Visual guidance workpiece placing device and method based on feature recognition

Abstract

The invention discloses a visual guidance workpiece placing device and method based on feature recognition, and the device comprises a tool bottom plate which is provided with a workpiece placing station, and the front end and the rear end of the workpiece placing station are each provided with an image collection module; the intelligent guide vehicle travels on the tool bottom plate and is used for carrying vehicle parts, positioning identification assemblies are arranged at the front end and the rear end of the intelligent guide vehicle, and when the intelligent guide vehicle travels to the workpiece placing station, positioning columns in the positioning identification assemblies are arranged opposite to the image acquisition module, so that the image acquisition module acquires target position information of the positioning columns; the part placing robot is arranged on the side face of the tool bottom plate and used for placing the vehicle parts on the intelligent guide vehicle. And the control module is in communication connection with the intelligent guide vehicle, the image acquisition module and the workpiece placing robot and is used for calculating offset information according to the target position information and the preset position information and generating a workpiece placing track according to the offset information so as to control the workpiece placing robot to accurately place workpieces.

Description

Technical Field

[0001] This invention relates to the field of feature recognition technology, and specifically to a visually guided placement device and method based on feature recognition. Background Technology

[0002] Intelligent guided vehicles, including Automated Guided Vehicles (AGVs) and Intelligent Guided Vehicles (IGVs), are core material handling equipment in the field of industrial automation. They do not require manual driving and can accurately complete tasks such as material handling and cargo transfer in preset scenarios (such as factory workshops, warehouses, and logistics parks). They are one of the key pieces of equipment for realizing "unmanned factories" and "intelligent logistics".

[0003] In the automotive manufacturing industry, when robots pick up automotive assembly components and place them onto the accompanying tooling of intelligent guided vehicles, the positioning of the intelligent guided vehicles themselves has an error of ±5mm, while the fit clearance between the positioning pins of the accompanying tooling and the positioning holes of the vehicle body is only 0.2mm. If the robot relies solely on a fixed trajectory to perform the placement operation, it is very easy to cause problems such as deformation of the positioning holes, wear of the positioning pins, or abnormal noise from the pins, which seriously affects the lifespan of the equipment and the assembly quality.

[0004] To overcome the aforementioned problems, visual guidance technology is needed to achieve precise part placement. The vision system identifies existing positioning pins on the intelligent guided vehicle's tooling as visual features, calculates their deviation from a standard template, and then corrects the robot's part placement trajectory in real time. However, the visual feature recognition devices in this process often employ a Z-axis circular hole combination structure, i.e., circular holes are set at the bottom of the intelligent guided vehicle. Because the circular hole structure easily accumulates welding slag, cleaning and maintenance costs are high. By installing a dust cover above the circular hole structure and using a matching servo cylinder and linkage mechanism to drive the dust cover, welding slag contamination of the circular hole can be prevented from affecting recognition. However, this results in a complex system structure, high cost, and limited reliability. Furthermore, the circular hole structure has poor imaging stability and low recognition reliability, leading to inaccurate part placement. Summary of the Invention

[0005] To address the problem of inaccurate robot placement in existing technologies, this invention provides a vision-guided placement device and method based on feature recognition, aiming to solve the problems of low reliability and complex system structure of visual feature recognition, and improve the quality and efficiency of automated assembly in the automotive manufacturing industry.

[0006] This invention discloses a visually guided placement device based on feature recognition, comprising: The tooling base plate is equipped with a part placement station, and image acquisition modules are installed at both the front and rear ends of the part placement station. The intelligent guided vehicle travels on the tooling base plate and is used to transport vehicle parts. The front and rear ends of the intelligent guided vehicle are equipped with positioning mark components, which include positioning posts. When the intelligent guided vehicle travels to the placement station, the positioning posts are set opposite to the image acquisition module so that the image acquisition module can collect the target position information of the positioning posts. The placement robot is located on the side of the tooling base plate. When the intelligent guide car moves to the placement station, the placement robot places the vehicle parts on the intelligent guide car according to the placement trajectory. The control module is connected to the intelligent guided vehicle, the image acquisition module, and the placement robot. The control module is used to calculate the offset information based on the target position information and the preset position information, and to generate the placement trajectory based on the offset information in order to control the placement robot to place the part.

[0007] Preferably, the positioning marker component includes a positioning base, a positioning bottom plate, and a preset number of positioning posts; The positioning base plate and the positioning base are detachably connected, and the plane of the positioning base plate is perpendicular to the bottom surface of the tooling base plate; The positioning column is fixedly connected to the positioning base plate, and the extension direction of the positioning column is perpendicular to the positioning base plate.

[0008] Preferably, the positioning base includes an adjustment mechanism and an L-shaped frame. The adjustment mechanism includes a first threaded component, a second threaded component, and an adjustment base plate. The adjustment base plate is connected to the first end of the L-shaped frame based on the first threaded component, and the adjustment base plate is connected to the positioning base plate based on the second threaded component. The second end of the L-shaped frame is detachably connected to the intelligent guided vehicle. The first threaded component is used to adjust the position of the positioning base plate in the first direction, which is the extension direction of the positioning post; The second threaded component is used to adjust the position of the positioning base plate in the second direction, which is parallel to the plane where the positioning base plate is located.

[0009] Preferably, the positioning marker assembly further includes a shim; the shim is disposed at the bottom of the second end of the L-shaped frame, and the shim is used to adjust the height of the positioning marker assembly.

[0010] Preferably, the positioning post has a cylindrical structure, and the preset number of positioning posts is 3.

[0011] This invention discloses a visually guided placement method based on feature recognition, applied to the aforementioned visually guided placement device based on feature recognition. The visually guided placement method includes: In response to the intelligent guided vehicle moving to the placement station of the tooling base plate, the target position information of the positioning marker component is obtained based on the image acquisition module; Acquire preset location information and calculate offset information based on the preset location information and target location information; the preset location information is used to characterize the calibration position of the positioning marker component; The placement trajectory is generated based on the offset information, and the placement robot is controlled to place the vehicle parts on the intelligent guided vehicle according to the placement trajectory.

[0012] Preferably, before obtaining the location information of the positioning marker component based on the image acquisition module, the method further includes: The position of the positioning column in the positioning mark component is determined based on the preset tooling digital model; Adjust the position of the positioning mark component so that the flatness of the positioning base plate in the positioning mark component is the same as the flatness of the tooling base plate.

[0013] Preferably, the target location information of the positioning marker component is obtained based on the image acquisition module, including: A laser line is emitted towards the positioning post of the positioning marker component using an image acquisition module, which is a 3D point cloud camera; Acquire light spot images on the surface of the positioning column; Based on the triangulation algorithm, the three-dimensional point cloud data of the light spot image is calculated to obtain the target location information.

[0014] Preferably, the offset information is calculated based on the preset location information and the target location information, including: Align the target location information with the preset location information in terms of coordinates; Calculate the translation vector and rotation matrix between the target position information and the preset position information; The actual pose deviation of the intelligent guided vehicle at the part placement station is calculated based on the translation vector and rotation matrix to generate offset information, which includes position offset and angular offset.

[0015] Preferably, generating the placement trajectory based on the offset information includes: Obtain the preset placement trajectory based on the category of vehicle component; Generate the transformation matrix based on the offset information; The transformation matrix is ​​applied to each target point of the preset placement trajectory to obtain the placement trajectory.

[0016] Compared with existing technologies, the beneficial effects of this invention are as follows: By setting a positioning marker component on the intelligent guided vehicle and configuring an image acquisition module at the placement station, when the intelligent guided vehicle travels to the placement station, the positioning post and the image acquisition module are positioned relative to each other, allowing the image acquisition module to accurately capture the light spot image of the positioning marker component, thereby determining the target position information of the positioning marker component. Furthermore, using the positioning post as a positioning marker facilitates precise recognition by the image acquisition module, improving the accuracy and precision of positioning. By calculating the actual pose deviation of the intelligent guided vehicle through preset calibration data, a precise placement trajectory is generated, effectively improving the placement accuracy of vehicle component assembly. Attached Figure Description

[0017] Figure 1 A schematic diagram of the structure of the feature recognition-based visual guidance device for placing components provided by the present invention; Figure 2 This is a schematic diagram of the structure of the positioning marker component provided by the present invention; Figure 3 This is a flowchart illustrating the visually guided placement method based on feature recognition provided by the present invention.

[0018] Figure label: 1. Intelligent guided vehicle; 2. Positioning and marking component; 21. L-shaped frame; 22. Positioning base plate; 23. Positioning column; 24. Adjustment mechanism; 241. First threaded part; 242. Second threaded part; 243. Adjustment base plate; 25. Shim; 3. Tooling base plate; 4. Image acquisition module; 5. Part placement robot; 6. Control module. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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, 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.

[0020] The present invention will now be described in further detail with reference to the accompanying drawings.

[0021] This invention provides a feature-recognition-based visual guidance device for placing vehicle components onto an intelligent guided vehicle, enabling the intelligent guided vehicle to transport the vehicle components to a designated location. The vehicle component can be a three-part assembly, namely, a vehicle body frame module formed by welding together the front, floor, and rear core frame components of the vehicle body.

[0022] like Figure 1As shown, in this embodiment of the invention, the visually guided placement device based on feature recognition includes an intelligent guided vehicle 1, a tooling base plate 3, a placement robot 5, and a control module 6. The tooling base plate 3 has a placement station, with image acquisition modules 4 at both the front and rear ends. The intelligent guided vehicle 1 travels on the tooling base plate 3 to transport vehicle parts. Positioning marker components 2 are located at both the front and rear ends of the intelligent guided vehicle 1. Each positioning marker component includes a positioning post 23. When the intelligent guided vehicle 1 reaches the placement station, the positioning post 23 is positioned opposite the image acquisition module 4, and the image acquisition module 4 acquires the target position information of the positioning post 23. The placement robot 5 is located on the side of the tooling base plate 3. When the intelligent guided vehicle 1 moves to the placement station, the placement robot 5 places the vehicle parts on the intelligent guided vehicle 1 according to the placement trajectory. The control module 6 is communicatively connected to the intelligent guided vehicle 1, the image acquisition module 4, and the placement robot 5. The control module 6 is used to calculate the offset information based on the target position information and the preset position information, and to generate the placement trajectory based on the offset information in order to control the placement robot 5 to place the parts.

[0023] In this way, the positioning marker component 2 set on the intelligent guided vehicle 1 and the image acquisition module 4 set on the tooling base plate 3 can accurately obtain the actual position information of the intelligent guided vehicle 1 and compare it with the preset position information to calculate the offset information. The control module 6 generates the placement trajectory in real time based on the offset information to ensure that the placement robot 5 can accurately place the vehicle parts on the intelligent guided vehicle 1.

[0024] In this embodiment of the invention, the positioning marker component 2 can adjust its position and height according to specific circumstances, so that the image acquisition module 4 can completely and accurately acquire the position information of the positioning marker component 2. For example... Figure 2 As shown, the positioning marker component 2 includes a positioning base, a positioning base plate 22, and a preset number of positioning posts 23. The positioning base plate 22 is detachably connected to the positioning base, and the plane of the positioning base plate 22 is perpendicular to the bottom surface of the tooling base plate 3. The positioning posts 23 are fixedly connected to the positioning base plate 22, and the extension direction of the positioning posts 23 is perpendicular to the positioning base plate 22. Thus, when the intelligent guided vehicle 1 travels to the placement station of the tooling base plate 3, the positioning posts 23 are positioned opposite to the image acquisition module 4.

[0025] Furthermore, the positioning base includes an adjustment mechanism 24 and an L-shaped frame 21. The adjustment mechanism 24 includes a first threaded component 241, a second threaded component 242, and an adjustment base plate 243. The adjustment base plate 243 is connected to the first end of the L-shaped frame 21 via the first threaded component 241, and the adjustment base plate 243 is connected to the positioning base plate via the second threaded component 242. The second end of the L-shaped frame 21 is detachably connected to the intelligent guided vehicle 1. The first threaded component 241 is used to adjust the position of the positioning base plate 22 in a first direction, which is the extension direction of the positioning post 23. The second threaded component 242 is used to adjust the position of the positioning base plate 22 in a second direction, which is parallel to the plane where the positioning base plate 22 is located.

[0026] In this embodiment of the invention, the positioning marker component 2 further includes a pad 25, which is disposed at the bottom of the second end of the L-shaped frame and is used to adjust the height of the positioning marker component.

[0027] For example, the overall height of the positioning marker assembly 2 can be finely adjusted by adding or removing the shims 25, ensuring the vertical alignment accuracy between the positioning post 23 and the image acquisition module 4. Combined with the threaded parts of the adjustment mechanism 24, the spatial position of the positioning base plate 22 can be precisely calibrated in both horizontal and vertical dimensions, thereby improving the installation flexibility and repeatability of the positioning marker assembly 2.

[0028] In this way, through the cooperation of the adjustment mechanism 24 and the first threaded component 241 and the second threaded component 242, the position of the positioning base plate 22 in the first and second directions can be flexibly adjusted, thereby precisely controlling the orientation of the positioning column 23 relative to the image acquisition module 4, ensuring that the image acquisition module 4 can acquire the position information of the positioning mark component 2 at the optimal viewing angle and distance. At the same time, the setting of the shim 25 further enhances the convenience and accuracy of the height adjustment of the positioning mark component 2, enabling the entire device to better adapt to different working environments and workpiece placement requirements, effectively improving the adaptability and accuracy of the visual guidance placement device.

[0029] Specifically, the positioning post 23 has a cylindrical structure, and the preset number of positioning posts 23 is 3.

[0030] It should be noted that the present invention does not specifically limit the shape and number of positioning posts. For example, the shape of positioning post 23 can also be a cuboid structure, and the preset number of positioning posts 23 can be 4.

[0031] In this embodiment of the invention, the image acquisition module 4 is a 3D point cloud camera. The 3D point cloud camera is used to acquire the point cloud outline of the positioning column 23 in the positioning marker component 2 in order to obtain the target position information.

[0032] like Figure 3As shown, this embodiment of the invention provides a visually guided placement method based on feature recognition, applied to the aforementioned visually guided placement device based on feature recognition. The visually guided placement method based on feature recognition includes the following steps.

[0033] S1. In response to the intelligent guided vehicle traveling to the placement station of the tooling base plate, the target position information of the positioning marker component is obtained based on the image acquisition module.

[0034] In this embodiment of the invention, positioning marker components are disposed at the front and rear ends of the intelligent guided vehicle, and image acquisition modules are disposed at the front and rear ends of the placement station. After the intelligent guided vehicle travels to the designated placement station, the image acquisition module starts working, accurately capturing the point cloud contour data of the positioning posts in the positioning marker components using a 3D point cloud camera. This module utilizes high-precision scanning technology to convert the three-dimensional spatial coordinate information of the positioning posts into digital signals, and transmits them in real time to the central processing unit for analysis and processing, thereby obtaining precise numerical values ​​of the target position information.

[0035] Before obtaining the position information of the positioning marker component based on the image acquisition module, it is also necessary to determine the position of the positioning column in the positioning marker component based on the preset tooling model, and adjust the position of the positioning marker component using the adjustment mechanism of the positioning base so that the flatness of the positioning base plate in the positioning marker component is the same as the flatness of the tooling base plate.

[0036] Based on the above method, the optimal matching state between the positioning marker component and the tooling base plate can be ensured, providing an accurate and reliable foundation for subsequent visually guided placement. By pre-setting the tooling digital model, the system can know the ideal position of the positioning post in the design in advance. Combined with the adjustment mechanism of the positioning base, the positioning marker component can be finely adjusted. This adjustment not only considers the spatial position of the positioning post, but also ensures the consistency of flatness between the positioning base plate and the tooling base plate, thereby greatly improving the accuracy and stability of placement.

[0037] Furthermore, the process of obtaining the target location information of the positioning marker component based on the image acquisition module includes: emitting a laser line to the positioning post of the positioning marker component using the image acquisition module; acquiring a light spot image on the surface of the positioning post; and calculating the three-dimensional point cloud data of the light spot image based on a triangulation algorithm to obtain the target location information.

[0038] Specifically, when using the image acquisition module to emit laser lines to the positioning posts of the positioning marker component, it is necessary to ensure that the angle and intensity of the laser lines are appropriate to form a clear and accurately identifiable light spot on the surface of the positioning post. During the acquisition of the light spot image on the positioning post surface, the image acquisition module must possess high resolution and high sensitivity to capture the subtle features of the light spot and avoid errors in subsequent calculations due to image blurring or noise interference. When calculating the 3D point cloud data of the light spot image based on the triangulation algorithm, the algorithm must undergo precise calibration and optimization to accurately calculate the coordinates of the light spot in 3D space, thereby obtaining accurate target position information and providing reliable data support for the entire visually guided placement process.

[0039] S2. Obtain preset position information and calculate offset information based on preset position information and target position information.

[0040] In this embodiment of the invention, the preset position information is used to characterize the calibration position of the positioning marker component, that is, the theoretical three-dimensional coordinates of the positioning marker component (three cylindrical pins) on the accompanying tooling of the intelligent guided vehicle under ideal conditions. This data comes from the CAD model of the tooling, or is the "gold standard" data collected and stored in the robot control system by high-precision measuring instruments (such as laser trackers) during installation and debugging.

[0041] Furthermore, the target position information is aligned with the preset position information, and the translation vector and rotation matrix between the target position information and the preset position information are calculated. Based on the translation vector and rotation matrix, the actual pose deviation of the intelligent guided vehicle at the placement station is calculated to generate offset information, which includes position offset and angle offset.

[0042] Specifically, during coordinate alignment, the Iterative Closest Point (ICP) registration algorithm is used to precisely match the real-time acquired target position information with the preset calibration position, eliminating deviations caused by coordinate system differences or installation errors. By calculating the translation vector, the positional offset of the intelligent guided vehicle in the horizontal and vertical directions can be determined; simultaneously, the rotation matrix is ​​used to characterize the rotation angle of the intelligent guided vehicle relative to the ideal posture, i.e., the angular offset. This offset information provides crucial basis for the subsequent correction of the placement trajectory, ensuring that the placement robot can accurately place vehicle parts on the intelligent guided vehicle, thereby improving the accuracy and efficiency of the entire assembly process.

[0043] In this way, by calculating the offset information, the difference between the current actual position and the ideal preset position of the positioning marker component can be accurately obtained. This difference is presented in the form of a rotation matrix and a translation vector, providing a crucial adjustment basis for subsequent visually guided placement operations.

[0044] S3. Generate a placement trajectory based on the offset information, and control the placement robot to place the vehicle parts on the intelligent guided vehicle according to the placement trajectory.

[0045] In this embodiment of the invention, a preset placement trajectory is obtained based on the category of vehicle components, a transformation matrix is ​​generated based on offset information, and the transformation matrix is ​​applied to each target point of the preset placement trajectory to obtain the placement trajectory.

[0046] Specifically, during the actual placement process, the control module can correct the preset placement trajectory in real time based on the offset information, ensuring that the placement action can be completed accurately and without error, greatly improving the precision and efficiency of placement, and effectively avoiding problems such as placement errors or product damage caused by position deviation.

[0047] As described above, this invention provides a visually guided part placement device and method based on feature recognition. By setting positioning marker components at the front and rear ends of an intelligent guided vehicle and using a 3D point cloud camera to collect the point cloud contours of the positioning posts, the deviation of the intelligent guided vehicle is calculated, achieving accurate identification of the intelligent guided vehicle's position. Furthermore, the positioning marker components used in this invention have higher imaging stability and recognition reliability, effectively avoiding the impact of welding slag accumulation on the recognition effect and reducing cleaning and maintenance costs. Simultaneously, the position and flatness of the positioning marker components can be easily adjusted through the adjustment mechanism of the positioning base, further improving the system's adaptability and stability. During the part placement process, the control module calculates offset information based on the target position information and preset position information, and generates a part placement trajectory, enabling the part placement robot to accurately place vehicle parts on the intelligent guided vehicle, thereby significantly improving the quality and efficiency of automated assembly in the automotive manufacturing industry.

[0048] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A feature recognition based vision guided put device, characterized in that, The utility model relates to a kind of vehicle component placing system, including: Tooling base plate, the tooling base plate is equipped with the piece placing station, the front end and rear end of the piece placing station are equipped with image acquisition module; Intelligent guided vehicle, travel on the tooling base plate, for handling vehicle components, the front end and rear end of the intelligent guided vehicle are equipped with positioning mark component, the positioning mark component includes positioning column, when the intelligent guided vehicle travels to the piece placing station, the positioning column is opposite with the image acquisition module, so that the image acquisition module acquires the target position information of the positioning column; Piece placing robot, the piece placing robot is arranged on the side of the tooling base plate, when the intelligent guided vehicle moves to the piece placing station, the piece placing robot places the vehicle component on the intelligent guided vehicle according to piece placing track; Control module, the control module is respectively connected with the intelligent guided vehicle, the image acquisition module and the piece placing robot, the control module is used to calculate offset information according to the target position information and preset position information, and generates the piece placing track according to the offset information, to control the piece placing robot piece.

2. The feature recognition based visual guidance putter device of claim 1, wherein, The positioning mark component includes a positioning base, a positioning bottom plate, and a preset number of positioning columns. The positioning bottom plate is detachably connected with the positioning base, and a plane on which the positioning bottom plate is located is perpendicular to a bottom surface of the tooling base plate. The positioning columns are fixedly connected with the positioning bottom plate, and an extension direction of the positioning columns is perpendicular to the positioning bottom plate.

3. The feature recognition based visual guidance putter device of claim 2, wherein, The positioning base includes an adjusting mechanism and an L-shaped frame. The adjusting mechanism includes a first screw, a second screw, and an adjusting bottom plate. The adjusting bottom plate is connected with a first end of the L-shaped frame based on the first screw. The adjusting bottom plate is connected with the positioning bottom plate based on the second screw. A second end of the L-shaped frame is detachably connected with the intelligent guided vehicle. The first screw is used to adjust a position of the positioning bottom plate in a first direction, and the first direction is the extension direction of the positioning columns. The second screw is used to adjust a position of the positioning bottom plate in a second direction, and the second direction is parallel to a plane on which the positioning bottom plate is located.

4. The feature recognition based visual guidance putter device of claim 3, wherein, The positioning mark component further includes a gasket. The gasket is arranged at a bottom of the second end of the L-shaped frame, and is used to adjust a height of the positioning mark component.

5. The feature recognition based visual guidance put-in-place device of any of claims 2-4, wherein, The positioning columns are in a cylindrical structure, and the preset number of the positioning columns is three.

6. A feature recognition based visual guidance placing method applied to the feature recognition based visual guidance placing device according to any one of claims 1-5, characterized in that, The method includes the following steps: In response to the intelligent guided vehicle traveling to the piece placing station of the tooling base plate, acquiring target position information of the positioning mark component based on the image acquisition module; Acquiring preset position information, and calculating offset information based on the preset position information and the target position information; The preset position information is used to represent a calibration position of the positioning mark component; Generating a piece placing track based on the offset information, and controlling the piece placing robot to place the vehicle component on the intelligent guided vehicle according to the piece placing track.

7. The feature recognition based visual guidance put-in method according to claim 6, characterized in that, Before the step of acquiring the position information of the positioning mark component based on the image acquisition module, the method further includes the following steps: Determining positions of the positioning columns in the positioning mark component based on a preset tooling number model; Adjust the position of the positioning mark assembly so that the flatness of the positioning base plate in the positioning mark assembly is the same as the flatness of the tooling base plate.

8. The feature recognition based visual guidance put-in-place method of claim 7, wherein, The target position information of the positioning mark assembly is obtained based on the image acquisition module, and the image acquisition module comprises: The image acquisition module is a 3D point cloud camera, and the image acquisition module emits a laser line to the positioning column of the positioning mark assembly; The image acquisition module acquires the light spot image on the surface of the positioning column; Based on the triangulation algorithm, the three-dimensional point cloud data of the light spot image is calculated to obtain the target position information.

9. The feature recognition based visual guidance put part method according to claim 6, characterized in that, The offset information is calculated based on the preset position information and the target position information, and the offset information comprises: The target position information is aligned with the preset position information in coordinates; The translation vector and the rotation matrix between the target position information and the preset position information are calculated; Based on the translation vector and the rotation matrix, the actual pose deviation of the intelligent guide vehicle at the part placing station is calculated to generate the offset information, and the offset information comprises a position offset and an angle offset.

10. The feature recognition based visual guidance put-in-place method of claim 9, wherein, The part placing trajectory is generated based on the offset information, and the part placing trajectory comprises: A preset part placing trajectory is obtained based on the category of the vehicle component; A transformation matrix is generated based on the offset information; The transformation matrix is applied to each target point of the preset part placing trajectory to obtain the part placing trajectory.

Images