y2=5000+(v2-v0) / 10

[0078] Specifically, the control system calculates the position data (x3, y3) of the center of the line connecting the two features and the angle data θ3 of the line in the world coordinates based on the world coordinate data (x1, y1) and (x2, y2) of the features, i.e., the position and posture data (x3, y3, θ3) of the body to be tightened. The basic conversion formula is as follows:

[0079] x3=(x1+x2) / 2

[0080] y3=(y1+y2) / 2

[0081] tagθ3=(y2-y1) / (x2-x1)

[0082] Specifically, assuming that the position and posture data of the sample tightening object are (x0, y0, θ0), the deviation value is:

[0083] Δx=x3-x0

[0084] Δy=y3-y0

[0085] Δθ=θ3-θ0

[0086] Specifically, based on the position and posture deviation values, the control system controls and corrects the worktable of the automatic assembly mechanism to an appropriate lifting height, controls and corrects the pallet of the worktable to move an appropriate distance in the X and Y directions, and controls and corrects the pallet to rotate an appropriate angle along the Z-axis center of the worktable.

[0087] Accordingly, the present invention also provides an automatic tightening method based on binocular vision recognition and positioning, such as Figure 2 As shown, the method includes the following steps:

[0088] Step 100: Start.

[0089] Step 101: When the lifting device carrying the vehicle body reaches the top of the automatic assembly mechanism, the control system collects a vehicle body arrival signal and starts a working cycle.

[0090] Step 102: The main control device collects vehicle body distance data and performs camera position correction.

[0091] Specifically, affected by the environment, the data output by the vehicle body distance detection sensor fluctuates within a certain range; the main control device of the embodiment of the present invention collects 5 distance data with a collection period of 100ms, and generates stable distance data after filtering.

[0092] Specifically, the main control device compares the stable distance data with the calibrated photographing distance sample data to generate the displacement correction value required by the camera adjustment device.

[0093] Specifically, the main control device controls the camera adjustment device to move to a suitable photographing position according to the correction value.

[0094] Step 103: The main control device collects camera group data and processes the data to generate world coordinate data.

[0095] Specifically, after controlling the camera to the appropriate photographing position, the main control device collects the visual coordinate data of each camera of the binocular vision intelligent camera group with an acquisition cycle of 100ms. The embodiment of the present invention collects 5 visual coordinate data for each camera and generates two determined visual coordinate data after filtering processing.

[0096] Specifically, for the determined visual coordinate data, data processing is performed to generate the world coordinate data of the feature. The basic conversion algorithm is as follows:

[0097] After calibration before application, it is known that the range conversion scale factor is 10 pixels / mm, and the visual coordinates of the center of the field of view of the smart camera are (u0, v0).

[0098] Specifically, the visual coordinate position data (u1, v1) output by camera 1 and the world coordinates of the feature are (x1, y1). The basic conversion formula is as follows:

[0099] x1=5000+(u1-u0) / 10

[0100] y1=5000+(v1-v0) / 10

[0101] Specifically, the visual coordinate position data (u2, v2) output by the second camera is the world coordinate of the feature (x2, y2). The basic conversion formula is as follows:

[0102] x2=5000+(u2-u0) / 10

[0103] y2=5000+(v2-v0) / 10

[0104] Step 104: The main control device processes the world coordinate data of the two features to generate position and posture data of the vehicle body.

[0105] Specifically, in the embodiment of the present invention, the center of the line connecting the two features of the vehicle body is taken as the center of the vehicle body, and its world coordinates are (x3, y3, θ3).

[0106] Specifically, the main control device calculates the position data (x3, y3) of the center of the line connecting the two features based on the world coordinate data (x1, y1) and (x2, y2) of the features. The basic conversion formula is as follows:

[0107] x3=(x1+x2) / 2

[0108] y3=(y1+y2) / 2

[0109] Specifically, the main control device calculates and obtains the world coordinate angle data θ3 of the feature line based on the world coordinate data (x1, y1) and (x2, y2) of the feature. The basic conversion formula is as follows:

[0110] tagθ3=(y2-y1) / (x2-x1)

[0111] Step 105: Calculate the position and attitude deviation value of the vehicle body.

[0112] Specifically, the sample body of the embodiment of the present invention is calibrated by the main control device. The position and posture data of the sample body are known to be (x0, y0, θ0), and the world coordinates of the body to be tightened are (x3, y3, θ3). The deviation value is:

[0113] Δx=x3-x0

[0114] Δy=y3--y0

[0115] Δθ=θ3-θ0

[0116] Step 106: The main control device controls and corrects the position and posture of the workbench.

[0117] Specifically, according to the displacement correction value of the camera adjustment device in step 102 and the basic assembly lifting height data, the main control device controls the working table of the automatic assembly mechanism to a suitable lifting height.

[0118] Furthermore, according to the position and posture deviation value of the vehicle body, the main control device controls the pallet of the workbench to move an appropriate distance in the X and Y directions, and controls the pallet to rotate an appropriate angle along the Z axis center of the workbench.

[0119] After the correction, the position and posture of the workbench carrying the battery pack are consistent with the position and posture of the vehicle body.

[0120] Step 107: Start the tightening process.

[0121] Specifically, after the workbench's position and posture are corrected, the centerline of each bolt on the battery pack is aligned perpendicularly with the centerline of the corresponding mounting hole on the underbody, or within the allowable deviation range. The main control unit then synchronizes and tightens all axes, visualizing the tightening data. Automatic tightening is complete, and a work cycle concludes.

[0122] Step 108: End.

[0123] The present invention can adopt PLC as the main control device and adopt intelligent camera to directly output the visual data of features. After simple data processing, the position and posture data of the vehicle body are obtained. After comparison with the sample vehicle body position and posture data, the position and posture deviation value of the vehicle body is obtained, and the position and posture of the battery pack are automatically corrected, thereby realizing the automatic assembly process of the battery pack and the vehicle body, facilitating on-site debugging, thereby improving the flexibility, stability and safety of the equipment and improving the production efficiency.

[0124] In summary, it includes a main control device, a detection part, and an execution part, wherein the execution part includes: a camera adjustment device, an automatic assembly mechanism, and an electric tightening device; the detection part includes: a vehicle body in position detection sensor, a vehicle body distance detection sensor, and a binocular vision intelligent camera group. The detection part and the execution part are both electrically connected to the main control device; the main control device detects the start signal output by the vehicle body in position detection sensor, controls and collects the distance data of the fixed object output by the vehicle body distance detection sensor according to the start signal, and then controls the camera adjustment device based on the distance data, so as to accurately collect the visual coordinate data obtained by the binocular vision intelligent camera group, and generates the position and posture data of the fixed object after the visual data is processed, so that the main control device can automatically adjust the workbench position and posture of the automatic assembly mechanism accordingly, and start the electric fixation to implement the tightening and fixing operation. The hardware architecture of the present invention is simple, easy to debug, and highly maintainable, so as to realize the flexible production of the platform product vehicle production line.

[0125] In the embodiment of the present invention, "at least one" refers to one or more, and "more" refers to two or more. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent the existence of A alone, the existence of A and B at the same time, and the existence of B alone. A and B can be singular or plural. The character " / " generally indicates that the previous and next associated objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b and c can be represented by: a, b, c, a and b, a and c, b and c, or a, b and c, where a, b, c can be single or multiple.

[0126] The above describes in detail the structure, features and effects of the present invention based on the embodiments shown in the drawings, but the above is only a preferred embodiment of the present invention. It should be noted that the technical features involved in the above embodiments and their preferred modes can be reasonably combined and matched into a variety of equivalent schemes by those skilled in the art without departing from or changing the design ideas and technical effects of the present invention; therefore, the scope of implementation of the present invention is not limited to what is shown in the drawings. Any changes made in accordance with the concept of the present invention, or modifications to equivalent embodiments with equivalent changes, which still do not exceed the spirit covered by the description and drawings, should be within the scope of protection of the present invention.

Claims

1. An automatic tightening control system based on binocular vision recognition and positioning, characterized in that: include: The detection device includes a position detection sensor for the tightening object, a distance detection sensor, and two sets of smart cameras, and the two sets of smart cameras are arranged diagonally in the rectangular tooling area in a horizontal plane; An automatic assembly mechanism, comprising a camera shift servo drive, a lifting platform servo motor, an X-direction servo motor for moving the tightened object, a Y-direction servo motor for rotating the tightened object around the Z-axis, and a multi-axis electric tightening device; A main control device controls the execution of signal acquisition, data acquisition, data preprocessing, automatic visual positioning, automatic adjustment and automatic tightening; the main control device determines that the sling carrying the tightening object has arrived above the automatic assembly mechanism carrying the assembly part based on the signal output by the in-place detection sensor, and starts a working cycle; after entering the working cycle, the main control device collects the tightening object distance data output by the distance detection sensor, and after comparing the data with the calibrated photographing distance sample data, generates the displacement correction value required by the camera shift servo driver, and the main control device controls the camera shift servo driver to move to the corresponding photographing position based on the correction value; the main control device filters the visual coordinate data collected by each of the two sets of the smart cameras, and generates visual coordinate position data, and processes the visual coordinate position data into world coordinate data; based on the two world coordinate data, the position and posture data of the tightening object are obtained; Furthermore, the main control device is further used to control and correct the position and posture: controlling and correcting the worktable of the automatic assembly mechanism to a desired height, controlling and correcting the tray of the worktable to move a desired distance in the X and Y directions, and controlling and correcting the tray to rotate a desired angle along the Z axis center of the worktable; wherein the worktable carries a battery pack, and the position and posture of the battery pack relative to the worktable are determined; The multi-axis electric tightening device is started to perform synchronous tightening operations on all axes, and the working cycle ends.

2. The automatic tightening control system based on binocular vision recognition and positioning according to claim 1 is characterized in that: The control system further comprises: a human-machine interaction unit connected to the main control device, wherein the main control device displays the distance to the tightening object, visual coordinate data, and the position and posture data of the tightening object through the human-machine interaction unit; The main control device determines whether the visual data is normal based on the range of the visual coordinate data and displays it on the human-computer interaction unit; if it is abnormal, the main control device alarms and stops running.

3. The automatic tightening control system based on binocular vision recognition and positioning according to claim 1 is characterized in that: The main control device is also used to correct the position and posture of the automatic assembly mechanism before starting the multi-axis electric tightening device: the main control device calculates the position and posture correction value of the automatic assembly mechanism based on the position and posture data of the tightening object, the position and posture data of the pre-calibrated tightening sample and the camera displacement correction value.

4. The automatic tightening control system based on binocular vision recognition and positioning according to claim 3 is characterized in that: The acquisition of the camera displacement correction value includes: selecting the image features of the tightening sample by setting the visual parameters of the two sets of smart cameras, and saving the selected features in the smart cameras.

5. The automatic tightening control system based on binocular vision recognition and positioning according to claim 3 is characterized in that: The method for calibrating the position and posture of the tightening sample includes: the main control device collects the visual coordinate data of the two sets of smart cameras to generate the position and posture data of the tightening sample.

6. The automatic tightening control system based on binocular vision recognition and positioning according to claim 3 is characterized in that: The main control device obtains photographing distance data of the tightening sample through a distance detection sensor.

7. The automatic tightening control system based on binocular vision recognition and positioning according to claim 1 is characterized in that: The method for determining the world coordinate data includes: the main control device uses the center point of the field of view of one set of smart cameras as a reference point of the world coordinate system, measures the physical distance from the center of the field of view of another set of smart cameras to the reference point, and measures the angle of the line connecting the two field of view centers relative to a preset center line of the automatic tightening control system, thereby obtaining the position data of the field of view centers of the two sets of smart cameras in the world coordinate system.

8. The automatic tightening control system based on binocular vision recognition and positioning according to claim 1 is characterized in that: Determining a range conversion parameter for converting a coordinate position into world coordinate data includes: adjusting the tightening object to move to a set displacement along the X and Y directions at a set photographing distance, generating a change value of the visual coordinate position data, and using the ratio of the change value to the set displacement as the range conversion parameter.

9. The automatic tightening control system based on binocular vision recognition and positioning according to claim 8, characterized in that: The data processing method of the main control device also includes performing coordinate data conversion based on the world coordinate value of the center of the field of view and the range conversion parameter: based on the center of the connection between the two world coordinate data, calculating the position data of the connection center and the angle data of the connection between the two in the world coordinate system, thereby representing the position and posture data of the tightening object.

Images