Method for correcting a spherically mounted retroreflector when resetting a distance meter

0001The present invention generally relates to a method of measuring a spherically mounted retroreflector (SMR), specifically a method of obtaining surface coordinates and a three-dimensional distance based on SMR measurements.

0002There is a class of measuring instruments that measure the coordinates of a point by sending a laser beam to the target of the retroreflector that touches the point. The measuring instrument obtains the coordinates of the point by measuring the distance to the target and the two angles. Distance is measured with a distance measuring device such as an ADM or an interferometer. The angle is measured by an angle measuring device such as an angle encoder. A gimbal-type beam steering mechanism inside the measuring instrument directs the laser beam to the point of interest.

0003A laser tracker is a particular type of coordinate measuring device that tracks a retroreflector target with one or more laser beams it emits. There is another category of measuring instrument known as a retroreflector or a total station or tachymeter that measures points on a diffusely scattered surface. Laser trackers generally have an accuracy similar to 1 micrometer or 2 micrometers, on the order of 1/1000 inches under certain circumstances, and are usually much more accurate than total stations. The broad definition of laser tracker includes total stations and is used throughout this application.

0004Normally, the laser tracker sends a laser beam to the target of the retroreflector. The target of a common type of retroreflector is SMR. In most cases, the term SMR applies to corner-cube retroreflectors embedded inside metal spheres. However, the term SMR can also be applied to the outer spherical portion of a metal-embedded cat-eye retroreflector. Such a cat-eye retroreflector can be configured in the shape of a sphere or the shape of two adjacent hemispheres. The corner cube retroreflector contains three mutually perpendicular reflectors. The apex, which is the common point of the intersection of the three reflectors, is near the center of the sphere. In normal tracking mode, the laser tracker sends a ray from the tracker to a location near the apex of the SMR. As long as the ray hits the apex, the returning ray retraces the path of the outgoing ray to the tracker. If the ray is slightly away from the apex of the SMR, the ray should return without an exact match, albeit parallel to the outgoing ray. The servo system inside the tracker adjusts its direction to return the light beam emitted by the tracker back to the center, which allows the light beam to track the moving retroreflector. Since the vertices are approximately aligned with the SMR's spherical center, the vertical distance from the vertices to any surface on which the SMR is located remains nearly constant, even when the SMR is rotating. Therefore, the laser tracker can measure the 3D coordinates of the surface with relatively high accuracy by tracking the position of the SMR as it moves over the surface. In this alternative, the laser tracker only needs to measure three degrees of freedom (one radial distance and two angles) to characterize the 3D coordinates of the surface.

0005SMR can also be used to measure the distance between two nests. A particularly useful type of nesting is dynamic nesting, which has the property that SMRs can be placed repeatedly in nesting. One type of nest makes three points of contact with the surface of the SMR. Some types of nesting are magnetic nesting that ensures that the SMR is in place with respect to the contact points of the nesting.

0006Some laser trackers have the ability to measure 6 degrees of freedom (DOF), which can include 3 translations such as x, y, and z and 3 rotations such as pitch, roll, and yaw. An exemplary 6DOF laser tracking system is incorporated herein by reference, Bridges, et. It is described in US Pat. No. 7,800,758 ('758) of al.). The probe disclosed in the '758 patent holds a corner-cube retroreflector with a mark placed on it. The corner-cube retroreflector is illuminated by a laser beam from the laser tracker, and the marks on the corner-cube retroreflector are captured by the orientation camera inside the laser tracker. Three degrees of freedom in orientation, for example pitch, roll, and yaw angle, are calculated based on the image acquired by the orientation camera. The laser tracker measures the distance to the apex of the corner cube retroreflector and the two angles. The distance and two angles that give the vertices three translational degrees of freedom are in place with respect to the vertices of the corner-cube retroreflector when combined with the three orientation degrees of freedom taken from the orientation camera image. The location of the probe tip placed in can be found. Such probe tips can be used, for example, to measure the coordinates of "hidden" features outside the line of sight of the laser beam from the laser tracker.

0007As explained above, the vertices of the corner-cube retroreflector in the SMR are ideally located at the exact center of the sphere in which the corner-cube is incorporated. In reality, the vertices are off the center of the sphere by a few thousandths of an inch. In some cases, the difference between the position of the apex and the position of the center of the sphere is known with high accuracy, but this data is not used to correct the tracker readings. For accurate measurements with a laser tracker, this error in centering the corner-cube retroreflector on the sphere may be greater than the error from the rangefinder and angle meter in the laser tracker. Therefore, there is a need for a method of correcting this centering error.

0008Most SMRs in use today include open-air corner-cube retroreflectors. There are some SMRs that use glass corner cube retroreflectors, but in most cases these are of limited accuracy. Due to the bending of the light entering such a glass corner cube, the light appears to travel in a direction other than the true direction within the corner cube. Therefore, SMRs made with glass corner cubes tend to be made very small, which reduces errors and tends to be used in applications where the highest accuracy is not required. A method of using a 6DOF laser tracker to minimize this error is shown in US Pat. No. 8,467,072, the content of which is incorporated by reference.

0009In a 3D coordinate measuring device such as a laser tracker, a procedure is executed to obtain a reference distance at a fixed position. The SMR may then be customarily placed in a fixed position nest and the rangefinder of the device can be set to a fixed position reference distance. However, the home reference distance is ideally given to the center of the SMR, while the measured distance is given to the apex of the SMR. The reset procedure needs to be corrected to show the distance to the vertices of the SMR, not the distance to the center of the sphere of the idealized SMR.

<p num="0010"><patcit num="1"><text>U.S. Pat. No. 7,800,758</text></patcit></p>

<p num="0011"> According to one aspect of the present invention, there is provided a method of resetting a distance meter in a three-dimensional (3D) coordinate measuring device when measuring a spherically mounted retroreflector (SMR) in a fixed nest. Is a step of preparing an SMR including a body, a retroreflector, and a reference point, where the body has a spherical outer portion with a spherical center, the body contains a cavity, and the cavity is retrograde. Dimensioned to hold the reflector, the cavity is open to the outer region of the body, the retroreflector is at least partially disposed in the cavity, and the retroreflector is An open-air corner-cube retroreflector, the retroreflector has a set of three lines and a set of three mutually perpendicular and flat reflectors that intersect at a common apex, and the cavity is ... Contains an air-filled region inside the reflective surface of the set of three flat reflectors, the retroreflector has an axis of symmetry with respect to the set of three lines, and the SMR is on the axis of symmetry. It has an SMR runout plane that is perpendicular to the center of the sphere, the SMR has an SMR intersection, the SMR intersection is the intersection of the SMR runout plane and the axis of symmetry, and the SMR is from the apex. It has an error vector that extends to the center of the sphere, the SMR error vector has an SMR depth error vector component and an SMR runout error vector component, and the SMR depth error vector component is from the apex to the intersection of the SMRs. A step in which the SMR runout error vector component is a vector extending from the intersection of the SMRs to the center of the sphere, and the depth error vector component of the SMR has a magnitude equal to the depth error of the SMR. And the step of preparing a 3D coordinate measuring device including a base, a first motor, a second motor, a first angle measuring device, a second angle measuring device, a distance meter, and a fixed position nest, which is 3D. The coordinate measuring device has the coordinate system of the device and is configured to emit light rays to the SMR, and the retroreflector returns a part of the emitted light rays as reflected light rays, and the first motor. And the second motor together guide the emitted rays in the direction they are emitted and are emitted.The direction is determined by the first rotation angle around the first axis and the second rotation angle around the second axis, the first rotation angle is generated by the first motor and the second rotation. The corners are generated by the second motor, the first and second axes are configured to rotate around the gimbal point, with the first and second axes rotating relative to the base. The gimbal point is on the line that coincides with the emitted light beam, the first angle measuring device measures the first rotation angle, and the second angle measuring device measures the second rotation angle. The distance meter measures the target distance from the device to the retroreflector, at least partially based on the reflected portion of the reflected light received by the device, and a in-situ nest is attached to the base. A step that is configured to receive the outer part of the sphere, a step that prepares the processor and memory, and a step that prepares a computer-readable medium with a computer-readable instruction to determine the reset distance when executed by the processor. The step of finding the reference distance of the fixed position, which is the distance from the gimbal point to the spherical center of the SMR in the nest of the fixed position, the step of finding the depth error of the SMR, and the reference distance of the fixed position and the depth error of the SMR. , The step of recording as a set of recorded data, the step of reading the set of recorded data by the processor, the step of placing the SMR in a fixed nest, and from the device to the SMR in a fixed nest. , The step of sending the first emitted ray, the step of executing computer-readable instructions on the processor to determine the reset distance based at least in part on the reference distance in place and the depth error of the SMR, and the distance meter. Includes a step of resetting to read a distance equal to the reset distance.It is configured to rotate on a computer, the first and second axes are rotated with respect to the base, the gimbal point is on the line that coincides with the emitted light beam, and the first angle measurement. The device measures the first rotation angle, the second angle measuring device measures the second rotation angle, and the distance meter is at least partially based on the reflected portion of the reflected light received by the device. , Measure the target distance from the device to the retroreflector, and prepare a processor and memory, as well as a step where a fixed nest is attached to the base and configured to receive the outer part of the sphere. And the step of preparing a computer-readable medium with a computer-readable instruction to find the reset distance when executed by the processor, and the fixed-position reference distance, which is the distance from the gimbal point to the spherical center of the SMR in the fixed-position nest. A step of finding, a step of finding the depth error of SMR, a step of recording a reference distance at a fixed position and a depth error of SMR as a set of recorded data, and a processor reading a set of recorded data. A step, a step of placing the SMR in a fixed nest, a step of sending a first emitted ray from the device to the SMR in the fixed nest, and a computer-readable instruction on the processor to perform the fixed. It includes the step of finding the reset distance based at least in part on the reference distance of the position and the depth error of the SMR, and the step of resetting the distance meter to read a distance equal to the reset distance.It is configured to rotate on a computer, the first and second axes are rotated with respect to the base, the gimbal point is on the line that coincides with the emitted light beam, and the first angle measurement. The device measures the first rotation angle, the second angle measuring device measures the second rotation angle, and the distance meter is at least partially based on the reflected portion of the reflected light received by the device. , Measure the target distance from the device to the retroreflector, and prepare a processor and memory, as well as a step where a fixed nest is attached to the base and configured to receive the outer part of the sphere. And the step of preparing a computer-readable medium with a computer-readable instruction to find the reset distance when executed by the processor, and the fixed-position reference distance, which is the distance from the gimbal point to the spherical center of the SMR in the fixed-position nest. A step of finding, a step of finding the depth error of SMR, a step of recording a reference distance at a fixed position and a depth error of SMR as a set of recorded data, and a processor reading a set of recorded data. A step, a step of placing the SMR in a fixed nest, a step of sending a first emitted ray from the device to the SMR in the fixed nest, and a computer-readable instruction on the processor to perform the fixed. It includes the step of finding the reset distance based at least in part on the reference distance of the position and the depth error of the SMR, and the step of resetting the distance meter to read a distance equal to the reset distance.The step of preparing the processor and memory, the step of preparing a computer-readable medium with a computer-readable instruction to obtain the reset distance when executed by the processor, and the spherical center of the SMR in the nest in place from the gimbal point. A step of finding the reference distance of the fixed position, which is the distance to, a step of finding the depth error of the SMR, and a step of recording the reference distance of the fixed position and the depth error of the SMR as a set of recorded data. A step of reading a set of recorded data by a processor, a step of placing an SMR in a fixed nest, and a step of sending a first emitted ray from the device to the SMR in a fixed nest. The processor executes computer-readable instructions to determine the reset distance based at least in part on the reference distance in place and the depth error of the SMR, and resets the distance meter to read a distance equal to the reset distance. Including steps.The step of preparing the processor and memory, the step of preparing a computer-readable medium with a computer-readable instruction to obtain the reset distance when executed by the processor, and the spherical center of the SMR in the nest in place from the gimbal point. A step of finding the reference distance of the fixed position, which is the distance to, a step of finding the depth error of the SMR, and a step of recording the reference distance of the fixed position and the depth error of the SMR as a set of recorded data. A step of reading a set of recorded data by a processor, a step of placing an SMR in a fixed nest, and a step of sending a first emitted ray from the device to the SMR in a fixed nest. The processor executes computer-readable instructions to determine the reset distance based at least in part on the reference distance in place and the depth error of the SMR, and resets the distance meter to read a distance equal to the reset distance. Including steps.</p><p num="0012"> The system for measuring the three-dimensional (3D) coordinates of a spherical mount retroreflector (SMR) with a coordinate measuring device is an SMR containing a main body and a retroreflector, and the main body is a spherical outer part having a spherical center. The body contains a cavity, the cavity is sized to hold a retroreflector, the cavity is open to the outer region of the body, and the retroreflector is inside the cavity. At least partially disposed, the SMR contains a bar code, the bar code is configured to hold the SMR information, and the SMR information is at least one of the SMR compensation parameters or the SMR serial number. , SMR and base, 1st motor, 2nd motor, 1st angle measuring device, 2nd angle measuring device, distance meter, position detector, control system, fixed position nest, camera, and processor A coordinate measuring device that includes and is configured to send a first ray to the SMR, where the target of the retroreflector returns a portion of the first ray as a second ray. A first motor and a second motor that guide a first ray together in a first direction, with the first direction being the first rotation angle and the second axis around the first axis. Determined by the second rotation angle around, the first rotation angle is generated by the first motor, the second rotation angle is generated by the second motor, and the first and second axes are A first motor and a second motor configured to rotate with respect to the base, a first angle measuring device for measuring the first rotation angle, and a second for measuring the second rotation angle. Measure the first distance from the coordinate measurer to the retroreflector target, at least in part, based on the angle measurer and the first part of the second beam received by the first light detector. A position that receives the second part of the distance meter and the second beam and is configured to generate a first signal in response to the position of the second part on the position detector. A detector and a second signal that is at least partially based on the first signal is sent to the first motor, and a third signal that is at least partially based on the first signal is sent to the second mode.A control system that sends to the camera, configured to adjust the first direction of the first beam to track the SMR, and attached to the base configured to hold the SMR. A camera with a fixed position nest and a photosensitive array and a lens with a focal length that has a first distance between the lens and the photosensitive array to receive and respond to the first image. It is configured to supply the camera's electrical signal to the processor so that when the SMR is placed in a in-position nest, it will focus on the bar code on the SMR on the first image. In cameras where the distance is selected at least partially based on the focal length, the camera has sufficient resolution to clearly decompose the bar code, and the bar code is at least partially based on the camera's electrical signal. A processor configured to process the image, extract SMR information, and provide the 3D coordinates of the retroreflector target, where the 3D coordinates are the first distance, the first rotation angle, It includes a second rotation angle, and a processor that is at least partially based on SMR information.It comprises a processor whose dimensional coordinates are at least partially based on first distance, first angle of rotation, second angle of rotation, and SMR information.It comprises a processor whose dimensional coordinates are at least partially based on first distance, first angle of rotation, second angle of rotation, and SMR information.</p><p num="0013"> The embodiments are then described by way of example only, with reference to the accompanying drawings, which are meant to be exemplary rather than limiting, and in some figures similar elements are numbered similarly. ..</p>

0014<figref num="1">It is a perspective view of the laser tracker and SMR by one Embodiment.</figref><figref num="2">FIG. 5 is a diagram of a laser tracker, an auxiliary unit, and an external computer according to an embodiment.</figref><figref num="3">It is a block diagram which shows the element of the payload of the laser tracker by one Embodiment.</figref><figref num="4">It is a perspective view of the element used to duplicate a corner cube retroreflector.</figref><figref num="5A">FIG. 6 is a perspective view of an SMR including an open-air corner cube slug incorporated within a spherical surface according to an embodiment.</figref><figref num="5B">FIG. 6 is a cross-sectional view of an SMR including an open-air corner cube slug incorporated within a spherical surface according to an embodiment.</figref><figref num="5C">FIG. 6 is a front view of an SMR including an open-air corner cube slug incorporated within a spherical surface according to an embodiment.</figref><figref num="6A">It is a perspective view of the SMR with the reflection area added according to the embodiment.</figref><figref num="6B">FIG. 3 is a perspective view of an SMR with a barcode pattern or serial number added, according to an embodiment.</figref><figref num="6C">FIG. 6 is a perspective view of an SMR with an RF identification tag added according to an embodiment.</figref><figref num="7A">FIG. 3 is a front view of an SMR with a corner cube retroreflector that is not completely centered within a sphere.</figref><figref num="7B">It is a side sectional view of an SMR having a corner cube retroreflector that is not completely centered in a sphere.</figref><figref num="8A">It is a perspective view of a part of a corner cube retroreflector having an axis of symmetry.</figref><figref num="8B">FIG. 5 is an enlarged view of an error vector and an enlarged view of the apex and spherical center of an SMR having an error vector component according to an embodiment.</figref><figref num="9A">It is a perspective view of a part of a corner cube retroreflector including a reference mark and a runout reference plane of SMR.</figref><figref num="9B">It is an enlarged view of the apex of SMR and the center of a spherical surface which shows the runout reference ray of SMR and the runout reference angle of SMR according to one Embodiment.</figref><figref num="9C">It is a perspective view of a part of a corner cube retroreflector including a reference mark and a runout reference plane of a ray.</figref><figref num="9D">FIG. 5 is an enlarged view of the apex and the center of the spherical surface showing the run-out reference ray of the ray and the run-out reference angle of the SMR according to one embodiment.</figref><figref num="10A">It is a perspective view of SMR which provides a connector socket.</figref><figref num="10B">FIG. 5 is a cross-sectional view of an SMR showing a temperature sensor configured to electrically connect with a connector cable, which is also shown, according to one embodiment.</figref><figref num="10C">It is a front view of SMR which provides a connector socket.</figref><figref num="10D">It is a figure which shows the element of FIG. 10C which added the built-in temperature measurement and communication unit.</figref><figref num="10E">It is a schematic diagram of the electrical component in a temperature measurement and communication unit.</figref><figref num="10F">It is a hand-held diagram of SMR with a temperature measurement and communication unit attached to the hand.</figref><figref num="11">FIG. 5 is a diagram of an SMR fitted with an interface unit containing a battery and antenna used with a temperature sensor.</figref><figref num="12">It is a schematic diagram of the error caused by the misalignment of the light ray from the 3D measuring device and the axis of symmetry of SMR.</figref><figref num="13A">FIG. 6 is a schematic perspective view of an alternative method for aligning a ray from a 3D measuring device with the axis of symmetry of the SMR according to one embodiment.</figref><figref num="13B">FIG. 6 is a schematic perspective view of an alternative method for aligning a ray from a 3D measuring device with the axis of symmetry of the SMR according to one embodiment.</figref><figref num="13C">FIG. 6 is a schematic perspective view of an alternative method for aligning a ray from a 3D measuring device with the axis of symmetry of the SMR according to one embodiment.</figref><figref num="13D">FIG. 6 is a schematic perspective view of an alternative method for aligning a ray from a 3D measuring device with the axis of symmetry of the SMR according to one embodiment.</figref><figref num="13E">FIG. 6 is a schematic perspective view of an alternative method for aligning a ray from a 3D measuring device with the axis of symmetry of the SMR according to one embodiment.</figref><figref num="13F">FIG. 6 is a schematic perspective view of an alternative method for aligning a ray from a 3D measuring device with the axis of symmetry of the SMR according to one embodiment.</figref><figref num="14A">It is a figure which shows the dynamic nest and the support shaft.</figref><figref num="14B">It is a figure which shows the dynamic nest which receives SMR.</figref><figref num="15">It is a figure which shows the mathematical model of the dynamic nest of FIG. 14A.</figref><figref num="16A">It is a front view of the SMR held by the dynamic nest.</figref><figref num="16B">It is a side view of the SMR held by the dynamic nest.</figref><figref num="16C">It is a figure which shows that the error of the radius of SMR does not have a significant influence on the measured length.</figref><figref num="17A">FIG. 5 is a side view of an SMR held by a dynamic nest having one of the nests perpendicular to the desired measurement direction.</figref><figref num="17B">It is a figure which shows the case where a serious error may occur.</figref><figref num="18">It is a diagram showing how to perform accurate 3D measurement of the spherical center of SMR for one device placed in two stations while performing SMR alignment in a single station.</figref><figref num="19A">In one embodiment, the direction of the maximum runout error vector component that can be adjusted to a reference point on the SMR can have a significant effect on the measurement error.</figref><figref num="19B">In one embodiment, the direction of the maximum runout error vector component that can be adjusted to a reference point on the SMR can have a significant effect on the measurement error.</figref><figref num="19C">It is a figure which shows the general method which minimizes a measurement error by adjusting a runout error vector component.</figref><figref num="19D">It is a figure which shows the general method which minimizes a measurement error by adjusting a runout error vector component.</figref><figref num="20A">It is a figure which shows the distance measurement performed for SMR of the foreground mode.</figref><figref num="20B">It is a figure which shows the distance measurement performed for SMR of the rear view mode.</figref><figref num="20C">It is a figure which shows the distance measurement performed with respect to a fixed position.</figref><figref num="21A">It is a figure which shows the method of measuring the distance between two spherical mount retroreflectors.</figref><figref num="21B">It is a figure which shows the method of obtaining the offset error of the measured distance.</figref><figref num="21C">It is a figure which shows the method of finding the axis offset value about a coordinate measuring apparatus.</figref><figref num="22">It is a figure which shows the method of setting the distance to the apex of SMR at a fixed position.</figref><figref num="23">It is a figure which shows the electronic device and the processor in the laser tracker by one Embodiment.</figref>

0015FIG. 1 shows an exemplary laser tracker 10. The exemplary gimbal beam steering mechanism 12 of the laser tracker 10 includes a Zenith carriage 14 mounted on an azimus base 16 and rotating about an azimus axis 20. The payload 15 is mounted on the Zenith carriage 14 and rotates around the Zenith shaft 18. The Zenith mechanical rotation axis (not shown) and the Azimas mechanical rotation axis (not shown) intersect vertically inside the tracker 10 at the gimbal point 22, which is generally the origin for distance measurement. The laser beam 46 substantially passes through the gimbal point 22 and is directed orthogonally to the Zenith axis 18. In other words, the laser beam 46 is in a plane perpendicular to the Zenith axis 18. The laser beam 46 is directed in a desired direction by a motor in a tracker (not shown) that rotates the payload 15 around the Zenith axis 18 and the azimus axis 20. The Zenith angle encoder and azimuth angle encoder inside the tracker (not shown) are attached to the Zenith mechanical shaft (not shown) and the azimuth mechanical shaft (not shown) to indicate the rotation angle with relatively high accuracy. .. The laser beam 46 travels to an external retroreflector 26, such as the SMR described above. By measuring the radial distance between the gimbal point 22 and the retroreflector 26 and the angle of rotation around the Zenith axis 18 and the azimuth axis 20, the position of the retroreflector 26 is found in the tracker's spherical coordinate system.

0016The azimuth base 16 to which the coordinate system 30 of the device of the laser tracker 10 is fixed is generally stationary with respect to the periphery of the tracker. The coordinate system 30 of the device can be represented by various coordinate systems. The coordinate system 30 of the device may be represented by a Cartesian coordinate system with three vertical axes x', y', and z'. The device coordinate system 30 may be represented in a spherical coordinate system, where point 74 has a radial distance of 73 (r), a first (Zennis) angle of 72 (θ), and a second (azimuth) angle of 71 (φ). ). The angle θ is obtained by using the projection of point 74 on the z-axis. The angle φ is obtained by using the projection of point 74 onto the x'-y'plane. The laser tracker 10 is originally measured in a spherical coordinate system using one rangefinder for measuring r and two angle encoders for measuring θ and φ. However, points measured in polar coordinates can easily be converted to Cartesian coordinates. In one embodiment, the gimbal point 22 is selected as the origin. Other coordinate systems are possible and may be used.

0017The laser tracker 10 also includes a rotating coordinate system. One of the rotating coordinate systems is the payload coordinate system 35, which directs the ray 46 towards the SMR26. The payload coordinate system rotates around axis 20 and axis 18. It should be understood that the term payload coordinate system can mean the final ray-transmitting portion of any type of ray-transmitting system, not just the payload 15 in Figure 1. For example, payload 15 can be replaced by a mirror that reflects ray 46 towards SMR 26. Payload coordinate system 35 can be shown in a Cartesian coordinate system with three vertical axes x ", y", and z "as shown in FIG. 1. In an exemplary payload coordinate system, the x" axis is output. The y "axis is in the direction of the Zenith axis 18 and the z" axis is in the direction perpendicular to the x "axis and the y" axis. The y "-z" plane is perpendicular to the direction x "of the laser beam. In one embodiment, the gimbal point 22 is the origin of the payload coordinate system 35.

0018The SMR26 has an SMR coordinate system 40. The SMR coordinate system can be represented, for example, in a Cartesian coordinate system with three vertical axes x, y, and z. The x-axis, y-axis, and z-axis of the SMR coordinate system 40 move with the SMR26 and, as a whole, are not parallel to the corresponding x "axis, y", and z "axis of the payload coordinate system 35. In an embodiment, the apex of the corner cube retroreflector in the SMR is the origin of the SMR coordinate system. The SMR 26 may be placed in contact with the surface 61 of the workpiece at point 63. 3D (3D) at point 63. To find the coordinates, the first tracker finds the 3D coordinates of the SMR26 apex using the distance and the two angles it measures. The first tracker then finds the 3D coordinates of the SMR apex on the surface. Shifts towards SMR by an amount equal to the spherical radius of the SMR. SMR26 can also be used to measure 3D coordinates when placed on a dynamic nest, as described further below.

0019The laser beam 46 may include one or more laser wavelengths, or the beam 46 may be a ray separate from the laser beam. For clarity and simplicity, the following discussion envisions the types of steering mechanisms shown in Figure 1. However, other types of steering mechanisms are possible. For example, it would be possible to reflect a laser beam from a mirror that rotates around the azimus and zenith axes. The techniques described herein are applicable regardless of the type of steering mechanism.

0020In the exemplary laser tracker 10, the locator cameras 52, 56 and the light source 54 are placed on the payload 15. The light source 54 illuminates the target 26 of one or more retroreflectors. In one embodiment, the light source 54 is an electrically driven LED that repeatedly emits pulsed light. Each locator camera 52 and 56 includes a photosensitive array and a lens placed in front of the photosensitive array. For example, the photosensitive array may be a CMOS or CCD array. In one embodiment, the lens of the locator camera 52 has a relatively wide field of view, for example 30 degrees or 40 degrees. In contrast, the lens of the locator camera 56 is relatively narrow, for example to allow clear reading of the SMR26 bar code or serial number held by the in-position nest 17, as discussed further below. It may have a field of view. The purpose of the lenses of cameras 52, 56 is to form an image of a target in the field of view of the lens on a photosensitive array. The image on the photosensitive array is sent to an electronic device, which may be inside or outside the photosensitive array, to supply an electrical signal to the processor. The electrical signal is evaluated by a processor to extract relevant information such as images, text, array positions, and so on. Typically, at least one light source 54 is located near the locator camera 52, so that light from the light source 54 is reflected from the target 26 of each retroreflector onto the locator camera 52. In this way, the retroreflector image is easily distinguished from the background on the photosensitive array because the spots in the retroreflector image are brighter and pulsed than the background object. In one embodiment, there are two locator cameras 52 and two light sources 54 arranged around a line of laser beams 46. By using the two locator cameras 52 in this way, the principle of triangulation can be used to find the three-dimensional coordinates of any SMR within the field of view of the locator camera. In addition, the 3D coordinates of the SMR can be monitored as the SMR moves from point to point.

0021For the locator camera 56 designed to read barcodes, the light source is the camera, as the bright flash of back-reflected light can prevent the camera from reading the much darker lines of the barcode. Not placed near. Instead, a light source 54 may be used to illuminate the barcode, optionally far enough from the locator camera 56 that the light from the retroreflector target 26 is not back-reflected to the locator camera 56. ..

0022As shown in FIG. 2, the auxiliary unit 70 may be a part of the laser tracker 10. The purpose of the auxiliary unit 70 is to power the laser tracker body and, in some cases, to provide the system with computing and timing capabilities. By transferring the function of the auxiliary unit 70 to the tracker body, it is possible to remove all the auxiliary unit 70. In most cases, the auxiliary unit 70 is attached to the general purpose computer 80. Application software loaded on the general purpose computer 80 can provide application capabilities such as reverse engineering. It is also possible to eliminate the general purpose computer 80 by building the computing power of the general purpose computer 80 directly in the laser tracker 10. In this case, a user interface that optionally provides keyboard and mouse functionality may be built within the laser tracker 10. The connection between the auxiliary unit 70 and the computer 80 may be wireless or may be via an electric cable. The computer 80 may be connected to the network, and the auxiliary unit 70 may also be connected to the network. Multiple devices, such as multiple measuring devices or actuators, may be connected to each other via either the computer 80 or the auxiliary unit 70. In one embodiment, the auxiliary unit 70 is omitted and the laser tracker 10 and the computer 80 are directly connected.

0023The laser tracker 10 measures the distance r using either an interferometer or an ADM. The laser tracker 10 uses an angle encoder to measure the azimuth angle φ and the zenith angle θ. Thus, although the laser tracker measures in a spherical coordinate system, the coordinate values for any measured point can be transformed into the coordinates of any other desired coordinate system, for example Cartesian coordinate system 30 in FIG.

0024It should be similarly understood that three translational degrees of freedom mean three independent translational degrees of freedom. Another way to say this is that the three directions corresponding to the three translational degrees of freedom form the basis set in three-dimensional space. In other words, each of the three directions corresponding to the translational degrees of freedom has components that are orthogonal to each of the other two directions.

0025FIG. 3 shows an embodiment of the electro-optical assembly 400 of the laser tracker 10. Light source elements 405 and 410 indicate a light source and, in some cases, additional electro-optic components. For example, the light source element 410 may represent a red helium neon laser in combination with an interferometer. Light source element 405 may represent an infrared laser in combination with ADM. Alternatively, the system may have only an interferometer or only an ADM. One of the light source elements 405 or 410 may have only a light source without a rangefinder. In addition to those contained within the light source elements 405 and 410, there may be additional light sources (not shown). The light source elements 405 and 410 may be located on the payload 15 or on one of the other parts of the tracker, such as the Zenith carriage 14 or the azimus base 16. The light source may be located in one part, such as the azimuth base, and the rangefinder may be located in another part, such as the payload 15. Light may be sent from one position to another by an optical fiber as described in the '758 patent. Alternatively, the light from the light source may be reflected from a mirror steered around the Zenith axis and a rangefinder held on the azimus base 16. The light source may include a laser, a super light emitting diode, a light emitting diode, or the like.

0026Light from the light source element 410 passes through the beam splitter 420. Light from light source element 405 is reflected from mirror 415 and beam splitter 420. When the light source elements 405 and 410 contain light of different wavelengths, the beam splitter 420 advantageously transmits the wavelength of the light emitted by the light source element 410 and reflects the wavelength of the light emitted by the light source element 405. , A dichroic beam splitter may be used.

0027Most of the light from the beam splitter 420 passes through the beam splitter 425. A small amount of light is reflected off the beam splitter 425 and lost. The light passes through the beam expander 435, which expands the size of the light beam at the exit from the tracker. Enlarging a ray helps because it can propagate a longer distance with smaller changes in ray size. The laser beam 440 leaving the tracker 10 travels to target 26 of the retroreflector. This part of the laser beam is reflected from the retroreflector 26 and returned to the tracker. The beam expander 435 reduces the size of the rays returning to the tracker.

0028A portion of the returning light travels to the beam splitter 425. Most of the light goes to elements 405 and 410, but a small amount is split and hits the position detector 450. In some cases, light may pass through the lens after being reflected from the beam splitter 425 and before hitting the position detector 450. The position detector 450 may be of several types, for example, a position-sensing detector, a photosensitive array, and the like. The detector capable of detecting the position may be, for example, a side effect detector or a quadrant detector. For example, the photosensitive array may be a CMOS or CCD array. The position detector responds to the position of the returning ray. Motors mounted on the Azimas mechanical shaft and motors mounted on the Zenith mechanical shaft are adjusted by a control system within the tracker 10 to keep the returning rays as close to the center of the position detector 450 as possible.

0029The body of the SMR26 has a spherical outer portion and a retroreflector. The outer portion of the sphere contains a cavity sized to hold the corner-cube retroreflector, which is at least partially disposed within the cavity. The outer portion of the sphere has a spherical center. The corner cube retroreflector may be an open-air corner cube or a glass corner cube. An open-air corner-cube retroreflector has an air interior, and a glass corner-cube retroreflector has a glass interior.

0030The corner-cube retroreflector includes three flat reflectors that are perpendicular to each other. The three flat reflectors intersect at a common vertex, which is ideally a point. Each flat reflector has two cross-joints, each cross-joint is shared with an adjacent flat reflector, and there are a total of three cross-joints in the corner-cube retroreflector. .. The interior of a corner-cube retroreflector is an area of space surrounded by reflectors that are flat on three sides. In the case of the open-air corner-cube retroreflector, which is the subject of this application, the cavity includes an air-filled portion inside three flat reflectors, three cross-joints, and vertices. The cavity is open to the outside of the body, which provides a means for light to be sent to and reflected from the retroreflector.

0031There are at least three common methods for making open-air corner-cube retroreflectors: the replication process, the mirror insertion process, and the ECM process. Figure 4 shows the replication process. The master element 510 is carefully machined to provide the desired properties in the finished replicated retroreflector. For example, the master element 510 may be machined so that each of the three flat reflector surfaces 512 is just perpendicular to the two adjacent 512s. The three flat reflector surfaces 512 of the master element 510 can be perpendicular to each of the adjacent reflectors with an angular error of up to 1 or 2 seconds. The master element 510 is coated with a reflective material such as gold. Corner cube slag 520 includes a machined void 522 coated with a thin adhesive layer of material such as epoxy resin. The corner cube slug 520 is brought into contact with the master element 510. The epoxy layer is then adapted to the shape of the master element 510. After the epoxy is cured and the slag 520 is lifted from the master element 510, a gold layer is attached to the epoxy resin, which results in a corner cube slag 520 with a reflective coating.

0032A second common method for making open-air corner-cube retroreflectors is a mirror insertion process in which a mirror panel joined to a corner-cube assembly is inserted into a cavity in a spherical outer portion. The three mirror panels are joined together so that they are perpendicular to each other.

0033A third common method for making open-air corner-cube retroreflectors is the electrochemical polishing (ECM) process. In some cases, the retroreflector and the spherical outer part are integrated into a single unit. Such SMRs can be made, for example, by removing cavities using a combination of conventional machining and ECM. Such an ECM process can be used to create three flat, smooth, perpendicular surfaces. Such surfaces may be covered with a reflective coating such as gold or silver to provide three reflective surfaces.

0034SMRs with open-air corner cube retroreflectors are shown in Figures 5A-5C. The SMR 700 shown in FIG. 5A includes a spherical outer portion 720, an open-air corner-cube retroreflector 710, a collar 905, and a reference mark or reference feature 930. In one embodiment, the cavity of the spherical outer portion 720 is sized to receive the corner cube retroreflector 710. The corner-cube retroreflector 710 is at least partially disposed within the spherical outer portion 720, optionally with an adhesive. The collar 905 provides protection against the corner cube retroreflector 710, providing a favorable grip. Reference marks or reference features 930 can be used to establish the orientation of SMRs in space, as discussed in more detail below. The reference feature 930 may be a structural feature such as a depression or ridge. The reference feature 930 may be a serial number, a reflection area (such as 610 in FIG. 6A), a barcode (such as 630 in FIG. 6B), a radio frequency identification (RFID) tag, or other feature. In this case, the reference mark may, by convention, be selected, for example, in the center, left side, or right side of a given feature. The reference feature may be any feature that allows the user or reader to identify the orientation of the retroreflector 710. FIG. 5B shows a cross section obtained through the center of the SMR700. This cross section is a duplicate of the open-air corner cube 710, but a corner cube retroreflector formed with three mirror panels or directly using ECM is equally convenient. It clarifies that it can be used. FIG. 5C shows a front view of the SMR700.

00356A-6C show three embodiments of SMR. In FIG. 6A, the SMR700 includes a spherical outer portion 720, a corner cube retroreflector 710, and a collar 905. The reflective material region 610 is located in front of the collar 905 of FIG. 6A. The area 610 of the reflective material is illuminated by light from the laser tracker and its position is determined by a locator camera disposed on the tracker. For example, light may be supplied by a light source 54, and an illuminated SMR image may be captured by one or more locator cameras 52 as shown in FIG. The location of region 610 can be used to find the orientation of SMR700, as further described below. In FIG. 6B, the SMR 700 contains the same elements as those in FIG. 6A, except that the reflective material region 610 is replaced by a barcode pattern 630. The barcode pattern 630 may be a one-dimensional barcode pattern or a two-dimensional barcode pattern. The two-dimensional bar code pattern is sometimes referred to as a matrix bar code, a 2D bar code, a 2D code, or a name such as QR indicating a specific form of the code. The barcode may serve to provide an identifier for the SMR, or it may serve to store one or more parameters of the SMR, as described in more detail below. The barcode 630 can also serve as a reference mark for the orientation of the SMR700 (useful for the function of 930 in Figure 5A), or as an area of reflective material. If desired, the barcode pattern may extend around the entire circumference of the color lip, not just a portion of the lip as shown in FIG. 6B. If desired, a type of bar code pattern known as a radial pattern can be used. In FIG. 6C, the SMR 700 contains the same elements as those in FIG. 6A, except that the reflective material region 610 is replaced by the RF identification chip 650. This chip may be questioned by the RF transmitter / receiver, the RF transmitter / receiver is about the SMR700 It can be, for example, a handheld unit or a unit placed on the laser tracker 10 for obtaining information. This information may be the SMR700 serial number or one or more parameters.

00367A and 7B show a front view and a side sectional view of the SMR 700. The SMR includes a spherical outer portion 720 and a retroreflector 710. The retroreflector 710 is a corner-cube retroreflector having three cross-joints 810A, 810B, and three flat reflectors 825AB, 825BC, and 825AC that intersect at 810C and vertex 820. In an ideal SMR, the three intersections are intersection lines and the vertices are points. In this application, the term line is often used to refer to a crossed joint, even if the joint is not perfectly straight and not sharp. Similarly, the term vertex is often used to describe a point in an area where three faces intersect, even if the actual intersection area is not a perfect point. The spherical center 860 of the spherical outer portion 720 is generally a different point in space from the apex 820.

0037The axis of symmetry 840 is symmetric with respect to the three intersecting lines 810A, 810B, and 810C. The angle between the axis of symmetry and any of the three intersecting lines is<maths num="1"><img id="000003" he="10" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>Is. The runout plane 865 passing through the center of the sphere 860 is drawn perpendicular to the axis of symmetry 840. The axis of symmetry 840 intersects the runout plane 865 at the intersection 870. The SMR error vector 885 extending from the apex 820 to the sphere center 860 has the SMR depth error vector component 880 extending from the apex 820 to the intersection 870 and the SMR runout extending from the intersection 870 to the sphere center 860. It is decomposed into two vector components with the error vector component 890. The depth error vector component 880 of the SMR has a length equal to the depth error of the SMR, and the runout error vector component 890 of the SMR has a length equal to the runout error of the SMR. FIG. 7B, showing component 880 of the SMR depth error vector and component 890 of the SMR runout error vector, is a side view of a cross section drawn through line AA in the front view of FIG. 7A. The SMR has an SMR coordinate system 730 fixed relative to the SMR. An example of a possible SMR coordinate system is shown in FIGS. 7A and 7B. All elements of the SMR, including the SMR error vector, the SMR depth error vector, and the SMR runout error vector, are fixed to the SMR coordinate system. The SMR coordinate system can conveniently be located at either the vertex 820 or the spherical center 860.

0038The elements of FIGS. 7A and 7B are shown in perspective views in FIGS. 8A and 8B. Part of the SMR700 shown in Figure 8A includes three flat reflectors 825AB, 825BC, 825AC, three cross-joints 810A, 810B, 810C, apex 820, and axis of symmetry 840. FIG. 8B shows an enlarged view of the region near the center of the SMR, with vertex 820 shown aligned with axis of symmetry 840. The runout plane 865 is perpendicular to the axis of symmetry 840 and passes through the center of the sphere 860. The axis of symmetry intersects the runout plane at SMR intersection 870. The SMR error vector 885 extends from the vertex 820 to the center of the sphere 860. The SMR depth error vector 880 extends from the vertex 820 to the SMR intersection 870. The SMR runout error vector 890 extends from the SMR intersection 870 to the spherical center 860.

0039A reference feature 910 is affixed to the color 905 contained in the SMR portion 900 shown in FIG. 9A, which can be, for example, a serial number or barcode. In this example, the reference mark 930 is customarily selected in the center of the reference feature 910. In another embodiment, the reference mark is line 930 written on color, as in Figure 5A. In another embodiment, the reference mark is affixed directly to the spherical outer portion 720. A reference point 932 is associated with the reference mark 930. The SMR reference plane 920 surrounds the reference point 932 and the axis of symmetry 840.

0040Figure 9B shows an enlarged perspective view near the center of the SMR. Vertex 820 is shown in both Figure 9A and Figure 9B. The SMR reference ray 940 is a ray that coincides with the line of intersection between the SMR reference plane 920 and the SMR runout plane 865, and is the SMR reference plane that begins at the SMR intersection 870 and contains the reference point 932. Guided along half of the 920. The angle between the SMR reference ray 940 and the SMR runout error vector 890 is the SMR runout reference angle 950. The numerical value of the runout reference angle of a particular SMR is a characteristic of that SMR. In one embodiment, the SMR depth error, SMR runout error, and SMR runout reference angle are determined for each SMR by performing measurements as discussed below.

0041FIG. 9C shows the same part 900 of SMR as FIG. 9A. The ray 46 from device 10 intersects vertex 820. The ray depth error vector 962 has a magnitude equal to the SMR depth error 880, from vertex 820 to ray intersection 970, as shown in Figure 9D, an enlarged perspective view near the center of the SMR. It extends along the direction of the ray 46. The ray runout plane 965 contains the ray intersection 970 and is perpendicular to the ray depth error vector 962. The ray reference plane 975 surrounds the reference point 932 and the ray depth error vector 962.

0042The reference ray 972 of a ray is a ray that coincides with the intersection line between the reference plane 975 of the ray and the runout surface 965 of the ray, starting at the intersection 970 of the ray and containing the reference point 932. Guided along half of. The angle between the ray reference ray 972 and the ray runout error vector 976 is the ray runout reference angle 982. Unless the ray 46 is aligned with the axis of symmetry 840, there will be a difference between the calculated 3D coordinate 978 of the sphere center and the actual 3D coordinate 860 of the sphere center. This difference is further discussed with reference to FIGS. 12A-12E.

0043Measurements are performed for each SMR to determine the position of the center of the sphere with respect to the vertices. The results of such measurements can be described in several different ways. One such description of the center of the sphere with respect to the vertices includes an SMR depth error of 880, an SMR runout error of 890, and an SMR runout reference angle of 950. A different but equivalent description includes the length of the component of the SMR error vector, using Cartesian coordinates. For example, the length of such a component can be given along the Cartesian axes x, y, z inside the SMR coordinate system, such as the coordinate system 730 of FIGS. 7A, 7B. The following are some alternative descriptions. Other coordinate systems are possible, as will be apparent to those skilled in the art.

0044The coordinate system 40 shown in FIG. 9B is fixed to the SMR, as shown in FIG. Assuming that vertex 820 is the origin of coordinate system 40, the 3D coordinates of the center of the sphere (C) are (x) in coordinate system 40._C, y_C, z_C), X_CIs the depth error of SMR, y_C, Z_CIs the coordinates of the SMR runout error vector on the SMR runout plane. In this case, the z-axis is taken along the direction of the SMR reference ray 940. Alternative methods<maths num="2"><img id="000004" he="8" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>Some use a coordinate system 988 where the axes are aligned at the position of the maximum runout, that is<maths num="3"><img id="000005" he="9" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>The axes are aligned with the SMR runout error vector 890. in this case,<maths num="4"><img id="000006" he="9" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>Since is zero, the 3D coordinates of the center of the sphere (C) are given by the following equation.<maths num="5"><img id="000007" he="10" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>Coordinate system 988 can be achieved by aligning the reference mark 930 to the position of the maximum runout of the retroreflector.

0045The Cartesian coordinate system can also be applied to the runout plane 965 of a ray. Figure 9D shows the coordinate system 996. Taking vertex 820 as the origin of coordinate system 996, the 3D coordinates of the sphere center 978 calculated based on the direction of the ray (B) are (X)._B, Y_B, Z_B), X_BIs a ray (X_BThe depth error of the SMR along the direction of (axis), Y_B, Z_BIs the coordinates of the runout error vector of the ray on the runout plane of the ray. Ingredient Y_B, Z_BThe size of component Y_C, Z_CIs the same size as, only shifted from surface 865 to surface 965. Alternative methods<maths num="6"><img id="000008" he="8" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>It uses a coordinate system 998 whose axes are aligned with the ray runout error vector 976. in this case,<maths num="7"><img id="000009" he="9" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>Since is zero, the calculated 3D coordinates of the center of the sphere (B) are given by the following equation.<maths num="8"><img id="000010" he="9" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>

0046There are many ways to find the position of the sphere center 860 with respect to the apex 820. One method is to measure SMR using a Cartesian coordinate measuring machine (CMM). Using this method, the position of the vertices with respect to the center of the sphere can be found up to a fraction of a micrometer. For example, with a very good Cartesian CMM, the extended uncertainty of the center's position with respect to the vertices is better than 0.4 micrometers along each of the three Cartesian axes x, y, and z. possible. This error is much smaller than the typical SMR centering error of 0.0005 inches = 12.7 micrometers along the axes x, y, z. For a very good SMR, the component of the specified centering error can be as small as 0.0001 inch = 2.54 micrometers.

0047Another way to measure the depth of the spherical center 860 with respect to the apex 820 is to utilize an absolute interferometer or other type of precision ADM. In one embodiment, a dynamic nest configured to repetitively center the spherically shaped target is arranged to push the SMR upwards. To find the SMR error vector 885, the reference SMR is measured using an accurate Descartes CMM. A reference SMR is placed in the nest and an absolute interferometer is used to measure the distance from the absolute interferometer to the apex of the SMR along a horizontal line. The test SMR is then placed in the nest and the measurements are repeated. The depth error of the reference SMR in the direction from the distance measuring device to the SMR E_DepthRefSMRIs generally taken as a plus. Then, following the procedure above, the depth error of the test SMR is E_DepthTestSMR= d_TestSMR-d_RefSMR+ E_DepthRefSMR (1) And d_TestSMRIs the measured distance of the test SMR, d_RefSMRIs the measured distance of the reference SMR.

0048Another way to measure the SMR runout error and the SMR runout reference angle (or equivalently the SMR runout surface decarte error) is to rotate the SMR under a microscope according to methods well known in the art. There is. A description of this method is given in the Appendix B section of the Performance Evaluation of Laser-Based Sperical Coordinate Measurement Systems, ASME Standard B89.4.19-2006. Shown in B-2.1, which is incorporated herein by reference. Using this method, the test SMR is placed on a dynamic nest resting on a microscope stand. The light source illuminates the frame of the microscope. The focus is adjusted to observe small pieces of dust (or other small items) on the frame of the microscope. The operator rotates the SMR around the center of the sphere in the dynamic nest and observes the radius of the runout circle under a microscope. Divide the observed radius by 4 to get the SMR runout error, which is the magnitude of the SMR runout error vector 890. The procedures discussed in this paragraph are used as a way to determine if an SMR meets its runout (centering) standard. For example, the manufacturer may provide a specification that the centering error of the SMR is less than 0.0005 inches. This would be interpreted to mean that the SMR depth error is less than 0.0005 inches and the SMR runout error is less than 0.0005 inches. So far, depth and runout errors have been measured for SMRs, but to correct the readings of device 10 based on SMR compensation parameters that can contain information on the vector components of SMR depth and runout errors. No preparation has been made so far for the processor to use these values.

0049The observed position of the SMR runout error relative to the reference mark on the SMR can be used to determine the SMR runout reference angle of 950. In one embodiment further discussed with respect to FIGS. 19A, 19B, reference mark 930 is aligned to the position where the largest runout was observed during the test procedure. The effect is that the SMR runout reference angle is zero. In one embodiment, the calibration inspector places the reference mark 930 in the position of maximum runout. In other words, the inspection technician places the reference point 932 within the reference plane 920.

0050In addition to the SMR centering error (the vector component of the SMR depth error and the SMR runout error), there is also the SMR radius error. In other words, the SMR radius is not always specified in the manufacturer's specifications. As an example, some high quality SMRs are produced from grade 25 steel balls with a radial dimensional tolerance of ± 0.0001 inch = ± 2.54 micrometers, which corresponds to a radial dimensional tolerance of ± 1.27 micrometers. In other words, for SMRs produced using this type of steel ball, the actual SMR radius is ± the nominal (specified) radius, at least in the spherical part not very close to the cavity holding the retroreflector. Expected to be in the range of 1.27 micrometers.

0051There are several ways to measure the SMR radius. A Cartesian CMM can be used to accurately measure the radius of the test SMR. Radius error is found by taking the difference between the measured radius and the reference or nominal radius.

0052Absolute interferometers can also be used to measure the radius error of the test SMR. In this method, the reference sphere is measured using a coordinate measuring device such as a Cartesian CMM or a Talyrond® roundness measuring device to find the radius of the reference sphere (sphere). In one embodiment, a dynamic nest configured to repetitively center the spherically shaped target is arranged to push the spherical surface upward. A reference sphere is placed within the nest and the absolute interferometer focuses on the surface of the sphere along a line perpendicular to the surface. The absolute interferometer measures the distance from the interferometer to the surface. Next, the test SMR is placed in the nest. The SMR is rotated so that the light from the absolute interferometer can be in line with the normal vector to the spherical outer portion 720, and the distance is measured using the absolute interferometer. The difference between the measured distance to the test SMR and the measured distance to the test sphere is the radius error.

0053As the temperature of the SMR changes, the position of the apex 820 with respect to the spherical center 860 may also shift. Such changes in SMR temperature include (1) changes in the ambient temperature of the air surrounding the SMR, (2) heating of the SMR by the operator, and (3) in-situ nesting of the laser tracker 10 in Figure 1. It can result from SMR contact with relatively warm objects such as.

0054For some types of SMR, the effect of temperature can be relatively small. For example, as discussed above, some types of SMRs are made by etching three perpendicular surfaces using ECM into an integrally formed steel ball. In this type of SMR, the change in SMR depth error due to temperature is the product of the coefficient of thermal expansion (CTE), the SMR temperature change, and the initial SMR depth error. If the initial SMR depth error is 0.001 inch = 25.4 micrometer and the SMR is made of steel with a coefficient of thermal expansion of 11.5 micrometer / meter / ° C, then at 30 ° C at the SMR temperature. The change in SMR depth error with respect to change is (25.4 × 10)^-6) (11.5) (30) micrometers = 0.009 micrometers, which is a negligible amount.

0055For some other types of SMR, the effect of heat is greater. For example, consider the type of SMR containing aluminum slag to be inserted into the spherical outer portion of steel with a diameter of 1.5 inches (38.1 mm). It is assumed that the aluminum slag extends 10 mm below the apex and sticks to the steel part at that position. Considering the point of contact where the axis of symmetry intersects the outer part of the sphere, ignoring the thermal expansion of the adhesive, the relative changes in vertex position with respect to temperature are the difference in CTE values between aluminum and steel and the 10 mm elongation. , The product of changes in temperature. Steel CTE is 11.5 μm / m / ° C, aluminum CTE is 23 micrometer / m / ° C, depth extension is 10 mm, and temperature change is 30 ° C. Suppose. The change of vertices with respect to the center of the sphere is (0.01) (23-11.5) (30) micrometers = 4.45 micrometers.

0056Temperature sensors such as thermistors or RTDs may be incorporated within the SMR to compensate for the movement of the SMR's vertices as a result of thermal expansion. In one embodiment shown in FIGS. 10A, 10B, 10C, a small connector socket 1030 is located above or within the SMR1000. The connector socket is attached to the temperature sensor 1034 through a set of wires 1032 (typically two or three wires). The temperature sensor 1034 may be a thermistor, RTD, thermocouple, or other device. The sensor cable 1040 includes a second set of wires 1044 with one end of the cable 1040 connected to the first connector 1042 and the other end of the cable connected to the second connector 1046. The first connector 1042 is connected to the connector socket 1030 and the second connector 1046 is a temperature measuring system located on top of a laser tracker 10, a computer 80, an accessory box 70, a thermometer, or other device. It is connected to the. Many types of temperature measurement systems can be used. A simple temperature measurement system (not shown) may include an electronic device based on a Wheatstone bridge with three internal resistors. The first internal resistor may be connected to a voltage source, the second internal resistor may be connected to ground, an external temperature sensor 1034 and cable 1044.The wire gives the bridge a fourth outer resistance leg. Such temperature measurement systems are well known in the art. The temperature measurement system converts the observed voltage into the temperature of the SMR. The coefficients stored in memory can then be used to correct the position of the vertices with respect to the center of the sphere. The memory may be contained within a laser tracker 10, a computer 80, an accessory box 70, or other device. Measurements are performed prior to a particular type of SMR obtaining information that describes the thermal expansion characteristics of the SMR. In a simple example, a single CTE value or a CTE value multiplied by the SMR depth error may be sufficient to explain the movement of the vertices with respect to the center of the sphere. For example, in the example shown above, where the vertices move by 4.45 micrometers with respect to 30 ° C, a single parameter that gives a value of 0.1483 for the CTE depth error multiple may be used, and this value. Has a unit of micrometers / ° C. In another embodiment, a table of values or coefficients is provided.

0057Using the embedded temperature sensor 1034 with the sensor cable 1040 has the advantage of eliminating the need for batteries, temperature circuits, or wireless communication systems within the SMR. In this system, the SMR may be inspected, for example, when the SMR returns to its home position on the tracker.

0058As shown in Figure 10D, one possibility is to connect cable 1040 to the temperature electronic module 1050 via socket 1052. In one embodiment shown in FIG. 10E, the battery 1054 powers the embedded temperature sensor 1034, the temperature processing electronic device 1056, and the wireless communication electronic device 1058 through the wire 1062. In one embodiment, the temperature processing electronic device 1056 includes a Wheatstone bridge, a resistor, and a microprocessor for supplying a digital signal to the wireless communication electronic device 1058 through wire 1064. The digital signal represents the temperature measured by the temperature sensor 1034. In one embodiment, the wireless communication electronics 1058 delivers a digital representation of the measured temperature through the antenna 1059. In one embodiment, the wireless communication electronic device is configured to transmit a digital representation through antenna 1059 at regular intervals, for example every 5 minutes. The interval time may be selected to provide adequate temperature information while conserving battery power. The wireless signal from antenna 1059 may be received by the electronic device in device 10 shown in FIG. 1, and the received temperature value is used to improve the compensation of the 3D coordinates of the spherical center of the SMR. In one embodiment, the temperature electronic module 1050 may be located in the operator's pocket, such as a shirt pocket or trouser pocket.

0059In one embodiment shown in FIG. 10F, the temperature processing electronics 1050 may be attached to the operator's wrist. In one embodiment, the glove 1070 worn by the operator incorporates a temperature processing electronics 1050 and a cable 1040. In one embodiment, the glove is an insulated glove that minimizes heat transfer from the operator's hand to the SMR1000. In another embodiment, the glove is not an insulating glove, but has a finger opening to allow the operator to handle the SMR directly. In another embodiment, the operator does not wear gloves. The temperature processing electronics 1050 are not attached directly to the operator's wrist, but by, for example, a strap or rubber band.

0060FIG. 11 shows the interface element 1120 mounted on the SMR1100. Interface element 1120 may include a plurality of optional elements. Interface element 1120 may be connected to a temperature sensor mounted within SMR1100. Antenna 1130 can be used for transmitting and / or receiving wireless data in the form of radio frequency signals. Such an antenna can be mounted on a small circuit board powered by a small battery 1128 mounted inside interface element 1120. Small circuit boards can be made of rigid flex material that allows very compact circuits to be sealed inside interface elements. By providing a temperature sensor and a battery-powered wireless communication system, the temperature of the SMR can be known at any time, resulting in the most accurate 3D coordinate measurement of the SMR1100.

0061As a way to enable more accurate 3D measurements, the numbers used to compensate for SMR imperfections are called SMR compensation parameters. SMR compensation parameters are generally stored in the memory of the measuring device or computer device. For example, the SMR compensation parameters may be stored in the memory of the laser tracker 10 of FIG. The processor in the laser tracker 10 can be used to correct the 3D coordinates of the spherical center of the SMR26. Alternatively, a processor, accessory device, or networked computer device in the external computer 80 accesses the compensation values stored in memory and uses these values to correct the 3D coordinates of the center of the sphere. Can be used as Such compensation may reflect the difference in the position of the vertices with respect to the center of the sphere. Such compensation may reflect the effects of spherical radius error using the methods described below with respect to FIGS. 14-17.

0062In some cases, the compensation calculation is performed inside the processor of the measuring device such as a laser tracker. For example, a tracker processor identifies an SMR used for a particular measurement, automatically performs a compensation operation on that SMR, and converts the 3D coordinates of the measured vertices to the 3D coordinates of the center of the sphere. In other cases, compensation is provided by the application software. For example, the application software may take into account the position and orientation of the SMR when making multiple measurements and applying compensation operations to eliminate errors in the radius of the SMR. In general, the term processor may be understood to mean the processor of a device such as a tracker, the processor of an external computer, or both processors.

0063There are several ways in which SMR compensation parameters can be entered into memory. In one embodiment, the SMR shipped from the factory with device 10 comes with compensation parameters preloaded in memory. In another embodiment, the user is provided with a list of numerical compensation values for each SMR. The application software embedded in the device 10 provides a means for the user to enter a numerical value. Such a value is stored in the memory in the device 10 and needs to be input only once. In another embodiment, SMR compensation parameters are provided by a flash drive, CD ROM, or other component read by device 10, computer 80, or other component for automatic storage in memory. In another embodiment, the compensation parameter numbers are downloaded to the tracker over the network. In another embodiment, the SMR compensation parameters are encoded into a one-dimensional or two-dimensional barcode 630. The barcode reader may read the SMR compensation values on the barcode and automatically download them into the memory of device 10.

0064In one embodiment, when the device 10 locks (starts tracking) to the SMR in one of the in-position nests 17, the short range camera 56 shown in FIG. 1 is directed directly at the SMR. In device 10 of FIG. 1, the camera 56 is placed above the aperture from which the light beam 46 is emitted. To ensure that the camera can clearly read the SMR barcode, the camera assembly may be directed towards the center of the SMR located in position 17. In other words, for the example shown in FIG. 1, the camera's optic axis may be directed slightly downward rather than parallel to the ray 46. In addition, in one embodiment, the camera 46 is configured to be clearly focused on the SMR in the in-position nest 17. This is generally different from the camera 52, which is usually designed to focus on a retroreflector that is relatively far from the tracker. In one embodiment, the camera 56 has a sufficiently small field of view to provide adequate resolution for barcodes or serial numbers. For example, the field of view of such a camera may be about 20 degrees. A small field of view and clear focus in place can be obtained by accurately selecting the focal length and adjusting the position of the lens in the camera 56. Combining the optimal pointing direction (towards fixed position 17) with the in-focus state at fixed position 17 and a relatively small field of view, a relatively inexpensive camera 56, for example with a one-dimensional or two-dimensional barcode. It is guaranteed that the marks on the possible barcode 610 can be clearly resolved. In addition, any light near the SMR is blocked during the operation of the camera 56 to ensure that the camera's photosensitive array is not blinded by retroreflected light. To prevent this loss of visibility, any light should normally be separated from the edge of the retroreflector by at least one diameter of the retroreflector. In other words, if the retroreflector is an open-air corner-cube retroreflector with a circular cross section and a diameter of 1 inch, any light will be less from the outer edge of the retroreflector. At least one inch should be separated. Since the laser beam 46 from the tracker is collimated and hits the center of the retroreflector, there should be little light scattered back to the camera 56. However, if the laser beam 46 is too bright, the laser beam 46 can be turned off while reading the barcode on the SMR. In one embodiment, the camera 56 is used in combination with Optical Character Recognition (OCR) software to read the serial number.

0065Since the SMR distance is mechanically reset in the home position, the bar code measurement on the SMR can be done in conjunction with the home position measurement without the need for extra steps on the part of the operator. In the exemplary method, the SMR is placed in place. The tracker first turns on the camera, examines the barcode and extracts the SMR parameters. The tracker then turns the ray on and sends it to the apex, measures the distance, and resets the distance to the home reference value. In one embodiment, in order to reset the rangefinder to the home reference value in the most accurate and possible way, the software in the tracker sets the applied distance value to the spherical center of the SMR rather than the apex. This is important because it ensures that the home reference value is applied correctly to all SMRs regardless of the SMR depth error. In order to determine the center of the sphere as accurately as possible, the SMR should be placed in the preferred orientation within the in-situ nest 17, or the camera 56 should automatically orient the SMR in the in-situ nest. Should be used as desired. These methods of establishing the position of the SMR in a fixed nest are discussed in more detail below. In Figure 17, there are three different in-position nests sized to accommodate three separately sized SMRs. In one embodiment, the in-situ nest is sized to accommodate SMRs with diameters of 1.5 inches, 0.875 inches, and 0.5 inches. Separate home reference values are provided for each of these home nests.

0066In one embodiment, the SMR compensation parameters are encoded into RFID tag 650. The encoded information is recovered using an RFID reader, which may be attached to the payload 12 instead of the short range camera 56. The recovered serial number can be used to access SMR data from a database of information stored inside a device such as a network system (cloud) or a tracker or computer.

0067In a typical example, a 3D coordinate measuring device, such as the Laser Tracker 10, obtains the measured 3D coordinates of the vertices of the SMR, for example by measuring the distance to the SMR and the two angles. To obtain the 3D coordinates of the center of the sphere, the SMR compensation factor is applied to the 3D coordinates of the measured vertices using the method described here. For measurement, the operator holds the reference point 932 so that the SMR reference plane 920 is oriented in a preferred orientation. The preferred orientation can be selected in several different ways, depending on the measurement objective.

0068In general, the operator should try to align the axis of symmetry in the direction of the light beam from the device. This is discussed above with reference to FIGS. 12 and 13. As used herein, the term preferred orientation means adjusting the position of a reference point as an additional step in the alignment procedure. A way to align the axes of symmetry with respect to the rays while achieving the preferred orientation is to align the axes of symmetry with respect to the rays and then rotate the SMR around the axes of symmetry. Since the SMR reference ray is perpendicular to the ray and is in the reference plane containing the reference point, the operator can adjust the SMR reference ray by rotating the axis of symmetry to obtain the desired orientation. I can say.

0069There are some special cases where it is not possible to align the axis of symmetry with the ray. For example, with the SMR in place 17, it would be impossible to align the axes of symmetry with respect to the rays due to mechanical constraints. Therefore, in this special case, the meaning of the term preferred orientation is changed to allow the necessary changes in alignment. Special cases for in-position alignment are discussed below.

0070In the first type of preferred orientation, the SMR reference plane 920 is aligned with the x "-z" plane of the payload coordinate system 35 as shown in FIG. The x "axis corresponds to the direction of the ray, and the z" axis is perpendicular to the x "axis and the y" (Zenith) axis. The z vector (in the device coordinate system 30), which is always contained in the x -z plane, points to the top of the device 10. For the device 10 in its normal upright position, the SMR reference plane is in the preferred orientation. Includes gravity vector as well as ray 46. However, when the gravity vector may not be appropriate, for example when the laser beam points straight up with respect to the tracker, the more general x -z plane can be used. In the preferred orientation of the first type, the operator only holds the SMR so that the spherical center of the SMR and the reference mark of the SMR are placed in the plane containing the ray 46 and the axis z'. As discussed above with respect to FIGS. 9C and 9D and below with respect to FIGS. 12 and 13, the operator should also attempt to align the axis of symmetry 840 of the SMR in the direction of the ray 46, preferably the SMR. While held in orientation, the software in the processor uses the 3D coordinates of the measured SMR vertices and the SMR compensation parameters to determine the SMR spherical center, as described above for FIGS. 9A-9D. calculate.

0071In the second type of preferred orientation, the SMR reference plane 920 is aligned with the y "-z" plane of the payload coordinate system 35 as shown in FIG. The y "-z" plane always contains the Zenith axis 18 (y "axis) of device 10. If device 10 is arranged in an upright orientation as shown in FIG. 1, the azimuth axis 20 is vertical. Oriented, the Zenith axis 18 is in a horizontal plane and rotates around the azimuth axis. In the second type of preferred orientation, the operator strikes the SMR spherical center and SMR reference mark, ray 46. And hold the SMR so that it is placed in the plane containing the Zenith axis 18 (x axis). If the reference mark 930 points radially outward on the collar 905 as shown in FIG. 9A, the reference mark should be kept in the horizontal plane for the second preferred orientation. For the second type of preferred orientation, the preferred position of the SMR reference mark 930 with respect to the spherical center 860 must also be given, for example, to the right of the spherical center. As mentioned above, the operator should attempt to align the SMR's axis of symmetry 840 in the direction of ray 46.

0072In the third type of preferred orientation, the SMR reference ray 940 is aligned with the measurement line in a predetermined manner to measure the dimensional characteristics of the measurement line. This is further discussed below with reference to FIGS. 19A and 19B.

0073Next, with respect to FIGS. 12A to 12E, the error related to the misalignment of the axis of symmetry 840 with respect to the light ray from the device 10 will be discussed. The ray 1210 received by the SMR in FIG. 12A has a ray center 1212 and a ray width 1214 and is clipped by the SMR collar 905. The angle 1215 is the light receiving angle of the SMR with respect to the ray 1210. For a 1.5 inch diameter SMR that receives a typical red light beam from the laser tracker 10, the light receiving angle 1215 is typically about 25 degrees.

0074The ray 1240 received by the SMR in Figure 12B coincides with the axis of symmetry 840. For FIG. 12B, the enlarged region 1230 near the apex 820 and the spherical center 860 is shown in enlarged view 1 of FIG. 12C. Since the ray 1240 is aligned with the axis of symmetry 840, the accuracy at which the center of the sphere is determined depends on the measurement error in finding the 3D coordinates of the vertices of the SMR, in other words the distance measured by device 10 and the error of the two angles. Only limited.

0075The SMR 720 in Figure 12D is rotated by an angle of 10 degrees around its spherical center. As a result of this rotation, the vertex 820 shifts by an amount of 1250 to the new vertex position 820R. The axis of symmetry, which is the axis of symmetry for the three cross-joints, also shifts from line 840 to line 840R by 10 degrees. As a result, for example, when the SMR is placed on a dynamic nest, rotating the SMR 720 around its center will apply compensation parameters to the SMR depth error vector and the SMR runout error vector. , Provides a misalignment error vector 1250 for the center position. As shown in FIG. 12E, the misalignment error 1250 includes component 1252 along axis of symmetry 840R and component 1254 on a plane perpendicular to axis 840R of symmetry.

0076During normal operation, the operator can keep the 1.5 inch diameter SMR axis of symmetry 840 aligned and held within a range of typically about 10 degrees with respect to the direction of the rays 1240. As shown in Figure 12D, the misalignment error vector 1250 that results from rotating the SMR around the spherical center 860 by an angle of 10 degrees is relatively small compared to the SMR error vector 885R. .. When the 3D coordinates of the SMR are measured by a laser tracker, ray 1240 is ray 46 as shown in FIG.

0077Next, with reference to FIGS. 13A to 13F, a method of improving the alignment of the axis of symmetry 840 with respect to the ray 46 and thereby minimizing the magnitude of the misalignment error vector 1250 will be described. In the first method shown in FIG. 13A, the alignment cap 1310 is placed on top of the collar 905. In the center of the opaque cover 1314 mounted over the ring 1312 contained in the alignment cap is a marked circle with a diameter approximately identical to the marked crosshair 1317 or the ray 46 from the 3D coordinate measuring device 10. There are 1316. To align the axis of symmetry 840 with respect to the incoming ray 46, the operator first blocks the ray with an alignment cap 1310 to stop the ray from tracking the SMR700. The operator places the alignment cap 1310 on top of the collar 905 and rotates the SMR so that the incident ray 1318 is centered on the marked circle 1316. The operator removes the alignment cap 1310 to lock the rays onto the SMR. At that time, the ray 46 is aligned with the axis of symmetry 840. With this method, alignment accuracy of 2 degrees or better can be expected for 1.5 inch SMR.

0078In the second method shown in FIG. 13B, a window 1320 that allows light to pass through the retroreflector is placed relative to the collar 905 so that the outer marked circle 1324 is aligned with the outer edge of the collar. The operator locates the incident light 1328 scattered from the window 1320 with respect to the crosshair 1317 or the inner marked circle 1326 and rotates the SMR to match the incident light 1328 with the inner marked circle 1326. Let me. At that time, the incident ray 46 is aligned with the axis of symmetry 830.

0079FIG. 13C shows a method of aligning SMRs by observing a portion 1332 of an incident ray with, for example, a reflective strip 1320, which can be an elongated cardboard. By moving the strip with respect to the opening of the SMR, the operator can visually determine the direction in which the SMR should be rotated to improve alignment.

0080Figure 13D is similar to Figure 13C, except that the finger 1340 is used instead of the reflective strip 1330 to observe part 1342 of the incident light. By moving the fingers relative to the opening of the SMR, the operator can visually determine in which direction the SMR should be rotated to improve alignment.

0081In the method shown in FIG. 13E, a reflective cap 1350 is placed on top of the collar 905, which is hidden from the view of FIG. 13E. The mirror 1354 mounted on top of the ring 1352 contained in the reflective cap is designed to receive light 1352 and reflect light 1353 from its front. Optionally, the crosshair 1317 or the marked circle 1356 is centered on the mirror 1354. The diameter of the circle is approximately equal to the diameter of the light beam from the 3D coordinate measuring device. To align the axis of symmetry 840 with respect to the incoming ray 1352, the operator first blocks the ray with a reflective cap 1350 to stop the ray from tracking the SMR700. The operator places a reflective cap on top of the collar 905 and, if circle 1356 is present, rotates the SMR to place the ray approximately in the center of the circle. The operator observes the position of the reflected light on or near the 3D coordinate measuring device and rotates the SMR700 if necessary to further improve the alignment, and the reflected light is emitted from the 3D coordinate measuring machine. Move it closer to the light emitting point. In FIG. 1, the light emitting point of the laser tracker 10 is the point where the light beam 46 departs from the laser tracker 10. When the reflected light beam is close enough to the emission point, the operator removes the mirror cap 1350 to lock the light beam onto the SMR. A typical high quality mirror has a wedge angle of only a few minutes, and the resulting angular shift in reflected light is negligible. By reflecting the light beam so that it hits a coordinate measuring device near the emission point, the axis of symmetry 830 can be aligned within a fraction of a degree with respect to the light beam. For example, if the SMR700 with a reflective cap 1350 is 10 meters from the laser tracker 26 and the reflected laser beam is within about 175 mm of the emission point, or about 7 inches, then the axis of alignment 840 emits. Aligned within 1 degree with respect to the light beam 46.

0082FIG. 13F shows how a 3D coordinate measuring device, such as the Laser Tracker 10, emits a ray of rotating pattern 1360 with a diameter equal to the diameter of the SMR, for example the SMR Color 905 or similar features. The operator aligns the SMR by rotating the SMR with respect to the rotating light beam.

0083Previously, with reference to FIGS. 1 and 6, a method for combining position measurement with bar code reading using the device's camera was taught. Part of this method was to set the rangefinder to the home reference value at the center of the sphere rather than the apex of the SMR used. In order to transform the 3D coordinates of the vertices with respect to the center of the sphere, it may be important to know the orientation of the SMR in place, as explained below with reference to FIG. Next, two methods for establishing the orientation of SMR are shown.

0084As explained above, establishing a preferred orientation has two aspects. Normally, the axes of symmetry of the SMR are aligned as well as possible with respect to the rays. In the second aspect, the SMR is rotated around an axis of symmetry to place the reference mark in the preferred orientation.

0085In some cases, it is not possible to perfectly align the axis of symmetry with respect to the ray. For example, at home position 17, in some types of nesting, there may be mechanical constraints to rotate the SMR to precise alignment with the ray. In place, another type of alignment criterion may be given. For example, the manufacturer may specify that the SMR collar should be placed approximately 2 mm above the nest. The software of device 10 may then calculate the error vector based on the given SMR parameters.

0086If the axis of symmetry is not aligned with the ray, the orientation moves the reference point to the desired orientation (eg to a vertical plane or a plane containing the azimuth axis) and then aligns the axis of symmetry as specified. It is easily obtained by aligning with respect to the ray direction (eg moving the collar 2 mm above the nest). Another possibility is to use the camera 56 to determine the orientation of the SMR. With the camera pointed at a point in place, the SMR orientation is easily found from the position of the barcode (or any other marker). Using the camera in this way gives accurate results and is easy for the operator.

0087This camera concept can also be applied when the SMR is at a distance of many meters from device 10 if a suitable zoom camera is installed inside the camera. The '758 patent discussed above and incorporated by reference describes an example of such a zoom camera. Reflective marks, such as mark 610 in FIG. 6A of the present application, can be used with such cameras to provide automatic determination of SMR orientation. This method is applicable when a 6DOF tracker is used with a 3DOF SMR.

0088SMR can be used to measure the coordinates of a surface point by contacting the surface with the SMR at multiple points. SMR can also be used to measure the coordinates associated with dynamic nesting. In both of these types of measurements, errors in the SMR radius can result in errors in measuring surface or nested points. As described above, a method for accurately measuring the radius of the SMR is obtained. The difference in radius defined as the measured radius minus the nominal radius may be stored in memory for later access. The measured radius in this case is measured by one of the exact methods described above, for example by using a Cartesian CMM, Talyrond®, or an absolute interferometer. The radius difference may also be stored in an information storage device such as a barcode or RFID chip attached to each SMR and read by the reader of the 3D coordinate measuring device in one embodiment.

0089For device 10 used to measure 3D coordinates on a surface using SMR, the SMR is moved to several positions on the surface and a set of 3D coordinates of the SMR's spherical center with respect to the corresponding surface contact point. Is obtained. 3D center coordinates close to each other are used to obtain a vector perpendicular to the set of 3D points. The normal vector is projected from one point in the set of contact points, or from one point near the set but not in the set of points. The 3D coordinates of the contact point corresponding to that one point are obtained by projecting a normal vector from the center coordinates of the 3D sphere towards the surface being measured.

0090The error in the radius of the SMR does not change the overall shape of the plane. However, errors in the SMR radius can lead to errors in other situations. This can be easily understood by considering the case where SMR is used to measure the distance between two planes, one plane on the left side of the SMR and one plane on the right side of the SMR. If the true diameter is larger than the nominal or reference diameter and all other measurements are complete, then the measured distance is twice as small as the radius error of the SMR.

0091An error in the radius of the SMR can also result in measurement errors for supports such as dynamic nesting. An example of the type of dynamic nest 1400 shown in FIG. 14A, in this case, includes three spherical balls 1420, each attached to the top surface 1412 of the base 1410 and separated by 120 degrees from the other balls. Support shaft 1430 is perpendicular to the plane intersecting the centers of the three spheres 1420. Also, the support shaft 1430 is equidistant from the centers of the three spheres 1420 at any point along the support shaft. Figure 14B shows a mechanism 1450 in which the spherical outer portion of the SMR700 is located on the dynamic nest 1400. As used herein, the term "kinematic" means that an SMR can be removed from or returned to a nest after its 3D coordinates move in much the same state as before it moved. As the SMR700 rotates, the center of the sphere remains fixed in place on the support shaft 1430. In some cases, a magnet is included in the base 1410 to hold the spherical outer portion of the SMR steel firmly against the dynamic nest 1400 and prevent it from falling out of the nest.

0092One way to describe the support shaft 1430 in mathematical terms is as a locus at the center of the sphere with respect to spheres of various sizes mounted on the nest. In other words, both a sphere with a relatively small radius and a sphere with a relatively large radius are on the support shaft 1420. Of course, there are some minimum and maximum sphere sizes that fit the three spheres 1420 well, but within the sphere size tolerance, the locus of the center of the sphere will be on the support axis 1430.

0093The SMR is provided by the manufacturer with a nominal radius. The application software performs calculations based on the SMR nominal radius, which is a number provided by the SMR manufacturer. As further discussed below, the difference between the nominal radius value and the actual radius value can cause an error.

0094Nests, such as Nest 14A, generally do not have a well-defined reference point. A convenient and stable reference point for a particular nest and a particular SMR combination is the "ideal support position", which is the position on the support axis 1430. The ideal support position is defined as the position on the support axis 1430 at the center of the sphere with a radius exactly equal to the nominal radius. An SMR with a radius different from the nominal radius will have a spherical center at a different position on the support axis, called the actual support position.

0095Neither the measurement of the radius nor the correction of the 3D coordinates based on the radius error relies on the internal reflection mechanism of the retroreflector, so the method of correcting the radius discussed herein is an open-air corner cube. It is not limited to SMRs having a retroreflector. This method is similarly applicable to SMRs that include a glass corner cube retroreflector, which is a retroreflector that includes a glass prism with three vertical reflecting surfaces. This method is used on cat-eye retroreflectors with glass optics that are molded as either a single sphere or two hemispheres joined together and at least partially located within the cavity of the outer portion of the sphere. Is also applicable. SMRs that include glass prisms or cateyes are well known in the art.

0096As mentioned above, the radius difference is the measured radius minus the nominal radius. The direction of the support axis 1430 from the surface connecting the centers of the three spheres to the center of the sphere held by the nest is positive. In other words, in Figure 14A, the right direction along the support shaft 1430 is considered positive. Therefore, the ideal support position is to measure the 3D coordinates of the spherical center of the SMR in the nest 1400, and move the 3D coordinates in the minus direction along the support axis by a distance equal to the difference in radius, not precisely Can be found by moving. The exact amount of movement also depends on the dimensions and shape of the nest 1400, as described below with reference to FIG. An equation can be used to find the ideal support position as a function of radius error or difference. Compensation values, such as coefficients, may be stored in memory so that they can be used, for example, by a processor in device 10 or a separate computer 80 to determine the 3D coordinates of the ideal support position.

0097Next, for the case of the SMR700 supported by a dynamic nest 1400 with three spheres 1420 separated by 120 degrees, the amount of shift of the spherical center 860 is calculated. FIG. 15 shows the line segments corresponding to the elements of Nest 1400 and SMR700 in FIG. 14B. Each of the three spheres 1420 has a center 1510, and any two centers 1510 are separated by a base distance of 1525 given the mathematical symbol a. A cross section drawn through the center of each sphere shows contour 1515. Each of the spheres contacts the spherical outer portion 720 of the SMR 700 at contact 1520. A line drawn from the sphere center 860 through the contact point 1520 passes through the center 1510 of the sphere 1420. The SMR has a reference or nominal radius R and a radius error e, so that the actual radius of the SMR is R + e. The three lines 1525 form the base of the triangular pyramid 1540. A line 1530 extending from the center of the sphere 1510 to the center of the sphere 860 forms the sides of the triangular pyramid. The radius of each sphere is r, so that the sides of the triangular pyramid have a length R + e + r. A line segment 1550 at the height of a triangular pyramid with a height h extends from the base center 1545 above the plane connecting the three sphere centers 1510 to the spherical center 860. Each of the base vertices 1510 has an angle of 60 degrees, so the length of the line drawn from the sphere center 1510 to the base center 1545 is<maths num="9"><img id="000011" he="9" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>Is. According to the Pythagorean theorem, the height of the line segment of height is<maths num="10"><img id="000012" he="9" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>Is. R the length of the side for radius error e = 0₀And the height is h₀And. Using the binomial approximation, the change in height h as a result of error e is δh = eR₀/ h₀Can be shown with excellent accuracy. Height h₀Is the side length R₀Since it is smaller than, the height change δh is somewhat larger than the radius error e. For example, for a radius error of 1 micrometer, the height change δh is 1.048 micrometer. The reason the height change is greater than the radial error is that the larger SMR700 intersects the sphere 1420 slightly far from the base connecting the sphere center 1510.

0098In general, the height change δh depends on the structure (dimension shape) of the SMR nest 1400 and is different for different types of nests. In order to compensate for the effect of radius error on height changes as accurately as possible, it is necessary to know both the amount of radius error and the type of nesting used with SMR. The radius error e can be used without taking into account the dimensional shape of the nest to obtain an approximate correction for the effect of the radius error on the change in height. For the example shown in the previous paragraph, the relative height error calculated without taking into account the dimensions and shape of the nest 1400 was 4.8%.

0099An example of the effect of SMR radius error on measurements made on SMRs in nests is shown below. FIG. 16A is a front view of the first SMR700A held by the first nest 1400A, and FIG. 16B is a side view of the second SMR700B held by the second nest 1400B. Ideally, the radii of both 700A and 700B are equal to the reference radius, the length between the centers of both SMRs is indicated by arrow 1672 in Figure 16C. Next, consider the case where the spherical radius of the first SMR700A is equal to the reference radius and the spherical radius of the second SMR700B is larger than the reference radius. For the SMR700B, the center of the sphere is pushed upwards, as indicated by arrow 1674. Therefore, the measured 3D coordinates are vector 1676. The length of the vector 1672 is L₁And if the length of vector 1674 is δh, the difference between lengths 1672 and 1676 is<maths num="11"><img id="000013" he="10" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>And when the binary approximation is applied, ΔL = δh²/ 2L₁Excellent accuracy is shown. As an example, δh = 2 micrometers, L₁If = 2 meters, the error is 10^-6It becomes a micrometer and can be ignored.

0100In contrast, in Figures 17A and 17B, the first SMR700A and the first nest 1400A are aligned as shown in Figure 16B, while the second nest 1400B connects the spherical centers of 700A and 700B. It shows the situation where the line is oriented perpendicular to the line. Ideally, the radii of both 700A and 700B are equal to the reference radius, the length between the centers of both SMRs is indicated by arrow 1752 in Figure 17B. Next, consider the case where the spherical radius of the first SMR700A is equal to the reference radius and the spherical radius of the second SMR700B is larger than the reference radius. For the SMR700B, the center of the sphere is pushed to the left, as indicated by arrow 1754. Therefore, the measured 3D coordinates are vector 1756, which is too short by the offset value 1754. As explained above, the offset value 1754 depends on both the difference (ie, the radius error) and the dimensional shape of the nest. In this case, if the spherical radius of 700B is too large by 2 micrometers and the nest dimensions are the same as those shown in the previous example, the measured distance between the two SMRs is 2.096 micrometers smaller. It will be too much. In other words, for the situation in Figures 16A and 16B, the measured length error between the two SMRs is not affected by the diameter error of one of the SMRs, while for the situation in Figures 17A and 17B. , The measured length error between both SMRs has a direct effect due to the diameter error of one of the SMRs.

0101It is clear from the above discussion that in many situations the orientation of the support shaft 1430 of the two nests can have a significant effect on the uncorrected distance measured between the SMRs placed within the nest. is there. There are several ways in which the orientation of the support shaft 1430 can be obtained. First, the operator may indicate the orientation of the support axis in the application software. Second, the nest may be mounted directly on a target with a surface known in the CAD model, from which the orientation of the support shaft 1430 can be determined. Third, measurements may be made for an inspection plan that includes the orientation of the support shaft 1430 for each nest used in the inspection. Fourth, the operator may measure the nested features to orient the support axis 1430. For example, in the case of Nest 1400 in Figure 14A, the operator may use SMR with device 10 to measure the 3D coordinates of a point on top 1412. The 3D coordinates of these points are applied to the plane where the vertical support axis 1430 is found.

0102In general, both the magnitude of the directional and radial differences of the support shaft 1430 are required to correct for errors in measurements made using SMRs placed within the nest. As discussed with reference to FIG. 15, more accurate correction is possible if the dimensions and shape of the nest are also taken into consideration.

0103The procedure described with reference to FIG. 18 is useful whenever the position of the same SMR must be measured in situations where the tracker is being moved to multiple positions. For example, the calibration procedure may need to measure the distance between two SMR700A and SMR700B using trackers set at two different positions 10A, 10B away from SMR, as shown in FIG. .. In other cases, the tracker may need to measure three or more nests from multiple positions as a way to put tracker measurements made from different positions on a common coordinate system.

0104In the first step of the method, two or more SMR700A, 700B are rotated in the support 1400A, 1400B, and the axis of symmetry 840 of each SMR is set to 3D coordinates (x) in the coordinate system 1810.₁, y₁, z₁) To the 3D measuring instrument placed in the first position. In this first position, the tracker is given reference number 10A. This alignment can be achieved using, for example, one of the alignment methods of FIGS. 13A-13F. The tracker at 10A measures the SMR700A, the SMR700B, and at least one additional retroreflector 700C that does not need to be aligned with the tracker 10A. In the second step, 3D coordinates (x₂, y₂, z₂The same 3D measuring instrument is shown as 10B when placed in a second position with). In this step, the SMR700A, 700B are kept in their initial orientation rather than being rotated further. The tracker 10B measures the 3D coordinates of the vertices of 700A, 700B, and 700C. The 3D coordinates of the vertices of SMR700A, 700B measured by the 3D measuring device 10B are mathematically corrected to reflect the misalignment of vertices 820 with respect to each spherical center 860 of SMR700A, 700B.

0105The mathematical method for doing this is easy to understand. Tracker 10B is added to the tracker coordinate system at 10A using the values measured for the apex of the retroreflector at 700A, 700B, 700C. This is a tracker 10B with 6 degrees of freedom (eg x, y, z, pitch, roll, yaw) until the 3D coordinates of 700A, 700B, and 700C measured by trackers 10A and 10B match as precisely as possible. Is adjusted using an optimization technique. The usual optimization method is to minimize the sum of the squares of the residual errors.

0106The transformation matrix required to transform the 3D coordinates measured by 10B into a 10A coordinate system transforms the 700A, 700B SMR error vectors at 10A into the 700A, 700B error vectors at 10B. Can be used for. In this way, the step of rearranging the SMR700A and 700B prior to the measurement by the tracker 10B can be omitted.

0107It should be understood that the method of establishing a transformation matrix by measuring the 3D coordinates of the targets of at least three retroreflectors from at least two positions can be performed without SMR. In other words, a corner cube retroreflector or other type of retroreflector can be attached to any type of target. It should also be understood that the methods described herein can be used to correct the 3D coordinates of a single SMR or multiple SMRs when observed from more than one station.

0108As described above, the preferred type of SMR orientation is such that the SMR reference ray 940 aligns with the measurement line in a predetermined manner as a way to minimize the measurement error of the dimensional characteristics associated with the measurement line. Is to be done. Next, consider the case where the dimensional characteristic is the length between two points.

0109In the measurements shown in FIGS. 19A, 19B, nests 1400A, 1400B have a zero SMR runout reference angle of 950. The Nest 1400A is mounted directly on top of the Nest 1400B and is a 3D coordinate measuring device such as the Tracker 10 for measuring the vertical distance between the SMR700A located within the Nest 1400A and the SMR700B located within the Nest 1400B. Is used. In FIG. 19A, the SMR700A is rotated within the nest 1400A so that its runout reference line 1910 is placed vertically above the SMR700A. The SMR700B is rotated within the Nest 1400B so that its runout reference line 1920 is placed vertically below the SMR700B. The true distance between the spherical center of the SMR700A and the spherical center of the SMR700B is shown by line 1902. The SMR 700A SMR runout error vector is indicated by vector 1904, and the SMR700B SMR runout error vector is indicated by vector 1906. The resulting measurement of the length between the centers of the spheres is removed by an amount equal to the sum of the magnitudes of the vectors 1904 and 1906. These errors appear in the measured length represented by length 1908.

0110In Figure 19B, the SMR700C is identical to the SMR700A, but the SMR700A is in a horizontal position where the runout reference line is perpendicular to the line connecting the center of the SMR in the nest 1400A and the center of the SMR in the nest 1400B. It is rotated to place the runout reference line. The SMR700D is identical to the SMR700B, but the SMR700D is rotated to position the runout reference line horizontally. The true distance between the spherical center of the SMR700C and the spherical center of the SMR700D is shown by line 1952. The SMR 700C SMR runout error vector is indicated by vector 1954, and the SMR700D SMR runout error vector is indicated by vector 1956. The resulting length 1958 has a length close to the true length 1952. The error resulting from the calculation of length 1958 is often known as the negligible cosine error, as shown in the example shown for FIG.

0111The error in the measurements in Figure 19A should be smaller if both runout reference lines for 700A and 700B are aligned up or down, but the result is generally that the runout reference line is for two SMRs. It is much better if it is rotated perpendicular to the line connecting the centers of the sphere. This is true in all directions, not just perpendicular to the line connecting the two SMRs.

0112To simplify the SMR runout alignment, it is convenient for the operator to know the direction of the maximum runout error vector component. This is easily achieved if the reference points are aligned with respect to the maximum runout error vector component, in other words, if the SMR runout reference angle is set to zero. Alternatively, the reference point can be set to another easily understood value, such as 180 degrees or +90 degrees or -90 degrees. This angle is a preferred angle in the sense that each SMR produced by the manufacturer has a reference point at the same position with respect to the maximum runout error vector.

0113This method of aligning reference marks can also be applied to minimize errors in dimensional measurements in addition to length measurements. An example of measuring the size of a small displacement in the direction x'at the position of the first SMR1982,1983. In this case, the orientation of the reference mark 932 shown in FIG. 19C is still accurate. In other words, if you want to sensitively measure small displacements along the direction x'of either the first nest or the second nest, the nest reference marks should be oriented as shown in Figure 19C. Is. On the other hand, if the purpose is to sensitively measure small displacements along the direction z'of either the first nest or the second nest, the optimum position of the reference mark 932 should be different. In this case, the runout error component of the SMR should be perpendicular to the direction of measurement. As shown in Figure 19D, the SMR reference ray 940 should be oriented perpendicular to the ray direction in the x'-y'plane. In this case, there are two possible orientations, on the left side of the ray, such as 1987 or on the right side of the ray, such as 1988. The SMR criteria may be oriented to minimize errors in measurements of other dimensions, as will be apparent to those skilled in the art.

0114FIG. 20A shows a schematic representation 3000 of the azimus axis 20 and the Zenith axis 18 described in FIG. The azimus mechanical shaft (axle) 20 and the zenith mechanical shaft (axle) 23 corresponding to the azimus shaft 20 and the zenith shaft 18, respectively, are also shown. The azimuth mechanical shaft 24 rotates around the azimuth shaft 20 by an angle of 21. The azimus shaft 20 corresponds to the center line 27 of the azimus mechanical shaft 24. The zenith mechanical shaft 23 rotates around the zenith shaft by an angle of 19. A vertical line 28 is drawn between the azimuth axis and the Zenith axis at the closest point of the two axes. The vertical line 28 intersects the center line 27 at point 22. The length 29 of the vertical line between the azimuth axis 20 and the Zenith axis 18 is the axis offset (AXOF) length 29. In the mechanical device of FIG. 1, the intersection 22 is the device of FIG. 1 because the payload 15 rotates around the Zenith mechanical shaft and the Zenith mechanical shaft rotates around the azimuth mechanical shaft attached to the fixed base. It is stationary with respect to the coordinate system 30 of. For this, point 22 may be considered as the gimbal point of device 10, i.e., around point 22, the azimus and zenith mechanical shafts rotate. Note, however, that the Zenith axis 18 does not rotate exactly around the gimbal point 22. However, mathematical compensation can be made so that all rotations are referenced to the gimbal point. After a small compensation for ray offset and ray tilt, ray 46 appears to come out of the point of rotation of the ray on the Zenith axis 18.

0115In summary, it is mathematically convenient to select the gimbal point 22 as the fixed point within device 10 and refer to the 3D coordinates measured by device 10 for this point, which is customary in the art. Is. This means that mathematical compensation is made to reflect the non-ideal aspect of the actual mechanical axis. One such non-ideal aspect is the axis offset 29. Another non-ideal aspect of the mechanical shaft is the non-squareness of the shaft. In an ideal mechanical system, the azimuth axis and the Zenith axis are exactly perpendicular. In a real system, there is a small deviation from the vertical, which is an angle value called the non-squareness value of the axis. Mathematical methods are used to correct the distance and the two angles measured by device 10 to obtain the measured 3D coordinates and refer to these values for gimbal point 22.

0116In a similar manner, in the ideal device 10, the ray 46 substantially passes through the gimbal point 22. In a real system, the rays are initially offset from gimbal point 22 by an axis offset distance of 29. In addition, the ray 46 is offset from the point of rotation of the ray (on the Zenith axis 18). The offset is a small distance value of y "and z" in the y "-z" plane of the payload coordinate system 35 shown in FIG.

0117In an ideal device 10, the direction in which the rays 46 exit the device is perpendicular to both the azimuth axis and the Zenith axis. In the actual device 10, the ray 46 comes out slightly off this ideal angle. Mathematically reflect these effects and refer to all measurements with respect to the origin in order to move the measured distance and the two angles relative to the coordinates with the tracker coordinate system, with the gimbal point as the origin. Parameters can be used to do this. For example, in the type of beam steering mechanism described in device 10 of FIG. 1, the parameters TX, TY may reflect the ray offset (relative to the ideal state) and the parameters RX, RY may reflect the ray tilt (relative to the ideal state). May be reflected. Other beam steering mechanisms, such as mirror-based mechanisms, may have various methods for compensating for mechanical and optical imperfections in the system, but are common as discussed herein. It should be understood that the principle is applicable to all such beam steering mechanisms.

0118It is necessary to give a reference distance to the rangefinder in the device 10. Since the rays from the rangefinder appear to come out of the rotation point of the rays, we need to provide a way for any measured distance to refer to this point. How to do this is discussed in detail below.

0119The normal operating mode of the device 10 is the foreground mode. The device shown in FIG. 20A is in forward vision mode. In the figure shown, the Zenith axis 18 is in front of the azimus axis 20, but it can also be on the opposite side (rear) of the azimus axis. The term foreground mode simply indicates that this is the normal mode of operation defined by the device manufacturer.

0120Another mode of operation of this device is the rear view mode. To start in front-view mode and enter back-view mode, the payload is rotated 180 degrees around the azimus axis and around the Zenith axis until the point where the ray 46 is returned to the retroreflector target. As shown in FIG. 20B, the effect of putting device 10 in rear view mode is to place the Zenith mechanical shaft on the opposite side of the Azimas mechanical shaft. In other words, if the axis offset parameter is positive in the foreground mode, it is negative in the rearview mode, and vice versa. When the apex 820 of the SMR700 is aligned with the closest line 28 as shown in FIGS. 20A and 20B, the axis offset value is the rear-view distance 47'minus the front-view distance 47. Equal to twice.

0121FIG. 20C shows a schematic representation 3040 of device 10 that sends light rays to the SMR 700 located within the fixed position measurement nest 17 of FIG. The ray 46 is emitted from the point of rotation of the ray. In one embodiment, the rangefinder is set to have a zero distance at point 18 in front-view mode. The rangefinder has the SMR700 in place. When placed in nest 17, it reads distance 3042, however, the distance that device 10 reads to gimbal point 22 (after compensation) is the distance 48 from gimbal point 22 to apex 820.

0122Next, another procedure is performed to establish the rotation point of the ray as the zero point of the rangefinder, as described with reference to FIGS. 21A-21C. In the first step shown in the schematic representation 3100 of FIG. 21A, the distance between the center positions of the two SMRs is measured. One way to measure this distance is to align the laser tracker gimbal point 3112 with the line connecting the centers of the two SMRs. The spherical center of the SMR3120 when placed within the nest 3130 can be referred to as the first nest center for this discussion. The spherical center of the SMR3120'when moved to the second nest 3130'can be referred to as the second nest center. The line connecting the first nest center and the second nest center can be referred to as the spherical center line. The portion of the spherical center line between the first spherical center and the second spherical center is referred to as the inner portion of the spherical center line, and the portion other than the inner portion of the spherical center line is the outer portion.

0123The gimbal point 3112 of device 3110 is located at a point on the outer side. The distance 3124 of the ray 3114 from the gimbal point 3112 to the apex 3122 of the SMR in the first nest is measured. SMR is moved to the second nest 3130'. The distance 3126 of the ray 3116 from the gimbal point to the apex 3122'of the SMR in the second nest is measured. The distance 3126 minus the distance 3124 is the distance between vertices 3122 and 3122. However, due to the dimensional shape of Figure 21A, the SMR depth error is a common mode and cancels out for the two length measurements. Therefore, the difference between the distances 3126 and 3124 is also the distance between the spherical center of SMR3120 and the spherical center of SMR3120'.

0124Two additional steps shown in Figures 21B and 21C are performed to set the distance 3042 so that the rangefinder is set to have a distance of zero at the point of rotation of the ray (on the Zenith axis 18). May be done. In the first step shown in schematic representation 3140, the device in front-view mode is arranged such that the gimbal point 3112'of device 3110'is the inner portion. The range finder measures the distance 3142 from the gimbal point 3112'to the apex of the SMR 3120'and the distance 3144 from the gimbal point 3112' to the apex of the SMR 3120'. In the step, device 3110 is placed in rear view mode. Measurements for the two SMRs are repeated to obtain distances 3172 and 3174.

0125If the rangefinder is set to exactly zero with respect to the rotation point of the ray (on the Zenith axis 18), the distance 3126 should be equal to the sum of the distances 3142 and 3144. The formula offset distance = (distance 3126- (distance 3142 + distance 3144)) / 2 is used to calculate the offset distance to compensate for any differences from the ideal state. This offset distance is added to the reading for each distance. Similarly, the offset distance can be considered to be set to set the rotation point of the ray to zero.

0126There are two different types of rangefinders, absolute rangefinders and incremental rangefinders, which have somewhat different reset methods. For incremental rangefinders such as interferometers, the distance is set to a known value at a given point. For example, a laser beam sent to an SMR sent to a in-situ nest could be expected to have a distance reading of 0.167 meters, based on factory or laboratory measurements. For incremental rangefinders, the rays are set to be sent to the SMR in place and read a value of 0.167 meters. The device then counts the number of wavelength shifts in the light wave and multiplies the count by the wavelength of the light of interest (in the local air medium) to obtain a total distance change. This change in distance is added to the original distance to obtain the distance at any later time.

0127In the case of an absolute rangefinder (ADM), the rays may be sent back to the rangefinder, but in this case the distance read by the rangefinder is compared to a known distance, which can also be 0.167 meters. One or more parameters related to the ADM measurement, such as the phase offset parameter, are adjusted to give the expected reading of 0.167 meters. The adjusted parameters are then continued to be used to correct the ADM readings.

0128The method of resetting the rangefinder is somewhat different in the two cases, such as resetting the distance for the incremental rangefinder and resetting the parameters for the ADM, but both prior to making the necessary compensation. It is necessary to send a ray from the device to the apex of the SMR.

0129FIG. 21C is a schematic representation 3170 showing how to obtain the axis offset value 29. The rangefinder measures distances 3172 and 3174 in retrospective mode. The axis offset value is calculated using the formula: axis offset value = ((distance 3142 + distance 3144)-(distance 3172 + distance 3174)) / 4. The axis offset value is used to convert the measured distance and the two angles into the 3D coordinates of the coordinate system 30 centered on the gimbal point 22.

0130This method for setting a zero distance value for the rangefinder in device 10 discussed above showed the distance measured by the rangefinder to vertex 820 of the SMR. However, each SMR has its own depth error, which means that the method in Figure 21 should give different results depending on one or more SMRs used. One way around this problem is to use SMR compensation parameters to measure up to the center of the sphere instead of the vertices of the SMR. Figure 21B shows the case where the axes of symmetry of the SMR 3120'and 3120' can usually be aligned with respect to the rays from device 3110', in which case the effects of runout error are negligible, for example. Assume that the SMR3120'runout error is 12 micrometers in a situation where the distance from gimbal point 3112'is equal to 2 meters. The measured length error is<maths num="12"><img id="000014" he="9" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>Is a negligible value. As long as the axes of symmetry of the SMR 3120'and 3120' can be well aligned with respect to the rays from the device 3110', the measured distances 3142, 3144 can only be increased or decreased to reflect the SMR depth error. By making this correction, the method described above with reference to FIGS. 21A-21C should be accurate regardless of the depth error of the SMR.

0131Several variants are possible in the procedure described with reference to FIGS. 21A-21C. The method of FIG. 21 can be modified by acquiring reference artifacts with a known distance between nests. The distance between the nest 3130 and the sphere placed on the 3130'can be measured using a single Cartesian CMM or interferometer. The distance between nests should not change, especially if the test is performed in a constant temperature environment of about 20 ° C, especially if the artifacts are made of a material with a low coefficient of thermal expansion (CTE). For example, artifacts may be made of carbon fiber composites with low CTE or Invar or Super Invar.

0132In another example, instead of moving one SMR between nests, another SMR may be placed within the nest 3130 and within the nest 3130'. Using two SMRs in this way can save time in automated procedures. In this case, the depth error of each SMR is reflected separately by setting the rangefinder to zero at the rotation point of the light beam.

0133The discussion of Figures 21A-21C was about how to set the rangefinder to read accurate values, or how to apply compensation or correction values to distance readings. An important aspect of these corrections is to reflect the SMR depth error so that the rangefinder compensation is accurate regardless of the SMR depth error. For device 10 with a fixed-position nest 17, the final result of the procedure in FIGS. 21A-21C is a number called the fixed-position reference distance, which is in-position from the rotation point (on the Zenith axis 18). It is defined as the distance to the center of the SMR (of a given diameter) placed within the nest.

0134For devices without in-position nesting, the final result of the procedure in FIGS. 21A-21C is a correction to the rangefinder itself. In other words, the rangefinder's processing is modified following a procedure that equalizes the sum of the distance readings 3142 and 3144 to the difference between the distance readings 3126 and 3124. This reset distance reading can then be used to set the distance to the retroreflector fixed to the device. Such a fixed retroreflector may always be measured as needed to remove drift from the rangefinder.

0135As mentioned in the previous paragraph, for devices with nesting that holds the SMR, the final result of the procedure in FIGS. 21A-21C is a home reference distance, a numerical value. In one embodiment, this number is made more accurate by correcting for the SMR depth error. The reference distance of the fixed position is stored in the memory of the device.

0136In the stylized usage of device 10, the fixed-position reference distance is used to correct the distance readings of the SMRs located in the fixed-position nest 17. Such stylized corrections can be useful in correcting rangefinder drift, which can occur over time as a result of temperature changes and mechanical shocks. As illustrated in FIG. 20C and shown again in FIG. 22, the ray 46 may be sent to vertex 820 of the SMR 700 within the in-position nest 17. If possible, the axis of symmetry of the SMR is aligned with respect to the rays from the device. When this is possible, the rangefinder may set the distance 3042 to the vertex 820 based on the reference distance in place and the depth error of the SMR without taking into account the runout error vector component of the SMR. For example, suppose the SMR's spherical apex 820 is 0.167 meters from the point of rotation of the ray on the Zenith axis 18, and the SMR's spherical radius is 0.75 inches = 19.05 mm. Further assume that the SMR runout error is 0.0005 inches = 12.7 micrometers. The error caused by the SMR runout error at the measurement distance is<maths num="13"><img id="000015" he="9" wi="144" file="JP2017503160A_D0001.tif" img-format="tif" img-content="drawing" /></maths>It is something that can be ignored.

0137In some cases, it may not be possible to align the SMR700's axis of symmetry in the ray direction 46. For example, in FIG. 22, collar 905 cannot contact the nested surface and be aligned with the axis of symmetry. Such contact should be prevented as it can raise the SMR from the nesting point of contact, which causes an error.

0138In this case, it is desirable to reflect the influence of the SMR depth error and the SMR runout error vector component. This can be done in two steps. In the first step, the SMR is placed in the nest with the collars separated from the nesting surface by a certain distance. For example, the collar may be lowered until it touches the surface of the nest, and then the nest may be raised by 2 mm. Assuming that the operator can adjust the collar within 1mm of the target value for an SMR with a radius of 19.05mm, an alignment within about ± 3 degrees is obtained. In the second step, the SMRs are aligned to place the reference point 932 in the specified orientation. For example, it may be a rule that the reference point 932 is set to the highest SMR position. The calculations discussed above for FIGS. 9C, 9D and 12A, 12D can then be performed to correct the distance readings to reflect the vector error. In other words, the home reference distance intended for the value of the SMR spherical center within the home nest 17 sets the distance to the apex of the SMR where the vector error is the actual point measured by device 10. Is adjusted to. An alternative method of performing the second step is to use the camera 56 of FIG. 1 to determine the orientation of the SMR, as discussed above.

0139FIG. 23 shows the internal and external electrical and computer components 2000 of the laser tracker 10 representing the equipment used to measure SMR. It should be understood that these electrical and computer components are merely typical and other configurations are possible. The master processor 2070 sends and receives data messages to and from the processor in the laser tracker. These messages may be sent through the wired device bus, the optical device bus, or the wireless device bus 2030. The process can be performed independently with respect to the function within the laser tracker 10. For example, Position Detector Processor 2012, Azimus Encoder Processor 2014, Zenith Encoder Processor 2016, ADM Processor 2020, Interferometer Processor 2022, Locator Camera Processor 2024, Indicator Light Processor 2018, Temperature Electronics Processor 2025, Azimas (AZ) and Zenith ( ZE) There can be motor processors, as well as RFID and wireless processors 2028. RFID and radio processor 2028 may be connected to antenna 2029 for radiating or receiving radio frequency (RF) signals. As used herein, the term processor refers to computer devices, which may include microprocessors, FPGAs, and DSPs, as well as electronic circuits that perform the functions required to regulate signals sent to computer devices or memory. Is also intended to include. Such electronic circuits may include, for example, an analog-to-digital converter or a temperature indexing electronic device. The master processor 2070 may be sealed in a box such as the interface box 70 of FIG. Alternatively, the master processor 2070 may be incorporated into an electronic device inside the tracker body. The signal from the master processor may go to the external computer 25 or be connected to networks 2044, 2042.

0140The electrical memory components inside the computer component 2000 may be used to store information. Such information may be used by the processor or may be transmitted to the remote processor by wired or wireless means. The information may include serial number and SMR parameters as discussed above.

0141Although the above description refers to a particular embodiment of the invention, it will be appreciated that many modifications can be made without departing from the spirit of the invention. The appended claims are intended to include modifications that fall within the true and spiritual scope of the invention.

0142Therefore, the currently disclosed embodiments should be considered in all respects as exemplary and not limiting, and the scope of the invention is indicated by the appended claims rather than the aforementioned description. And therefore, all modifications that fall within the equivalent meaning and scope of the claims are intended to be incorporated therein.