"A SYSTEM FOR CONTROLLING THE SHAPE AND ALIGNMENT OF WHEELS AND A METHOD FOR CONTROLLING THE SHAPE AND ALIGNMENT OF WHEELS"
The present invention relates to a system used in controlling the shape and alignment of vehicle wheels in general, and to a method of controlling shape and alignment, applied to vehicle wheels in general.
Description of the Prior Art
The steel wheel of an automotive vehicle is usually composed of a rim and a disc. These components are produced by a lamination process and a stamping process, after which the wheel is assembled and welded. For the most part, the shaping processes used in manufacturing the wheel cause an imprecision in the wheel alignment and/or shape.
Wheels are one of the most important components in an automotive vehicle, and so they should have safety margins even when under severe driving conditions. Their dimensions should be precisely controlled, and so narrow ranges of alignment-deviation tolerance are permitted. For example, for passenger vehicle the ranges of axial and radial tolerance for alignment deviation are approximately between 0.5 and 0.8 mm. On the other hand, the ranges of axial and radial tolerance for alignment deviation for light-load automotive vehicles are of about 1.0 mm.
The system employed at present for measuring and/or controlling these tolerances in alignment deviation consists in using calibrators, for example, a comparator clock. For this purpose, the wheel is secured to a measuring support, and the calibrators are adjusted close to the measurement points, these points being defined by the contact between the 16 mm diameter sphere positioned on the calibrator and the wheel surface. Then, the wheel is manually turned about its axle, and at this moment the alignment deviation is measured. This process, however, requires a manual initial adjustment, and the quality achieved in the measurements depends upon the operator's skill to handle and read the comparator clock. Another alternative system used in this measurement is based on the contact between the cylindrical rollers and the wheel. However, this system presents imprecision in the measured data due to the impregnation of dust on the roller surface, which causes a bad seating of the roller when it rests on the wheel surface.
Objectives of the Invention
An objective of the present invention is to provide a system capable of controlling the shape and alignment of the wheels of a vehicle by using computational processing means. Another objective of this invention is to provide a method for controlling the shape and alignment of wheels, in order to optimize the quality control and the safety limits imposed on the wheels of automotive vehicles in general.
Brief Description of the Invention
The invention has the objective of providing a system for controlling the shape and alignment of wheels, which comprises:
(i) a wheel-image capturing device;
(ii) a system for processing image and generating contours; and
(iii) a control-data generating system, the system for processing image and generating contours comprising a processing element associated to the image-capturing device and to the data generating system.
One also describes a method of controlling the shape and alignment of wheels, which comprises the following steps: a) calibrating the system by means of a calibrator; b) capturing an image of the wheel by means of an image-capturing device; c) processing the wheel image by means of an image-processing element, generating and viewing wheel contours by means of a contour-generating element; d) generating control data.
Brief Description of the Drawings
The present invention will now be described in greater detail with reference to an embodiment represented in the drawings. The figures show:
- Figure 1 is a schematic view of the system for measuring the axial deviation alignment of the prior art;
- Figure 2 is a schematic view of the system for measuring the radial
deviation alignment of the prior art;
- Figure 3 is a block diagram of the system for controlling the alignment of wheels of the present invention;
- Figure 4 is a block diagram representing a first preferred embodiment of the association of the image-capturing device with the image- processing and contour-generating system of the present invention;
- Figure 5 is a block diagram representing a second preferred embodiment of the association of the image-capturing device with the image- processing and contour-generating system of the present invention; - Figure 6 is a block diagram representing the association of the image- processing and contour-generating system with the control-data generating system of the present invention;
- Figure 7 is a schematic view of the quotas to be controlled by the system for controlling shape and alignment of wheels of the present invention; - Figure 8 is a schematic view of the calibrator of the system of the present invention;
- Figure 9 is a schematic view of the means for fixing the calibrator and the wheel;
- Figure 10 is a schematic view of the fixing of the calibrator to the means for fixing the calibrator and the wheel;
- Figure 11 is a view of the image of the orifice of the calibrator, captured by the image-capturing system;
- Figure 12 is a view of the image of the contour of the orifice perimeter of the calibrator, obtained by means of the Sobel algorithm; - Figure 13 is a view of the image of the orifice of the calibrator and of the reference straight lines, obtained by means of the Hough algorithm;
- Figure 14 is a view of the image of the reference straight lines adjusted according to the mathematical method of minimum squares;
- Figure 15 is a schematic view of the fixing of the wheel to the means of fixing the calibrator and the wheel;
- Figure 16 is a view of the image of the wheel, captured by the image- capturing system;
- Figure 17 is a view of the image of the contour of the wheel, generated with an error by means of the Sobel algorithm;
- Figure 18 is a view of the image of the contour of the wheel, generated without error by means of the Canny algorithm; - Figure 19 is a view of the overlapping of the image of the wheel with the contour generated by the contour-generating system;
- Figure 20 is a view of the overlapping of the image of the wheel with the measurement straight lines; and
- Figure 21 is a schematic view of the points of control of shape and alignment of the wheel.
Detailed Description of the Figures
Figures 1 and 2 illustrate the system for controlling shape and alignment known from the prior art. This system consists in fixing the wheel 1 to a seating base 7 capable of turning the wheel 1 , while a thickness-gauge-clock-type calibrator 5, adjusted in contact with the wheel 1 , records the variations in shape and alignment of this wheel 1.
Figure 1 illustrates the calibrator 5 positioned so as to record the variations in shape and alignment in the axial direction of the wheel 1 , while figure 2 illustrates the calibrator 5 positioned so as to record the variations in shape and alignment in the radial direction of the wheel 1.
According to a preferred embodiment and as can be seen in figure 3, the system for controlling the shape and alignment of wheels 1 of the present invention comprises:
(i) an image-capturing device 10 of the wheel 1; (ii) an image-processing and contour-generating system20; and
(iii) a control-data generating system 30.
According to figure 4, the image-capturing device 10 has the function of capturing the image of the wheel 1 and sending it to the image-processing and contour-generating system 20. In this case, the image-capturing system 10 may be an analog-type or digital-type photographic camera or else an analog or digital camera.
In a preferred embodiment, the image-capturing device 10 is
positioned on a fixed base, while the wheel 1 is positioned on a rotating base. The wheel 1 is then turned at predetermined and adjustable angles, and its images are successively captured by the device 10. It is also foreseen to place the wheel 1 on a fixed base and the image-capture device 10 on a rotating base. In this case, the device 10 is moved around the wheel 1 at predetermined and adjustable angles, while it captures images of the wheel.
The image-processing and contour-generating system 20 has the function of receiving the images captured by the image-capturing device 10, changing them into digital signals, if necessary, and then processing these images by means of mathematical algorithms, to transform them into measurable contours of the surface of the wheel 1. This obtained contour is then viewed by the user of the system.
Thus, as shown in figure 4, the image-processing and contour- generating system 20 is formed by an image-processing element 22, associated to the image-capturing device 10 and to a contour-generating element 23, which in turn is associable to a viewer 24. Additionally, the image-processing and contour- generating system 20 may also comprise a conversion circuit 21 associated to the image-processing element 22, as illustrated in figure 5.
This conversion circuit 21 corresponds to a plate for converting analog signals into digital signals, and its function is to convert the analogically captured images into digital signals.
In this way, the image-capturing device 10 will be associated to the conversion circuit 21, in case this device 10 captures analog images (figure 5). On the other hand, the image-capturing device 10 will be directly associated to the image-processing element 22 when it generates images in digital signals (figure 4).
According to figure 6, the control-data generating system 30 is formed by a data-processing element 32, associated to the contour-generating element 23, to a keyboard 31 and to a data-generating element 33, which in turn is associated to the view 24. The control-data generating system has the function of receiving the contour of the wheel 1 , generated by means of the contour-generating element 23 and, upon a command by the user, transmitted by means of a command input means 31 , generating the requested control data, so that the user can view them through the viewer 24. The control data generated by the control-data generating
system may be qualitative and/or quantitative data. The command input means 31 may be, for example, a keyboard or a mouse.
As schematized in figure 7, the control of the shape and alignment of the wheels 1 is initiated with the control of the diameter D of the wheel 1 and of the width L of the wheel 1 in function of its geometric center. From the intersection point between the straight lines that define the diameter D and the width L of the wheel 1 , one traces the tire seat A, the α-angle of which is obtained with respect to the axial straight line of the diameter D of the wheel 1. Then, one controls the radius of concordance R between the width L of the wheel 1 and the straight line of the tire seat A.
In this way, starting from the straight lines that define the tire seat A and the width L of the wheel 1 , it is possible to control the oscillation of the wheel 1.
Thus, with the system and method for controlling shape and alignment of wheels 1 described, it is possible to achieve straight lines that define the tire seat A and the width L of the wheel 1 , which enables one to control the quotas illustrated in figure 7. This control method comprises the following steps: a) calibrating the system by means of a calibrator 3; b) capturing an image of the wheel 1 by means of an image-capturing device 10; c) processing the image of the wheel 1 by means of an image- processing element 22, generating and viewing contours of the wheel 1 by means of a contour-generating element 23; d) generating control data.
Before using the system for controlling shape and alignment of the wheels 1 , this system has to be calibrated.
This calibration consists in positioning and fixing a calibrator 3 to the fixing device 2, as shown in figure 10. According to figure 9, the fixing means 2 has a seating surface 6, where a conical pin 8 is provided and, additionally, an expansive bushing 9. In turn, according to figure 8, the calibrator 3 comprises a first end 15, where there is a fixing bore 17 and a second free end 16 having a quadratic calibration bore 4, positioned in accordance with the coordinates Li, H1 F L2, H2, which are pre-established in function of the symmetry axis 10 of the fixing means 2.
As can be seen in figure 10, the fixing of this calibrator 3 is effected by fitting or associating the fixing bore 17 with the conical pin 8 and the expansive bushing 9 of the fixing means 2.
Once the calibrator 3 has been fixed to the means 2, one initiates the step of calibrating the system by capturing the image of the calibrator 3 by means of the image-capturing device 10. In figure 11, one can observe that the captured image is focussed essentially in the calibration bore 4 of the calibrator 3. In this step, this captured image is not perfectly aligned.
After capturing the image of the calibrator 3, one initiates the step of determining the perimeter of the bore 4 of this calibrator 3 (figure 12) by means of mathematical algorithms, preferably the Sobel algorithm. In this way, one detects the points that form the border or perimeter of the bore 4 of the calibrator 3, which means that still there are no straight lines or curves of this image, as shown in figure 12. Follow, according to figure 13, one initiates the step of obtaining the reference straight lines r1 τ r2, r3 and r4, which may be obtained by employing the Hough algorithm. However, the conventional application of this algorithm foresees the analysis of all the points of the perimeters by means of the Equation 1 , θ ranging from -90° to 90°, and obtaining the corresponding value of p.
x cos θ + y sen θ - p Equaςao 1
However, making use of the information from the direction of the perimeter obtained during its detection and by the fact that the orientation of the straight lines are a known datum, one can reduce the effort in this processing by imposing a restriction. In this way, the processing will be restricted to separating the points that present the direction of the perimeter at θ = 90°, -90°, 0° and 180°, to obtain the straight lines r1 f r2l r3 and r4, respectively, and further to obtain the values of p for the selected points and to look for points that represent the greatest occurrence.
Thus, by making the straight lines as pairs (θ, p ), we will have: η (90°, 395); r2 (-90°, 265); r3 (0°, 379) and r4 (180°, 515).
However, this resolution presents perfectly aligned straight lines, as shown in figure 3, which do not correspond to reality.
Thus, in order to obtain a better approximation of the straight lines, that
is to say, a resolution closer to reality, one takes the selected points and adjusts the straight lines by the mathematical method of the squared minimums. In this way, besides obtaining the better straight lines for the points, one will now work with a resolution smaller than a pixel, as illustrated below.
Starting from the adjustment of the reference straight lines .■[, r2, r3 and r4, we will have the straight lines η (91 ,093°, 403,75); r2 (-89,831°, 266,39); r3 (0,7996°, 374,80) and r4 (180,39°, 512,66).
These straight lines, illustrated in figure 14, allow the system to be more precisely calibrated.
After adjusting the reference straight lines, one will start the obtention of points pt^ pt2, pt3 and pt» from the intersection of the reference straight lines and r3; ^ and r4; r2 and r3; r2 and r4, respectively. Considering these already adjusted straight lines, the coordinates of these points will be: ptt (380,37; 396,56); pt2 (515,37; 393,98); pt3 (378,54; 265,27) and ptj (514,49; 264,87).
As soon as the points pt^ pt2, pt3 and ptt have been obtained, one may determine the point ptm by means of the average of the position of the points pt,, pt2, pt3 and ptj; therefore, the coordinates of ptm will be (447,19; 330,17).
Once the reference straight lines and the image points have been obtained and having the values of reference parameters of the bore 4 of the calibrator 3 (L^ Hi, L2, H2), one determines the correction factors, * real real
Ax. and an inclination angle θim for the images captured later. Both the correction values and the inclination angle mentioned above are determined by mathematical processing.
Considering an example of calibration, we have:
Table 1 Parameters of the bore of the calibrator
Table 2 Parameters of the reference straight lines of the image of the bore of the calibrator
Table 3 Parameters of the reference points of the image of the caliber
The inclination of the image will given by the inclination of the reference straight lines. In this way, the average of the inclination of the reference straight lines is calculated in according to the Equation 2:
_ (ΘΛ - 90°)+ (flr2 + 90ή+(θri - 0°)+ (θr4 -180°) θ; = Equation 2
4
Then , we will have, for this example, inclination angle of:
θ. = 0.61' Equation 3
The reference x of the image will be obtained by means of the distance of intersection points of a horizontal straight line that passes through ptm with the reference straight lines r3 e r4. In this way, we will have:
Equation 4
Δx
;„ = -330,17 -tg(l80,39)+
77)
+ 359'
74 |
δ ;
'
. cos(o
577)J
Equation 6
Axim = 510,42 -370,23 = 140,20 [ ixe/] Equation 7
With regard to the calibrator, the distance between the reference straight lines r3 and r4 is given by:
Δ* calibre ~ " 2 " l Equation 8
^cα re = 39,992 - 29,941 = 10,051 [mm] Equation 9
However, the image is rotated from θm. Therefor, the correction factor at x of the image will given by:
calibre/
Δ real cos fo. ) Equation 10
Δx,„ Δx,.
Equation 11
Analogously, the reference y of the image will be obtained by means of the distance of the intersection points of a vertical straight line that passes by pt
m with the reference straight lines r
n e r
2. Therefore, we will have:
Ay™ = yrι (x P,m )- yrχ (xptm ) Equation 12
( 447,19 266,39 ^ ( 447,19 - + , - -403,75 tyim = ■ + - Equation 14 tg{- 89,83) sen(- 89,83) J ^ tg(91,09) sen(91,09)
Ay.m = -265,06 + 395,31 = 130,25 [pixel] Equation 15
With regard to the calibrator, the distance between the reference straight lines η e r2 is given by:
& cαlibre ~ Equation 16
Cali re = 195,481 - 185,563 = 9,918 [mm] Equation 17
Due to rotation of the image, we will have:
calibre / , .. ψim ) Equation 18
pixel] Equation 19
Therefore, we will have the following parameters presented in the table below.
Table 3 Summary of the relationship between the image and what is real
Once the calibration of the system is completed and before initiating the step of capturing the wheel image, the steps of fixing this wheel 1 to the fixing device 2 begins, with the aid of the expansive pincers 9, as illustrated in figure 15.
For a precise control, the innermost portions of the wheel rim should remain in contact with the seating surface 6 of the device 2.
Then, one initiates the step (b) with by adjusting a lighting system (not shown) and capturing the image of the wheel 1 by means of the image-capturing device 10, as illustrated in figure 16. The captured image is then sent to the image- processing and contour-generating system 20, and then the step (c) begins. In this step (c), a contour of the wheel 1 is generated from the processing of the image of this wheel 1 by means of mathematical algorithms. In this way, for the processing of the image of the wheel 1 and generation of its contours, one has initially appraised the utilization of the Sobel algorithm. However, this algorithm did not present good results, since it generated both defined contours and little-defined contours (figure 17) for different captured images of the wheel 1. This is due to the fact that the wheel 1 is cylindrical and is made from a reflexive material, which requires a better controlled illumination.
It was then found that, by working with the Canny algorithm, one generates clearer and constant contours, as can be seen in figure 18.
Once the contour of the wheel 1 has been generated and viewed, one initiates the step (d), in which the user may request control data of his interest by means of a command-input device 31. In this regard, one selects the contour points that present a normal direction of the straight line of the wheel seat A and of the straight line of the side that defines the width L of the wheel 1 , according to figure 19.
Once the points of interest of the contour have been determined, one adjusts the measurement lines 27 to the points, according to figure 20, by using the mathematical method of the squared minimums. In this step, both the contour of the wheel 1 and the measurement lines 27 may be viewed by the user through the viewer 24.
Thus, with the measurement lines 27, the commands received by the command-input device 31 are sent, together with the generated contour, to the data- processing element 32, which, by means of mathematical algorithms, will process the requested data and send them to the control-data generating element 33, which generates data requested by the user and makes the viewable through the viewer 24. In this regard, considering the preceding example, we have the following data for this section of the wheel:
Table 5 Data of the straight lines of interest from the image of the wheel:
In addition, the generated control data may be arranged in the form of graphs.
Then, the fixing means 2 is turned at predetermined and adjustable angles, and a new image of the wheel 1 is captured by the image-capturing device 10, whereby one initiates the process again, as described above, that is to say, in
general, with each turn of a predetermined adjustable angle, an image of the wheel 1 is captured and processed, its contour is generated and the data requested by the user are viewed through the viewer 24.
Moreover, the system does not need the user's interaction in order to function and presents the options: supervised or independent work. If the system is working independently, it is capable of approving or rejecting the product and/or taking the necessary actions depending upon the specification of the product after the information has been processed.
Besides measuring the alignment of the wheel 1 , this system and process enables one to measure the diameter of the wheel, the axial oscillation of the wheel and further enable one to achieve the shape of the wheel, among other possibilities.
For measuring the radial oscillation of the wheel, one will consider a point at a determined distance d from the side flap L of the wheel 1; and for measuring the axial oscillation one will consider a point at a distance d from the tire seat A, as shown in figure 21.
Thus, starting from the example proposed before and allotting the value of 8.00 mm to the distances d, it is necessary to obtain the straight line rd displaced 8.00 mm from the side flap L of the wheel 1. For this purpose, we have the following characteristics of this wheel :
- width = 152.4 mm
- distance from the center of application of load onto the wheel = 40.00 mm
- distance of tire seat A to the point ptm = 34,967 mm Therefore, with the distance from the pint ptm, of the bore 4 of the calibrator 3 to the straight line rdL will be given by:
("152,4
Preal ~~ ' -40-8 . + 34,967 = 6,767 mm Equation 20
Converting into the image unit, considering the rotation thereof of θin = 0,61°, we will have:
r im Equation 21
pim = 94,319[pixel] Equation 23
Therefore, the straight line rdι_ will present the parameters presented in the table below.
Table 6 Parameters of the straight liner r^L
In the same way, it is necessary to obtain the straight line rdA, displaced 8,00 mm from the tire seat A, which is the same as:
D
+ h Equation 24
wherein D is the distance from the wheel and h is obtained by the equation: h = r ■ tg(β) Equation 25
wherein ris the radius of the sphere and β is obtained by the equation:
90° -α β = Equation 26
wherein α is the inclination of the seat, in this case, equal to 5° (figure 7). Thus, the distance of the rdA is determined by:
D 90° ■ a
■ + r -q Equation 27
Therefore, the distance from the straight line rdA to the point ptm will be:
Preo/ = 7, 1 [mm] Equation 30
Converting into the image unit, considering the rotation of the same of θim = 0,61 °, we will have:
pim = 95,995 [pixel] Equation 32
In this way, the straight line rdA will present the parameters presented in the following table:
Table 7 Parameters of the straight line rdA
Once both straight lines rdL and r A are determined, one can obtain the intersection point I between the straight line of the wheel seat rA and the straight line rdL displaced of the side wing of the wheel 1 , which allows the measure of the radial oscillation of the wheel 1.
The intersection point I2 between the side wing straight line rL and the
straight line rdA displaced of the wheel seat allows the measure of the axial oscillation of the wheel.
In this way, the axial oscillation of the wheel will be obtained by the variation of the intersection point I2 between the straight line of the side wing of the wheel r and the straight line rdA, on the axis direction, wherein it is measured in many sections of the wheel 1 completing 360°.
Now, the radial oscillation of the wheel will be obtained by the variation of the intersection point /, between the wheel seat straight line rA and the straight line rd ) in the radio direction, wherein it is measured in many sections of the wheel 1 completing 360°.
The algorithms used on image processes, in the obtaining of contours and in obtaining of control data are, preferably, math algorithms found in computer math programs, as for example MATLAB®.
A preferred embodiment having been described, it should be understood that the scope of the present invention embraces other possible variations, being limited only by the contents of the accompanying claims, which include the possible equivalents.