WO2002063240A1

WO2002063240A1 - Method and apparatus for mapping system calibration

Info

Publication number: WO2002063240A1
Application number: PCT/GB2002/000467
Authority: WO
Inventors: Barbara Lynn Jones
Original assignee: Snap On Technologies Inc
Current assignee: Snap On Inc
Priority date: 2001-02-02
Filing date: 2002-02-04
Publication date: 2002-08-15
Anticipated expiration: 2003-08-02

Abstract

Method and apparatus for calibration of optical or other three-dimensional mapping systems suitable for automotive repair and diagnostics provides a system to avoid the X axis (camera viewing direction) calibration errors arising from calibration based on two-dimensional emitter arrays in the Y and Z plane. This is achieved by providing twin planar emitter arrays separated in the X axis direction and simultaneously within shot of the triple camera assembly.

Description

METHOD AND APPARATUS FOR MAPPING SYSTEM CALIBRATION

This invention relates to a method and apparatus for mapping system calibration. The method and apparatus is particularly but not exclusively applicable to the calibration of apparatus for the three-dimensional mapping of automotive vehicles for diagnostic and/or repair purposes. A particular application of the embodiments of the invention described below relates to the provision of a system whereby optical three-dimensional mapping apparatus which has been factory-calibrated for initial use can be readily and relatively quickly validated from time to time in terms of its compliance with initially- established calibration criteria so as to ensure that ongoing mapping tasks can be performed with the requisite level of dimensional accuracy despite the rigours of automotive service and diagnostic equipment usage which can involve shock loadings and impacts which are detrimental to the maintenance of accurate calibration. In this connection, the embodiments of the invention provide a simple method and apparatus enabling the relatively rapid and straightforward validation of the original calibration accuracy.

According to the invention there is provided a method and apparatus as defined in the accompanying claims. An object of the present invention is to provide a method and apparatus adapted for use in relation to the calibration of optical three-dimensional mapping systems and offering improvements in relation to system calibration in accordance with one or more of the factors discussed herein, or indeed improvements generally.

In the embodiments, the apparatus requirements are extremely modest in the sense that there is a requirement merely for the provision of dual energy emitter means at structurally-defined positions in the y, z axes plane and separated in the x-axis plane (which extends in the direction of the linear distance between the camera and the dual energy-emitting calibrational targets) .

Also in the embodiments, adoption of calibration apparatus in which mechanically defined energy emitters are provided at locations providing the necessary degree of precision position definition in not only the plane generally transverse to the camera viewing direction but also in the x axis plane (which provides a measure of the distance separating the camera from the article to be mapped in use, and indeed which separates in the lengthwise direction separate points on the article to be mapped) .

It is explained below with supporting data that, with respect to the normal three-dimensional mapping mode of use of an optically-based camera mapping system, that there is a difference in terms of a significant variation in magnitude as between the calibrational errors which can occur between the three axial directions defining the space mapped by an optical mapping system. Previous (possibly unpublished) proposals for calibration methods applicable to such optical mapping systems have been based upon the obviously relevant and direct and simple approach of providing accurately-defined calibrational emitter positions only in the plane which extends at right angles to the camera viewing direction. While such an approach does provide a basis for satisfactory calibration of the system with respect to the two axes lying in that particular plane, this nevertheless leaves out the dimensionally more important third axis for calibration purposes, thereby leading to a significant risk that one of the coordinates with respect to any given map point may lack the requisite accuracy provided by routine calibration at relevant intervals during service use of the equipment . In the described embodiments, the approach to the provision of the position-defined energy emission locations in each of the relevant three axes is as follows. The embodiment provides two planar assemblies of energy emitters, these assemblies themselves being separated by an acurately defined distance in the x, z plane which separates the camera from the article to be mapped, in use. In this way, each of the two planar assemblies of emitters provides (by virtue of its n emitters) , not only n defined emission locations in the transverse y, z plane but also n position defined emission locations for each of the corresponding emitters on the other planar assembly of emitters, and separated therefrom by a defined distance in the requisite third plane. By insuring that the two planar assemblies are simultaneously "visible" to the camera assembly to be calibrated, this approach provides, in the embodiments, the means for fulfilling the desired criteria discussed above.

Embodiments of the invention will now be described, by way of example, with reference to the accompanying drawings in which: Fig 1 shows a diagrammatic representation of a three dimensional mapping system employing a three-module camera assembly to determine by triangulation the coordinates of a pointer at a location to be mapped, the Figure including identification of the x, y and z axes therefor; Fig 2 shows error curves dx, dy, dz relating to the mapping of a point x,y,z in the mapping system of Fig 1 ;

Fig 3 shows a plot of calibrational errors arising from local origin sources (two dimensional calibration) as against global origin sources (three dimensional calibration) , the divergence between the curves illustrating the need for a third dimension of calibrational checking; and

Fig 4 shows an assembly of two planar mounting planes separated in the x axis direction (the viewing direction) as employed in an embodiment of the invention for calibration purposes.

The specific embodiments of this invention are adapted to provide validation of the calibration of an optical mapping system incorporating a camera having triple energy sensing modules. Such a camera system is of course calibrated in- factory at the time of manufacture but subsequent robust usage in the automotive repair and diagnostics industry can lead to a need for validation and/or adjustment of the calibration settings. The embodiments described below may be employed with optical three-dimensional mapping apparatus of the kind employing a triple module camera assembly, for example the camera assembly described in GB 0102420.7 filed 31 January 2001 (our reference P350219GB) . Such camera apparatus may be factory calibrated by the method and apparatus described in GB 0102651.7 filed 2 February 2001 (our reference P350444GB) .

The camera apparatus disclosed in the above GB 0102420.7 application is of substantial construction and adapted to withstand significantly better than presently available apparatus the shock loadings and the like which arise during use of automotive diagnostic and servicing equipment in everyday usage. Nethertheless, it has to be recognised that such usage can give rise to the need for recalibration since the dimensional effect of a very minor relative movement of, for example, the sensor means within one or more camera modules is inherently magnified very substantially by the geometry of the in-use camera-to- camera mapping point distances, as indicated in Fig 1. It should be noted that for reasons of simplicity and clarity of illustration in Fig 1, the camera-to-mapping point distances have been somewhat reduced in proportion to the camera dimensions themselves, whereas in fact the apparatus may well be used for mapping points at distances from the camera very significantly greater than the camera end-to- end longitudinal dimension.

Considering the significance of calibrational errors, it is to be understood that in the mapping of a given point in three dimensional space the mapping operation is performed by measuring the angles subtended at three known locations, the camera module locations, and knowing also the linear distances between the camera modules. These geometrical relationships are shown in Fig 1 in which mapping apparatus 10 comprises a camera assembly 12 comprising three camera modules 14, 16 and 18 of which module 14 is taken as being at the origin 20 for the mapping system and thus the coordinates of the camera modules 14, 16 and 18 in the X, Y and Z directions

(indicated at 22) are shown at 24, 26 and 28, namely 0,0,0 (for 24) and 0,d,0 (for 26) and 0,2d,0 (for 28).

In the system illustrated in Fig 1, the camera assembly 22 is mapping a pointer 30 having a light-emitting diode (LED) 32 located at a point in space which may be designed point X,Y,Z and identified by arrow 34. It is to be understood that the data processing apparatus associated with mapping apparatus 10 is adapted to calculate the linear dimensional offset between LED 32 and the point 36 of pointer 30 so as to map the actual location of point 36 rather than of LED 32, and this is achieved by providing a second LED (not shown) spaced lengthwise from LED 32 on the shaft or body of the pointer, and causing the system to map simultaneously both locations whereby the location of pointer 30 itself can be readily determined.

It will now be readily seen that since the linear distances between camera modules 14, 16, 18 are known

(distance d between each pair) and since the camera units themselves can determine the angles øl, ø2 , ø3 , whereby the coordinates of X,Y,Z at 34 can be determined on the basis of the equations for the interception of three planes as shown in Fig 1, namely: y = 2d / ( 1 + tan ( ø₁ ) /tan ( ø₃ ) )

Z = xtan (ø₂)

In the embodiment, each of the camera modules 14, 16 and 18 comprises a linear array of 3648 pixies of pitch 8 microns and located at a distance of 35mm from a 300 micron camera slit. The separation between modules 14, 16 and 18 is 400mm in each case. The resultant error curves for a perfect system, errors dx, dy, dz are shown in Fig 2 which is a plot of the error in mm against the X axis in mm. It will be seen from the divergence between the error curves for the dx,dy and dz axis that the X axis error potential is significantly greater than the corresponding y and z axes error potentials in this geometrical situation, whereby the unacceptability of providing only two dimensional calibration apparatus becomes apparent.

In further investigation of the limitations of two dimensional calibration, we have plotted in Fig 3 firstly the error data arising from two dimensional calibrational plotting, and secondly the error data relating to the third axis namely the X axis. These two plots are identified in Fig 3 as, respectively, the dy, dz plot and the dx plot.

In the dy, dz plot this is based on the use of a simple two dimensional plate-type calibration device carrying a number of LEDs in which one of the LEDs is used as the axial origin and the two perpendicular axes are located in the plane of the plate. All other LEDs are mapped with respect to the origin LED. As Fig 3 shows, from the very shallow slope of the dy, dz curve, although such an arrangement can record errors, they are too small to identify a small calibrational change requirement (although they can identify a major catastrophic change to the camera dimensions) . In short, the two dimensional plate calibration system fails to identify error in the x axis which in fact is the largest contributor to systematic errors in mapping single points.

Further evidence of the significance of X axis calibrational errors is evident from the following computation.

Considering an article to be mapped comprising a simple bar of 500mm length with an LED at each end (such bar can be considered as located at the pointer 30 in Fig 1) the following tabulation shows typical error simulation for known camera dimensions when the bar is parallel to the x axis and parallel to the y axis and parallel to the z axis respectively, as shown:

The above tabulated data clearly shows the substantially greater X axis errors as compared with the y and z axis errors, which likewise confirms the related data disclosed herein in relation to Figs 2 and 3 hereof.

Turning now to the embodiment of the invention as illustrated in Fig 4, this provides calibration apparatus 100 for the calibration of optical mapping apparatus for three dimensional coordinate determination, for example such apparatus adapted for automotive crash repair and diagnostics and the like. Such mapping apparatus is of the kind discussed above and illustrated as mapping apparatus 10 in Fig 1 hereof.

Calibration apparatus 100 comprises optical coordinate data evaluation apparatus including optical transmitter means 102, and optical receiver means comprising camera assembly 12 of Fig 1 hereof. Apparatus 100 further comprises data processing means (not shown) in the form of computer apparatus adapted to process data derived from the transmission of an optical energy signal between the transmitter and receiver means so as to determine information with respect to the three dimensional coordinates of the transmitter means with respect to the receiver means. Such data processing means is well known in relation to optical mapping systems of this kind and no further disclosure thereof is deemed necessary.

In use of apparatus 100, a series of data evaluation calibration steps are carried out with the apparatus in which the transmitter means is provided at a series of known locations while an energy signal is transmitted and the receiver means is located at one or more fixed and known locations relative to the transmitter locations throughout the data-receiving calibration steps.

As shown in Fig 4, optical transmitter means 102 comprises a series of individual transmitter means 104, each in the form of a light -emitting diode or LED positioned at known spaced locations on a pair of mounting planes 106, 108 each in the form of a linear and planar board, the mounting planes being disposed in parallel planes at a spacing 110 extending in the X-axis direction, this spacing being fixed and determined at a precision definition thereof by means of mechanical structure (not shown) interconnecting the mounting planes so that they are held in their defined relative positions (and relative to camera assembly 12) with the requisite degree of mechanical precision.

Thus, the calibration apparatus 100 is employed in relation to the camera assembly 12 of Fig 1 as follows.

Calibration apparatus 100 is positioned in the manner illustrated in Fig 1 in relation to the camera assembly 12 at a location corresponding, approximately to the location 34 of point X, Y, Z so that the calibration apparatus is easily within shot of the camera assembly. In addition, although such is not needed in relation to the normal mapping use of camera assembly 12, the calibration apparatus 100 is positioned at a precisely-defined distance and attitude with respect to the camera assembly. This is achieved by structure (not shown) interconnecting the calibration apparatus 100 and camera assembly 12.

Accordingly, camera assembly 12 can determine the angles ø_1# ø₂ and ø₃ as shown in Fig 1. In addition the X axis distances from the camera assembly to mounting planes 106 and 108 are known and likewise the locations of LEDs 104 on the mounting planes whereby the internal calibrational data relating to camera assembly 12 can be readily determined since these are now the only unknowns in the mapping coordinate determination equations discussed above. Moreover, because mounting planes 106, 108 are at a precisely-defined X axis spacing from each other, and likewise the LEDs 104 on each such plane, there is conveniently contained a calibrational coordinate mapping determination for each of these which provides multiple calibrational data in all three x,y,z axes so that averaged calibrational settings can be obtained for greater accuracy.

Claims

1. A method of calibration of apparatus for three- dimensional coordinate determination adapted for automotive crash repair and diagnostics and the like, the method comprising : a) the step of providing coordinate data evaluation apparatus comprising transmitter means and receiver means and data processing means adapted to process data derived from the transmission of an energy signal between said transmitter and receiver means to determine information with respect to the three-dimensional coordinates of one of said transmitter means and said receiver means with respect to the other thereof; and b) the step of carrying out a series of data evaluation calibration steps with said apparatus in which said transmitter means is provided at a series of known locations while said energy signal is transmitted and said receiver means is located at one or more fixed or known locations relative to said first-mentioned known locations throughout said data receiving calibration steps; characterised by c) said step of providing said transmitter means at said series of known locations comprising providing same mounted at a series of known spaced locations on at least two mounting planes spaced-apart in the forward viewing direction of energy reception by said received means; and d) said step of causing said receiver means to be located at one or more fixed or known locations relative to said known locations comprising mounting same at a known distance in said forward viewing direction from said mounting planes and at a known attitude thereto and such as to enable the effective transmission of said energy signal therebetween .

2. A method of calibration of apparatus for three- dimensional coordinate determination characterised by the step of providing transmitter means at a series of known locations by mounting same on at least two spaced apart mounting planes, and further characterised by the step of causing receiver means to be located at one or more fixed locations relative to said known locations by mounting same at a known distance from and attitude to said mounting planes and causing the effective transmission of an energy signal therebetween.

3. A method according to claim 1 or claim 2 characterised by providing said transmitter means comprising a series of optical LED emitters and said receiver means comprising a camera module.

4. Calibration apparatus for the calibration of mapping apparatus for three-dimensional coordinate determination adapted for automotive crash repair and diagnostics and the like, the mapping apparatus comprising: b) coordinate data evaluation apparatus comprising transmitter means and receiver means and data processing means adapted to process data derived from the transmission of an energy signal between said transmitter and receiver means to determine information with respect to the three- dimensional coordinates of one of said transmitter means and said receiver means with respect to the other thereof; and c) said mapping apparatus being adapted to carry out a series of data evaluation calibration steps in which said transmitter means is provided at a series of known locations while said energy signal is transmitted and said receiver means is located at one or more fixed or known locations relative to said first-mentioned known locations throughout said data receiving calibration steps; characterised by d) said calibration apparatus being adapted to provide said transmitter means at said series of known locations by providing same mounted at a series of known spaced locations on at least two mounting planes spaced- apart in the forward viewing direction of energy reception by said receiver means whereby, in use said receiver means located at one or more fixed or known locations relative to may be mounted at a known distance in said forward viewing direction from said mounting planes and at a known attitude thereto and such as to enable the effective transmission of said optical energy signal therebetween.

5. Apparatus for the calibration of mapping apparatus for three-dimensional coordinate determination characterised by said calibration apparatus comprising transmitter means provided at a series of known locations by mounting same on at least two spaced apart mounting planes, whereby, in use, receiver means may be located at one or more fixed locations relative to said known locations by mounting same at a known distance from and attitude to said mounting planes thereby to permit the effective transmission of an energy signal therebetween.

6. Apparatus according to claim 4 or claim 5 characterised by said transmitter means comprising a series of optical LED emitters and said receiver means comprising a camera module .

7. A method of calibration substantially as herein before described with reference to the accompanying drawings .

8. Apparatus for calibration substantially as herein before described with reference to the accompanying drawings .