WO2023178713A1 - 预制构件拼装面智能扫描和误差自动标识系统及方法 - Google Patents

预制构件拼装面智能扫描和误差自动标识系统及方法 Download PDF

Info

Publication number
WO2023178713A1
WO2023178713A1 PCT/CN2022/083579 CN2022083579W WO2023178713A1 WO 2023178713 A1 WO2023178713 A1 WO 2023178713A1 CN 2022083579 W CN2022083579 W CN 2022083579W WO 2023178713 A1 WO2023178713 A1 WO 2023178713A1
Authority
WO
WIPO (PCT)
Prior art keywords
measurement
scanning
point
boundary
component
Prior art date
Application number
PCT/CN2022/083579
Other languages
English (en)
French (fr)
Inventor
杨秀仁
林放
廖翌棋
黄美群
李天升
彭智勇
Original Assignee
北京城建设计发展集团股份有限公司
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 北京城建设计发展集团股份有限公司 filed Critical 北京城建设计发展集团股份有限公司
Priority to JP2023523264A priority Critical patent/JP2024519633A/ja
Publication of WO2023178713A1 publication Critical patent/WO2023178713A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/30—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces
    • G01B11/303—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces using photoelectric detection means
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G01B11/25—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object
    • G01B11/2518—Projection by scanning of the object
    • 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
    • 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
    • G01N2021/8883—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 involving the calculation of gauges, generating models
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Definitions

  • the invention relates to the technical field of intelligent scanning of prefabricated components of prefabricated structures, and in particular to a system and method for intelligent scanning and automatic error identification of assembly surfaces of prefabricated components.
  • prefabricated above-ground structures such as box beams and T-beams
  • most prefabricated components do not need to be assembled, and the requirements for assembly quality are not high.
  • the requirements for production accuracy can be appropriately relaxed, and traditional methods such as direct measurement, Manual measurement methods such as rulers and feeler gauges can meet the requirements for measuring the manufacturing accuracy of prefabricated components.
  • the prefabricated components usually have the characteristics of large volume, thick structure, different and irregular shapes.
  • prefabricated structures have high requirements for waterproofing, joint width, and load loading.
  • the quality of prefabricated components has a significant impact on the construction technology, assembly quality, and waterproof performance of prefabricated structures.
  • the unevenness of the assembly surface of prefabricated components will cause the assembly of this surface to be out of synchronization, and locally large pressurized loads may easily cause damage to the components.
  • the precast component production department is paying more and more attention to the production quality of precast components, and quality inspection is required before components leave the factory.
  • the conventional prefabricated component measurement method using fixtures and rulers cannot be fully implemented on large components.
  • There are often problems such as the insufficient size of traditional calipers and the inability to measure the component belly, which cannot be fully detected by manual operations, and the measurement accuracy and purpose cannot be achieved.
  • the above-mentioned prefabricated structures have high requirements for the flatness of the assembly surfaces of prefabricated components, current technology has no way to accurately measure the manufacturing accuracy of the assembly surfaces of large components.
  • early large-scale prefabricated underground structure prefabricated components were deformed due to mold deformation after leaving the factory.
  • Manufacturing errors in the assembly surface of prefabricated components mainly refer to errors that affect the assembly of prefabricated structures, such as bulging areas of components, which can easily lead to problems such as lax joints and loose tension on the assembly surface. Although there are local bulges on the assembly surface, the edge contour of the entire component splicing surface remains accurate. This can be used as the benchmark condition and basic step for intelligent scanning to further determine whether the component is bulging and accurately find the bulging area. Due to the characteristics of large-scale prefabricated components such as large volume, heavy weight, and irregular boundary contours, it is impossible to accurately control the hoisting and placement angle of the components on the intelligent scanning platform.
  • the surface of the precast concrete components may appear pitted or pockmarked. Honeycomb, holes, particles and other phenomena. This type of defect is different from the manufacturing error of the assembly surface of concrete prefabricated components of prefabricated structures. It does not affect the fit and splicing of prefabricated components at the assembly site of the prefabricated structure and is not hindered in use. However, it requires the use of intelligent scanning to detect the assembly surface of prefabricated components.
  • the detection device may be placed outdoors. Affected by temperature changes, humidity changes and other factors, it may cause problems such as deformation of the large-scale scanning device, resulting in errors in the measurement results under different climatic conditions, making it impossible to Meet the needs of practical on-site applications. It is necessary to develop special testing equipment that can detect the manufacturing accuracy of the assembly surface of prefabricated components of large-scale prefabricated structures and can adapt to different environmental conditions without affecting the equipment detection accuracy.
  • the designer of the present invention has researched and designed a system and method for intelligent scanning and automatic error identification of prefabricated component assembly surfaces through painstaking research and design, combined with long-term experience and achievements in related industries. Overcome the above shortcomings.
  • the object of the present invention is to provide a system and method for intelligent scanning and automatic error identification of prefabricated component assembly surfaces, which has a simple structure and is easy to operate. It can effectively overcome the shortcomings of the existing technology and automatically identify and scan prefabricated components with high accuracy. Faster and more innovative.
  • the present invention discloses a method for intelligent scanning and error automatic identification of manufacturing accuracy of prefabricated component assembly surfaces, which is characterized by including the following steps:
  • Step 1 Preparation step: hoist the target scanning component to the measurement area on the component scanning platform;
  • Step 2 Start the step, open the electric rolling shutter door of the warehouse, and start the scanning gantry to open the scanning detection;
  • Step 3 Adjust the height of the vertical traveling tray so that the light of the laser measurement device is close to the height of the surface of the component scanning table;
  • Step 4 The scanning gantry walks forward from the warehouse along the horizontal walking track. During the walking process, the laser measurement value of the laser range finder and the scanning gantry temperature and humidity sensor, vertical track temperature and humidity sensor, and horizontal track temperature and humidity sensor are collected in real time. monitoring value;
  • Step 5 Collect the laser measurement value of the laser rangefinder in real time while walking, and determine whether the light spot of the laser rangefinder hits the target scanning component. If so, stop the horizontal advancement of the scanning gantry;
  • Step 6 Determine the boundary contour of the measurement surface in the target scanning component, and all boundary points form the boundary point data set Q 0 ;
  • Step 7 Complete the data collection of the measurement surface and form the laser point cloud data set Q 1 ;
  • Step 8 Correct the placement posture of the target scanning component, and correct the target scanning component to the ideal placement posture
  • Step 9 Analyze the measurement data of the measurement surface of the target scanning component after correction, and eliminate abnormal measurement values of the measurement surface of the target scanning component;
  • Step 10 Determine the manufacturing accuracy of the assembly surface of the target scanning component, and spray mark the error area on the surface of the target scanning component;
  • Step 11 Generate an inspection report and complete all inspections of the target scanned components.
  • step five is as follows: when the laser rangefinder light spot hits the target scanning component, the measurement value of the left laser rangefinder L l + the measurement value of the right laser rangefinder L r ⁇ L, the detection is determined Scan component 2 to the target, and record the current measurement values of the two laser rangefinders as and
  • step 6 the stop point of step 5 is used as the starting point, and the roundabout half and stepwise approximation methods are used to search for the boundary point in the horizontal direction of the target scanning component corresponding to the current height, which is recorded as the first boundary point P 0 of the target scanning component.
  • the forward direction of the horizontal walking track is the positive direction of the x-axis
  • the upward direction of the vertical walking track is the positive direction of the y-axis
  • the measurement value of the laser rangefinder is the z-axis coordinate value
  • the first boundary point P 0 of the target scanning component is As the coordinate origin, an xyz coordinate system is established.
  • the remaining boundary points of the perimeter contour of the measurement surface of the target scanning component are captured in the clockwise direction.
  • the theoretical data model of the target scanning component is combined with the rapid iteration method to continuously Reduce the number of searches for subsequent boundary points, speed up the capture of remaining boundary points, and finally form all boundary points into a boundary point data set Q 0 .
  • Step 6.1 Every time the scanning gantry takes a step forward, the measurement values of the laser rangefinders on both sides of the current position are collected and recorded as and Calculate the change rate of the laser rangefinder measurement values on both sides of the current position and the previous position.
  • the change rate of the measurement value on the left side is The change rate of the measured value on the right side is Compared and size:
  • the scanning gantry 1 continued to walk forward horizontally along the horizontal walking track 5 with a step length S until
  • Step 6.2 Move the laser rangefinder corresponding to the target measurement surface in Step 6.1 to continue walking one step upward along the vertical walking track with a step length S, and determine the rate of change of the measurement value in the y-axis direction;
  • Step 6.3 The mobile laser rangefinder returns to the first measurement point in the current measurement plane, and starts to acquire the first scanning boundary point P 0 in a roundabout way;
  • Step 6.4 After obtaining the first scanning boundary point P 0 , take P 0 as the origin of the coordinates, take the horizontal forward direction of the scanning gantry as the positive direction of the x-axis, take the vertical upward direction of the laser rangefinder scanning as the positive direction of the y-axis, and take The measurement value of the laser rangefinder is the z-axis coordinate value, establish the xyz coordinate system, and obtain the coordinates of the P 0 point (x 0 , y 0 , z 0 ), that is, (0,0,0);
  • Step 6.5 Based on the first scanned boundary point, perform subsequent boundary point capture and complete the capture of all boundary points of the target scanned component.
  • step eight The specific steps of step eight are as follows:
  • Step 8.1 Place the theoretical model in the coordinate system established in step 6, and make the left corner point P′ 0 at the bottom of the measurement surface of the theoretical model coincide with the P 0 point on the measurement surface of the target scanning component.
  • the bottom boundary of the measurement surface of the theoretical model coincides with x
  • the axes coincide, the left boundary of the theoretical model measurement surface coincides with the y-axis, and the theoretical model measurement surface data set Q′ 1 is obtained;
  • Step 8.2 Couple the measurement surface in the actual placement with the measurement surface of the theoretical model, complete the attitude inversion correction of the target scanning component, and obtain the corrected laser point cloud data set Q 1_3 and boundary point data set Q 0_3 .
  • step nine the specific steps of step nine are as follows:
  • Step 9.1 Traverse the corrected laser point cloud data set Q 1_3 and boundary point data set Q 0_3 , and compare the measured values in the laser point cloud data set Q 1_3 and boundary point data set Q 0_3 with the theoretical model data set Q′ 1 Whether the measured values of the same xy coordinate position are the same, the coordinate points whose difference exceeds the minimum accuracy of measurement points ⁇ S 0 are stored in the abnormal point data set Q ⁇ 0 ;
  • Step 9.2 Exclude reasonable abnormal points in the abnormal point data set Q ⁇ 0 , and define the abnormal point data set after eliminating reasonable abnormal points as Q ⁇ 1 ;
  • Step 9.3 Traverse the abnormal point data set Q ⁇ 1 after eliminating reasonable abnormal points, take out the measurement points in the data set one by one, and take the current point as the center, use the correlation search method to compare the current point with the surrounding measurement points on the measurement surface. Measurement value relationship; obtain the measurement values of all measurement points within the 5*5 point area of the current point in the corrected laser point cloud data set Q 1_3 , and compare the measurement values of the measurement points in the surrounding area with the measurement values of the current measurement point in sequence In comparison, if the measurement values of the surrounding measurement points are all greater or smaller than the measurement value of the current measurement point, the current measurement point is determined to be a local defect point.
  • an intelligent scanning and automatic error identification system for the assembly surface of prefabricated components, which includes a workbench, a warehouse, a scanning gantry, a horizontal walking track and a component scanning platform, and is characterized by:
  • the warehouse is located at one end of the workbench and is provided with a storage space for a scanning gantry and an electric gate.
  • the horizontal walking track is two tracks and extends along both sides of the workbench, with one end extending into the warehouse.
  • the frame scans
  • the platform is located on the workbench and is set in the middle of the horizontal walking track for placement of target scanning components.
  • the scanning gantry is slidably set on the horizontal walking track.
  • the scanning gantry is a door-shaped walking mechanism that walks on the scanning components.
  • On the horizontal walking rails on both sides of the platform there is a vertical walking rail on both inner sides of the scanning gantry, and there is a vertical walking tray on each vertical walking rail to move up and down along the vertical walking rail.
  • the walking pallet is equipped with mounting bases for laser measurement equipment and telescopic arms of the point-drawing inkjet printing machinery.
  • the laser measurement equipment is fixed on the upper end of the mounting base and is a single-point laser range finder
  • the point-drawing mechanical telescopic arm is fixed on the lower end of the mounting base and can be controlled to print color spots on the designated surface through telescopic control.
  • a scanning gantry bottom beam on each side of the bottom of the scanning gantry, and a horizontal walking limiter is provided at both ends of the scanning gantry bottom beam to limit the traveling range of the scanning gantry on the horizontal walking track.
  • the middle area of the scanning gantry bottom beam is equipped with a horizontal traveling motor and a horizontal traveling encoder to drive the scanning gantry forward and backward and record the position of the scanning gantry.
  • a vertical travel motor and a vertical travel encoder are provided at the lower end of the connection point between the vertical travel track and the scanning gantry bottom beam to drive the vertical travel pallet to move up and down and record the position of the vertical travel pallet.
  • the gantry adopts a gantry-type walking structure.
  • the gantry is loaded with multi-directional moving motors and tracks, which can complete the measurement of any shape of measurement surface.
  • the gantry is also equipped with various sensors to detect the tracks, gantry and air.
  • the temperature and humidity are monitored in real time, and the measurement results are corrected through the temperature and humidity monitoring results to ensure that the system can work under complex climate conditions without affecting accuracy.
  • the measurement surface of the target scanning component may have various factors such as concave and convex grooves, rubber channels, bubble pits, stone particles, slight edge damage, etc. that affect the production accuracy of the scan.
  • This patent is based on The theoretical model of the target scanning component is used as a reference.
  • Various influencing factors must be processed from the data, and the measurement data must also be processed using algorithms.
  • algorithmic methods such as correlation search algorithm, theoretical model reference method, and overall trend deduction method to eliminate influencing factors step by step to complete the production accuracy of the assembly surface of prefabricated components. calculation.
  • the system calculates the coordinate data set of the boundary points of the bulge area, and controls the gantry to carry the inkjet printing device to the corresponding position of the boundary point of each bulge area one by one, and inkjet the boundary points of the bulge on the surface of the component. It is convenient for workers to compare the inspection report and quickly find the location of the bulge during later polishing and repair, which greatly reduces the difficulty and workload of polishing.
  • Figure 1 shows a schematic diagram of the intelligent scanning and automatic error identification system for prefabricated component assembly surfaces of the present invention.
  • Figure 2 shows a perspective view of the scanning gantry of the present invention.
  • Figure 3 shows a side view of the scanning gantry of the present invention.
  • Figure 4 shows a front view along the horizontal traveling direction of the scanning gantry during the scanning process of the target scanning component of the present invention.
  • Figures 5A, 5B and 5C show examples of applications of the present invention to various types of special-shaped prefabricated components.
  • Figure 6 shows the position of the target scanning component of the present invention in the coordinate system in the ideal posture.
  • Figure 7 shows a bitmap of intensive scanning data points of the target scanning component of the present invention.
  • Figure 8 shows a schematic diagram of the position of the target measurement surface in the coordinate system after being corrected to the xy-plane and translated.
  • Figure 9 shows a front view of the dotting point position and the measurement surface during the process of determining the target measurement surface according to the present invention.
  • Figure 10 shows a schematic diagram of measurement when the target scanning component of the present invention is extremely biased to one side.
  • Figures 11A and 11B respectively show a top view of the ideal placement and extreme tilt placement of the target scanning component of the present invention in the measurement area.
  • Figure 12A, Figure 12B and Figure 12C respectively show the target scanning component of the present invention during the laser rangefinder scanning process, under different placement posture scenarios, from the laser rangefinder light spot not hitting the component to the laser measurement on both sides.
  • Figure 13 shows a schematic diagram of the relationship between the projection positions of actual boundary points and estimated boundary points on the xy-plane during the process of capturing boundary points in the present invention.
  • Figure 14 shows a flow chart of the method of the present invention.
  • the prefabricated component assembly surface intelligent scanning and error automatic identification system includes a workbench, a warehouse 4, a target scanning component 2, intelligent scanning equipment, a horizontal walking track 5 and a component scanning platform 3.
  • the warehouse 4 is located at One end of the workbench can accommodate the scanning gantry 1 of the intelligent scanning equipment and is equipped with an electric gate.
  • the horizontal walking track 5 is two parallel tracks and extends along both sides of the workbench. One end of the horizontal walking track 5 extends to In the warehouse 4, the frame scanning platform 3 is located on the workbench and is set in the middle of the horizontal walking track 5 for placing the target scanning component 2.
  • the scanning gantry 1 is set on the horizontal walking track 5 and can walk along the horizontal track.
  • the scanning gantry 1, horizontal walking track 5 and component scanning platform 3 of the intelligent scanning equipment are often set up in an open-air environment where components are stacked.
  • the scanning gantry 1 is a precision The equipment cannot be exposed to wind and rain for a long time. When it is not working, it needs to be placed in the warehouse 4 for protection. When working, open the electric gate of the warehouse 4, and the scanning gantry 1 can automatically drive out of the warehouse 4 and start working.
  • the scanning gantry 1 is a door-shaped walking mechanism, which runs on the horizontal walking rails 5 on both sides of the component scanning platform 3. See Figures 2, 3 and 4.
  • the two inner sides of the scanning gantry 1 each have A vertical traveling track 6 has a vertical traveling tray 8 on each of the vertical traveling tracks 6.
  • the vertical traveling trays 8 can move up and down along the vertical traveling track 6.
  • the vertical traveling pallets 8 are provided with A mounting base for a laser rangefinder 7 and a dot-drawing inkjet printing mechanical telescopic arm 9.
  • the laser rangefinder 7 is fixed on the upper end of the mounting base
  • the dot-drawing inkjet printing mechanical telescopic arm 9 is fixed on the lower end of the mounting base and is telescopic. Inkjet color spots, etc. on the designated surface.
  • each scanning gantry bottom beam 101 there is a scanning gantry bottom beam 101 on both sides of the bottom of the scanning gantry 1.
  • Horizontal travel limit devices are provided at both ends of each scanning gantry bottom beam 101.
  • the horizontal travel limit device includes a horizontal travel front limiter.
  • the position 110 and the horizontal travel rear limiter 111 are used to limit the travel range of the scanning gantry 1 on the horizontal travel track 5.
  • a horizontal travel motor 106 and a horizontal travel encoder 107 are provided in the middle area of the bottom beam 101 of the scanning gantry. , to drive the scanning gantry 1 forward and backward and record the exact position of the scanning gantry 1.
  • a vertical traveling motor 108 and a vertical traveling encoder 109 are provided at the lower end of the connection point between the vertical traveling track 6 and the scanning gantry bottom beam 101 to drive the vertical traveling pallet 8 to move up and down and record the vertical movement.
  • vertical travel limiting devices are provided at the upper and lower ends of the vertical traveling track 6.
  • the vertical traveling limiting devices include an upper vertical traveling limit 112 and a lower vertical traveling limit 105. , to limit the traveling range on the vertical traveling track 6 of the vertical traveling pallet 8 .
  • a control cabinet 10 is provided on the vertical structure on one side of the scanning gantry 1.
  • the control cabinet 10 is provided with a horizontal traveling motor 106, a vertical traveling motor 108, a horizontal traveling front limiter 110, a horizontal traveling rear limiter 110, and a horizontal traveling motor 106.
  • the outside of the vertical structure of the scanning gantry 1, the side of the vertical walking track 6, and the side of the horizontal walking track 5 are respectively provided with a scanning gantry temperature and humidity sensor 102, a vertical track temperature and humidity sensor 103, and a horizontal track temperature and humidity sensor.
  • the sensor 104 is used to measure temperature and humidity changes at the location, compensate for measurement and control results, and improve measurement and control accuracy.
  • the target scanning component 2 can be a prefabricated component of the same type with any shape and any size, as shown in Figure 5A, Figure 5B and Figure 5C.
  • the target scanning component 2 is placed on the component scanning platform 3. Within the measurement area, make the component as close as possible to the ideal placement position (that is, place the target scanning component in the middle of the measurement platform, with the splicing surfaces on both sides close to the direction of the horizontal walking track), and the target scanning component 2 can be provided with a target scanning component Zhang
  • the holes 23 and the target scan component grooves 24 are drawn.
  • the two tracks of the horizontal walking track 5 are horizontal and parallel, and a rack track is provided on the inside thereof to provide the advantages of low noise, fast moving speed, and high walking accuracy through the rack track.
  • the upper surface of the component scanning platform 3 and the horizontal walking rails 5 on both sides are in a parallel state.
  • the upper surface of the component scanning platform 3 is provided with a rectangular measurement area. Before scanning, the target scanning component 2 is placed in the measurement area. Test within.
  • the present invention also discloses a method for intelligent scanning of manufacturing accuracy and automatic identification of errors for the assembly surface of prefabricated components.
  • the method includes the following steps:
  • Step 1 Preparation step: hoist the target scanning component 2 to the measurement area on the component scanning platform 3. During the hoisting process, keep the measurement surface 21 of the target scanning component 2 as close to the direction parallel to the horizontal walking track 5 as possible.
  • Step 2 Start the step. You can select the type of target scanning component 2 to be scanned (at this time, the system automatically loads the theoretical model data corresponding to the target scanning component 2), and configure the scanning accuracy according to the production quality requirements of the assembly surface of the target scanning component 2. and other parameters (self-configured or system default). Open the electric rolling shutter door of warehouse 4, start scanning detection with one click, and start the intelligent scanning equipment (run the scanning gantry 1 to walk horizontally along the horizontal walking track 5 in the direction of the target scanning component 2, and at the same time turn on the laser range finder 7).
  • Step 3 Adjust the height of the vertical traveling tray 8 so that the light of the laser rangefinder 7 is close to the height of the upper surface of the component scanning platform 3 .
  • Step 4 The scanning gantry 1 starts from the warehouse 4 and walks forward along the horizontal walking track 5.
  • the deformation data of the steel structure in different temperature and humidity environments can be used to compensate for the laser measurement value and walking distance to ensure measurement and control accuracy in complex climate environments.
  • Step 5 Collect the laser measurement value of the laser rangefinder 7 in real time while walking, and determine whether the light spot of the laser rangefinder 7 hits the target scanning component 2. If so, stop the horizontal advancement of the scanning gantry 1. Specifically, as shown in Figure 10, since the distance between the paired laser rangefinders 7 is L, the sum of the measurement values of the pair of laser rangefinders 7 must be greater than the distance between them. Therefore, , when the light spot of the laser rangefinder 7 does not hit the target scanning member 2, the left laser rangefinder measurement value L l + the right laser rangefinder measurement value L r >L.
  • the measurement light emitted by the two laser rangefinders should be in the same straight line. That is, no matter how the target scanning component is placed, when one of the laser rangefinders’ measurement light hits the When the object is on the component, another measurement light will also hit the component. Therefore, when the light spot of the laser rangefinder hits the target scanning component, the measurement value L l of the left laser rangefinder + the measurement value L r of the right laser rangefinder ⁇ L is used to determine whether the target scanning component 2 is detected. And record the current measurement values of the two laser rangefinders as and
  • the light spots of the two laser rangefinders may be hit on their corresponding measurement surfaces 21, or only one side of them may be hit.
  • the light spot of the laser rangefinder hits the corresponding measurement surface 21, and the light spot of the laser rangefinder on the other side hits the front end surface 20 as shown in Figures 11A and 11B.
  • Step 6 Determine the boundary contour of the measurement surface 21 in the target scanning component 2 (shown as the measurement surface boundary 22 in Figure 2). Taking the current position as the starting point, the horizontal boundary point of the target scanning component 2 corresponding to the current height is searched using methods such as roundabout half-turning and gradual approximation, which is recorded as the first boundary point P 0 of the target scanning component 2 .
  • the forward direction of the horizontal walking rail 5 be the positive x-axis direction
  • the upward direction of the vertical walking rail 6 be the positive y-axis direction
  • the measurement value of the laser rangefinder 7 be the z-axis coordinate value
  • the boundary point P 0 is the origin of the coordinates, and the xyz coordinate system is established.
  • the remaining boundary points of the peripheral contour of the measurement surface boundary 22 of the target scanning component 2 are captured in a clockwise direction.
  • the theoretical data model of the target scanning component 2 is combined, and a rapid iteration method is used to continuously reduce the search for subsequent boundary points. times to speed up the capture of other boundary points, and finally all boundary points form a boundary point data set Q 0 .
  • Step 6.1 Every time scanning gantry 1 takes a step forward, the measurement values of the laser range finders on both sides of the current position are collected and recorded as and Calculate the change rate of the laser rangefinder measurement values on both sides of the current position and the previous position.
  • the change rate of the measurement value on the left side is The change rate of the measured value on the right side is Compared and size:
  • the laser rangefinder light spots on both sides hit their respective corresponding measurement surfaces 21 for two measurements. At this point, the detection of the target measurement surface is completed.
  • Scanning gantry 1 continues to walk forward horizontally along the horizontal walking track 5 with a step length S until
  • the change rate of the measurement value of the measuring surface in the corresponding x-axis direction can be determined. Based on this change rate, during the horizontal walking process of scanning gantry 1, we can continuously calculate whether the change rate of the measurement value of the laser rangefinder is consistent with If the change rates are equal, it can be judged that the measurement surface of the component is hit.
  • the change rate of the measurement value of the measurement surface in the y-axis direction can also be determined as needed to further determine whether the laser rangefinder 7 hits the component measurement surface when walking vertically. The specific method is It is similar to the x-axis direction and will not be described again here.
  • Step 6.2 Move the laser rangefinder corresponding to the target measurement surface in step 6.1 one step further upward along the vertical walking track 6 with a step length S, and determine the rate of change of the measurement value in the y-axis direction. Among them, at the end of step 6.1 Record it as the reference change rate of the measured value in the horizontal walking direction of the target measurement surface.
  • the current measurement surface as the right measurement surface in Figure 12 as an example, collect the measurement value of the laser rangefinder at the current position and record it as Calculate the base change rate of the measured value in the vertical walking direction of the current measurement surface
  • Step 6.3 The mobile laser rangefinder 7 returns to the first measurement point in the current measurement plane, and starts to acquire the first scanning boundary point P 0 in a roundabout way.
  • the specific judgment method is as follows:
  • Step 6.3.1 When the mobile laser rangefinder 7 returns to the first measurement point in the current measurement plane, the current laser rangefinder 7 is located in Figure 9 Corresponding to the position, first move the laser rangefinder 7 along the vertical walking track 6 and continue to walk vertically downward one step with a step length S until it reaches Corresponding position, that is, in step 6.1 The position of the laser rangefinder 7 at this time. All measurement values numbered between 1 and n-1 after target scanning component 2 is detected are compared one by one with pass-through Calculate the change rate, where i is the measurement value serial number, that is, the measurement value subscript number, and the value range is 1 ⁇ (n-1).
  • the scanning gantry is controlled to walk n-i steps backward along the horizontal walking track with a step length S.
  • the current light point position of the laser rangefinder 7 is the first measurement point in the current measurement plane.
  • Step 6.3.3 Obtain the measurement value of the laser rangefinder 7 at the current position and record it as L temp , and calculate the change rate of the measurement value between the position and the measurement point P 00
  • Step 6.3.4 Determine whether K temp is equal to
  • Step 6.4 After obtaining the first scanning boundary point P 0 , take P 0 as the coordinate origin, take the horizontal forward direction of the scanning gantry 1 as the positive x-axis direction, and take the vertical upward direction of the laser rangefinder 7 scanning as the positive y-axis direction. , using the measurement value of the laser rangefinder as the z-axis coordinate value, establish the xyz coordinate system, and obtain the coordinates of the P 0 point (x 0 , y 0 , z 0 ), that is, (0,0,0).
  • Step 6.5 Based on the first scanned boundary point, perform subsequent boundary point capture and complete the capture of all boundary points of target scanned component 2.
  • the search range can be greatly reduced, and the capture speed of subsequent boundary points can be gradually accelerated, and finally the capture of all boundary points of the target scanning component 2 can be completed.
  • the specific implementation method is as follows:
  • Step 6.5.1 Starting from the coordinate origin, capture the second boundary point.
  • the y-axis coordinate value of P′ 1 is the y-axis coordinate value of P 0 +S, that is, y 0 +S ;
  • Calculate the intersection point between y y 0 +S and the theoretical target measurement surface model (the closest point in the clockwise direction of P 0 point, such as the P′ 1 position in Figure 14), which is the P′ 1 point, and the coordinates corresponding to the xy-plane are ( x′ 1 ,y′ 1 ), its z-coordinate value is
  • next boundary point is in the horizontal backward direction of the current boundary point
  • first control the scanning gantry 1 to walk one step backward along the horizontal walking track 5 with a step length S
  • control the laser rangefinder 7 to walk one step backward along the vertical walking track 6 with a step length of S.
  • the length S temp y′ 1 -y 0 continues to walk one step and starts to obtain the actual scanning boundary points.
  • Step 6.5.1.1 Obtain the measurement value of the laser rangefinder at the current position and record it as L temp , and calculate the change rate of the measurement value between the position and the theoretical boundary point P′ 1
  • Step 6.5.1.2 Determine whether K temp is equal to
  • Step 6.5.2 Starting from the second boundary point P 1 , capture subsequent boundary points.
  • the first two boundary points are at The function formed by the xy-plane is a straight line (that is, the vertical boundary of the measurement surface estimated through the captured boundary points is projected 212 on the xy-plane).
  • the function formed by the captured boundary points on the xy-plane may be Curves and curve functions can be obtained through the theoretical measurement surface model. According to this method, the theoretical position P′ n of the subsequent boundary point can be obtained.
  • the coordinates corresponding to the xy-plane are (x′ n ,y′ n ), and its z-axis coordinate value is
  • Step 6.5.2.1 Obtain the measurement value of the laser rangefinder at the current position and record it as L temp , and calculate the change rate of the measurement value between the position and the theoretical boundary point P′ n Among them: L n-1 is the laser measurement value corresponding to the last captured boundary point P n-1 .
  • the search range can be continuously narrowed by the above-mentioned shadow-following method to achieve the purpose of accelerating the capture speed. Until all boundary points of the current measurement surface are found, a boundary point data set Q 0 is formed.
  • Step 7 Complete the data collection of the measurement surface and form the laser point cloud data set Q 1 .
  • the scanning gantry 1 is controlled to walk at equal intervals in the x-axis direction. Each time it walks, the vertical walking pallet 8 is controlled to walk along the vertical walking track 6 at equal intervals to complete the measurement of the measurement surface 21 of the target scanning component 2 corresponding to the current vertical line. Take measurements. Walk step by step in this way to complete the scanning of the measurement surface 21 of the target scanning component 2 (all measurement points form a grid-like distribution as shown in Figure 7).
  • the laser rangefinder 7 is controlled in the vertical direction to move step by step on the vertical walking track 6 with the system set step length, and at each step Collect laser measurements after movement. Until the laser rangefinder 7 moves to the upper boundary point of the target scanning component 2, the scanning of the vertical line on the coordinate origin is completed; then the scanning gantry 1 is controlled to walk one step along the horizontal walking track 5 with the system set step length to complete the current horizontal position. Scan the corresponding vertical lines; in this way, complete the scanning of all vertical lines within the boundary range of the entire component measurement surface, form dense scanning data, and record all scanning results to form a laser point cloud data set Q 1 .
  • Step 8 The spatial attitude inversion correction algorithm of prefabricated components based on laser scanning point cloud can be used to correct the placement posture of the target scanning component.
  • the target scanning component 2 When the target scanning component 2 is placed, it cannot be guaranteed that its measurement surface 21 is completely parallel to the horizontal walking track 5 (i.e., the ideal placement posture in Figure 6).
  • the ideal placement posture in Figure 6
  • spatial geometry transformation is used. , correct the target scanning component 2 with unknown placement posture to the ideal placement posture (that is, project the measurement value of the measurement surface in the current placement posture to the plane of the measurement surface in the ideal placement posture).
  • Step 8.1 Before starting scanning, the theoretical model of the target scanning component 2 has been imported into the system, and the theoretical model has been placed in the coordinate system established in step 6, and the bottom left corner point P' of the theoretical model measurement surface in Figure 6 has been imported into the system. 0 coincides with the P 0 point of the measurement surface 21 of the target scanning component 2 in the actual measurement in Figure 7, the bottom boundary of the theoretical model measurement surface coincides with the x-axis, and the left boundary of the theoretical model measurement surface coincides with the y-axis, and the theoretical model measurement surface data set is obtained Q′ 1 .
  • Step 8.2 Couple the measurement surface in the actual placement with the measurement surface of the theoretical model to complete the attitude inversion correction of the target scanning component 2, so as to facilitate the judgment of abnormal points in subsequent steps.
  • the target scanning component 2 cannot be guaranteed to be placed in the direction of the coordinate axis exactly like the theoretical model before measurement, and angles may occur in the three directions of x, y, and z.
  • Step 8.2.1 Extract the maximum and minimum points in the x and y axis directions and the origin of the coordinates from the boundary point data set Q 0 , and take three points that are not on the same straight line as feature points.
  • Three features The coordinates of the points are A (x 1 , y 1 , z 1 ), B (x 2 , y 2 , z 2 ), and C (x 3 , y 3 , z 3 ) respectively.
  • the equation of the feature point surface is Equation 1:
  • Step 8.2.2 First correct the measurement surface to the xy-plane, that is, Equation 2:
  • Equation 3 the intersection line L between the feature point surface and the xy-plane is obtained as Equation 3:
  • a, b, c are function constants
  • the angle ⁇ between the two planes can be obtained from the feature point surface, xy-plane, and plane intersection L;
  • Step 8.2.3 Correct all measurement points in the laser point cloud data set Q 1 and boundary point data set Q 0 to the xy-plane.
  • the corrected plane is recorded as plane ABC′.
  • Pi Take the measurement point data from the laser point cloud data set Q 1 and the boundary point data set Q 0 in sequence and record it as the measurement point Pi .
  • Pi makes a perpendicular line to the intersection line L, and then uses the intersection line L as the rotation axis, and takes the vertical line as the axis of rotation.
  • the foot is the center of the rotation circle , and the angle ⁇
  • the corrected feature point surface is a plane ABC′, at this time plane ABC′ already coincides with the xy-plane.
  • Step 8.2.4 Through step 8.2.3, the corrected plane ABC′ has coincided with the xy-plane, but in the actual measurement, the attitude of the target scanning component 2 is unknown, and the left bottom corner point of the measured surface after correction may not be the coordinate origin P 0 (0,0,0) points coincide. Extract the corrected left bottom corner point P angle l (x l , y l , z l ) of the measurement surface from the corrected boundary point data set Q′ 0 , and translate the point P angle l to coincide with the coordinate origin point P 0 , That is, point P and angle l are translated x l along the x-axis and y l along the y-axis.
  • all points in the corrected measurement surface laser point cloud data set Q 1_1 and measurement surface boundary point data set Q 0_1 are translated x l along the x axis and y l along the y axis.
  • the translated laser point cloud data set Q 1_1 and boundary point data set Q 0_1 are recorded as Q 1_2 and Q 0_2 respectively.
  • Step 8.2.5 Through step 8.2.4, the translated measurement surface is already on the same plane as the measurement surface in the ideal placement posture, and the left bottom corners of the two measurement surfaces coincide. At this time, the left and right bottom corners of the measurement surface are There may still be an angle ⁇ between the connection line and the x-axis (as shown in Figure 8).
  • the plane ABC′ can be centered on the coordinate origin point P 0 and rotated along the z-axis so that the target scanning component will move the two bottom corner points. coincident with the x-axis.
  • the left bottom corner point P angle l (0,0,0) of the measurement surface extracted from the translated boundary point data set Q 0_2 (has been translated to the coordinate origin) and the right bottom corner point P angle r (x r ,y r , z r ), the slope K 1 of the straight line where the two bottom corner points are relative to the x-axis is
  • Step 8.2.6 Rotate the measured point values in the translated laser point cloud data set Q 1_2 and boundary point data set Q 0_2 one by one around the z-axis direction with ⁇ as the rotation angle, and take the left bottom corner point P of the component measurement surface as angle l. Rotate the center to the x-axis to obtain the final correction value. Repeat the above operation until all measurement points complete the final correction. Obtain the rotationally corrected laser point cloud data set Q 1_3 and boundary point data set Q 0_3 to complete the attitude reflection of the target scanning component 2. Perform correction.
  • Step 9 The local defect identification method of the prefabricated component assembly surface based on dense scanning data can be used to analyze the measurement data of the measurement surface 21 of the target scanning component 2 after correction, and eliminate small bubbles, pits and stone particles on the measurement surface 21 of the target scanning component 2. Measurement values corresponding to other influencing factors. Specific steps are as follows:
  • Step 9.1 Traverse the corrected laser point cloud data set Q 1_3 and boundary point data set Q 0_3 .
  • the theoretical model comparison method can be used to compare the measured values in each set with the same xy coordinate position in the theoretical model data set Q′ 1 . Whether the measured values are the same, and the coordinate points whose difference exceeds the minimum measurement precision ⁇ S 0 (according to the scanning accuracy requirements of the target scanning component 2, the user can set it in step 2) are stored in the abnormal point data set Q ⁇ 0 , until the corrected laser point cloud data set Q 1_3 and boundary point data set Q 0_3 are traversed once.
  • Step 9.2 Exclude the reasonably existing abnormal points in the abnormal point data set Q ⁇ 0 .
  • These reasonable abnormal points include the areas corresponding to the tension holes 23 of the target scanning component and the concave and convex tenons 24 of the target scanning component as shown in Figure 6. measuring points, the specific operations are as follows:
  • the corresponding xy coordinate range is obtained. Traverse all the measurement points in the abnormal point data set Q ⁇ 0 , and eliminate the data in the above-mentioned reasonable abnormal point area without analysis.
  • the abnormal point data set after eliminating reasonable abnormal points is defined as Q ⁇ 1 .
  • Step 9.3 Traverse the abnormal point data set Q ⁇ 1 after eliminating reasonable abnormal points, take out the measurement points in the data set one by one, and take the current point as the center, use the correlation search method to compare the current point with the surrounding measurement points on the measurement surface. Measurement value relationships. Obtain the measurement values of all measurement points in the 5*5 point area of the current point in the corrected laser point cloud data set Q 1_3 , and compare the measurement values of the measurement points in the surrounding area with the measurement values of the current measurement point in turn. If the surrounding If the measurement values of the measurement points are all greater than or less than the measurement value of the current measurement point, the current measurement point is determined to be a local defect point, and the point is removed from the abnormal point data set Q ⁇ 1 after traversing and eliminating reasonable abnormal points. Repeat this step until all local defect point data of the target scanning component 2 are eliminated, and the abnormal point data set Q ⁇ 1 after the local defect points are eliminated is recorded as the final abnormal point data set Q ⁇ .
  • the local depression measurement matrix is as follows:
  • the local bump measurement matrix is as follows:
  • Step 10 Determine the manufacturing accuracy of the assembly surface of the target scanning component, and spray mark the error area on the surface of the component.
  • the assembly surface manufacturing errors are divided into two forms: bulges and dents.
  • the dent area does not affect the assembly of prefabricated components, while the bulge area may cause the prefabricated components to fail to be assembled during the assembly process, and the joints are not tightened, etc., which affects When prefabricated components are used on-site, the bulging area needs to be polished.
  • Eliminate non-bulging points in the final abnormal point data set Q ⁇ calculate the bulging area boundary and bulging height in the final abnormal point data set Q ⁇ , and spray mark the corresponding bulging area in the target component. The steps are as follows:
  • Step 10.1 Compare the theoretical model to eliminate non-bulging points in the final abnormal point data set Q ⁇ that do not affect the assembly of components.
  • the heights of all bulge points in the current measurement surface are calculated and stored in the bulge height data set Q ⁇ h of the current measurement surface.
  • Step 10.2 Traverse the final abnormal point data set Q ⁇ after eliminating non-bulging points (after eliminating non-bulging points in step 10.1, only bulging abnormal points remain in the data set), use the correlation search method to center on the current point, and search for the current
  • Step 10.3 Control the scanning gantry 1 and the vertical traveling pallet 8 to transport the point tracing and printing mechanical telescopic arm 9 to the corresponding position on the edge of the bulge area of the measurement surface, control the point tracing and printing mechanical telescopic arm 9 to extend to the measurement surface 21 of the target scanning component 2, complete Delineate a point, walk through the entire error area boundary one by one, and complete the labeling of the error area.
  • Step 11 Follow the above steps to complete the detection of all measurement surfaces of the target scanning component 2. Generate an electronic inspection report, clearly identify the bulging area of the measurement surface that affects component assembly and the data in this area, and complete all inspections of the target scanning component 2.
  • the gantry adopts a gantry-type walking structure.
  • the gantry is loaded with multi-directional moving motors and tracks, which can complete the measurement of any shape of assembly surface.
  • the gantry is also equipped with various sensors to measure the tracks, gantry and air.
  • the temperature and humidity are monitored in real time, and the measurement results are corrected through the temperature and humidity monitoring results to ensure that the system can work under complex climate conditions without affecting accuracy.
  • the assembly surface of the target scanning component may have various factors such as concave and convex mortises, rubber channels, bubble pits, stone particles, slight edge damage, etc. that affect the production accuracy of scanning.
  • This patent is based on The theoretical model of the target scanning component is used as a reference.
  • Various influencing factors must be processed from the data, and the measurement data must also be processed using algorithms.
  • algorithmic methods such as correlation search algorithm, theoretical model reference method, and overall trend deduction method to eliminate influencing factors step by step to complete the production accuracy of the assembly surface of prefabricated components. calculation.
  • the system calculates the coordinate data set of the boundary points of the bulge area, and controls the gantry to carry the inkjet printing device to the corresponding position of each boundary point one by one, and inkjet the boundary points of the bulge on the surface of the component to facilitate later stages.
  • the system calculates the coordinate data set of the boundary points of the bulge area, and controls the gantry to carry the inkjet printing device to the corresponding position of each boundary point one by one, and inkjet the boundary points of the bulge on the surface of the component to facilitate later stages.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Signal Processing (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Length Measuring Devices With Unspecified Measuring Means (AREA)

Abstract

一种预制构件拼装面制作精度智能扫描和误差自动标识系统及方法，采用门架式行走结构，门架上装载有多方向移动电机和轨道，能够完成对任意形状的测量面的测量工作，并且门架上装载有各类传感器，对轨道、门架以及空气的温度和湿度进行实时监测，通过温湿度的监测结果对测量结果进行校正，保障系统可以在复杂气候条件下工作而不会影响精度，而且，以目标扫描构件（2）理论模型为参照，既从数据上处理各种影响因素，也利用算法对测量数据做出处理。先利用逐步逼近、如影随形等方法快速找出目标扫描构件（2）的边界数据，再利用关联搜索算法、理论模型参照法、整体趋势推演法等算法手段一步步剔除影响因素，完成对预制构件拼装面制作精度的计算。

Description

预制构件拼装面智能扫描和误差自动标识系统及方法 技术领域
本发明涉及装配式结构预制构件智能扫描的技术领域，尤其涉及一种预制构件拼装面智能扫描和误差自动标识系统及方法。
背景技术
针对一些装配式地上结构，如箱型梁、T型梁，绝大多数预制构件不需要拼装情况下，对拼装质量要求不高，可以适当放宽制作精度的要求，沿用传统方法如直接尺量、靠尺、塞尺等人工进行量测的方法对预制构件制作精度进行测量能满足要求。对于大型装配式结构，特别如装配式地铁车站结构、大尺寸盾构管片等，其预制构件通常具有体积大、结构厚、形状不一且不规则等特点。另外，装配式结构对防水、拼缝宽度以及加载荷载要求高，预制构件质量对装配式结构工程施工工艺、拼装质量、防水性能等影响重大。而且，预制构件拼装面的不平整会导致该面拼装无法同步，局部较大的加压荷载容易造成构件损伤。
预制构件生产部门越来越关注预制构件生产质量，构件出厂前需进行质量检测。而常规预制构件利用卡具和靠尺测量方法不能在大构件上完整实施，常常出现包括传统卡尺尺寸不够、构件腹部量测不到等人工操作无法全面检测的问题，达不到测量精度和目的。在上述装配式结构对预制构件拼装面平整性要求很高情况下，当前技术没有办法对大型构件的拼装面制作精度进行精确量测，以至于早期大型装配 式地下结构预制构件出厂之后因为模具变形起鼓等原因影响拼装面的制作精度，造成制作误差，在多个工程拼装现场出现预制构件拼不上、接缝拉不严等情况，导致构件到现场无法使用。因此，急需研发一种预制构件智能化检测系统及方法进行构件拼装面制作误差检测，并对误差区域做出标识。
预制构件拼装面制作误差主要指影响装配式结构拼装的误差，如构件鼓包区域，易导致拼装面接缝拼不严、拉不紧等问题。虽然拼装面有局部鼓包，但是整个构件拼接面的边缘轮廓依旧保持精准，可依此作为智能扫描的基准条件和基础步骤，进一步判定构件是否鼓包并精确找到鼓包区域。由于大型预制构件具有体积大、重量大、边界轮廓不规则等特点，无法精准控制构件置于智能扫描平台上的吊装摆放角度。因此，在构件摆放姿态不确定、边界轮廓不规则的情况下，常规的识别方法用在大体积构件上速度过慢，已无法满足需求，亟需一种可以适应于任意摆放，且高效的边界捕获方法，以及一种理论模型与实际摆放姿态相耦合的算法。
另外，混凝土预制构件生产时，因模板接缝不严、模板表面未清理干净、模板脱模时发生黏连、振捣不充分、气泡未排出等原因，可能出现预制混凝土构件表面出现麻面、蜂窝、孔洞、颗粒等现象。此类缺陷不同于装配式结构混凝土预制构件拼装面制作误差，不影响装配式结构拼装现场预制构件之间的贴合拼接，于使用无碍，但在利用智能扫描的方式来检测预制构件拼装面制作精度时产生较大的影响，从而导致分析数据的不准确甚至误导整体分析结果，称之为局部混凝 土缺陷。对于局部缺陷点，直接从测量数据集剔除即可。有效剔除局部缺陷是确保通过智能扫描方式快速准确高效完成混凝土预制构件拼装面检测工作的关键。
另外，由于大型预制构件尺寸庞大且结构复杂，检测装置可能置于室外，受温度变化、湿度变化等因素影响会引起大型扫描装置变形等问题，导致测量结果在不同的气候条件下产生误差，无法满足现场实际应用的需要。需要研发专用检测装备，能够对大型预制装配式结构预制构件拼装面制作精度进行检测，并能够适应不同环境条件而不影响装备检测精度。
可见，拼装面制作精度检测目前存在上述装备和系统算法难题，而保证拼装面制作精度，是能够实施好装配式地下结构施工的关键。
为此，本发明的设计者有鉴于上述缺陷，通过潜心研究和设计，综合长期多年从事相关产业的经验和成果，研究设计出一种预制构件拼装面智能扫描和误差自动标识系统及方法，以克服上述缺陷。
发明内容
本发明的目的在于提供一种预制构件拼装面智能扫描和误差自动标识系统及方法，其结构简单，操作方便，能有效克服现有技术的缺陷，对预制构件进行自动识别扫描，且精度高，效率快，更具创新性。
为实现上述目的，本发明公开了一种预制构件拼装面制作精度智能扫描和误差自动标识方法，其特征在于包含如下步骤：
步骤一：准备步骤，将目标扫描构件吊装到构件扫描台上的测量 区域内；
步骤二：启动步骤，打开仓房的电动卷闸门，启动扫描龙门开启扫描检测；
步骤三：调节步骤，调节竖向行走托盘的高度，使激光测量设备的光线贴近于构件扫描台上表面高度；
步骤四：扫描龙门沿水平行走轨道从仓房中出发向前行走，行走过程中实时采集激光测距仪的激光测量值和扫描龙门温湿度传感器、竖直轨道温湿度传感器、水平轨道温湿度传感器的监测值；
步骤五：行走中实时采集激光测距仪的激光测量值，判断激光测距仪的光点是否打在了目标扫描构件上，如果是，停止扫描龙门的水平前进；
步骤六：确定目标扫描构件中测量面的边界轮廓，所有边界点形成边界点数据集Q 0；
步骤七：完成对测量面的数据采集，形成激光点云数据集Q 1；
步骤八：校正目标扫描构件的摆放姿态，将目标扫描构件校正到理想摆放姿态；
步骤九：对校正后目标扫描构件测量面的测量数据进行分析，剔除目标扫描构件测量面的异常测量值；
步骤十：确定目标扫描构件的拼装面制作精度，并在目标扫描构件表面对误差区域喷涂标识；
步骤十一：生成检测报告，完成对目标扫描构件的全部检测。
其中：步骤五的具体方法如下：当激光测距仪光点打在目标扫描 构件上时左侧激光测距仪测量值L l+右侧激光测距仪测量值L r≤L时，判定检测到目标扫描构件2，并记录两个激光测距仪当前测量值分别为
Figure PCTCN2022083579-appb-000001
和
Figure PCTCN2022083579-appb-000002
其中：步骤六中以步骤五的停止点为起点，利用迂回折半、逐步逼近方法搜索当前高度对应的目标扫描构件在水平方向的边界点，记为目标扫描构件的第一个边界点P 0，以水平行走轨道前进方向为x轴正方向、以竖向行走轨道向上方向为y轴正方向、以激光测距仪测量值为z轴坐标值、以目标扫描构件的第一个边界点P 0为坐标原点，建立xyz坐标系，从坐标原点P 0出发，沿顺时针方向捕获目标扫描构件测量面边界周边轮廓其余边界点，过程中结合目标扫描构件的理论数据模型，利用快速迭代的方法不断减少后续边界点的搜索次数，加快其余边界点捕获速度，最终将所有边界点形成边界点数据集Q 0。
其中还包含如下步骤：
步骤6.1：扫描龙门每向前行走一步，采集当前位置两侧激光测距仪测量值，分别记录为
Figure PCTCN2022083579-appb-000003
和
Figure PCTCN2022083579-appb-000004
计算当前位置与上一步位置两侧激光测距仪测量值的变化率，左侧测量值变化率为
Figure PCTCN2022083579-appb-000005
右侧测量值变化率为
Figure PCTCN2022083579-appb-000006
对比
Figure PCTCN2022083579-appb-000007
和
Figure PCTCN2022083579-appb-000008
大小：
当
Figure PCTCN2022083579-appb-000009
时，两侧的激光测距仪光点前后两次测量均打在了各自对应的测量面上，至此完成目标测量面的检测；
当
Figure PCTCN2022083579-appb-000010
时，其中一侧的激光测距仪光点前后两次测量并未全部打在了对应的测量面上；扫描龙门1沿水平行走轨道5以步长S继续水平向前行走，直到
Figure PCTCN2022083579-appb-000011
步骤6.2：将步骤6.1中目标测量面对应的激光测距仪沿竖向行走轨道以步长S继续向上行走一步，确定测量值在y轴方向上的变化率；
步骤6.3：移动激光测距仪退回到当前测量面内第一个测量点，并以迂回折半行走方式开始获取第一个扫描边界点P 0；
步骤6.4：获取第一个扫描边界点P 0后，以P 0为坐标原点，以扫描龙门水平前进方向为x轴正方向，以激光测距仪扫描竖直向上方向为y轴正方向，以激光测距仪测量值为z轴坐标值，建立xyz坐标系，得到P 0点坐标(x 0,y 0,z 0)即(0,0,0)；
步骤6.5：以第一个扫描边界点为基础，进行后续边界点捕获，并完成目标扫描构件所有边界点的捕获。
其中：步骤八的具体步骤如下：
步骤8.1：将理论模型置于步骤六中建立的坐标系中，并将理论模型测量面底部左角点P′ 0与目标扫描构件测量面的P 0点重合，理论模型测量面底边界与x轴重合，理论模型测量面左边界与y轴重合，得到理论模型测量面数据集Q′ 1；
步骤8.2：将实际摆放中的测量面与理论模型的测量面相耦合，完成目标扫描构件的姿态反演校正，得到校正后的激光点云数据集Q 1_3和边界点数据集Q 0_3。
其中：步骤九的具体步骤如下：
步骤9.1：遍历校正后的激光点云数据集Q 1_3和边界点数据集Q 0_3，比对激光点云数据集Q 1_3和边界点数据集Q 0_3中测量值与理论模型 数据集Q′ 1中相同xy坐标位置的测量值是否相同，差距超过测量打点最小精度ΔS 0的坐标点存入到异常点数据集Q △0中；
步骤9.2：将异常点数据集Q △0中合理存在的异常点进行排除，剔除合理存在异常点后的异常点数据集定义为Q △1；
步骤9.3：遍历剔除合理存在异常点后的异常点数据集Q △1，逐个取出该数据集中的测量点，并以当前点为中心，利用关联搜索法对比当前点与测量面上周边测量点的测量值关系；获取当前点在校正后的激光点云数据集Q 1_3中方圆5*5个点区域内所有测量点的测量值，将周边区域测量点的测量值依次与当前测量点的测量值对比，如果周边测量点的测量值全部大于或小于当前测量点的测量值，则判定当前测量点为局部缺陷点。
还公开了一种预制构件拼装面智能扫描和误差自动标识系统，包括工作台、仓房、扫描龙门、水平行走轨道和构件扫描台，其特征在于：
所述仓房位于工作台的一端且设有容纳扫描龙门的容纳空间以及电动闸门，所述水平行走轨道为两条轨道且沿工作台的两侧延伸，其一端延伸至仓房内，所述构架扫描台位于工作台上且设置于水平行走轨道的中间，以供目标扫描构件的放置，扫描龙门可滑动的设置于水平行走轨道上，所述扫描龙门为门形的行走机构，其行走在构件扫描台两侧的水平行走轨道上，所述扫描龙门的两内侧各有一条竖向行走轨道，在竖向行走轨道上各有一个竖向行走托盘以沿竖向行走轨道上下移动，所述竖向行走托盘上设有激光测量设备和描点喷绘机械伸 缩臂的安装座。
其中：所述激光测量设备固定于安装座的上端且为单点激光测距仪，所述描点喷绘机械伸缩臂固定于安装座的下端且可通过伸缩控制在指定表面喷绘色斑。
其中：所述扫描龙门的两侧底部各有一条扫描龙门底梁，在扫描龙门底梁的两端各设有一个水平行走限位装置以限制扫描龙门在水平行走轨道上的行走范围，所述扫描龙门底梁的中间区域设置有一个水平行走电机和一个水平行走编码器以驱动扫描龙门前进后退和记录扫描龙门所处位置。
其中：所述竖向行走轨道和扫描龙门底梁的连接点下端设有一个竖向行走电机和一个竖向行走编码器以驱动竖向行走托盘上下移动和记录竖向行走托盘所处的位置。
通过上述内容可知，本发明的预制构件拼装面智能扫描和误差自动标识系统及方法具有如下效果：
1、采用门架式行走结构，门架上装载有多方向移动电机和轨道，能够完成对任意形状的测量面的测量工作，并且门架上装载有各类传感器，对轨道、门架以及空气的温度和湿度进行实时监测，通过温湿度的监测结果对测量结果进行校正，保障系统可以在复杂气候条件下工作而不会影响精度。
2、在实际测量过程中，目标扫描构件的测量面上可能会有凹凸榫槽、橡胶槽道、气泡凹坑、石子颗粒、边缘轻度破损等各种影响制作精度扫描的因素，本专利以目标扫描构件理论模型为参照，既要从 数据上处理各种影响因素，也要利用算法对测量数据做出处理。先利用逐步逼近、如影随形等方法快速找出目标扫描构件的边界数据，再利用关联搜索算法、理论模型参照法、整体趋势推演法等算法手段一步步剔除影响因素，完成对预制构件拼装面制作精度的计算。
3、在完成一块构件的扫描测量之后，系统计算出鼓包区域的边界点坐标数据集，控制门架携带喷绘装置逐个行走到各个鼓包区域的边界点对应位置，将鼓包边界点喷绘在构件表面，方便后期打磨修复时工人比对检测报告可以快速的找出鼓包位置，极大降低打磨难度和工作量。
本发明的详细内容可通过后述的说明及所附图而得到。
附图说明
图1显示了本发明的预制构件拼装面智能扫描和误差自动标识系统的示意图。
图2显示了本发明的扫描龙门立体图。
图3显示了本发明的扫描龙门侧视图。
图4显示了本发明的目标扫描构件扫描过程中沿扫描龙门水平行走方向正视图。
图5A、图5B和图5C显示了本发明应用于各种类型异形预制构件示例。
图6显示了本发明的目标扫描构件理想摆放姿态下在坐标系中的位置
图7显示了本发明的目标扫描构件密集扫描数据点位图。
图8显示了本发明的目标测量面校正到xy-平面并平移后在坐标系中的位置示意图。
图9显示了本发明确定目标测量面过程中打点点位与测量面的正视图。
图10显示了本发明的目标扫描构件在极限偏向一侧时测量示意图。
图11A和图11B分别显示了本发明的目标扫描构件在测量区域内的理想摆放和极限倾斜摆放的俯视图
图12A、图12B和图12C分别显示了本发明的目标扫描构件在激光测距仪扫描过程中，不同摆放姿态场景下，从激光测距仪光点未打到构件上到两侧激光测距仪光点均打在构件测量面上的位置示意图。
图13显示了本发明捕获边界点过程中实际边界点与推算边界点在xy-平面的投影位置关系示意图。
图14显示了本发明的方法流程图。
附图标记：
1：扫描龙门；101：扫描龙门底梁；102：扫描龙门温湿度传感器；103：竖向轨道温湿度传感器；104：水平轨道温湿度传感器；105：竖向行走下限位；106：水平行走电机；107：水平行走编码器；108：竖向行走电机；109：竖向行走编码器；110：水平行走前限位；111：水平行走后限位；112：竖向行走上限位；2：目标扫描构件；20：前端面；202：目标扫描构件底面；21：被测拼装面，即测量面；211： 理想摆放姿态时测量面竖向边界在xy-平面投影；212：通过已捕获边界点推测的测量面竖向边界在xy-平面投影；213：实际摆放姿态测量面竖向边界在xy-平面投影；22：测量面边界；23：目标扫描构件张拉孔洞；24：目标扫描构件凹凸榫；25：理论模型测量面边界；26：左测量面；27：右测量面；3：构件扫描台；4：仓房；5：水平行走轨道；6：竖向行走轨道；7：激光测距仪；71：左侧激光测距仪；72：右侧激光测距仪；73：激光测距仪测量光线；8：竖向行走托盘；9：描点喷绘机械伸缩臂；10：控制柜。
具体实施方式
参见图1至图6，显示了本发明的预制构件拼装面智能扫描和误差自动标识系统。
如图1所示，所述预制构件拼装面智能扫描和误差自动标识系统包括工作台、仓房4、目标扫描构件2、智能扫描装备、水平行走轨道5和构件扫描台3，所述仓房4位于工作台的一端，可容纳智能扫描装备的扫描龙门1且配有电动闸门，所述水平行走轨道5为两条平行轨道且沿工作台的两侧延伸，所述水平行走轨道5的一端延伸至仓房4内，所述构架扫描台3位于工作台上且设置于水平行走轨道5的中间，以供目标扫描构件2的放置，扫描龙门1设置于水平行走轨道5上，可沿水平轨道行走，由于目标扫描构件2通常为大型预制构件，其占地较大，智能扫描装备的扫描龙门1、水平行走轨道5和构件扫描台3常设置于堆放构件的露天环境中，但是扫描龙门1属于精密设 备，不能长时间处于风雨环境，在不工作时需要将其放置在仓房4内保护，工作时打开仓房4的电动闸门，扫描龙门1能自动驶出仓房4开始工作。
其中，所述扫描龙门1为门形的行走机构，其行走在构件扫描台3两侧的水平行走轨道5上，参见图2、图3和图4，所述扫描龙门1的两内侧各有一条竖向行走轨道6，在竖向行走轨道6上各有一个竖向行走托盘8，所述竖向行走托盘8可以沿竖向行走轨道6上下移动，所述竖向行走托盘8上设有激光测距仪7和描点喷绘机械伸缩臂9的安装座，其中，所述激光测距仪7固定于安装座的上端，所述描点喷绘机械伸缩臂9固定于安装座的下端且可伸缩的在指定表面进行喷绘色斑等。
其中，所述扫描龙门1的两侧底部各有一条扫描龙门底梁101，在各扫描龙门底梁101的两端设有水平行走限位装置，所述水平行走限位装置包含水平行走前限位110和水平行走后限位111，用以限制扫描龙门1在水平行走轨道5上的行走范围，所述扫描龙门底梁101的中间区域设置有一个水平行走电机106和一个水平行走编码器107，以驱动扫描龙门1前进后退和记录扫描龙门1的准确位置。
其中，所述竖向行走轨道6和扫描龙门底梁101的连接点下端设有一个竖向行走电机108和一个竖向行走编码器109，用以驱动竖向行走托盘8上下移动和记录竖向行走托盘8所处的位置，所述竖向行走轨道6的上下两端设有竖向行走限位装置，所述竖向行走限位装置包含竖向行走上限位112和竖向行走下限位105，以限制竖向行走托 盘8的竖向行走轨道6上的行走范围。
其中，所述扫描龙门1一侧的竖向结构上设有一个控制柜10，所述控制柜10内设有水平行走电机106、竖向行走电机108、水平行走前限位110、水平行走后限位111、竖向行走上限位112、竖向行走下限位105、水平行走编码器107、竖向行走编码器109、描点喷绘机械伸缩臂9的控制器。
其中，所述扫描龙门1的竖向结构外侧、竖向行走轨道6的侧面、水平行走轨道5的侧面，分别设置有扫描龙门温湿度传感器102、竖向轨道温湿度传感器103、水平轨道温湿度传感器104，用以测量所在位置的温湿度变化，补偿测量和控制的结果，提高测控精度。
其中，所述目标扫描构件2能为同类型的任意形状、任意尺寸的预制构件，如图5A、图5B和图5C所示，测量前，先将目标扫描构件2放置在构件扫描台3上的测量区域内，使构件尽量靠近理想摆放位置(即将目标扫描构件放置在测量台中间位置，两侧拼接面贴近于水平行走轨道方向)，且目标扫描构件2上可设有目标扫描构件张拉孔洞23和目标扫描构件凹凸榫24。
其中，所述水平行走轨道5的两条轨道水平且平行，其内侧设有齿条轨道，以通过齿条轨道提供低噪声、移动速度快、行走精度高等优点。
其中，所述构件扫描台3的上表面和两侧的水平行走轨道5处于平行状态，所述构件扫描台3的上表面设有矩形的测量区域，扫描前将目标扫描构件2放置在测量区域内进行检测。
如图7至图14所示，本发明还公开了一种预制构件拼装面制作精度智能扫描和误差自动标识方法，该方法包含如下步骤：
步骤一：准备步骤，将目标扫描构件2吊装到构件扫描台3上的测量区域内，吊装过程中，尽可能使目标扫描构件2的测量面21靠近平行于水平行走轨道5的方向。
步骤二：启动步骤，可选择即将扫描的目标扫描构件2的类型(此时系统自动加载目标扫描构件2对应的理论模型数据)，并根据目标扫描构件2的拼装面生产质量要求，配置扫描精度以及其他参数(自行配置或系统默认)。打开仓房4的电动卷闸门，一键开启扫描检测，启动智能扫描装备(运行扫描龙门1沿水平行走轨道5向目标扫描构件2方向水平行走，同时开启激光测距仪7)。
步骤三：调节步骤，调节竖向行走托盘8的高度，使激光测距仪7的光线贴近于构件扫描台3上表面高度。
步骤四：扫描龙门1沿水平行走轨道5从仓房4中出发向前行走，行走过程中实时采集激光测距仪7的激光测量值和扫描龙门温湿度传感器102、竖向轨道温湿度传感器103、水平轨道温湿度传感器104的监测值，其中可利用钢结构在不同温湿度环境下的形变数据来补偿激光测量值和行走距离，保证在复杂气候环境下的测量和控制精度。
步骤五：行走中实时采集激光测距仪7的激光测量值，判断激光测距仪7的光点是否打在了目标扫描构件2上，如果是，停止扫描龙门1的水平前进。具体而言，如图10所示，由于成对设置的激光测距仪7之间距离为L，所以一对激光测距仪7的测量值之和必然大于 两者之间的距离，由此，在激光测距仪7的光点未打在目标扫描构件2上时，左侧激光测距仪测量值L l+右侧激光测距仪测量值L r＞L。
由于每一对激光测距仪都是正对安装，所以两个激光测距仪射出的测量光线应处于同一条直线，即无论目标扫描构件如何摆放，当其中一条激光测距仪测量光线打在构件上时，另外一条测量光线也会打在构件上。所以，当激光测距仪光点打在目标扫描构件上时左侧激光测距仪测量值L l+右侧激光测距仪测量值L r≤L，以判定是否检测到目标扫描构件2，并记录两个激光测距仪当前测量值分别为
Figure PCTCN2022083579-appb-000012
和
Figure PCTCN2022083579-appb-000013
其中：以图12A、12B和12C中各个场景为例，当检测到目标扫描构件2时，可能两个激光测距仪光点都打在了各自对应的测量面21上，或只有其中一侧激光测距仪光点打在了对应的测量面21上，另一侧的激光测距仪光点打在了如图11A、11B中所示的前端面20上。
步骤六：确定目标扫描构件2中测量面21的边界轮廓(如图2中测量面边界22所示)。以当前位置为起点，利用迂回折半、逐步逼近等方法搜索当前高度对应的目标扫描构件2在水平方向的边界点，记为目标扫描构件2的第一个边界点P 0。以水平行走轨道5前进方向为x轴正方向、以竖向行走轨道6向上方向为y轴正方向、以激光测距仪7测量值为z轴坐标值、以目标扫描构件2的第一个边界点P 0为坐标原点，建立xyz坐标系。从坐标原点P 0出发，沿顺时针方向捕获目标扫描构件2测量面边界22周边轮廓其余边界点，过程中结合目标扫描构件2的理论数据模型，利用快速迭代的方法不断减少后续边 界点的搜索次数，加快其余边界点捕获速度，最终将所有边界点形成边界点数据集Q 0。
具体而言，还可包含如下步骤：
步骤6.1：扫描龙门1每向前行走一步，采集当前位置两侧激光测距仪测量值，分别记录为
Figure PCTCN2022083579-appb-000014
和
Figure PCTCN2022083579-appb-000015
计算当前位置与上一步位置两侧激光测距仪测量值的变化率，左侧测量值变化率为
Figure PCTCN2022083579-appb-000016
右侧测量值变化率为
Figure PCTCN2022083579-appb-000017
对比
Figure PCTCN2022083579-appb-000018
和
Figure PCTCN2022083579-appb-000019
大小：
当
Figure PCTCN2022083579-appb-000020
时，两侧的激光测距仪光点前后两次测量均打在了各自对应的测量面21上。至此完成目标测量面的检测。
当
Figure PCTCN2022083579-appb-000021
时，其中一侧的激光测距仪光点前后两次测量并未全部打在了对应的测量面上。扫描龙门1沿水平行走轨道5以步长S继续水平向前行走，直到
Figure PCTCN2022083579-appb-000022
其中：从激光测距仪光点打到目标扫描构件2上开始，到完成目标测量面的检测最少再行走一步。如图12A、12B和12C中三个场景所示例，两侧激光测距仪每一步测量值与目标扫描构件2的测量面21在不同摆放姿态下的变化趋势。
其中：
Figure PCTCN2022083579-appb-000023
时，即可确定了测量面打点测量值在对应x轴方向上的变化率，可依此变化率为基准在扫描龙门1水平行走过程中，不断计算激光测距仪测量值的变化率是否与此变化率相等，即可判断打在了构件测量面上。在具体实施例中，还可根据需要确定测量面打点测量值在y轴方向上的变化率，以进一步确定激光测距仪7在竖向行走时是否打在了构件测量面上，其具体方法和x轴方向上类似，在 此不再进行累述。
步骤6.2：将步骤6.1中目标测量面对应的激光测距仪沿竖向行走轨道6以步长S继续向上行走一步，确定测量值在y轴方向上的变化率。其中，步骤6.1结束时
Figure PCTCN2022083579-appb-000024
将其记为目标测量面水平行走方向测量值基准变化率
Figure PCTCN2022083579-appb-000025
即
Figure PCTCN2022083579-appb-000026
以当前测量面为图12中右侧测量面为例，采集当前位置激光测距仪测量值，记录为
Figure PCTCN2022083579-appb-000027
计算当前测量面竖向行走方向测量值基准变化率为
Figure PCTCN2022083579-appb-000028
其中：测量值
Figure PCTCN2022083579-appb-000029
的编号与测量面21上打点测量点位对应关系如图9中所示。
其中：左侧测量面上述变化率计算将下标r换为l即可，本发明中后续步骤亦同。
步骤6.3：移动激光测距仪7退回到当前测量面内第一个测量点，并以迂回折半行走方式开始获取第一个扫描边界点P 0，具体判断方法如下：
步骤6.3.1：当移动激光测距仪7退回到当前测量面内第一个测量点时，当前激光测距仪7位于图9中
Figure PCTCN2022083579-appb-000030
对应位置，先将激光测距仪7沿竖向行走轨道6以步长S继续竖直向下行走一步，走到
Figure PCTCN2022083579-appb-000031
对应位置，即步骤6.1中
Figure PCTCN2022083579-appb-000032
时激光测距仪7所处的位置。将检测到目标扫描构件2之后编号为1到n-1之间的所有测量值逐个与
Figure PCTCN2022083579-appb-000033
通过式
Figure PCTCN2022083579-appb-000034
计算变化率，其中，i为测量值序号，即测量值下标编号，取值范围为1～(n-1)。从i＝n-1到1之间依次取出对应变化率
Figure PCTCN2022083579-appb-000035
判断
Figure PCTCN2022083579-appb-000036
是否等于
Figure PCTCN2022083579-appb-000037
记录最后一个
Figure PCTCN2022083579-appb-000038
的点位，该点即为 当前测量面内第一个测量点P 00，对应测量值记为L 00，记录当前i的值。
控制扫描龙门沿水平行走轨道以步长S向后行走n-i步，激光测距仪7当前光点位置即为当前测量面内第一个测量点。
步骤6.3.2：扫描龙门沿水平行走轨道5以步长S temp＝S/2水平向后行走一步；
步骤6.3.3：获取当前位置激光测距仪7的测量值记为L temp，计算该位置与测量点P 00之间测量值变化率为
Figure PCTCN2022083579-appb-000039
其中：当前测量位置相比上一个测量位置水平后退/竖向向下时，S temp取值为-S temp(即方向相反时取负值)。
步骤6.3.4：判断K temp是否等于
Figure PCTCN2022083579-appb-000040
如果否，扫描龙门1沿水平行走轨道5以步长S temp＝当前步长一半水平向前行走一步，回到步骤6.3.3；如果是，判断当前步长是否小于扫描龙门1的最小行走精度ΔS 0(根据目标扫描构件2的扫描精度要求，使用者可自行设定，ΔS 0远小于行走步长S)，如果是，成功获取第一个扫描边界点，记为P 0，记录对应测量值L 0，结束当前步骤；如果否，扫描龙门1沿水平行走轨道5以步长S temp＝当前步长一半水平向后行走一步，跳到步骤6.3.3。
步骤6.4：获取第一个扫描边界点P 0后，以P 0为坐标原点，以扫描龙门1水平前进方向为x轴正方向，以激光测距仪7扫描竖直向上方向为y轴正方向，以激光测距仪测量值为z轴坐标值，建立xyz坐标系，得到P 0点坐标(x 0,y 0,z 0)即(0,0,0)。
步骤6.5：以第一个扫描边界点为基础，进行后续边界点捕获，并完成目标扫描构件2所有边界点的捕获。
其中，可通过不断的数据累积学习，利用如影随形的方式，大幅度减小搜索范围，逐步加快后续边界点的捕获速度，最终完成目标扫描构件2所有边界点的捕获，具体实现方法如下：
步骤6.5.1：从坐标原点出发，捕获第二个边界点。
以目标扫描构件2为理想摆放姿态(以图6中目标扫描构件模型在xyz坐标系中的位置为例)为标准，计算P 0点上方相邻边界点P 1(即第二个边界点)的理论位置P′ 1(以图13中第二个边界点在竖向边界上为例)：P′ 1的y轴坐标值为P 0的y轴坐标值+S，即y 0+S；计算y＝y 0+S与理论目标测量面模型交点(P 0点顺时针方向最近的点，如图14中P′ 1位置)即为P′ 1点，对应xy-平面的坐标为(x′ 1,y′ 1)，其z向坐标值为
Figure PCTCN2022083579-appb-000041
将激光测距仪7移动到P′ 1点位置，即激光测距仪沿竖向行走轨道6以步长S继续向上行走一步，然后扫描龙门1再沿水平行走轨道5以步长S temp＝x′ 1-x 0水平行走一步(如果S temp＞0，则为向前行走；如果S temp＜0，则为向后行走；如果S temp＝0，则无需行走)，开始获取实际扫描边界点P 1。
其中：对于竖直边界，如果下一个边界点处于当前边界点的竖直向上方向，则先控制激光测距仪7沿竖向行走轨道6以步长S继续向上行走一步，然后扫描龙门1再沿水平行走轨道5以步长S temp＝x′ 1-x 0水平行走一步(如果S temp＞0，则为水平向前行走； 如果S temp＜0，则为水平向后行走；如果S temp＝0，则无需行走)，开始获取实际扫描边界点。如果下一个边界点处于当前边界点的竖直向下方向，则先控制激光测距仪7沿竖向行走轨道6以步长S继续向下行走一步，然后扫描龙门1再沿水平行走轨道5以步长S temp＝x′ 1-x 0水平行走一步。
其中：对于水平边界，如果下一个边界点处于当前边界点水平向前方向，则先控制扫描龙门1沿水平行走轨道5以步长S向前行走一步，然后再控制激光测距仪7沿竖向行走轨道6以步长S temp＝y′ 1-y 0继续行走一步(如果S temp＞0，则为竖直向上行走；如果S temp＜0，则为竖直向下行走；如果S temp＝0，则无需行走)，开始获取实际扫描边界点。如果下一个边界点处于当前边界点水平向后方向，则先控制扫描龙门1沿水平行走轨道5以步长S向后行走一步，然后再控制激光测距仪7沿竖向行走轨道6以步长S temp＝y′ 1-y 0继续行走一步，开始获取实际扫描边界点。整个边界点的搜索过程中皆遵循以上规则，下文中不再赘述。
步骤6.5.1.1：获取当前位置激光测距仪的测量值记为L temp，计算该位置与理论边界点P′ 1之间测量值变化率为
Figure PCTCN2022083579-appb-000042
步骤6.5.1.2：判断K temp是否等于
Figure PCTCN2022083579-appb-000043
如果否，扫描龙门1沿水平行走轨道5以步长S temp＝当前步长一半水平向前行走一步，回到步骤6.5.1.1；如果是，判断当前步长是否小于扫描龙门1的最小行走精度ΔS 0，如果是，成功获取第二个扫描 边界点，记为P 1(x 1,y 1,z 1)，记录对应测量值L 1，结束当前步骤；如果否，扫描龙门1沿水平行走轨道5以步长S temp＝当前步长一半水平向后行走一步，跳到步骤6.5.1.1。
步骤6.5.2：从第二个边界点P 1出发，捕获后续边界点。
计算P 1点上方相邻边界点P 2的理论位置P′ 2(如图13矩形构件竖向边界为例，分别显示了理想摆放姿态时测量面竖向边界在xy-平面投影211、通过已捕获边界点推测的测量面竖向边界在xy-平面投影212和实际摆放姿态测量面竖向边界213)：P′ 2的y轴坐标值为P 1的y轴坐标值+S，即y 1+S；计算y＝y 1+S与已捕获到的前两个边界点在xy-平面形成的函数交点为P′ 2，对于第三个边界点来讲，前两个边界点在xy-平面形成的函数为直线(即通过已捕获边界点推测的测量面竖向边界在xy-平面投影212)，对于其他同类型预制构件，已捕获的边界点在xy平面形成的函数可能为曲线，曲线函数可通过理论测量面模型求得。按此方法，可依此求得后续边界点理论位置P′ n，对应xy-平面的坐标为(x′ n,y′ n)，其z向坐标值为
Figure PCTCN2022083579-appb-000044
Figure PCTCN2022083579-appb-000045
将当前位置的激光测距仪7移动到P′ n点位置，即激光测距仪7沿竖向行走轨道6以步长S继续向上行走一步，然后扫描龙门1再沿水平行走轨道5以步长S temp＝x′ n-x n-1水平行走一步(如果S temp＞0，则为向前行走；如果S temp＜0，则为向后行走；如果S temp＝0，则无需行走)，开始获取实际扫描边界点P n。
步骤6.5.2.1：获取当前位置激光测距仪的测量值记为L temp，计 算该位置与理论边界点P′ n之间测量值变化率为
Figure PCTCN2022083579-appb-000046
其中：L n-1为上一个捕获到的边界点P n-1对应激光测量值。
步骤6.5.2.2：判断K temp是否等于
Figure PCTCN2022083579-appb-000047
如果否，扫描龙门1沿水平行走轨道5以步长S temp＝当前步长一半水平向前行走一步，回到步骤6.5.2.1；如果是，判断当前步长是否小于扫描龙门1的最小行走精度ΔS 0，如果是，成功获取第n个扫描边界点，记为P n(x n,y n,z n)，记录对应测量值L n，结束当前步骤；如果否，扫描龙门1沿水平行走轨道5以步长S temp＝当前步长一半水平向后行走一步，跳到步骤6.5.2.1。
其中：其中：随着已捕获边界点增加，测量面实际边界与通过已捕获边界点推测的理论边界越来越逼近，即S temp越来越小。且在理想摆放姿态时测量面竖向边界在xy-平面投影211捕获第二个边界点时，S temp＝S 1(即P 1和P′ 1在x轴方向的差值)，捕获第三个边界点是，S temp＝S 2(即P 2和P′ 2在x轴方向的差值)，S 2＜S 1，即S n≤S n-1。随着越来越多边界点被捕获，可以通过上述如影随形的方式不断缩小搜索范围的方法，达到加快捕获速度的目的。直至找到当前测量面所有边界点，形成边界点数据集Q 0。
步骤七：完成对测量面的数据采集，形成激光点云数据集Q 1。
控制扫描龙门1在x轴方向等间距行走，每行走一步，以等间隔步长控制竖向行走托盘8沿竖向行走轨道6行走，完成对当前竖线对应的目标扫描构件2测量面21的打点测量。以此方法逐步行走，完 成对目标扫描构件2测量面21的扫描(所有测量点形成如图7所示的网格状分布)。
其中：根据目标扫描构件2测量面21的边界数据，从P 0点出发，沿竖直方向控制激光测距仪7以系统设定步长在竖向行走轨道6上逐步移动，并在每一步移动后采集激光测量值。直至激光测距仪7移动到目标扫描构件2的上边界点，完成坐标原点上竖向线路的扫描；然后控制扫描龙门1沿水平行走轨道5以系统设定步长行走一步，完成当前水平位置对应的竖向线路的扫描；以此方法完成整个构件测量面边界范围内所有竖向线路的扫描，形成密集扫描数据，并记录全部扫描结果，形成激光点云数据集Q 1。
步骤八：可利用基于激光扫描点云的预制构件空间姿态反演校正算法校正目标扫描构件的摆放姿态。目标扫描构件2摆放时无法保证其测量面21与水平行走轨道5完全平行(即图6中的理想摆放姿态)，结合目标扫描构件2的理论模型数据和实际测量数据，利用空间几何转换，将未知摆放姿态的目标扫描构件2校正到理想摆放姿态(即将当前摆放姿态下测量面的测量值投影到理想摆放姿态下的测量面所在平面)。
具体而言，校正的详细步骤如下：
步骤8.1：开始扫描前已将目标扫描构件2的理论模型导入到系统中，并将理论模型置于步骤六中建立的坐标系中，并将图6中理论模型测量面底部左角点P′ 0与图7中实际测量中目标扫描构件2测量面21的P 0点重合，理论模型测量面底边界与x轴重合，理论模型测量面 左边界与y轴重合，得到理论模型测量面数据集Q′ 1。
步骤8.2：将实际摆放中的测量面与理论模型的测量面相耦合，完成目标扫描构件2的姿态反演校正，以便后续步骤对异常点的判断。
实际测量中目标扫描构件2在测量前无法保证像理论模型一样完全贴合坐标轴方向摆放，在x，y，z三个方向都可能产生夹角。
步骤8.2.1：从边界点数据集Q 0中提取在x、y轴方向取值最大和最小的点以及坐标原点，取其中不位于同一条直线上的三个点作为特征点，三个特征点的坐标分别是A(x 1，y 1，z 1)、B(x 2，y 2，z 2)、C(x 3，y 3，z 3)。利用这些特征点计算出目标扫描构件2测量面所在平面ABC即特征点面的方程为式1：
Figure PCTCN2022083579-appb-000048
步骤8.2.2：首先将测量面校正到xy-平面，即式2：
z＝0；   2)
结合步骤8.2.1中特征点面的方程，得到特征点面和xy-平面之间的交线L为式3：
ax+by+d＝0；      3)
其中：a、b、c为函数常量；
由特征点面、xy-平面、平面交线L可得到两个平面夹角α；
其中：平面ABC上取三个特征点中不位于交线L上的任意一点向平面交线L做垂线，垂足为点P ⊥，以点P ⊥为垂足，在xy-平面上做平面交线L的垂线，两垂线之间的夹角即为平面夹角α；
步骤8.2.3：将激光点云数据集Q 1和边界点数据集Q 0中所有测量 点校正到xy-平面，校正后平面记为平面ABC′。
从激光点云数据集Q 1和边界点数据集Q 0中依次取测量点数据并记为测量点P i，P i向交线L做垂线，然后以交线L为旋转轴，以垂足为旋转圆心，以α x为旋转角向xy-平面旋转，最终P i点旋转到xy-平面上的新点P i′即为P i点校正后的位置。
直至激光点云数据集Q 1和边界点数据集Q 0中所有的测量点全部完成校正形成校正后的激光点云数据集Q 1_1和边界点数据集Q 0_1，校正后的特征点面为平面ABC′，此时平面ABC′已经与xy-平面重合。
步骤8.2.4：通过步骤8.2.3，校正后的平面ABC′已经与xy-平面重合，但实际测量中目标扫描构件2姿态未知，校正后其测量面左底角点未必与坐标原点P 0(0,0,0)点重合。从校正后的边界点数据集Q′ 0提取校正后的测量面左底角点P 角l(x l,y l,z l)，平移点P 角l使其于坐标原点P 0点重合，即将点P 角l点沿x轴平移x l，沿y轴平移y l。依此方法，将校正后的测量面激光点云数据集Q 1_1和测量面边界点数据集Q 0_1中所有点均沿x轴平移x l，沿y轴平移y l，至此，平移后的测量面已经与理想摆放姿态下的测量面处于同一平面，并且两个测量面的左底角点重合。平移后的激光点云数据集Q 1_1和边界点数据集Q 0_1分别记为Q 1_2和Q 0_2。
步骤8.2.5：通过步骤8.2.4，平移后的测量面已经与理想摆放姿态下的测量面处于同一平面，并且两个测量面的左底角点重合，此时测量面左右底角点连线与x轴之间可能尚存在夹角β(如图8中所示)，可将平面ABC′以坐标原点P 0点为中心，沿z轴旋转使目标扫描构件将 两个底角点与x轴重合。从平移后的边界点数据集Q 0_2提取的测量面左底角点P 角l(0,0,0)(已平移到坐标原点)和右底角点P 角r(x r,y r,z r)，两个底角点所在直线相对于x轴的斜率K 1为
Figure PCTCN2022083579-appb-000049
由于目标扫描构件理想摆放时两个底角点所在直线相对于x轴的斜率K 2＝0，得到校正后的边界点数据集Q 0_2中两个底角点所在直线和理想摆放时两个底角点所在直线夹角为
Figure PCTCN2022083579-appb-000050
步骤8.2.6：将平移后的激光点云数据集Q 1_2和边界点数据集Q 0_2中测量点值逐个绕z轴方向以β为旋转角，以构件测量面左底角点P 角l为中心向x轴旋转，得到最终校正值，重复以上操作直至所有测量点完成最终校正，得到旋转校正后的激光点云数据集Q 1_3和边界点数据集Q 0_3，完成目标扫描构件2的姿态反演校正。
步骤九：可利用基于密集扫描数据的预制构件拼装面局部缺陷识别方法对校正后目标扫描构件2测量面21的测量数据进行分析，剔除目标扫描构件2测量面21的小气泡凹坑以及石子颗粒等影响因素对应的测量值。具体步骤如下：
步骤9.1：遍历校正后的激光点云数据集Q 1_3和边界点数据集Q 0_3，可利用理论模型比对法，比对逐个集合中测量值与理论模型数据集Q′ 1中相同xy坐标位置的测量值是否相同，差距超过测量打点最小精度ΔS 0(根据目标扫描构件2的扫描精度要求，使用者可在步骤二中自行设定)的坐标点存入到异常点数据集Q △0中，直至将校正后的激光点云数据集Q 1_3和边界点数据集Q 0_3遍历一遍。
步骤9.2：将异常点数据集Q △0中合理存在的异常点进行排除， 这些合理存在的异常点包含如图6所示的目标扫描构件张拉孔洞23和目标扫描构件凹凸榫24所对应区域的测量点，具体操作如下：
根据目标扫描构件2结构模型中的合理存在异常点(即目标扫描构件张拉孔洞23和目标扫描构件凹凸榫24)所处的区域，得出对应的xy坐标范围。遍历异常点数据集Q △0中所有测量点，将处于上述合理存在异常点区域内的数据进行剔除，不做分析，剔除合理存在异常点后的异常点数据集定义为Q △1。
步骤9.3：遍历剔除合理存在异常点后的异常点数据集Q △1，逐个取出该数据集中的测量点，并以当前点为中心，利用关联搜索法对比当前点与测量面上周边测量点的测量值关系。获取当前点在校正后的激光点云数据集Q 1_3中方圆5*5个点区域内所有测量点的测量值，将周边区域测量点的测量值依次与当前测量点的测量值对比，如果周边测量点的测量值全部大于或小于当前测量点的测量值，则判定当前测量点为局部缺陷点，将该点从遍历剔除合理存在异常点后的异常点数据集Q △1中剔除。重复此步骤，直至完成目标扫描构件2所有局部缺陷点数据的剔除，将剔除局部缺陷点后的异常点数据集Q △1记为最终异常点数据集Q △。
举例来说，局部凹陷测量矩阵如下：
Figure PCTCN2022083579-appb-000051
举例来说，局部凸起测量矩阵如下：
Figure PCTCN2022083579-appb-000052
步骤十：确定目标扫描构件的拼装面制作精度，并在构件表面对误差区域喷涂标识。
其中：拼装面制作误差分为鼓包和凹坑两种形式，其中凹坑区域不影响预制构件拼装，而鼓包区域可能导致预制构件在装配过程中拼不上、接缝拉不严等情况，影响预制构件现场使用，需将鼓包区域进行打磨。
剔除最终异常点数据集Q △中的非鼓包点，计算最终异常点数据集Q △中的鼓包区域边界以及鼓包高度，并对目标构件中对应的鼓包区域喷涂标识，其步骤如下：
步骤10.1：对比理论模型以剔除最终异常点数据集Q △中不影响构件拼装的非鼓包点。
从最终异常点数据集Q △中逐个选取点P n(x n,y n,z n)，根据当前点位在xy-平面中的位置，从理论模型测量面数据集Q′ 1中获取同等位置对应的理论点位P′ n(x′ n,y′ n,z′ n)，即：x n＝x′ n，y n＝y′ n。
判断异常点是否为非鼓包点(以图7中靠近xy-面的测量面为例)：对比z n和z′ n大小：当z n≥z′ n时，当前异常点为非鼓包点(凹坑点或平整)，否则为鼓包点。
如果是非鼓包点：将当前点数据从最终异常点数据集Q △中剔除，选取下一个测量点，继续执行步骤10.1；
如果是鼓包点：计算当前鼓包点的鼓包高度h n为h n＝z′ n-z n。
根据上述方法计算完当前测量面内所有鼓包点高度，并存入当前测量面鼓包高度数据集Q Δh。
步骤10.2：遍历剔除非鼓包点后的最终异常点数据集Q Δ(经过步骤10.1剔除非鼓包点之后，数据集内只剩下鼓包异常点)，利用关联搜索法以当前点为中心，搜索当前点在x轴和y轴方向九宫格内其他8个测量点，查看是否也为鼓包异常点(即其他八个点中存在最终异常点数据集Q Δ中的测量点)：
如果是，以新找出的鼓包异常点为中心，搜索其九宫格范围内是否存在鼓包异常点。重复此步骤，直到临近的测量点没有鼓包异常点为止，将此过程中所有鼓包异常点形成鼓包区域数据集。并记录该鼓包区域数据集中鼓包高度最大的点位P max以及鼓包高度值h max。
如果否，将当前鼓包异常点标记为干扰数据并从最终异常点数据集Q Δ中剔除。
重复当前步骤，直至完成对测量面的鼓包区域的搜索和记录。
步骤10.3：控制扫描门架1和竖向行走托盘8运送描点喷绘机械伸缩臂9到测量面的鼓包区域边缘对应位置，控制描点喷绘机械伸缩臂9伸长到目标扫描构件2测量面21，完成一个点的描绘，逐个走完整个误差区域边界，完成对误差区域的标注。
步骤十一：按照以上步骤，完成目标扫描构件2所有测量面的检测。生成电子检测报告，在报告中对影响构件拼装的测量面鼓包区域以及该区域的数据做出明确标识，完成对目标扫描构件2的全部检测。
由此可见，本发明的优点在于：
1、采用门架式行走结构，门架上装载有多方向移动电机和轨道，能够完成对任意形状的拼装面的测量工作，并且门架上装载有各类传感器，对轨道、门架以及空气的温度和湿度进行实时监测，通过温湿度的监测结果对测量结果进行校正，保障系统可以在复杂气候条件下工作而不会影响精度。
2、在实际测量过程中，目标扫描构件的拼装面上可能会有凹凸榫槽、橡胶槽道、气泡凹坑、石子颗粒、边缘轻度破损等各种影响制作精度扫描的因素，本专利以目标扫描构件理论模型为参照，既要从数据上处理各种影响因素，也要利用算法对测量数据做出处理。先利用逐步逼近、如影随形等方法快速找出目标扫描构件的边界数据，再利用关联搜索算法、理论模型参照法、整体趋势推演法等算法手段一步步剔除影响因素，完成对预制构件拼装面制作精度的计算。
3、在完成一块构件的扫描测量之后，系统计算出鼓包区域的边界点坐标数据集，控制门架携带喷绘装置逐个形走到各个边界点对应位置，将鼓包边界点喷绘在构件表面，方便后期打磨修复时工人比对检测报告可以快速的找出鼓包位置，极大降低打磨难度和工作量。
显而易见的是，以上的描述和记载仅仅是举例而不是为了限制本发明的公开内容、应用或使用。虽然已经在实施例中描述过并且在附图中描述了实施例，但本发明不限制由附图示例和在实施例中描述的作为目前认为的最佳模式以实施本发明的教导的特定例子，本发明的范围将包括落入前面的说明书和所附的权利要求的任何实施例。

Claims (10)

  1. 一种预制构件拼装面制作精度智能扫描和误差自动标识方法，其特征在于包含如下步骤：
    步骤一：准备步骤，将目标扫描构件吊装到构件扫描台上的测量区域内；
    步骤二：启动步骤，打开仓房的电动卷闸门，启动扫描龙门开启扫描检测；
    步骤三：调节步骤，调节竖向行走托盘的高度，使激光测量设备的光线贴近于构件扫描台上表面高度；
    步骤四：扫描龙门沿水平行走轨道从仓房中出发向前行走，行走过程中实时采集激光测距仪的激光测量值和扫描龙门温湿度传感器、竖直轨道温湿度传感器、水平轨道温湿度传感器的监测值；
    步骤五：行走中实时采集激光测距仪的激光测量值，判断激光测距仪的光点是否打在了目标扫描构件上，如果是，停止扫描龙门的水平前进；
    步骤六：确定目标扫描构件中测量面的边界轮廓，所有边界点形成边界点数据集Q 0；
    步骤七：完成对测量面的数据采集，形成激光点云数据集Q 1；
    步骤八：校正目标扫描构件的摆放姿态，将目标扫描构件校正到理想摆放姿态；
    步骤九：对校正后目标扫描构件测量面的测量数据进行分析，剔除目标扫描构件测量面的异常测量值；
    步骤十：确定目标扫描构件的拼装面制作精度，并在目标扫描构 件表面对误差区域喷涂标识；
    步骤十一：生成检测报告，完成对目标扫描构件的全部检测。
  2. 如权利要求1所述的预制构件拼装面智能扫描和误差自动标识方法，其特征在于：步骤五的具体方法如下：当激光测距仪光点打在目标扫描构件上时左侧激光测距仪测量值L l+右侧激光测距仪测量值L r≤L时，判定检测到目标扫描构件2，并记录两个激光测距仪当前测量值分别为
    Figure PCTCN2022083579-appb-100001
    和
    Figure PCTCN2022083579-appb-100002
  3. 如权利要求1所述的预制构件拼装面智能扫描和误差自动标识方法，其特征在于：步骤六中以步骤五的停止点为起点，利用迂回折半、逐步逼近方法搜索当前高度对应的目标扫描构件在水平方向的边界点，记为目标扫描构件的第一个边界点P 0，以水平行走轨道前进方向为x轴正方向、以竖向行走轨道向上方向为y轴正方向、以激光测距仪测量值为z轴坐标值、以目标扫描构件的第一个边界点P 0为坐标原点，建立xyz坐标系，从坐标原点P 0出发，沿顺时针方向捕获目标扫描构件测量面边界周边轮廓其余边界点，过程中结合目标扫描构件的理论数据模型，利用快速迭代的方法不断减少后续边界点的搜索次数，加快其余边界点捕获速度，最终将所有边界点形成边界点数据集Q 0。
  4. 如权利要求3所述的预制构件拼装面智能扫描和误差自动标识方法，其特征在于还包含如下步骤：
    步骤6.1：扫描龙门每向前行走一步，采集当前位置两侧激光测距仪测量值，分别记录为
    Figure PCTCN2022083579-appb-100003
    和
    Figure PCTCN2022083579-appb-100004
    计算当前位置与上一步位置两侧 激光测距仪测量值的变化率，左侧测量值变化率为
    Figure PCTCN2022083579-appb-100005
    右侧测量值变化率为
    Figure PCTCN2022083579-appb-100006
    对比
    Figure PCTCN2022083579-appb-100007
    和
    Figure PCTCN2022083579-appb-100008
    大小：
    当
    Figure PCTCN2022083579-appb-100009
    时，两侧的激光测距仪光点前后两次测量均打在了各自对应的测量面上，至此完成目标测量面的检测；
    当
    Figure PCTCN2022083579-appb-100010
    时，其中一侧的激光测距仪光点前后两次测量并未全部打在了对应的测量面上；扫描龙门1沿水平行走轨道5以步长S继续水平向前行走，直到
    Figure PCTCN2022083579-appb-100011
    步骤6.2：将步骤6.1中目标测量面对应的激光测距仪沿竖向行走轨道以步长S继续向上行走一步，确定测量值在y轴方向上的变化率；
    步骤6.3：移动激光测距仪退回到当前测量面内第一个测量点，并以迂回折半行走方式开始获取第一个扫描边界点P 0；
    步骤6.4：获取第一个扫描边界点P 0后，以P 0为坐标原点，以扫描龙门水平前进方向为x轴正方向，以激光测距仪扫描竖直向上方向为y轴正方向，以激光测距仪测量值为z轴坐标值，建立xyz坐标系，得到P 0点坐标(x 0,y 0,z 0)即(0,0,0)；
    步骤6.5：以第一个扫描边界点为基础，进行后续边界点捕获，并完成目标扫描构件所有边界点的捕获。
  5. 如权利要求3所述的预制构件拼装面智能扫描和误差自动标识方法，其特征在于：步骤八的具体步骤如下：
    步骤8.1：将理论模型置于步骤六中建立的坐标系中，并将理论模型测量面底部左角点P′ 0与目标扫描构件测量面的P 0点重合，理论模 型测量面底边界与x轴重合，理论模型测量面左边界与y轴重合，得到理论模型测量面数据集Q′ 1；
    步骤8.2：将实际摆放中的测量面与理论模型的测量面相耦合，完成目标扫描构件的姿态反演校正，得到校正后的激光点云数据集Q 1_3和边界点数据集Q 0_3。
  6. 如权利要求5所述的预制构件拼装面智能扫描和误差自动标识方法，其特征在于：步骤九的具体步骤如下：
    步骤9.1：遍历校正后的激光点云数据集Q 1_3和边界点数据集Q 0_3，比对激光点云数据集Q 1_3和边界点数据集Q 0_3中测量值与理论模型数据集Q′ 1中相同xy坐标位置的测量值是否相同，差距超过测量打点最小精度ΔS 0的坐标点存入到异常点数据集Q △0中；
    步骤9.2：将异常点数据集Q △0中合理存在的异常点进行排除，剔除合理存在异常点后的异常点数据集定义为Q △1；
    步骤9.3：遍历剔除合理存在异常点后的异常点数据集Q △1，逐个取出该数据集中的测量点，并以当前点为中心，利用关联搜索法对比当前点与测量面上周边测量点的测量值关系；获取当前点在校正后的激光点云数据集Q 1_3中方圆5*5个点区域内所有测量点的测量值，将周边区域测量点的测量值依次与当前测量点的测量值对比，如果周边测量点的测量值全部大于或小于当前测量点的测量值，则判定当前测量点为局部缺陷点。
  7. 一种预制构件拼装面智能扫描和误差自动标识系统，包括工作台、仓房、扫描龙门、水平行走轨道和构件扫描台，其特征在于：
    所述仓房位于工作台的一端且设有容纳扫描龙门的容纳空间以及电动闸门，所述水平行走轨道为两条轨道且沿工作台的两侧延伸，其一端延伸至仓房内，所述构架扫描台位于工作台上且设置于水平行走轨道的中间，以供目标扫描构件的放置，扫描龙门可滑动的设置于水平行走轨道上，所述扫描龙门为门形的行走机构，其行走在构件扫描台两侧的水平行走轨道上，所述扫描龙门的两内侧各有一条竖向行走轨道，在竖向行走轨道上各有一个竖向行走托盘以沿竖向行走轨道上下移动，所述竖向行走托盘上设有激光测量设备和描点喷绘机械伸缩臂的安装座。
  8. 如权利要求7所述的预制构件拼装面智能扫描和误差自动标识系统，其特征在于：所述激光测量设备固定于安装座的上端且为单点激光测距仪，所述描点喷绘机械伸缩臂固定于安装座的下端且可通过伸缩控制在指定表面喷绘色斑。
  9. 如权利要求7所述的预制构件拼装面智能扫描和误差自动标识系统，其特征在于：所述扫描龙门的两侧底部各有一条扫描龙门底梁，在扫描龙门底梁的两端各设有一个水平行走限位装置以限制扫描龙门在水平行走轨道上的行走范围，所述扫描龙门底梁的中间区域设置有一个水平行走电机和一个水平行走编码器以驱动扫描龙门前进后退和记录扫描龙门所处位置。
  10. 如权利要求9所述的预制构件拼装面智能扫描和误差自动标识系统，其特征在于：所述竖向行走轨道和扫描龙门底梁的连接点下端设有一个竖向行走电机和一个竖向行走编码器以驱动竖向行走托 盘上下移动和记录竖向行走托盘所处的位置。
PCT/CN2022/083579 2022-03-25 2022-03-29 预制构件拼装面智能扫描和误差自动标识系统及方法 WO2023178713A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2023523264A JP2024519633A (ja) 2022-03-25 2022-03-29 プレハブ部材組立面のインテリジェント走査及び誤差の自動識別システム並びに方法

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210300941.1 2022-03-25
CN202210300941.1A CN114719792B (zh) 2022-03-25 2022-03-25 预制构件拼装面智能扫描和误差自动标识系统及方法

Publications (1)

Publication Number Publication Date
WO2023178713A1 true WO2023178713A1 (zh) 2023-09-28

Family

ID=82240057

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2022/083579 WO2023178713A1 (zh) 2022-03-25 2022-03-29 预制构件拼装面智能扫描和误差自动标识系统及方法

Country Status (3)

Country Link
JP (1) JP2024519633A (zh)
CN (1) CN114719792B (zh)
WO (1) WO2023178713A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
TWI823567B (zh) * 2022-09-15 2023-11-21 財團法人工業技術研究院 物件材積量測系統及其方法
CN117092114B (zh) * 2023-10-16 2023-12-29 苏州德机自动化科技有限公司 一种基于ai的外观检测系统

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101666629A (zh) * 2008-09-04 2010-03-10 三星重工业株式会社 测量曲面的系统和方法
US20160083283A1 (en) * 2014-09-19 2016-03-24 Asahi Glass Company, Limited Ceramic member and defect test system
CN106127770A (zh) * 2016-06-27 2016-11-16 感知控股集团有限公司 轮廓测量方法及系统
CN106706666A (zh) * 2017-03-28 2017-05-24 温馨 一种大型水泥构件表面缺陷的检测与标示装置
CN106996758A (zh) * 2017-05-13 2017-08-01 朱嘉懿 一种大型预制构件表面缺陷检测与标示装置
CN108253896A (zh) * 2018-03-06 2018-07-06 大连理工大学 一种基于激光传感器的轿车尺寸测量装置
CN112097070A (zh) * 2020-07-22 2020-12-18 南京工业大学 移动龙门架式装配式建筑预制构件生产辅助装置与方法
CN113450379A (zh) * 2021-07-02 2021-09-28 湖南国天电子科技有限公司 一种异形工件断面轮廓线提取和分析方法和装置

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US10077061B2 (en) * 2015-03-12 2018-09-18 Mi-Jack Products, Inc. Profile detection system and method
JP6607228B2 (ja) * 2017-04-20 2019-11-20 Ｊｆｅスチール株式会社 校正片、校正方法、形状測定システムおよび形状測定方法
CN111536901B (zh) * 2020-05-21 2021-11-30 南京航空航天大学 一种基于现场龙门与机器人组合式三维激光扫描测量方法
CN215572719U (zh) * 2021-08-04 2022-01-18 深圳大学 一种竖向预制构件的检测装置

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101666629A (zh) * 2008-09-04 2010-03-10 三星重工业株式会社 测量曲面的系统和方法
US20160083283A1 (en) * 2014-09-19 2016-03-24 Asahi Glass Company, Limited Ceramic member and defect test system
CN106127770A (zh) * 2016-06-27 2016-11-16 感知控股集团有限公司 轮廓测量方法及系统
CN106706666A (zh) * 2017-03-28 2017-05-24 温馨 一种大型水泥构件表面缺陷的检测与标示装置
CN106996758A (zh) * 2017-05-13 2017-08-01 朱嘉懿 一种大型预制构件表面缺陷检测与标示装置
CN108253896A (zh) * 2018-03-06 2018-07-06 大连理工大学 一种基于激光传感器的轿车尺寸测量装置
CN112097070A (zh) * 2020-07-22 2020-12-18 南京工业大学 移动龙门架式装配式建筑预制构件生产辅助装置与方法
CN113450379A (zh) * 2021-07-02 2021-09-28 湖南国天电子科技有限公司 一种异形工件断面轮廓线提取和分析方法和装置

Also Published As

Publication number Publication date
CN114719792A (zh) 2022-07-08
CN114719792B (zh) 2022-12-13
JP2024519633A (ja) 2024-05-21

Similar Documents

Publication Publication Date Title
US11188688B2 (en) Advanced automated process for the wing-to-body join of an aircraft with predictive surface scanning
WO2023178713A1 (zh) 预制构件拼装面智能扫描和误差自动标识系统及方法
CN111268530B (zh) 电梯井道测量、定位及电梯安装方法和设备
CN108981580B (zh) 一种起重机轨道在线检测装置及方法
CN113009456B (zh) 车载激光雷达数据校准方法、装置及系统
CN108824816B (zh) 一种高空大跨网架滑移定位安装及监测方法
CN210310341U (zh) 轨道交通机车车辆巡检机器人
EP3567340A1 (en) Visual inspection arrangement
Yan et al. Multi-line laser structured light fast visual positioning system with assist of TOF and CAD
CN117361331B (zh) 一种基于激光雷达的集装箱吊具位姿检测方法
Durdevic et al. Lidar assisted camera inspection of wind turbines: experimental study
CN112884823A (zh) 面向实测实量应用的靠尺模拟点云选取与计算方法
JP2019056671A (ja) 壁面損傷検査装置
CN114646263B (zh) 预制构件拼装面制作精度智能扫描和误差自动标识系统
JP4387540B2 (ja) エレベータの寸法測定装置
CN114485462A (zh) 一种用于轨道交通的车辆外形轮廓检测系统及方法
CN115562284A (zh) 一种高铁箱梁巡检机器人实现自动巡检的方法
CN114705134A (zh) 一种电梯导轨垂直度与平行度自动检测装置
CN207866202U (zh) 一种叶轮平面测量装置
CN114663403B (zh) 基于密集扫描数据的预制构件拼装面局部缺陷识别方法
CN114674226B (zh) 预制装配式构件扫描快速捕获边界的方法
CN114674225B (zh) 预制装配式构件拼装面制作误差识别算法
CN114707100B (zh) 基于激光扫描点云的预制装配构件空间姿态反演校正方法
CN111366086B (zh) 面向自动化装车的车厢位姿测量系统及方法
WO2023112828A1 (ja) 作業装置および作業装置の制御方法

Legal Events

Date Code Title Description
WWE Wipo information: entry into national phase

Ref document number: 2023523264

Country of ref document: JP

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 22932774

Country of ref document: EP

Kind code of ref document: A1