real-world three-dimensional measuring of built environment with a portable wire-based coordinate-measuring machine.

[Abstract]
The long-term strategy of the forest products industry is to improve the refinement of products, thus improving the value of products.
This strategy applies to the primary and secondary processing of wood products.
Downstream of the value stream, in order to add value and efficiency in the supplier process, different types of knowledge are required.
In this study, the focus is onbuilt three-dimensional (3-D)
Sensing as a means to improve the level of product prefabrication when providing engineersto-
Order woodworking products from the construction industry. A 7-Three metres-
Portable wire shaft-
Based on coordinates-
Testing Machine (PWCMM)
Evaluation in the implementation of-built site-
Size verification in 3-D.
This is the necessary means to transfer the woodworking product accessories to the digital field during the design phase, thus improving the level of prefabrication and automation when providing engineersto-
Order woodworking products.
PWCMM has been used to replicate different construction sites for the following benefits
Space Information designed, manufactured and entered online as a supplier
On-site assembly process.
The evaluation results show that the accuracy of each coordinate position can be within millimeter range.
However, the problem can still meet the requirements of accuracy and ease of use.
On-site dimensional verification when supplying woodworking products.
The problem of error lever and low measurement resolution limits the practical possibility of reproducing the accuracy and detail level of-
Building environment.
The sawmill industry has a tradition of supplying large quantities of primary products
Processing and refining wood products with limited value in a vast geographical area.
This brings considerable export revenue to the world. export-
Major countries such as Canada, the United States, Sweden and Germany (
Sweden Forest Industry 2013).
However, these products face the challenge of demand. In this industry, a clear strategy is to increase the value of the products.
The wood processing industry is also struggling with this strategy.
The focus of this research is on supply engineersto-
Order woodworking products, hereinafter referred to as \"woodworking products \".
\"This is a secondary wood --
The processing industry with the construction industry as the main customer.
Worldwide, the construction industry is one of the most important elements in every economy and a major customer of most wood products.
Therefore, it is possible to reveal value by increasing interaction with the industry
Increase opportunities for wood products.
Highly refined woodworking products-of-a-
Entrance, glass partition, door, window, interior, cabinet accessories, special accessories, stairs and other wood products.
Designed to meet specific customer needs.
These components are designed and manufactured outside the factory, where they are packaged and transported to the construction site where they are assembled.
This is a process with two main parts :(1)
More efficient factory production, and (2)
Work on less efficient and/or labor-intensive construction sites.
Labor
The intensive work of the construction site depends on the level of product prefabrication and the degree to which the finished woodworking products match the expected location. Reliable as-
The building space data of the construction site is essential for the manufacture of fine woodworking products that can be assembled efficiently.
At present, assembly work usually consumes half of the carpentry
The product supply budget and the manual assembly of components are the main reasons for the consumption of Assembly working hours.
So by improving interaction with customers and improving on-
By reducing the spatial uncertainty of the surrounding environment of the woodworking products, the efficiency of on-site assembly is improved.
Use of Building Information Models (BIM)
More and more implementation in construction, engineering and construction (AEC)domain.
These semantic rich Threedimensional (3-D)
The model that stores the information in a single integration source was originally designed to enhance planning, visualization and communication during the design process, and to help identify errors during construction, process simulation, and space planning during management (Sacks et al.
2004, Akinci and others.
2006, Xiong man 2008, Xiong and others. 2013).
With the history created in the design process,-designed or as-
In the plan, BIMs is the dominant BIMs in the AEC field.
These BIMs may be very different from the actual current situation. built or as-
Conditions of facilities.
These differences come from a variety of different sources such as unlicensed design changes, unexpected errors in construction, and subsequent renovations (Xiong et al. 2013).
Therefore, it must be based on BIMs-built or as-is conditions. Hereafter as-
Built is used to refer to these two terms. Further, as-
Plan to include the following twoplanned and as-designed terms.
Generation and use of-
Architectural geometric conditions in buildings gain momentum in the research literature, covering quality-
Evaluation, progress and productivity monitoring of construction vehicles, material tracking and automatic wiring. (Tang et al.
2009, Huber and others.
2011, Turkan and others.
2012, Anil and others.
2013, Allegri-Fraga et al.
2013, Kim and others.
2013 Xiong and others.
2013, boscht and Guenet 2014, boscht and others. 2014).
The focus is on the automation of the process of acquiring-
Build geometry from different 3-
D. Sensing Technology (Anil et al.
2013, Kim and others.
2013 Xiong and others. 2013).
In the next work, I will focus on achieving the following goals
The geometry built from the construction site can be used as a means to increase the degree of automation in the process of supplying the woodworking products.
Uncertainty of-
Build geometric conditions with the as-method currently used
It has been proved that in the supply of the woodworking products, the built-in verification can cause multiple types of waste: unnecessary transportation, movement, waiting, over-processing, over-production and defects (Foreman et al. 2012).
Potentially, this waste can be eliminated through different automated operations based on BIMs, accurate-
The geometry of the build and the semantics of the build process.
Here are three examples of automated operations to eliminate waste :(1)
Early in the supply process, move the manual accessories from the end of the supply process to the digital environment in order to automate the product-to-
Room fittings and allow physical fittings of product components using CNC machinery; (2)
Because of the size limit of in-
On-site transport routes, design the dimensions of packages in the digital field to optimize on-site Transportsite delivery; (3)
By ensuring that the environment adjacent to the woodworking product is prepared for assembly, the supply process is synchronized with the construction process.
These examples show that the certainty of as-increases
The geometric conditions of construction have the potential to significantly increase the efficiency of supplying woodworking products to the construction industry.
The purpose of this study is to evaluate whether portable wires
Based on coordinates-
Testing Machine (PWCMM)
Accurate enough and available to perform actual operations
Building site verification in 3-3
D. During the design process, the installation of woodworking products can be carried out in the digital field.
Assuming PWCMM can eliminate the size uncertainty of-
Construction sites reach the same level as Carpentry
Meet the actual use requirements.
In the use of the PWCMM function, the sensor accuracy and the impact on the measurement uncertainty were studied, and precision and usability experience were demonstrated from four cases.
The original contribution to this article is about 3-
D. Use PWCMM to sense the building environment and generate real-world digital models that can be used to install woodworking products in the digital field during design.
Theoretical overview of-Built 3-
D measurement geometry in 3-
D is a widely used technology in many different industrial applications.
The overall objective is to achieve the following objectives
Geometric information of the object under test (
Pereira and Hocken 2007, Cuypers, etc.
2009, Barini and others. 2010).
The culture of ensuring the correct geometry of the components is the foundation of the industrialized process.
The use of Henry Ford and interchanged components is an example.
Three technical categories can be identified at present :(1)coordinate-
Measuring equipment; (2)laser scanners; (3)
Optical measurement technology.
These three methods are used in different industrial environments, and the resolution and scale are different depending on the application.
Below, the focus is on coordination-
Testing Machine (CMMs).
CMMs converts the position of the measuring probe to 3-
D coordinate system (Schwenke et al. 2008).
Two types of CMMs can be identified: traditional CMMs and portable CMMs.
The traditional CMMs are static and are widely used in the control stations of the manufacturing industry.
These are very accurate (0. 3 to 2. 0 pm)
, But they usually control known 3-
D model instead of describing an object and generating a 3-
D model observed from coordinates (Barini et al.
2010, Reze measurement 2014, Nikon measurement 2014).
The two most common types of portable CMMs, joint arm coordinates-
Measuring Machine (AACMMs)
And optical portable CMMs.
AACMMs is a measuring arm with five to seven rotating joints or shafts, measured with ASME b89. 4. 22 single-
Accuracy of 20 to 140 points 【micro]
M is within the scope of work of 1. 5-to 4. 5-m radius (
American Society of Mechanical EngineersASME]
2004, Sladek, etc.
2013, Hexagon measurement 2014).
Optical portable CMMs is an optical camera
Triangle measurement system based on handheld probe.
The probe location is perceived by a marker on the probe, which is compared to a set of reference markers. The ASME B89. 4. 22 single-
Point repeatability 37 to 95 【micro]
M, the scope of work is a coordinate system up to 17 [m. sup. 3](Cuypers et al.
2009, Creaform measurement solution 2014, Nikon measurement 2014).
Most of the research on CMMs involves traditional CMMs and mobilick-Miguel et al. (1996)
Claims that their measurements may be affected by various errors.
This is reflected in many subsequent studies, such as the uncertainty of coordinate measurement (Wilhelm et al. 2001)
Source of geometric errors (Schwenke et al. 2005, 2008)
Dynamic error of CMM (
Jin Wen and Yan Ling 2011)
Separation of machine and probe errors (Nafi et al. 2011).
Most of the content of this study is about understanding the uncertainty of measurement and shows that the level of uncertainty of traditional CMMs is small relative to the geometric accuracy of the construction industry.
However, the traditional CMMs are stationary and therefore measured-
Build geometry at the construction site.
The study of portable CMMs is limited, but there is still considerable effort in understanding the uncertainty of measurement and calibration and error correction.
Shimojima, etc. (2002)
A calibration method is recommended that is better than the accuracy specified by the manufacturer.
This is necessary because the traceability of the measuring machine is difficult because the calibration is done by the manufacturer and the unpublished method, similar to the PWCMM studied.
Other studies involve the identification and modeling of AACMM errors and propose correction processing for improved performance (
Santolaria, etc.
2008, Sladek, etc. 2013)
Or find the best measurement area (Zheng et al. 2012)
Or suggest a thermal error correction model that does not affect the calibration conditions (
Santolaria, etc. 2009).
Portable CMMs are not used in construction and carpentry
The production environment has been identified.
In construction-
Related research of-
Other 3-construction measurement
Laser scanning is the most commonly used technology.
In this study, the attention to the accuracy of the measurement is limited, but some quantitative research on the accuracy of the laser
The scanner data found that the removal of mixed pixels will lead to significant measurement errors (Tang et al. 2009).
In addition, it found laser
Scanner Resolution, distance to object, object color, object radius and laser-
Beam strength is the five variables that lead to the largest measurement error (Shen et al. 2013).
Pay little attention to tolerances and reduce the uncertainty of scanning to achieve the dimensional reliability and information required for \"production metering (Kunzmann et al. 2005).
Planning Guide 3-
D imaging of the build environment specifies the general level of accuracy and detail (
General Services Administration 2009)
, But there is no definition of accuracy for different industries of construction engineering.
Since the guilds that provide woodworking products account for a lower proportion of the research literature, this is also valid for the guilds.
Therefore, the required measurement accuracy depends largely on the situation, which hinders the-
Measured at construction sites. Methods 3-D sensing of-
Building site geometry with portable wires
Based on coordinates-
Measuring Machine, Proliner 8 (Prodim 2014)
The research was carried out in the context of supplying woodworking products.
Through mobile products, the accuracy and availability of the machine are analyzed from the perspective of improving the automation of the supply processto-
Rooms suitable for digital environment.
Machine capabilities were reviewed and data from four cases were obtained through interviews, direct observation, participation and control documents.
Notes, photos, and files in documents are the basis of the analysis.
In these cases, PWCMM is used for 3-D sensing of-
Construction-Site size.
The machine capability and case experience measured by PWCMM were evaluated for the potential to eliminate spatial uncertainty through the following criteria: 1. Accuracy, i. e.
, The opportunity to eliminate the spatial uncertainty of the construction site to a level comparable to the tolerance requirements claimed by the Woodworking product ([+ or -]1 mm)
It is better to comply with the \"measurement gold rule\" where the measurement uncertainty should not exceed a tenth or maximum of the tolerance requirement, so [+ or -]0. 1 to 0. 2 mm (Beckert et al. 2010). 2. Usability, i. e.
, The opportunity for technology to adapt to carpentry-
Process of product supplier
This involves measuring range, portability, quality of information, efficiency in performing measurements and necessary data processing, operation and reconstruction 3-
D geometry used for measurement purposes, quality improvement of project information communication and ways in which information quality may enhance manufacturing and operation
On-site assembly process.
The accuracy test of the PWCMM sensor Proliner 8 PWCMM registers the position of the stylus probe as the coordinates in the Descartes coordinate system.
The stylus probe is connected to the machine with a wire extracted from the measuring arm that can be rotated horizontally and vertically (
Janssen 2004, Prodim 2014).
The machine has three sensors, one for the pull line, the second for measuring the horizontal position of the arm, and the third for the vertical position of the arm.
The range of the wire is up to 7 m and the measuring arm can rotate 402 [degrees]
Level 104 [degrees]vertically.
Coordinate registration is performed with a stylus probe positioned on the object, and the user operates the remote control to command the machine to register the location (Fig. 1).
The measurement results are presented to the user on the screen of PWCMM.
The output data measured by PWCMM is stored as DXF files that can be transferred to most computers-aided design (CAD)software.
To test the accuracy of PWCMM, the random error of measurement registration when using PWCMM was measured.
This allows PWCMM users to understand the possible accuracy that can be obtained from measurements without special equipment.
By using only the components provided, any user of PWCMM can perform this test.
The experimental device uses PWCMM and four moving reference targets.
By sticking these reference targets to the ground, a fixed measurement registration can be performed.
Thanks to their support for the stylus probe for PWCMM.
Each measurement position in the test uses these when recording observations to fix the measurement probe.
The performed sensor tests use a completely random design that is repeated 30 times. Wire-
Extract sensor test settings. --
Setting of PWCMM wire accuracy test-
The extraction sensor uses four reference targets fixed on the floor
M range of the cable (Fig. 2). The wire-
The recorded extraction locations are 100, 280, 470 and 650 cm from the machine origin, respectively.
The position of the horizontal sensor is fixed, and the position of the vertical sensor is 11 [degrees], 5[degrees], 3[degrees], and 2[degrees].
In Descartes space, Z-
The direction represents the wire-
Extraction location, T-
Position of horizontal sensor indicated in direction, Z-
The orientation indicates the position of the vertical sensor.
Test settings for horizontal and vertical sensors. --
For the horizontal sensor accuracy test, PWCMM is placed horizontally on the floor and four reference targets are positioned along 402 of the measuring arm [1]degrees]
Level range (Fig. 3A).
Reference target positioning and 90 [degrees]
The interval between the start and the end of the horizontal range is-180[degrees], -90[degrees], 0[degrees], 90[degrees], and 180[degrees]positions. The 180[degrees]
Location reuse the same reference target location-180[degrees]recording.
Each position extracts 700mm of the wires from the machine origin.
In Descartes space, Z-
The direction is consistent with the axis between 180 [degrees]and 0[degrees]
Location, T-
Direction along the axis-90[degrees]and 90[degrees]
Location and Z-
The direction from the image plane.
Vertical-
The sensor accuracy test, PWCMM is placed vertically on the floor.
The four reference targets are tightly positioned on the floor, allowing the measuring arm to move vertically in the following positions20[degrees], 10[degrees], 40[degrees], and 70[degrees]
Along its 104 [location]degrees]range (Fig. 3B).
Each position extracts 300mm of the wires from the machine origin.
The horizontal movement of the measuring arm is not fixed;
The horizontal position is 93 [degrees], 92[degrees], 93[degrees], and 130[degrees].
In Descartes space, Z-
Direction and axis-20[degrees]and 70[degrees]
Location, T-
Vertical to Z-
Direction and Z-in image plane-
The direction from the image plane. Responses. --
In the accuracy test of the three sensors of the machine, the variability of the measured values in four positions was used.
Since the position of PWCMM is recorded in the Descartes coordinate system, the data is stored as three values of X, Y and Z.
The coordinate values of the four positions need to be compared between the measured positions.
The design selected for this comparison is to calculate the size of the response vector from the center of gravity of each measurement position.
Find the center of gravity by using the mean values of each X, Y, and Z coordinate of 30 duplicates.
Equation 1 shows the calculation of the position of gravity ,[X. sub. PG], for the X-
Coordinate values of one of the four test locations. [[bar. X]. sub. PG]= [summation][X. sub. n]/30 (1)where [X. sub. n]is the nth X-
Coordinate values of one of the four test locations.
This again for Y. and Z-
Coordinate values for each test location.
In this way, the center of gravity is established at each test location.
The response vector is then calculated using equation 2 as the center of gravity distance for each measurement record. [XYZ. sub. RV]= [Square root ()[([X. sub. RP]-[[bar. X]. sub. PG]). sup. 2]+ [([Y. sub. RP]-[[bar. X]. sub. PG]). sup. 2]+ [([Z. sub. RP]-[[bar. X]. sub. PG]). sup. 2])](2)where [X. sub. RP], [Y. sub. RP], and [Z. sub. RP]
Is the coordinate value of each measurement record, compared to the center of gravity of each test location.
When the center of gravity is considered as a reference value, the response vector represents the absolute value of the error recorded for each measurement.
It is now possible to represent the variability of PWCMM random errors.
Response vector [XYZ. sub. RV]
Used for all performance evaluation of PWCMM sensor accuracy.
Analysis of variance (ANOVA)
Significance tests were performed on the measurement error contribution of the three sensors of the machine.
Paired comparison of Tukey (Tukey 1953)
Is there a significant difference in the measurement error used to control the position of the test factor.
It is assumed that when the random error is low, the relative accuracy is high, thus ignoring the system error.
The test of the PWCMM leap function PWCMM has a function called leap that extends the measurement range by repositioning the machine while maintaining measurements before and after repositioning in the same coordinate system.
Four reference targets are measured before and after the machine is relocated, and the location of these targets is used to calculate the new location of the machine after the relocation (Fig. 4).
PWCMM leap function by measuring 88-m-
Long corridor wall of 9 machines (leaps).
By measuring the position of two fixed reference targets on the front and back walls of each jump, the mismatch error of each jump is tested (Fig. 4).
Each jump along 88-has a set of two reference targetsm distance.
The upper wall reference for each set of two-wall references is aligned with the horizontal line laser projection of Leica Lino L2 (
Leica geography 2014).
In size and direction, the single jump mismatch error was measured as the difference in position of the two Wall references before and after the machine was relocated.
This is compared to the mismatch information shown by PWCMM.
After each jump, the absolute mismatch error is measured as the distance from the registered position of the upper wall reference target to the horizontal laser reference line.
The absolute error is two-
Sense of dimension, because there is no three-
Size reference.
The horizontal accuracy of Leica Lino L2 line laser is [+ or -]1. 5 mm/5 m.
Two tests were conducted.
Four cases conducted case studies of varying degrees of complexity.
Measurements have been made using the tested PWCMM (Proliner 8.
Leica Lino L2 line lasers are used to create horizontal or vertical reference lines for controlling the direction of the Descartes coordinate system when modeling measurement data.
The measurement data is exported from PWCMM to the Work CAD software, where they are refined to 3-D models.
Case 1 is a room
The profile of the section measured to provide prefabricated walls and glass partitions (including doors) to industrial sites where the office environment is being rebuilt.
The measurements are carried out as two profile measurements of walls and glass partitions.
The two profiles are measured separately and aligned manually in CAD software.
The leap feature is not used.
Case 2 involves measuring the room wall surface of the indoor wood panel system and measuring a series of office profiles that will receive prefabricated walls, doors, and window partitions, in relation to the office corridor.
By defining the surface plane with three coordinates and then measuring the profile of these wall segments, the surface of the wall of the conference room is measured.
For a range of offices, the leap function of PWCMM is used to extend the scope to measure a range of office stalls.
When measuring the rectangular profile of the office, two coordinate registrations were performed on each side of the profile.
The leap function registered four reference targets before repositioning the machine.
After the relocation, the same reference target position was registered.
This makes it possible for the leap function to calculate the new machine position in order to maintain the new measured value in the original coordinate system.
Case 3 involves measuring a complex
Large shape objects, 12-
Story stairs, where are the Carpenters-
The product supplier is the development, manufacture and assembly of solid wood stair railing systems.
The interior profiles of all stair sections are measured as profiles of many small and large surface planes.
Each of the 12 floating stairs was measured separately with a single positioning of PWCMM.
The leap feature is not used here.
For each floor, a horizontal reference laser line is projected on the side of the floor part of the stairs.
Before conducting a comprehensive measurement, the measurement method is tested by manufacturing and assembling the three prototype railing sections based on PWCMM measurements.
Refine the measurement data to 3-D model, floor-
The height measurement in the model is compared with the manual steel strip measurement and drawing.
Case 4 involves measuring buildings with complex external and internal shapes beyond 90 [s]degrees]wall--Wall alignment.
The materials supplied include the shelf system, clothing wardrobe, reception desk, guest seat, wall panel, \"hidden\" door consistent with the wall panel system, post box, etc. (Fig. 5).
Two PWCMM measurements were performed using two different methods.
The first is the plane projection method, which defines the floor plane with three coordinates, then measures the position of the wall near the floor, and projects it onto the floor plane, and then squeezes the wall vertically.
On the second surface.
On the measurement method, the machine stylus probe is swept over the surface of the wall to record many coordinates.
Then the wall plane is defined by averaging the measured coordinates of each wall surface.
In this way, information about vertical alignment of walls is captured.
Corner between wall and wallto-
The floor is defined as the intersection between surface planes.
In both measurements, the range of PWCMM is insufficient, and the jumping function of the machine is used when one machine is repositioned.
In CAD software, different color models of two different measurement methods are compared with each other.
To illustrate the differences between each other, these models are superimposed on each other.
In addition, external contractors use the Leica scanning station CIO for Laser Scanning Measurements and compare PWCMM measurements with them.
Results The accuracy of PWCMM sensor is 3-
D scatter plot for each of the three sensors
The horizontal and vertical position sensor extraction sensor shows how the measurement records are distributed around the measurement center of gravity (Fig. 6). For the wire-
The position of the sensor is extracted, and the error is expanded [+ or -]0. 8 mm in the X-direction, [+ or -]0. 5 mm in the Y-Direction and [+ or -]1. 9 mm in the Z-direction (Fig. 6A).
The error of the position of the horizontal sensor is extended [+ or -]0. 95 mm in the X-and Y-Direction and [+ or -]0. 5 mm in the Z-direction (Fig. 6B).
For vertical sensors, the error expansion in all three directions is equal ,[+ or -]0. 25 mm (Fig. 6C). Note that X-, Y-, and Z-
Due to the different Descartes directions in the settings, the direction cannot be compared between the tested sensors.
According to the sensor position, the measurement error size that can be expected at the time of each PWCMM measurement registration is shown by the confidence interval diagram (Fig. 7). The wire-
The extraction sensor gives an absolute error in the range of 0. 27 to 0.
35mm in 100-
Cm position, 0. 78 to 1.
13mm in 650-
Cm range, 95% personal confidence in both (Fig. 7A). The horizontal-and vertical-
The contribution of the position sensor to the measurement error within its working range is more constant (Figs. 7B and 7C).
Note that measurements of horizontal and vertical sensors use different quantities of wire extraction, which explains the difference in the size of the average error between them. A one-
Way ANOVA shows a significant difference in the error size between wires
Extract the position of the sensor.
For level-and vertical-
Sensor position, there is no significant difference in the error size between different sensor positions.
Paired comparison of Tukey between wire-
The location of the extraction sensor shows that the measurement error at 100 cm is significantly lower than that of other wires-
Extract location.
The measurement error at 280 cm was significantly lower than 650 cm, but not significantly lower than 470. cm position.
There was no significant difference in error between 470 and 650 cm. Tested leap-
88-measurement of functional performancem-
The long corridor with a series of nine PWCMM repositioning or jumping shows that the measurement individual mismatch error for each jump is greater than the user information provided by the machine (Figs. 8A and 8B).
The user information of the machine shows a mismatch error in the range 0. 5 to 2 mm (CMM-Info)
The mismatch range of the measurement is from 0. 25 to 6. 5 mm (Ref1 and Ref2).
The mismatch error has an irregular direction.
With the continuation of the leap series, the absolute mismatch error measured is significantly greater than the cumulative single mismatch error (Figs. 8C and 8D).
The absolute error here reaches a value of several hundred millimeters.
The absolute error can also change the direction (Fig. 8C). Case results--
Accuracy and availability in Table 1 provides an overview of the accuracy and usability experience in the four cases presented here.
Case 1: repair results from factory to office. --
The first case was considered successful by the Woodworking Company.
Product supplier using processed measurement data in product design modeling (Fig. 9).
The woodworking products are assembled in-
Site without measurement
Related issues
However, some issues of accuracy and availability have been noted (Table 1).
Case 2 result: new supplier office. --
In the second case, the processing measurement model of the conference room shows the uncertainty, which becomes apparent when studying the way the corners and the meters are exposed to each other.
Of the 6 corners measured, there are 0 that do not match. 43, 1. 46, 2. 36, 3. 44, 5. 54, and 8. 68 mm (Fig. 10).
The measurement of the glass partition series office brings trouble to PWCMM to extend the measurement range and leap function.
In the three trials with the leap function, PWCMM responded with mismatch Information 5. 2, 1. 99, and 42.
Three machines are 05mm after each move.
Finally, in all three trials, PWCMM was unable to calculate a new location after relocation.
Ultimately, PWCMM measurements cannot contribute to the supply of a range of walls and glass partitions.
The case experience summary shows some accuracy and usability issues, but also shows advantages over manual measurement techniques (Table 1).
Case 3: Result of stair railing. --In case 3, a 12-
Story stairs were measured with PWCMM.
The process is field measurement, data processing of the measurement is 3-
D model, align the product model with the measured model, manufacture the product, and finally assemble it on site (Fig. 11).
First test measurement-
Manufacturing and assembly based on the prototype railing part was successful.
However, in the next 12-
The story staircase, which reveals some uncertainty about the measurement while processing the measurement data.
For example, some measured surface planes are not parallel or vertical as expected.
These deviations are usually due to a wrong coordinate registration or misutilization when defining a plane in the PWCMM model.
In addition, it is found that the small angle deviation of the surface plane is easily considered reliable, because the object under test may contain small irregularities, which may result in accuracy due to leverage
These are all recurring problems that affect the measurement of floor height and stair profile dimensions as defined in the building drawings.
Due to these uncertainties and handling 3-
D model based on measurement data is time-
Consumption, the supplier chooses to handle the complete 12-story 3-
D model based on building drawings.
The floor height measured by the PWCMM model and steel tape measure is different from the height specified in the building drawings, and is also different from other heights (Fig. 12).
Sometimes the height of the measured floor is close to each other and sometimes not, which indicates the uncertainty of the measurement.
However, these differences are still within the requirements of the Swedish building code (Hus AMA 1998). The case-
A summary of experience shows many accuracy and usability issues (Table 1).
Case 4 results: Office reception room. --
In the fourth case, two different PWCMM measurement methods-the plan-
Projection measurement method and surface-
Measurement method--
Give some different measurement data while the final 3-
Similar Model D (Fig. 13).
The model superposition of the two measurement methods confirms the similarities (Fig. 14).
However, the difference between the two models can be distinguished by example measurements shown as A1 to A3 and M1 to M5, as well as wall Wl.
Wall W1 shows the most obvious difference, which is explained by the fact that the surveyor measured it as a curved wall without a surface measurement method, and it is a continuous wall, neither of them has a measurement method.
In two different PWCMM measurement methods, A1 to A3 and M1 to M5 are measured from surface
The measurement method is closer to laser measurement. scanning 3-D model (Fig. 15).
The maximum deviation between the PWCMM measurement methods is 0. 07[degrees]and 7.
Measures 9mm of A1 to A3 and M1 to m5.
These differences are visually difficult to detect and can have a big impact on the carpentry
Product supplier.
In addition, the advantages of the surface-
The measurement method is that it captures the presence of a non-vertical wall at the construction site.
You can see how the surfaces of the overlay model intersect each other, because there are non-vertical walls in the surface model
Measurement Method (Fig. 16).
This is not mentioned in the plan
Projection measurement method.
Again, this case
A summary of experience shows many accuracy and usability issues (Table 1).
Analyze and discuss the expected result of PWCMM test is-
The building geometry of the construction site can be reduced to a level that allows the carpentry
Product accessories are transferred to the digital environment early in the supply process, rather than manually executed at the end of the supply process.
This has the potential to significantly increase the efficiency of supplying woodworking products to the construction industry.
In order to achieve this goal, the measurement error of the woodworking product is equivalent to the tolerance ([+ or -]1 mm)
The best golden rule of measurement must be achieved.
The test of sensor accuracy on three PWCMM sensors shows that the number of wires extracted is the most significant factor affecting the random error size of PWCMM.
For the error extracted from the wire, the maximum error contribution is in Z-direction (Fig. 6). The error in X-and T-
The direction is less than the total error of the horizontal sensor.
The test device used to test lead leads also includes measuring the vertical movement of the arm, which means Z-direction.
Therefore, although the effect of the vertical sensor on the measurement error is small, the error contribution is related to the number of wires extracted from PWCMM.
This shows that with the increase in the number of wires extracted from PWCMM, the error increases proportionally.
The error of the horizontal sensor may also be affected in a similar way.
Within a close range of 100 cm, the accuracy level is close to the requirements of the golden rule of measurement, for verification-
The geometry of the construction site when supplying woodworking products.
However, as experienced in the case studied, the normal situation is that measurements need to be made within the external scope of the PWCMM being tested.
Then the random error can be expected to be 0. 78 to 1.
13mm registered in a single coordinate.
This will be comparable to the tolerance requirements of the woodworking products --
Product supplier.
The analysis of the leap function shows that the use of the PWCMM leap function increases the uncertainty of the measurement.
The mismatch error shown on the machine may seem trivial, but the mismatch actually measured is up to three times larger (Figs. 8A and 8B).
In addition, the test shows that the absolute mismatch error will increase to a large proportion in each jump, significantly greater than the cumulative single mismatch error.
In addition, the mismatching direction is irregular.
Due to these cases, PWCMM users cannot predict the impact of mismatch errors when using the leap function for a series of jumps.
Case experience shows that 7-
The M range is usually a limitation of PWCMM availability.
The size of the object and the presence of obstacles make it necessary to use the jump function.
Therefore, the leap function is desirable, but the absolute mismatch error is greatly increased after several jumps at present, reducing the availability of the function.
It should be possible to further develop the leap function by using a method that averages duplicate measurement registration for reference locations.
If the absolute mismatch can be reduced to a few millimeters after a few jumps, the availability of the machine will be significantly improved in terms of the demand for carpentry
Product supplier.
Case analysis for the purpose of creating a representativebuilt 3-
D model available for digital products-to-
Room decoration of woodworking products, case reveals 3-D sensing of-
Build construction site dimensions using PWCMM.
Precision analysis. --
Case experience shows that the magnitude of the measurement error is significantly higher than the test reported by the sensor accuracy.
In the case studied, four factors affecting the measurement accuracy are mainly determined: * the accuracy of the measurement coordinates * the representative of the selected coordinates * the error first uses the * mapping function, the distance from PWCMM to the object being measured affects the accuracy of the measured coordinates.
This distance is always different when measured, so the accuracy is also different.
The results show that the error in the external range is three times larger than the error in the short range.
In cases where PWCMM often operates within its external scope, the accuracy is low, but still comparable to that of the joinery --
Product tolerance.
Nevertheless, in the case studied, the uncertainty of the measurement accuracy has emerged when processing the measurement data.
Second, the accuracy of measurement coordinates is about [+ or -]
1mm the following requirements are made for the representation of the measured coordinate position.
On the construction site, the contour line or surface usually has irregularities larger [+ or -]
Affects 1mm of the PWCMM measurement accuracy.
This has been experienced in all cases, but can be illustrated in case 2 where angular mismatch is due to low representation and error levers when defining surface planes (Fig. 10).
Similarly, in case 3, when the floor plane is defined with three coordinates, the uncertainty of the floor height is affected by this coordinate representation problem.
Therefore, the surface plane of the measurement becomes tilted.
Third, error utilization is an error contributor in most PWCMM measurements.
In small scale, error utilization is generated when measuring profiles with two coordinate positions on both sides, and then connecting the profiles they intersect.
This happens in almost every PWCMM measurement.
Another common mistake
When defining a surface plane with three coordinates, the leverage situation occurs, and these coordinates do not extend well in all Descartes directions, and then take measurements far away from the surface area --
Define coordinates.
These surfaces are slightly tilted due to normal measurement errors.
When the outer profile of that larger surface is subsequently measured and these measurements are projected onto the defined surface plane, errors from the first defined coordinate will be utilized.
This is the source of the significant error lever observed in case 2 and corner mismatch of floor height and profile
Size uncertainty in case 3.
The uncertainty in case 3 is less than the tolerance requirement for the floor in the Swedish building code
The plane height is very important for the stair railing system provided for installation in this case.
The case of the study shows that the errors generated by the representative lower coordinate measurement and error utilization affect each other, thus increasing the original error of PWCMM.
Reducing sensitivity to error utilization is beneficial to the availability of PWCMM.
To achieve this, the measurement should plan to register as many coordinates as possible and average these values when getting the following values
Construction information from the construction site. The surface-
Measurement method in case 4 (Fig. 14)
Is an example of how to use the tested PWCMM for average capabilities.
Found this method more reliable than planned
Measurement method, so can improve the accuracy of measuring-
Construction site with PWCMM.
However, since PWCMM data needs to be processed in CAD software, the average measurement strategy is applied to measure the rectangle-
The shape of the object is limited by the lack of lines
Accessory operation in CAD software.
At present, most CAD software is not suitable for importing measurement data that requires fitting operation, which affects the usability of testing PWCMM.
Fourth, extending the scope of operations using the PWCMM leap function is a very attractive feature, which unfortunately introduces major errors.
The use of the leap function was introduced in case 2 and case 3, but the general structure-
Field conditions prevent successful measurements using a series of leaps.
In case 4, the leap feature was successfully used.
Here, a jump was used and uncertainty was found in the range of 3mm.
In addition to the error size introduced by the jump, the problem is that the user cannot predict the direction of the introduced error.
By measuring the part of the object before and after the jump, the direction of the error can be evaluated when modeling the measurement information and may be compensated.
A more precise leap function requires higher PWCMM sensor accuracy, or, by means of repeated measurements of the reference target and the averaging of the measured position, a method for further developing the EDM function, used for calculation of PWCMM position after WMS.
In addition, another factor affecting PWCMM measurement is
The construction environment of the construction site usually has undesirable horizontal and vertical surface curvature.
They often affect the assembly of woodworking products.
Even if the accuracy of the PWCMM tested is sufficient to measure these bad surface curvature with an average method, repeated grid measurement strategies are required.
Here, Skalski et al. Propose a repeated measurement of a full geometric recognition method. (1998)
Need. For large-
Construction sites and other large-scale objects, such as high
Density Grid measurement will be time-
The consumption of data collection needs to display supplementary equipment in grid mode.
The modeling of these data requires considerable processing time and improved software support.
In terms of availability, the tested PWCMM with manual probe positioning is not suitable for this high
Density measurement.
Due to these factors, PWCMM 3-
D. In the case of research, the measurement error is significantly greater than the tolerance of the woodworking product.
Because it is difficult to estimate the size and direction of the error, the reliability of the model based on PWCMM measurement is inconsistent with the carpentry
Product tolerance.
Therefore, the assumption of eliminating the size uncertainty of-
Construction sites reach the same level as Carpentry
Product tolerances are rejected.
Usability Analysis. --
In the case of the study, three major usability issues were experienced: * scope and scope * \"imaginary\" limits of construction site and its details * professional level required for accurate measurement * Data processing of measurements as measurable 3-
First of all, the scope and scope of the tested PWCMM are unique compared to other products on the market.
However, in the case studied, PWCMM often needs to work at the high end of the scope, otherwise the scope is insufficient.
In addition, the measurement of the wire of the machine needs to be within the line of sight.
Even a small ledge on the surface may be an obstacle to locating the measuring probe.
Therefore, the limitations of these ranges and ranges often limit the possibility of perceiving many locations that can improve the level of the building
Details of the website that can be described.
When trying to overcome these limitations with the leap function, there is a serious accuracy problem.
Therefore, the limitation of scope and coverage has always been a major usability issue.
Second, the limitation of \"depicting\" the construction site and its details means that PWCMM measurement and modeling can only provide simplified reconstruction of the construction site.
This simplification means space information that is important to the carpentry
Product suppliers may still be missing.
Experience in case and carpentry --
There are 3 suppliers of products involved-
There are very few models of construction sites.
So-
Validation of the build cannot be performed using only a few control coordinate locations;
The site needs to be described and reconstructed into an understandable model.
In fact, there is a limit on the level of detail that can be achieved.
In case 3, it is actually impossible to capture the understandable 3-D model.
Here, additional information from drawings is added in model reconstruction.
Because of this, the model is assembling the product to-
Building environment.
So the ability to describe the details of the site is limited, which is a serious usability issue.
Third, measuring construction sites with PWCMM requires a high level of expertise.
Case experience shows many potential handling errors.
There are many details to capture when measuring the construction site.
In addition, for accurate measurement, the measuring probe has an offset that the user needs to consider.
Displaying large objects on small screens makes it difficult for users to track measurement progress on PWCMM screens.
Therefore, before processing the measurement data in CAD software after measurement, it is difficult to judge whether sufficient information is captured.
These cases show that in rebuilding a 3-
D model of measurement data.
Experience at the construction site and understanding of the measurement process are essential.
Therefore, the reconstruction of the measurement data also requires a high level of expertise, and it is difficult for anyone to complete except for the person who did the original measurement.
So many mistakes can be made without high
PWCMM measurement and measurement data re-build 3-level expertiseD model.
Therefore, this is a critical usability issue, and therefore the assumption that the actual availability requirements are met is rejected.
Conclusion portable wire
Based on coordinates-
Testing Machine (PWCMM)
Already executing-
Provide on-site verification of building dimensions for the construction industry. Reliable as-
Construction site size 3-
D is a necessary condition for the transfer of the accessories for the woodworking products to the digital field, which means increasing the efficiency of the supply process.
To eliminate the size uncertainty of-
The level at which the construction site meets the equivalent of the product tolerance is considered a minimum requirement.
The random error of PWCMM is close to meeting the tolerance requirements of woodworking products. When the measured objects are small, the requirements of the golden rule of measurement are also close to meeting.
The cases studied show that the accuracy is more uncertain than the investigation of random errors.
The case analysis shows that due to the roughness and/or non-uniformity of the surface of the construction site and the role of error levers, the limitations of coordinate representation affect the actual accuracy.
With the increase in the density of coordinates and the increase in the possibility of the average application of the measured coordinates, these precision problems may be reduced.
The PWCMM measurement at the construction site often requires the equipment to work at or above the upper end of the work range, so the leap function is required, which further increases the inaccuracy.
This is the area where PWCMM produces the highest level of random error.
In addition, the structure can only be described with low resolution.
Because this description is performed manually, the skills of the surveyor are critical to their quality.
Automatic processing of PWCMM data is 3-
Because of the need to understand the measurement data, the D model is almost impossible.
In addition, many uncertainties in the resulting models hinder the availability of the supply process of the woodworking products and the automation of improvements.
Therefore, the assumption of eliminating the size uncertainty of-
Construction sites built with PWCMM are at par with carpentry
Product tolerances are rejected.
The author is a doctoral student in the engineering science department.
And math, Div. of Wood Sci.
College of Engineering, lulai Universityof Technol.
Sweden (Sweden)samuel. forsman@ltu. se [
Author of communications).
The document was published in September 2014. Article no. 14-00086. doi: 10. 13073/FPJ-D-14-
Akinci, B. 00086 references, F. Boukamp, C. Gordon, D. Huber, C. Lyons, and K. Park. 2006.
Formalism of active construction quality control using sensor system and integrated project model.
Automation construction. 15(2): 124-138. D01:10. 1016/j. autcon. 2005. 01.
American Society of Mechanical Engineers (ASME). 2004. B89. 4. 22.
Performance Evaluation method of joint arm coordinate measuring instrument (CMM). ASME, New York. Anil, E. B. , P. Tang, B. Akinci, and D. Huber. 2013.
Deviation analysis method for As-quality evaluation
Is to build an information model generated from point cloud data.
Automation construction. 35(0):507-516. DOI: 2013. 06. Allegri 003-Fraga, R. , C. Ordonez, S. Garcia-Cortes, and J. Roca-Pardinas. 2013.
Measurement planning of circular cross
Tunnels scanned using ground laser.
Automation construction. 31(0):1-9. DOI:10. 1016/j. autcon. 2012. 11. Barini, E. M. , G. Tosello, and L. De Chiffre. 2010.
Uncertainty Analysis of pointsby-
Point sampling of complex surfaces using touch probe CMMs: DOE verification of complex surfaces using CMM. Precision Eng. 34(1):16-21. DOI:10. 1016/j. precisioneng. 2009. 06.
009 Beckett, S. F. , E. P.
Paradini and L. V. O.
Valentine. 2010.
Acceptable deviation analysis of the measurement process established by MSA and VDA manuals.
Minutes of the world Engineering Conference, WCE 004 6 months 30 days-
London, July 2, 2010. Vol. III:2207-2212. Bosche, F. , M. Ahmed, Y. Turkan, C. T. Haas, and R. Haas. 2014.
Value of integrated scanning-to-BIM and scan-vs-
BIM technology for construction monitoring using laser scanning and BIM: Case for cylindrical MEP components.
Automation construction. DOI: doi. org/10. 1016/j. autcon. 2014. 05. 014 Boche, F. and E. Guenet. 2014.
Automated Surface flatness control using ground laser scanning and building information models.
Automation construction. 44(0):212-226. DOI: j. autcon. 2014. 03. 028 Cauchick-Miguel, P. , T. Kinga, and J. Davis. 1996.
CMM verification: a survey. Measurement 17(1):1-16. DOI:10. 1016/0263-2241(96)00001-
2 Creaform measurement solution. 2014.
Portable coordinate measuring instrument (CMM). www. creaform3d. com/en.
Visit January 14, 2015. Cuypers, W. , N. Van Gestel, A. Voet, J. Kruth, J.
Mino and P. Bleys. 2009.
Mobile and large optical measurement technology
Measurement of dimensions.
Optical laser47(3-4):292-300. DOI: 10. 1016/j. optlaseng. 2008. 03.
C. EastmanM. 2008.
BIM Manual: Building information modeling guide for owners, managers, designers, engineers and contractors.
Willie of Hoboken, New Jersey. Forsman, S. , N. Bjorngrim, A. Bystedt, L. Laitila, P. Bomark, and M. Ohman. 2012.
Innovative demand for supply engineersto-
Order woodworking products to construction: case study in Sweden.
Building Innov. Inf. Process Manag. 12(4):464-491.
Hexagon metering. 2014.
Absolute arm. www.
Hexagonmetroology. se/ROMER-Absolute-Arm_862. htm.
Visit December 19, 2014. Huber, D. , B. Akinci, A. Oliver, E. Anil, B. Okorn, and X. Xiong. 2011.
Automatic modeling and representation of-
Building Information Model is established.
Summary of the National Science Foundation\'s cmmi r & D innovation conference, January 4-7, Atlanta. Hus AMA. 1998. 27. F 11.
Material--
Och arbeteskrivning husbyggnadsarbeten.
RA 98 lysate and urine poisoning syndrome: Radian anvisningar does not go to husama 98 (1998).
Swensk byggtjanst, Stockholm. Janssen, A. J. 2004.
Includes measuring devices for movable measuring probes.
Patent 6,785,973 USAJinwen, W. and C. Yanling. 2011.
Geometric dynamic error of CMMs in rapid scanningprobing. Measurement 44(3):511-517. DOI: 10. 1016/j. measurement. 2010. 11. 003 Kim, C. , H. Son, and C. Kim. 2013.
Fully and automatically register 3D data into a 3D CAD model for project progress monitoring.
Automation construction. 35:75-82. DOI: 10. 1016/j. autcon. 2013. 01.
005 Kunzmann, H. , T. Pfeifer, R. Schmitt, H.
Schvenk and A. Weckenmann. 2005.
Production Measurement--
Add value to manufacturing. CIRP Ann. Manuf. Technol. 54(2): 155-168. DOI: 10. 1016/ S0007-8506(07)60024-
Leica geographic system. 2014. Leica lino L2. com/en/Leica-Lino-L2_80239. htm.
Visit December 19, 2014.
Leitz metering2014.
Visit December 19, 2014. Nafi, A. , J. R. R. Mayer, and A. Wozniak. 2011. Novel CMM-
Multi-Implementation based
A step-by-step method for the separation of machine and probe errors. Precision Eng. 35(2):318-328. DOI:10. 1016/j. precisioneng. 2010. 11.
007 Nikon metering. 2014.
K-coordinate meter
Optical CMM series.
Visit December 19, 2014. Pereira, P. H. and R. J. Hocken. 2007.
Representation and compensation of dynamic error of scanning coordinate measuring instrument. Precision Eng. 31(1):22-32. DOI:10. 1016/j. precisioneng. 2006. 01. 006 Prodim. 2014. Proliner 8CS.
Visit December 19, 2014. Sacks, R. , C. M. Eastman, and G. Lee. 2004.
Parametric 3D modeling in construction with examples of prefabricated concrete.
Automation construction. 13(3):291-312. DOI: S0926-5805(03)00043-
8 Santa Maria, J. , J. Aguilar, J. Yague, and J. Pastor. 2008.
Calibration and repeatability of joint arm coordinate measuring instrument improved motion parameter estimation technique. Precision Eng. 32(4):251-268. D01:10. 1016/j. precisioneng. 2007. 09.
J. Santoria 002, J. Yague, R. Jimenez, and J. Aguilar. 2009. Calibration-
Thermal error model based on joint arm coordinate measuring instrument. Precision Eng. 33(4):476-485. DOI:10. 1016/j. precisioneng. 2009. 01.
002 Shi Liberal Arts, H. , M. Franke, J.
Hanford and H. Kunzmann. 2005.
Error mapping of CMMs and machine tools is performed by a single tracking interferometer. CIRP Ann. Manuf. Technol. 54(1):475-478. DOI:10. 1016/S0007-8506(07)60148-6 schvenk, H. , W. Knapp, H. Haitjema, A. Weckenmann, R. Schmitt, and F. Delbressine. 2008.
Measurement and compensation of geometric error of machine-An update. CIRP Ann. Manuf. Technol. 57(2):660-675. DOI: 10. 1016/j. cirp. 2008. 09. 008 Shen, Z. , P. Tang, O. Kanaan, and Y. Cho. 2013. As-
Error modeling is established for effective 3D laser scanning at the construction site. Comput. Civil Eng. 2013:533-540. DOI:10.
1061/9780784413029.
067 Shimojima, K. , R. Furutani, K.
Takamasu and K. Araki. 2002.
Estimation method of uncertainty of joint coordinate measuring instrument. IEEE Int. Conf. Ind. Technol. 1:411-415. DOI:10. 1109/ICIT. 2002.
1189931 Skalski, K. , R. Filipowski, W.
Swieszkowski, K. Kedzior, A.
Dabrowski, J. Zawora. 1998.
Recognition and geometric modeling of complex shape surfaces using coordinate measuring instruments and CAD/CAM systems. J. Mater. Process. Technol. 76(1-3):49-55. Sladek, J. , K.
Ostrowska, and. Gaska. 2013.
Model and identify the error of the coordinate measuring arm using the metering model. Measurement 46(1):667-679. DOI:10. 1016/j. measurement. 2012. 08.
The forest industry in Sweden. 2013.
Facts and figures
The forest industry in Sweden. Tang, P. , B. Akinci, and D. Huber. 2009.
Quantification of edge loss of spatial discontinuous laser scanning data.
Automation construction. 18(8): 1070-1083. DOI: 10. 1016/j. autcon. 2009. 07. 001 Tukey, J. W. 1953.
Mathematics and Engineering: A quick and easy way to select statistical analysis. Trans. N. Y. Acad. Sci. 16(2 Series II):88-97. DOI:10. 1111/j. 2164-0947. 1953. tb01326. x Turkan, Y. , F. Bosche, C. T. Haas, and R. Haas. 2012.
Automatically track progress using 4D timeline and 3D sensing technology.
Automation construction. 22(0):414-421. DOI: 10. 1016/j. autcon. 2011. 10.
003 general affairs administration of the United States (US GSA). 2009.
Gsa bim Guide for 3D imaging.
GSA, Washington, D. C. C. Wilhelm, R. G. , R. Hocken, and H. Schwenke. 2001.
Task-specific uncertainty in coordinate measurement. CIRP Ann. Manuf. Technol. 50(2):553-563. DOI:10. 1016/S0007-8506(07)62995-3 Xiong, X. , A. Adan, B. Akinci, and D. Huber. 2013.
Automatically create semantic rich 3D building models from laser scanner data.
Automation construction. 31(0)325-337. DOI: j. autcon. 2012. 10. 006 Zheng, D. , C. Du, and Y. Hu. 2012.
Research on the best measuring area of flexible coordinate measuring instrument. Measurement 45(3):250-254. DOI:10. 1016/j. measurement. 2011. 12.