CN114720476A - Intelligent detection and polishing method for defects of automobile body paint surface - Google Patents

Intelligent detection and polishing method for defects of automobile body paint surface Download PDF

Info

Publication number
CN114720476A
CN114720476A CN202210297231.8A CN202210297231A CN114720476A CN 114720476 A CN114720476 A CN 114720476A CN 202210297231 A CN202210297231 A CN 202210297231A CN 114720476 A CN114720476 A CN 114720476A
Authority
CN
China
Prior art keywords
polishing
vehicle body
defects
detection
camera
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202210297231.8A
Other languages
Chinese (zh)
Inventor
李煌
罗兴锋
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hunan Shibite Robot Co Ltd
Original Assignee
Hunan Shibite Robot Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hunan Shibite Robot Co Ltd filed Critical Hunan Shibite Robot Co Ltd
Priority to CN202210297231.8A priority Critical patent/CN114720476A/en
Publication of CN114720476A publication Critical patent/CN114720476A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B24—GRINDING; POLISHING
    • B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
    • B24B27/00—Other grinding machines or devices
    • B24B27/0084—Other grinding machines or devices the grinding wheel support being angularly adjustable
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B24—GRINDING; POLISHING
    • B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
    • B24B49/00—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation
    • B24B49/12—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation involving optical means
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B24—GRINDING; POLISHING
    • B24B—MACHINES, DEVICES, OR PROCESSES FOR GRINDING OR POLISHING; DRESSING OR CONDITIONING OF ABRADING SURFACES; FEEDING OF GRINDING, POLISHING, OR LAPPING AGENTS
    • B24B49/00—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation
    • B24B49/16—Measuring or gauging equipment for controlling the feed movement of the grinding tool or work; Arrangements of indicating or measuring equipment, e.g. for indicating the start of the grinding operation taking regard of the load
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8854—Grading and classifying of flaws
    • G01N2021/8861—Determining coordinates of flaws
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8854—Grading and classifying of flaws
    • G01N2021/8874—Taking dimensions of defect into account
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
    • G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques

Abstract

The invention discloses an intelligent detection and polishing method for defects of a paint surface of an automobile body, which comprises the following steps: after the vehicle body is conveyed to the detection station, triggering a camera, collecting vehicle body point cloud and calculating the vehicle body pose; acquiring point clouds and images of the surface of the vehicle body in a specified range step by step according to the pose of the vehicle body and an offline planned track; positioning the defects and judging the types; after the vehicle body is conveyed to a polishing station, vehicle body point cloud is collected, the vehicle body pose is calculated, and defect coordinates are converted; and guiding the grinding mechanical arm to reach the defect position according to the defect coordinate, and repairing the defect by the force-controlled grinding head. The invention has the advantages of simple and compact principle, low realization cost, high intelligent degree and the like.

Description

Intelligent detection and polishing method for defects of automobile body paint surface
Technical Field
The invention mainly relates to the technical field of automobile body production and manufacturing intelligence, in particular to an intelligent detection and polishing method for automobile body paint surface defects.
Background
The coating process is an important link in the production and manufacturing of automobile bodies, and improves the rust resistance and corrosion resistance of the automobile bodies and the ornamental value of automobiles by spraying paint on the surfaces of the automobile bodies. However, vehicle body painting is a high-precision and difficult work, and the quality of the process is affected by the paint, the painting environment, the setting of the painting parameters and the like.
In the actual production and manufacturing process, the painted surface of the automobile body inevitably suffers some fouling or damage, such as concave-convex points caused by impurities on the surface in spraying, paint scratch and stain caused by improper operation in transportation, and the like, and the existence of these appearance defects directly influences the brand image of manufacturers and the automobile sales. In contrast, detecting and repairing the defects of the paint surface of the vehicle body before delivery is a necessary part of the coating process.
The painted surface of the automobile body is made of special materials and has high light reflection characteristics, so that the detection and repair of the defects of the painted surface are challenging.
At present, the detection and repair of the defects of the paint surface of the automobile body are mainly completed manually. By means of a special light source, a worker detects the defects in a multi-view observation and touch mode, and records information such as types, sizes and positions of the defects; then, the repair tool is held by hand, and the paint surface defects are repaired through operations of grinding, polishing, spraying and the like. Although the traditional manual mode can meet the requirements, the method depends on the experience and concentration of workers seriously, and has the problems of low detection and repair efficiency and unstable quality. Furthermore, under the intensity of high concentration work for a long time, the quality of the inspection and repair work may continue to deteriorate.
With the continuous development of visual imaging and robot technology, a mode of detecting and repairing paint surface detection by a machine vision guide robot is possible. In recent years, methods have emerged that use 2D cameras to detect paint defects under specially tailored light sources. However, this type of method cannot achieve automatic grinding because three-dimensional information of defects cannot be acquired. Some 3D cameras based on the phase deflection principle are used for obtaining high-precision three-dimensional point clouds of mirror-like vehicle body paint surfaces, so that defects and three-dimensional coordinate information of the defects are effectively detected, but the position information of the defects is not transmitted to a defect repairing robot to achieve automatic repairing.
For example, a practitioner has proposed a device and a method for detecting defects in a painted surface of a vehicle body (application No. 202110117975.2) in chinese patent application, which includes a servo controller, an industrial personal computer, a vehicle body conveying mechanism, a plurality of area-array cameras, a plurality of projectors, and a plurality of projection screens, wherein the attitude relationship between the coordinate system of the above components and the world coordinate system is calibrated in advance by a calibration board. The area-array camera, the projector and the projection screen jointly form an image acquisition system for acquiring and analyzing surface data of a vehicle to be detected, the industrial personal computer is connected with the projector, the projection screen is arranged on the periphery of the projector and corresponds to the projector, the projector can respectively project 4 horizontal and vertical sinusoidal phase shift stripe images onto the corresponding projection screen, the area-array cameras are distributed around the vehicle to be detected, each area-array camera shoots the stripes on the projection screen through surface reflection of a vehicle body, 4 horizontal and vertical sinusoidal phase shift stripe images can be shot in a single view of the vehicle body, and the total number of the 4 sinusoidal phase shift stripe images is 8. By designing the servo controller, the industrial personal computer and the vehicle body conveying mechanism to be matched with the image acquisition system for use, the sinusoidal phase shift stripe image related to the surface of the vehicle body can be conveniently and efficiently acquired, so that a specular reflection image, a diffuse reflection image, a glossiness image, a curvature image and the like are output, and further, the smudgy defects, the texture defects, the scratch and bruise defects and the concave-convex defects of the paint surface of the vehicle body can be detected through a traditional image processing algorithm and a deep learning algorithm.
However, the technical scheme is quite complex, high in cost, narrow in application range and not high enough in detection precision, and most importantly, the traditional technology can only complete detection and cannot be automatically linked with stations such as subsequent repair and the like. Particularly, a stable and reliable mode is not available, so that the two work of detecting and repairing the paint surface defects of the automobile body can be linked accurately, and intelligent and automatic processing operation is formed.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: aiming at the technical problems in the prior art, the invention provides the intelligent detection and polishing method for the paint surface defects of the automobile body, which has the advantages of simple and compact principle, low implementation cost and high intelligent degree.
In order to solve the technical problems, the invention adopts the following technical scheme:
an intelligent detection and polishing method for defects of a paint surface of an automobile body comprises the following steps:
after the vehicle body is conveyed to the detection station, triggering a camera to collect vehicle body point cloud and calculating the vehicle body pose; acquiring point clouds and images of the surface of the vehicle body in a specified range step by step according to the pose of the vehicle body and the track planned off-line; positioning defects and judging types;
after the vehicle body is conveyed to a polishing station, vehicle body point cloud is collected, the vehicle body pose is calculated, and defect coordinates are converted; and guiding the grinding mechanical arm to reach the defect position according to the defect coordinate, and repairing the defect by the force-controlled grinding head.
As a further improvement of the process of the invention: and at the grinding station, converting the defects under the coordinate system of the imaging camera in the detection station into the base coordinate system of the robot in the grinding station through defect coordinate conversion.
As a further improvement of the process of the invention: after the vehicle body arrives at the detection station, the vehicle body point cloud is collected through the detection camera, the vehicle body pose is calculated, the mechanical arms are adjusted to arrive at the reference positions of the paint surface areas respectively governed by the mechanical arms, then the paint surface data are collected in a multi-beat collection mode, the defect types are detected and positioned in a mode of combining deep learning and feature matching, and meanwhile, the outline coordinates of the defects are calculated.
As a further improvement of the process of the invention: and the vehicle body is conveyed to a polishing station, vehicle body point clouds in a field of view are collected through a camera and are registered with the point clouds of a collection camera of a detection station at a reference position, and then vehicle body pose deviation is calculated, so that the deviation of defect coordinates under different coordinate systems of the detection station and the polishing station is estimated, and further the defect coordinates under the camera coordinate system of the detection station are converted into the camera coordinate system of the polishing station.
As a further improvement of the process of the invention: and calibrating the pose relation between the polishing station camera and the polishing mechanical arm through an offline hand eye, and converting the defect coordinate under the coordinate system of the polishing station camera into the corresponding base coordinate system of the mechanical arm.
As a further improvement of the process of the invention: when the vehicle body to be detected is conveyed to a detection station, vehicle body positioning and defect coordinate guiding are carried out to collect vehicle body point cloud, the vehicle body point cloud is used as input, and vehicle body pose is estimated through registration with a reference parking space point cloud or a CAD model, so that parking space positioning is realized.
As a further improvement of the process of the invention: based on the position and the posture of the vehicle body, the acquisition track of the off-line planning is adjusted, and paint surface data including 3D point cloud and 2D images generated based on reflection are acquired in a multi-beat mode through a paint surface imaging acquisition assembly.
As a further improvement of the method of the invention: and detecting the type and location of the defect in a mode of combining deep learning and feature matching, and calculating the outer contour coordinate, the mass center and the normal direction of the defect.
As a further improvement of the method of the invention: the detection station comprises more than two detection mechanical arms, more than two paint surface imaging acquisition assemblies, a coordinate guide acquisition assembly and a defect detection control assembly, wherein the defect detection control assembly comprises a detection mechanical arm controller, an acquisition assembly controller and an image processing unit; the detection mechanical arm controller is connected with the detection mechanical arm, and the acquisition component controller and the image processing unit are connected with the paint surface imaging acquisition component and the coordinate guiding acquisition component.
As a further improvement of the process of the invention: the automatic polishing machine comprises a polishing station, a polishing machine body and a polishing machine body, wherein the polishing station comprises a vehicle body positioning camera, more than two sets of polishing mechanical arms, more than two sets of force control polishing heads and a polishing control assembly, the polishing control assembly comprises a collection camera control and image processing unit, a polishing mechanical arm controller and a force control polishing head controller, the vehicle body positioning camera is connected with the collection camera control and image processing unit, the polishing mechanical arms are connected with the polishing mechanical arm controller, and the force control polishing heads are installed at the tail ends of the polishing mechanical arms and are connected with the force control polishing head controller.
Compared with the prior art, the invention has the advantages that:
1. the intelligent detection and polishing method for the paint surface defects of the automobile body is simple and compact in principle, low in implementation cost and high in intelligent degree, can solve the problem of conversion from local defect coordinates to a polishing system base coordinate system through a 3D imaging scheme of local positioning and global guiding, and meanwhile realizes automatic polishing and repairing of the defects by combining a robot motion control technology. The invention can accurately link the two work of detecting and repairing the paint surface defect of the automobile body in a stable and reliable mode to form intelligent and automatic processing operation.
2. According to the intelligent detection and polishing system and method for the defects of the paint surface of the automobile body, local camera defect coordinates are converted to a polishing robot base coordinate system in a mode of combining multi-camera off-line calibration and point cloud registration, and automatic detection and polishing of the defects of the paint surface are achieved.
3. According to the intelligent detection and polishing system and method for the paint surface defects of the automobile body, the pose of the mechanical arm is moved a priori, the pose of the mechanical arm is calibrated offline, the pose transformation relation among cameras in a multi-shot acquisition area is accurately calculated, and the paint surface coordinate point clouds acquired at different beats are unified or spliced.
Drawings
FIG. 1 is a schematic flow diagram of the process of the present invention.
Fig. 2 is a schematic diagram of the structural principle of the system constructed in a specific application example of the invention.
FIG. 3 is a flow chart of the present invention in an embodiment.
FIG. 4 is a schematic illustration of the present invention in a particular application for vehicle body finish area allocation.
Fig. 5 is a schematic diagram illustrating the principle of coordinate transformation in a specific application example of the present invention.
Illustration of the drawings:
1. a defect detection control component; 2. detecting the mechanical arm; 3. a painted surface imaging acquisition assembly; 4. a coordinate guide acquisition assembly; 5. a vehicle body to be detected; 6. a grinding control component; 7. polishing the mechanical arm; 8. force-controlled polishing heads; 9. a body-positioning 3D camera; 10. a defect detection station; 11. and (5) defect polishing stations.
Detailed Description
The invention will be described in further detail below with reference to the drawings and specific examples.
As shown in fig. 1, fig. 2, fig. 3 and fig. 5, the intelligent detection and polishing method for the defects of the paint surface of the automobile body comprises the following steps:
after the vehicle body is conveyed to the detection station, triggering a camera to collect vehicle body point cloud and calculating the vehicle body pose; acquiring point clouds and images of the surface of the vehicle body in a specified range step by step according to the pose of the vehicle body and the track planned off-line; positioning the defects and judging the types;
after the vehicle body is conveyed to a polishing station, vehicle body point cloud is collected, the vehicle body pose is calculated, and defect coordinates are converted; and guiding the grinding mechanical arm to reach the defect position according to the defect coordinate, and repairing the defect by the force-controlled grinding head.
In a specific application example, at the grinding station, the defects under the coordinate system of the imaging camera in the detection station are converted into the base coordinate system of the robot in the grinding station through defect coordinate conversion.
In a specific application example, after a vehicle body reaches a detection station, a detection camera is used for collecting point cloud of the vehicle body and calculating the position and posture of the vehicle body, so that a mechanical arm is adjusted to reach a reference position of a paint surface area in the respective jurisdiction, paint surface data are collected in a multi-beat collection mode, defect types and positioning are detected in a mode of combining deep learning and feature matching, and meanwhile, the outline coordinates of the defects are calculated.
In a specific application example, a vehicle body is conveyed to a polishing station, vehicle body point clouds in a visual field are collected through a camera and are registered with point clouds of a collection camera of a detection station at a reference position, then vehicle body pose deviation is calculated, deviation of defect coordinates under different coordinate systems of the detection station and the polishing station is estimated, and then the defect coordinates under a camera coordinate system of the detection station are converted into those under a camera coordinate system of the polishing station.
In a specific application example, the pose relationship between the grinding station camera and the grinding mechanical arm is calibrated through an off-line hand eye, and the defect coordinate in the coordinate system of the grinding station camera is converted into the corresponding coordinate system of the mechanical arm base.
In a specific application example, when the vehicle body to be detected is conveyed to a detection station, vehicle body positioning and defect coordinate guidance are carried out to acquire vehicle body point cloud, the vehicle body point cloud is used as input, and vehicle body pose is estimated through registration with a reference parking space point cloud or a CAD model, so that parking space positioning is realized.
In a specific application example, the off-line planned acquisition track is adjusted based on the pose of the vehicle body, and paint data including 3D point cloud and 2D images generated based on reflection are acquired in a multi-beat mode through the paint imaging acquisition assembly.
In a specific application example, the defect type and the defect location are detected in a mode of combining deep learning and feature matching, and the outline coordinates, the mass center and the normal direction of the defect are calculated.
In a specific application example, the detection station comprises more than two sets of detection mechanical arms, more than two paint surface imaging acquisition assemblies, a coordinate guide acquisition assembly and a defect detection control assembly, wherein the defect detection control assembly comprises a detection mechanical arm controller, an acquisition assembly controller and an image processing unit; the detection mechanical arm controller is connected with the detection mechanical arm, and the acquisition component controller and the image processing unit are connected with the paint surface imaging acquisition component and the coordinate guiding acquisition component.
In a specific application example, the polishing station comprises a vehicle body positioning camera, more than two sets of polishing mechanical arms, more than two sets of force-controlled polishing heads and a polishing control assembly, the polishing control assembly comprises a collection camera control and image processing unit, a polishing mechanical arm controller and a force-controlled polishing head controller, the vehicle body positioning camera is connected with the collection camera control and image processing unit, the polishing mechanical arms are connected with the polishing mechanical arm controller, and the force-controlled polishing heads are arranged at the tail ends of the polishing mechanical arms and connected with the force-controlled polishing head controller.
Referring to fig. 2, taking a specific application of the present invention as an example, the present invention is applied as follows:
at the defect detection station 10, a defect detection unit is included for positioning defects and judging the types of the defects; the defect detection unit comprises more than two sets of detection mechanical arms 2, more than two paint surface imaging acquisition assemblies 3, a coordinate guide acquisition assembly 4 and a defect detection control assembly 1, wherein the defect detection control assembly 1 comprises a detection mechanical arm controller, an acquisition assembly controller and an image processing unit; the detection mechanical arm controller is connected with the detection mechanical arm 2, and the acquisition component controller and the image processing unit are connected with the paint surface imaging acquisition component 3 and the coordinate guiding acquisition component 4.
At defect polishing station 11, including automobile body location camera 9(3D camera), the arm 7 of polishing more than two sets, the power control of two sets of more than polishing head 8 and the control assembly 6 of polishing, the control assembly 6 of polishing is including gathering camera control and image processing unit, the arm controller of polishing and the power control of polishing head controller, automobile body location camera 9 links to each other with gathering camera control and image processing unit, the arm 7 of polishing is connected with the arm controller of polishing, the power control is polished 8 and is installed at the end of the arm 7 of polishing to be connected with the power control of polishing head controller.
After the car body is conveyed to the polishing station, the car body positioning camera 9 is triggered to collect car body point cloud, the image processing unit calculates the pose of the car body and converts defect coordinates, the controller guides the polishing mechanical arm 7 to reach the defect position according to the defect coordinates, and the force control polishing head 8 repairs the defects.
In this embodiment, four sets of detection mechanical arms 2 are provided; the painted surface imaging acquisition assembly 3 is four, a 3D camera for painted surface imaging is adopted, and the coordinate guiding acquisition assembly 4 is a 3D camera for vehicle body positioning and defect coordinate guiding.
And after the vehicle body is conveyed to the detection station, the camera is triggered to collect the point cloud of the vehicle body and calculate the pose of the vehicle body. According to the pose of the vehicle body, four paint surface imaging acquisition assemblies 3 (imaging cameras) which are respectively fixed at the tail ends of the detection mechanical arms 2 reach the appointed positions under the guidance of a detection mechanical arm controller, point clouds and images of the surface of the vehicle body in the appointed range are acquired step by step according to offline planned tracks under the action of the acquisition assembly controller and an image processing unit, and then the defects are positioned through the image processing unit and the types of the defects are judged.
A plurality of detection mechanical arms 2 are respectively managed in different areas (such as 4 areas), and because the field of view of a camera is limited, each area needs to be shot for multiple times to acquire the paint surface data of the whole vehicle body. As shown in FIG. 4, in the respective inspection robot arm 2 jurisdictionsThe region, the painted surface, is divided into a plurality of regions (specifically, depending on the area of the designated region and the field of view of the camera). During collection, the detection mechanical arm 2 moves to the reference area ArMoving to A according to the off-line planned path under the action of the mechanical arm controlleriAnd (i is 1,2, …, 8), completing multi-shot acquisition of the camera.
In this embodiment, the car body positioning camera 9 is a single 3D camera, but it should be understood that in other embodiments, there may be a plurality of cameras according to the length and width of the car body and the actual positioning requirement. Four sets of grinding mechanical arms 7 are provided, and correspondingly, four sets of force-controlled grinding heads 8 are provided.
In this embodiment, at the polishing station 11, according to the defect coordinate, the polishing robot controller guides the polishing robot arm 7 to reach the defect position, and the defect is polished and repaired by the force-controlled polishing head 8 under the action of the force-controlled polishing controller.
In the detection station 10, the system comprises 4 mechanical arms Ri(i ═ 1,2,3,4), 4-stage paint surface imaging camera BiAnd a controller, wherein the camera BiIs arranged on the mechanical arm RiThe execution tail end, the mechanical arm and the camera are connected with the controller; in the polishing station, the system consists of 4 mechanical arms Ri4 force-controlled polishing heads and 4 positioning cameras B ″iAnd the controller is formed, the polishing head is arranged at the execution tail end of the mechanical arm, and the camera is fixedly arranged beside the appointed working area so as to position the vehicle body and polish and guide. The concrete implementation is as follows: after the automobile body arrives at the detection parking space, the imaging camera BiThe method comprises the steps of collecting vehicle body point cloud, calculating vehicle body pose, adjusting mechanical arms to reach reference positions of paint surface areas respectively governed by the mechanical arms, collecting paint surface data in a multi-beat collection mode, detecting defect types and positioning in a deep learning and feature matching combined mode, and calculating outline coordinates of defects. After the process is finished, the vehicle body is conveyed to a grinding station BiThe camera collects the point cloud of the vehicle body in the view field and is connected with the point cloud BiPoint cloud registration at the reference position, and then calculating the pose deviation of the vehicle body, thereby estimating the defect coordinates at the camera BiB' with cameraiDeviation in the coordinate system, and further BiDefect coordinate conversion to B' under camera coordinate systemiUnder the camera coordinate system. Finally, off-line hand-eye calibration B ″iThe pose relation between the camera and the mechanical arm can change the B 'into the B' stateiDefect coordinates under camera coordinate system are converted into corresponding mechanical arm R ″iUnder a base coordinate system. Based on this, the controller reaches the defect position through guiding the arm, controls the head at power control controller finger gravitation and repairs the defect.
The above are only preferred embodiments of the present invention, and the scope of the present invention is not limited to the above examples, and all technical solutions that fall under the spirit of the present invention belong to the scope of the present invention. It should be noted that modifications and embellishments within the scope of the invention may be made by those skilled in the art without departing from the principle of the invention.

Claims (10)

1. The intelligent detection and polishing method for the defects of the paint surface of the automobile body is characterized by comprising the following steps of:
after the vehicle body is conveyed to the detection station, triggering a camera to collect vehicle body point cloud and calculating the vehicle body pose; acquiring point clouds and images of the surface of the vehicle body in a specified range step by step according to the pose of the vehicle body and the track planned off-line; positioning defects and judging types;
after the vehicle body is conveyed to a polishing station, vehicle body point cloud is collected, the vehicle body pose is calculated, and defect coordinates are converted; and guiding the grinding mechanical arm to reach the defect position according to the defect coordinate, and repairing the defect by the force-controlled grinding head.
2. The method of claim 1, wherein at the polishing station, the defects in the imaging camera coordinate system at the inspection station are transformed to the robot-based coordinate system at the polishing station by defect coordinate transformation.
3. The method for intelligently detecting and polishing the paint surface defects of the automobile body according to claim 2, is characterized in that after the automobile body reaches a detection station, a detection camera is used for collecting point cloud of the automobile body and calculating the pose of the automobile body, so that mechanical arms are adjusted to reach reference positions of respective paint surface regions, then paint surface data are collected in a multi-shot collection mode, the types and the positions of the defects are detected in a mode of combining deep learning and feature matching, and meanwhile, the outline coordinates of the defects are calculated.
4. The method as claimed in claim 3, wherein the vehicle body is transported to the polishing station, the point cloud of the vehicle body in the field of view is collected by the camera and registered with the point cloud of the collection camera of the inspection station at the reference position, and then the vehicle body pose deviation is calculated, so as to estimate the deviation of the defect coordinates under the different coordinate systems of the inspection station and the polishing station, and further convert the defect coordinates under the camera coordinate system of the inspection station to under the camera coordinate system of the polishing station.
5. The method as claimed in claim 4, wherein the pose relationship between the polishing station camera and the polishing mechanical arm is calibrated by off-line hand eyes, and the defect coordinates in the coordinate system of the polishing station camera are converted to the corresponding coordinate system of the mechanical arm base.
6. The method for intelligently detecting and polishing the paint defects of the automobile body according to any one of claims 1 to 5, wherein when the automobile body to be detected is conveyed to a detection station, the automobile body is positioned and the defect coordinates guide the collection of the point cloud of the automobile body, and the point cloud of the automobile body is used as input to estimate the position of the automobile body by registering the point cloud of the reference parking space or the point cloud of the CAD model, so that the parking space is positioned.
7. The intelligent detection and polishing method for the defects of the paint surface of the automobile body according to any one of claims 1-5, characterized in that an off-line planned acquisition track is adjusted based on the pose of the automobile body, and paint surface data including 3D point cloud and 2D image generated based on reflection are acquired in a multi-beat manner through a paint surface imaging acquisition assembly.
8. The intelligent detection and polishing method for the defects of the paint surfaces of the automobile bodies as claimed in any one of claims 1 to 5, wherein the types and positions of the defects are detected in a mode of combining deep learning and feature matching, and the coordinates, the mass center and the normal direction of the outer contour of the defects are calculated.
9. The intelligent detection and polishing method for the defects of the paint surface of the automobile body according to any one of claims 1-5, characterized in that a detection station comprises more than two sets of detection mechanical arms, more than two paint surface imaging acquisition assemblies, a coordinate guide acquisition assembly and a defect detection control assembly, wherein the defect detection control assembly comprises a detection mechanical arm controller, an acquisition assembly controller and an image processing unit; the detection mechanical arm controller is connected with the detection mechanical arm, and the acquisition component controller and the image processing unit are connected with the paint surface imaging acquisition component and the coordinate guiding acquisition component.
10. The intelligent detection and polishing method for the defects of the paint surface of the automobile body according to any one of claims 1 to 5, characterized in that a polishing station comprises an automobile body positioning camera, more than two sets of polishing mechanical arms, more than two sets of force-controlled polishing heads and a polishing control assembly, the polishing control assembly comprises an acquisition camera control and image processing unit, a polishing mechanical arm controller and a force-controlled polishing head controller, the automobile body positioning camera is connected with the acquisition camera control and image processing unit, the polishing mechanical arms are connected with the polishing mechanical arm controller, and the force-controlled polishing heads are installed at the tail ends of the polishing mechanical arms and connected with the force-controlled polishing head controller.
CN202210297231.8A 2022-03-24 2022-03-24 Intelligent detection and polishing method for defects of automobile body paint surface Pending CN114720476A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210297231.8A CN114720476A (en) 2022-03-24 2022-03-24 Intelligent detection and polishing method for defects of automobile body paint surface

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210297231.8A CN114720476A (en) 2022-03-24 2022-03-24 Intelligent detection and polishing method for defects of automobile body paint surface

Publications (1)

Publication Number Publication Date
CN114720476A true CN114720476A (en) 2022-07-08

Family

ID=82239246

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210297231.8A Pending CN114720476A (en) 2022-03-24 2022-03-24 Intelligent detection and polishing method for defects of automobile body paint surface

Country Status (1)

Country Link
CN (1) CN114720476A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116576787A (en) * 2023-04-14 2023-08-11 南京航空航天大学 Gap surface difference measurement method and measurement device based on mechanical arm cooperation

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116576787A (en) * 2023-04-14 2023-08-11 南京航空航天大学 Gap surface difference measurement method and measurement device based on mechanical arm cooperation
CN116576787B (en) * 2023-04-14 2024-01-30 南京航空航天大学 Gap surface difference measurement method and measurement device based on mechanical arm cooperation

Similar Documents

Publication Publication Date Title
CN114720475A (en) Intelligent detection and polishing system and method for automobile body paint surface defects
CN107884419B (en) Automatic detection equipment for automobile chassis and intelligent detection system for automobile
CN110524582B (en) Flexible assembly welding robot workstation
CN109671123B (en) Sole glue spraying equipment and method based on monocular vision
US10875592B2 (en) Automobile manufacturing plant and method
CN111192307A (en) Self-adaptive deviation rectifying method based on laser cutting of three-dimensional part
CN109483539A (en) Vision positioning method
CN110881748A (en) Robot sole automatic gluing system and method based on 3D scanning
CN109127194A (en) Automobile gluing vision follows detection method
US20210255117A1 (en) Methods and plants for locating points on complex surfaces
CN109731708A (en) Auto repair auto spray painting method based on image recognition
CN112010024A (en) Automatic container grabbing method and system based on laser and vision fusion detection
CN112319429B (en) Control system and control method for automatically cleaning front windshield of bullet train
CN114720476A (en) Intelligent detection and polishing method for defects of automobile body paint surface
CN111215800B (en) Maintenance amount detection device and detection method for welding maintenance robot
CN114654465A (en) Welding seam tracking and extracting method based on line laser structure optical vision sensing
CN115343289A (en) Automatic scratch detection system and method for whole automobile assembly pit package
CN112058559B (en) Coating robot and vehicle body coating method
CN113245094A (en) Robot spraying system and method for automobile brake drum
CN214583043U (en) Three-dimensional scanning system for workpiece coating
CN114434036B (en) Three-dimensional vision system for gantry robot welding of large ship structural member and operation method
CN204288242U (en) Based on the Control During Paint Spraying by Robot trajectory extraction device that curved three-dimensional is rebuild
CN110160455A (en) Clearance surface difference detection system
CN115318762A (en) Laser cleaning method and cleaning device for complex structure surface
CN116945005A (en) Automobile paint defect detection and polishing method and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination