DE102010039947A1 - Methods and devices for position determination - Google Patents

Abstract

Methods and devices for determining a position and / or orientation of an object (11) are described. For this purpose, values of a plurality of measurement variables are determined at different times and values of the plurality of measurement variables are estimated at a common determination time based on the determination of the plurality of measurement variables. The position and / or orientation of the object (11) are then calculated based on the estimated values.

Description

The present invention relates to methods and apparatus for determining a position and / or an orientation of an object in a spatial area. More particularly, the present invention relates to methods and apparatus for determining orientation and position of a multi-axis kinematics section.

A multi-axis kinematics is understood to mean a device in which movements can be realized by a plurality of axes coupled to one another. Examples of such multi-axis kinematics are robot arms, wherein at the ends of such robot arms or at other points of the robot arm encoders, tools and the like may be attached.

For measuring objects with such encoders or for editing objects with such tools, it is necessary depending on the encoder or tool to know the position and orientation of the portion of the multi-axis kinematics, on which the encoder or the tool is mounted to know exactly.

In such systems, it is desirable to determine the position in spaces of a few meters in length with an accuracy in the micrometer range. This poses a high technical challenge, especially when positions are to be determined at a high rate and short signal processing time in order to enable real-time position determination. A correspondingly accurate determination would also be desirable for orientation.

For location determination, in particular laser path length measuring devices are known in which a distance of a point whose position is to be determined from one or more fixed reference points determined by transit time measurement of light beams.

For example, is from the DE 101 18 392 A1 a system and method for determining a position of two objects to each other. The method uses the coherence properties of laser beams for distance detection, in which a plurality of light beams are coherently superimposed.

In K. Minoshima and H. Matsumoto, "High accuracy measurement of 240 m distance in an optical tunnel by use of a compact femtosecond laser", Applied Optics, Vol. 39, No. 30, pages 5512-5517 (2000) a distance measurement using frequency combs is described.

From the DE 10 2004 021892 A1 For example, a device is known in which, to determine a position and an orientation, six different retroreflectors are successively measured by means of a so-called laser tracker. However, such a system is relatively expensive metrology. Moreover, in such a device it may be difficult to measure the position and orientation of an object in motion at a sufficient rate.

It is therefore an object of the present invention to provide methods and apparatuses with which the position and / or orientation of an object can be determined at a sufficient rate, even with moving objects.

This object is achieved by a method according to claim 1 and an apparatus according to claim 10. The dependent claims define further embodiments.

In a method according to the invention for determining a position and / or orientation of an object, wherein a plurality of measured variables (ie at least two measured variables) depends on the position and / or orientation of the object, the plurality of measured variables are respectively determined at a plurality of times and Values of the plurality of measured variables are estimated at a common determination time on the basis of the values of the measured variable at the different times. A position and / or orientation of the object is then calculated based on the estimated values of the plurality of measured quantities at the common determination time.

By estimating the values, for example by interpolation, it is not necessary for all or part of the measured variables to be determined, in particular measured, at the time of determination. This reduces the requirement for the test setup.

In addition, lower requirements for the synchronization of the measurement data recording, since the sampling of different parameters can also be asynchronous and only the sampling must be known.

The different times may differ from each other, at least for some measures. For example, the measures may be grouped into at least two groups, and the times differ from group to group. For example, at least six measured variables are necessary for a determination of a three-dimensional position and a three-dimensional orientation, which can be determined for example in two groups of three measured variables. In In other words, it is not necessary in the method according to the invention to determine all measured variables simultaneously. As a result, a measurement is made possible, for example, with a reduced number of measuring devices, since, for example, two measured quantities can be determined in succession with one measuring device.

The measured variables may in particular comprise distances of the object to predetermined reference points or angles between the object and connecting lines to predetermined reference points and be measured optically, in particular by means of laser beams.

The measured variables may have a bandwidth limitation, in particular a low-pass characteristic. This is typically the case, for example, for measured variables which depend on a trajectory of an object moved by a robot. In this case, the measured quantities can be sampled at a rate greater than twice the bandwidth, which according to the sampling theorem enables a complete reconstruction of the measured quantities over time.

Corresponding measuring devices are also provided.

The invention will be explained in more detail with reference to the accompanying drawings with reference to embodiments. Show it:

1 a schematic diagram of an embodiment of a device according to the invention,

2 3 is a diagram for illustrating some embodiments of the present invention;

3 a flowchart of an embodiment of a device according to the invention, and

4 Graphs illustrating illustrative examples of measurements and position values in embodiments of the present invention.

In the following, embodiments of the present invention will be explained in more detail. In this case, features of various embodiments can be combined with each other, unless otherwise specified. On the other hand, a description of an embodiment having a plurality of features is not to be construed as requiring all of these features to implement the invention. In particular, other embodiments may have only a portion of the features illustrated.

In 1 an embodiment of a device according to the invention is shown. The illustrated device comprises a robot arm 10 , which is an example of multi-axis kinematics. At one end of the robot arm 10 is a meter or tool 11 attached, whose position and / or orientation is to be determined. Here, a total of six independent measured values are needed to determine position and orientation in a three-dimensional space. If only the position is to be determined, three independent measured values are needed.

For position and / or orientation determination is at the end of the robot arm 10 a measuring device 12 appropriate. The measuring device 12 is designed, distances and / or angles to fixed reference points 13 . 14 . 15 to eat. The measured values are then used by an evaluation unit 16 determines the position and / or orientation.

The measurement can be carried out, for example, by optical means, in particular by laser path length measurement. For this purpose, a running time of a laser beam from the measuring device 12 to the reference points 13 . 14 . 15 , which in this case may for example have retroreflectors, are measured. It should be noted that in another embodiment, a plurality of stationary measuring devices can be used, which is the distance to at least one point at the end of the robot arm 10 For example, at least one at the end of the robot arm 10 attached retroreflector, measure.

Distances and angles to the three reference points can, for example, be used to determine the six measured variables necessary for a complete position and orientation determination 13 . 14 . 15 be measured. In other embodiments, more than three reference points may be present. In yet other embodiments, the measuring device 12 three separate measuring devices include, for example, each measuring device distances to two of the reference points 13 . 14 . 15 measures.

In such an exemplary embodiment, for example, always three measured variables can be measured in parallel (one per measuring device) before a determination of the next three measured variables takes place. A rate of the measurements may be chosen such that it is greater than twice the bandwidth of the measured variables, which according to the sampling theorem allows a complete reconstruction of the time profile of the measured variables. In particular, measured variables which relate to robot arms such as the robot arm 10 of the 1 are measured, typically a low pass characteristic and are accordingly limited in bandwidth. The cutoff frequencies of such a low-pass characteristic are typically in the range of a few hertz to kilohertz. The measured values thus obtained for the measured variables can then be interpolated and / or extrapolated in order to obtain a complete set of six values of the measured variables for an arbitrary point in time, at which position and / or orientation.

It should be noted that the representation of the robot arm 10 by way of example only, and the methods and apparatus of the invention may be used generally to determine position and orientation.

In 2 a diagram is shown, which the above-mentioned embodiment, wherein a plurality of measuring devices in the measuring device 12 are present, explained in more detail.

In 2 is with 20 a measuring device, which comprises three measuring devices T ₁ , T ₂ and T ₃ . For example, each measuring device T ₁ , T ₂ and T _{3 may} comprise a laser or another suitable light source as well as a detector for detecting backscattered laser light. In this case, the laser beams emitted by the individual measuring devices T ₁ , T ₂ and T ₃ can differ, for example with regard to their wavelength or a modulation used, in order to be able to separate the light beams during the detection.

The measuring devices T ₁ , T ₂ , T ₃ can furthermore comprise movable mirrors, for example micromechanical mirrors, in order to direct the light beams in a desired direction, in particular to retroreflectors R ₁ , R ₂ and R ₃ , which form stationary reference points and generally with 21 are designated.

For determining the orientation and position of the measuring device 20 or an associated object are in the embodiment of the 3 Distances between the measuring devices T ₁ , T ₂ , T ₃ and the retroreflectors R ₁ , R ₂ , R ₃ , measured. For example, as by lines between the measuring device 20 and the retroreflectors 21 indicated the distances between the measuring device T ₁ and the retroreflectors R ₂ and R ₃ , the distances between the measuring device T ₂ and the retroreflectors R ₁ and R ₃ and the distances from the measuring device T ₃ to the retroreflectors R ₁ and R _{2 are} measured to obtain a total of six measured values from which the position and / or orientation can then be determined. However, other combinations of measuring devices and retroreflectors are possible. In addition, it is also possible to determine more than six measured values in order to be able to obtain an overdetermined system of equations and to be able to reduce measurement inaccuracies.

To target the individual retroreflectors, the measuring devices T ₁ -T ₃ as mentioned, for example, have movable mirrors such as micromechanical mirror. A used laser beam can be widened to form a cone of light, so that the position of the retroreflectors only about (depending on an angle which is illuminated) must be known.

In the illustrated embodiment, therefore, the distance to two different retroreflectors is measured by each of the measuring devices T ₁ , T ₂ , T ₃ . This can be done sequentially alternating. For example, at a first time the distances from T ₁ to R ₃ , from T ₂ to R ₁ and from T ₃ to R ₂ can be measured, and at a subsequent time the distances from T ₁ to R ₂ , from T ₂ to R ₃ and from T ₃ to R ₁ , etc. In order to then determine the position and orientation at a desired determination period, interpolation or extrapolation may be applied to estimate values for all six measurands, that is, every six distances in the case of 2 to receive at the time of determination. The determination time can coincide with one of the measurement times, but this need not be the case. The determination time can be determined, for example, by a trigger signal. At the end of the robot arm z. B. a tactile switching button be mounted, which sends a trigger signal when touching an object. In this way, the estimated coordinate at the time of determination corresponds exactly to the coordinate on the object. Alternatively, an optical sensor (eg, a laser line scanner) may emit trigger signals defining multiple determination times during a motion. Typical laser line scanners take z. B. 30 readings per second. In each recording, a trigger signal is sent in one embodiment, so that each measurement of the optical sensor with a position determined by the measuring device and / or orientation, and thus z. B. a coordinate can be linked to a workpiece. If the time profile of the signals to be measured-as is typically the case-is bandwidth-limited, in particular the rate of the respective measurements can be greater than twice the bandwidth (ie the distance of two measurements of the same measured variable is less than 1 / (2 × f _max ), where f _{max is} the bandwidth of the measured variable). As a result, according to the sampling theorem, the estimation by interpolation is substantially accurate.

An embodiment of a corresponding method, which, for example, in the devices of the 1 or 2 , but also in Other measuring devices can be implemented in 3 shown. To explain individual process steps is also on 4 Reference is made, which shows simplified examples of measured values of measured variables for illustration.

In step 30 Measured variables are determined at different times. In this case, different measured variables can be determined at least partially at different times.

This is exemplified in diagrams (a) and (b) of FIG 4 shown. In this case, six measured variables are determined, which are designated m1-m6.

In diagram (a) of 4 In this case, the measured values for the measured quantities m1, m2 and m3 are plotted over the time t, while in the graph (b) measured values for the measured quantities m4, m5 and m6 are plotted over the time t.

The measured quantities m1, m2 and m3 are determined at times t1, t3, t5, t7... While the measured variables m4, m5 and m6 are determined at times t2, t4, t6, t8. The measured variables m1-m6 can, for example, six distances from three measuring devices to three retroreflectors as with reference to 2 be explained, wherein between the times z. B. mirror can be switched in the measuring devices to measure another reflector.

In the example of 4 Thus, three measured quantities are always determined simultaneously, whereby two groups of three of measured variables are determined alternately. However, it is also possible a different division with other group sizes, the group size, for example, corresponds to a number of available measuring devices. For example, only a single measuring device can be provided, which - for example, by means of a movable mirror - alternately anmisst different reflectors.

In 3 then be in step 31 Values of the measured quantities are estimated at a common determination time, for example by interpolation. This is also in 4 shown schematically.

In the diagrams (a) and (b) show curves 40 - 45 an interpolation or extrapolation (in the respective dashed area) based on the measured values for the measured variables m1-m6, wherein curve 40 the measured variable m1, curve 41 the measured variable m2, curve 42 the measurand m3, curve 43 the measurand m4, curve 44 the measured variable m5 and curve 45 is associated with the measured variable m6.

In diagram (c) two determination times te1 and te2 are shown. In order to estimate the measured quantities at the time te1, the interpolated values of the measured quantities m1-m6 are determined according to the curves 40 - 45 As indicated by a dashed line, and for the determination time te2, which is outside the times t1-t10, to which was measured in the illustrated simplified example, corresponding extrapolated values are used. The interpolation / extrapolation can be done for example by means of splines or polynomial interpolation. In another embodiment, instead of an interpolation or extrapolation of measured values, an averaging over a plurality of measured values can also take place. For example, in the environment of the determination time, a model function can be adapted to a plurality of measured values and thereby the measured value can be estimated at the time of determination. Due to the averaging, the uncertainty of the estimated measured variable can be substantially smaller than that of the individual measurement. In the example of 4 The determination times te1, te2 differ from the measurement times t1-t10. In other embodiments, the determination times may also coincide with one of the measurement times. In this case, interpolation / extrapolation is only necessary for those measured quantities which are not measured at the time of determination. If, for example, te1 were equal to t3, the measured values m1-m3 could be used directly, while interpolated values are used correspondingly for the measured variables m4-m6.

It should also be noted that such an interpolation can also be used to estimate measured quantities at a time of determination if all measured variables are to be measured simultaneously, but the determination is to take place at a time not equal to a measuring instant.

In 3 will then be in step 32 the position and / or orientation of an object at the time of determination based on the in step 31 estimated measured variables. This is also exemplary in 4 , shown in diagram (c) in this case. In the example shown, two position variables, for example two coordinates t1 and t2, are determined at the times te1 and te2. In embodiments in which a complete position and orientation in three-dimensional space are required, correspondingly more position variables, for example three coordinates for the position and three angles for the orientation, would be determined. As by lines 46 and 47 indicated on the basis of the determined position variables then a course of the respective position size - for example, again by interpolation - be determined.

It should be noted that the measured quantities and values of the 4 merely serve as an illustrative example and are not to be regarded as actual readings.

It should be noted that while in the 2 and 4 in each case three measured variables are determined in parallel, the determination of the measured variables for different measuring devices can also be asynchronous, so that, for example, the measured variables are each measured at different times.

In some embodiments, the described interpolation and / or extrapolation can be improved by using additional information describing a relative change of the measured quantities. For example, in a device such as the device 1 additionally one or more initial sensors may be provided, from which an acceleration and thus a relative change of the coordinates can be derived. For example, in other embodiments, setpoints of control of the device such as the robotic arm of the 1 be used.

While in the example of the 2 six lengths are determined, in other embodiments, other measures, eg. B. angle determined. For example, three lengths and three angles, a length and five angles or other combinations of lengths and angles can also be determined. In particular, overdetermination is also possible, ie more independent measured variables are detected than are necessary for determining position and position. In this case, the accuracy of the extracted position and position information can be improved by compensation calculation. If only one position is determined, it may also be sufficient to perform three independent length measurements with a measuring device.

As is apparent from the above descriptions of variants and modifications, the invention is not limited to the illustrated embodiments.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• DE 10118392 A1 [0006]
• DE 102004021892 A1 [0008]

Cited non-patent literature

• H. Matsumoto, "High accuracy measurement of 240 m distance in an optical tunnel by use of a compact femtosecond laser", Applied Optics, Vol. 39, No. 30, pages 5512-5517 (2000) [0007]

Claims (17)

Method for determining a position and / or orientation of an object, wherein the position and / or orientation of the object ( 11 ) a plurality of measured quantities (m1-m6), comprising: determining values of the plurality of measured quantities (m1-m6) at a plurality of times (t1-t10), estimating values of the plurality of measured quantities (m1-m6) at least one common determination time (te1, te2) on the basis of the determined values, and calculating the position and / or orientation of the object in dependence on the estimated values of the plurality of measured variables (m1-m6) at the common determination time (te1, te2). The method of claim 1, wherein the plurality of times (t1-t10) for at least a part of the measured variables differs from measured variable to measured variable. The method of claim 2, wherein the measures are divided into at least two groups, wherein the plurality of times (t1-t10) differs from group to group. The method of any of claims 1-3, wherein the estimating comprises interpolation or extrapolation. Method according to one of claims 1-4, wherein the estimation comprises a parameter adaptation of a model function to values of the measured variables determined in the environment of the determination times. Method according to one of claims 1-5, wherein the measured variables include angles between the object and fixed reference points and / or distances between the object and fixed reference points. The method of claim 1, wherein a rate of determining the values of the plurality of measurands corresponding to the inverse of the distance of adjacent ones of the different times is greater than a double bandwidth of the respective measurand. Method according to one of claims 1-7, wherein the estimation takes place taking into account additional information about the plurality of measured variables. The method of claim 8, wherein the additional information includes information about a relative change of the measured quantities. The method of claim 1-9, wherein the determination time is defined by a trigger signal of a sensor. The method of claim 10, wherein the calculated position and / or orientation is linked to the measured value of the sensor to a coordinate on a workpiece. The method of claim 10 or 11, wherein the sensor comprises a tactile feeler and / or an optical sensor. Measuring device ( 12 - 16 ; 20 . 21 ) for determining a position and / or orientation of an object ( 11 ), wherein the position and / or orientation of the object ( 11 ) a plurality of measured quantities (m1-m6), the measuring device ( 12 - 16 ; 20 . 21 ) is configured to determine values of the plurality of measured variables (m1-m6) in each case at different times, wherein the measuring device ( 12 - 15 ; 20 . 21 ) an evaluation unit ( 16 ) which is configured to estimate values of the plurality of measured quantities (m1-m6) at least one common determination time (te1, te2) based on the determined values of the plurality of measured variables at the respective different times and position and / or orientation of the object ( 11 ) based on the estimated values. Apparatus according to claim 13, wherein the measuring device comprises a plurality of objects ( 11 ) to be attached measuring devices (T ₁ -T ₃ ), wherein the measuring means are adapted to emit a light beam and to detect a reflection of the light beam, and a plurality of stationary to be arranged reflectors (R ₁ -R ₃ ). Apparatus according to claim 14, wherein the measuring means comprise steering means for selectively directing the light beam towards one of the reflectors (R ₁ -R ₃ ) to measure a distance to the respective selected reflector. The apparatus of claim 13, wherein the measuring device comprises a plurality of stationary to be mounted measuring devices, wherein the measuring means are each adapted to emit a light beam and to detect a reflection of the light beam, and one or more to be mounted on the object reflectors. Device according to one of claims 13-16, wherein the device is arranged for carrying out the method according to one of claims 1-12.

Images