METHOD FOR TRACKING A RETROREFLECTOR AND DEVICE FOR CARRYING OUT SUCH A METHOD

Description

AREA OF TECHNOLOGY

[0001] The present invention relates to a method for tracking a retroreflector movable in space and a device for carrying out such a method. STATE OF THE ART

[0002] Laser trackers have long been used to determine three-dimensional coordinates in space and are employed, for example, in the measurement of large components or the calibration of machine tools. Such laser trackers typically include an interferometer for highly accurate distance measurement to a retroreflector that is movable in space and attached to the object being measured. The measurement beam emitted by the interferometer's light source is automatically tracked by a tracking actuator to follow the retroreflector, which reflects the light back in the opposite direction of incidence. For automated tracking of the retroreflector, a portion of the reflected measurement beam is coupled out via a beam splitter and directed to a position-sensitive detector.Depending on the position of the incident light beam on the detector, a tracking control signal is generated by a signal processing unit. This signal is used by the tracking actuator to follow the light beam towards the retroreflector. If the light beam is temporarily interrupted between the retroreflector and the detector, a tracking control signal can no longer be generated, and tracking of the retroreflector is no longer possible; this is also referred to as "losing" the retroreflector. Such a beam interruption can be caused, for example, by an obstacle in the beam path of the light beam. After the beam interruption, it is necessary to locate the retroreflector again as quickly as possible for further measurement; that is, the beam interruption should be as short as possible.

[0003] One solution to this problem is given, for example, in EP 2 118 682 B1. This proposes creating a model of the target's trajectory or distance based on ongoing measurements and using this model to estimate the target's current position. If the target is lost during the measurement, the estimated position is used to locate it. A disadvantage of this solution is that, in the event of a potentially prolonged beam interruption, the actual position of the retroreflector can deviate considerably from the estimated position, and therefore, the retroreflector cannot be reliably located again. SUMMARY OF THE INVENTION

[0004] The present invention is based on the objective of providing a method and a device for tracking a retroreflector moving in space, thereby ensuring a rapid and reliable relocation of the retroreflector in the event of a beam interruption.

[0005] This problem is solved according to the invention by a method having the features of claim 1.

[0006] Furthermore, the above-mentioned problem is solved according to the invention by a device having the features of claim 10.

[0007] Advantageous embodiments of the method and device according to the invention result from the measures listed in the respective dependent claims.

[0008] The method according to the invention serves to track a retroreflector movable in space. A light source emits a measuring light beam towards the retroreflector, from which it is reflected back in the opposite direction of incidence; at least part of the reflected measuring light beam falls onto a position-sensitive detector. A tracking control signal is generated by a signal processing unit, depending on the position of the measuring light beam on the detector. This tracking control signal is used by a tracking actuator to follow the measuring light beam to the retroreflector. The signal processing unit continuously monitors whether the measuring light beam can be tracked to the retroreflector by the tracking actuator or whether the retroreflector has been lost.In the event of a detected loss of the retroreflector, a search operation is initiated to estimate the trajectory based on measurements from the tracking actuators. This estimate is used to determine the retroreflector's trajectory for future times. Search areas around the estimated trajectory are then defined for future times, with the search areas varying for different times. A search pattern is then created that is adapted to the estimated trajectory and the defined search areas. The measurement light beam is then moved along a corresponding search path using tracking control signals derived from the estimated trajectory and the defined search pattern. The search operation is terminated as soon as a tracking control signal can again be generated via the position-sensitive detector.

[0009] Preferably, the spatial search areas around the estimated trajectory of the retroreflector are enlarged during the search.

[0010] It is possible that the measuring light beam is repeatedly moved along a spiral search pattern from a center of a search area to a boundary of the search area and back again.

[0011] In this process, a measuring light beam can be moved back and forth along a serpentine search pattern in a predetermined direction between opposite boundaries of a search area.

[0012] Furthermore, it can be provided that the measuring light beam is first moved along the estimated trajectory of the retroreflector at a first speed for a predetermined period of time, and after the predetermined period of time, the measuring light beam is moved oscillating around the estimated trajectory at a second speed that is higher than the first speed, with the oscillation amplitude increasing during the search.

[0013] Preferably, the trajectory estimation is performed using a Kalman filter, over which the actual trajectory of the retroreflector traveled until the loss of the retroreflector is extrapolated.

[0014] In this context, at least angle measurements from the tracking actuator can be used as measured values ​​for trajectory estimation, which are obtained from the rotation of the measuring light beam around two mutually perpendicular rotation axes of the tracking actuator.

[0015] Furthermore, it is possible to determine additional search areas around the estimated trajectory for a multiple of time points during the search using the Kalman filter.

[0016] Furthermore, it may be provided that, for the purpose of determining the state, the intensity of the measuring light beam incident on the detector is determined, compared with at least one predetermined intensity threshold, and the state determination is carried out depending on the comparison result.

[0017] The device according to the invention serves to track a retroreflector movable in space. A provided light source emits a measuring light beam in the direction of the retroreflector, which reflects this back in the opposite direction of incidence. At least a portion of the reflected measuring light beam falls onto a position-sensitive detector.

[0018] A further signal processing device is designed and configured to Depending on the position of the measuring light beam on the detector, a tracking control signal is generated, wherein the tracking control signal can be used by a tracking actuator to track the measuring light beam towards the retroreflector, and a status check is continuously performed to determine whether the measuring light beam can be tracked towards the retroreflector via the tracking actuator or whether the retroreflector has been lost, and in the event of a detected loss of the retroreflector, a trajectory estimation is performed in a search operation based on measured values ​​from the tracking actuator in order to determine an estimated trajectory of the retroreflector for future times, and search areas around the estimated trajectory for future times, whereby the search areas differ for different times, and a search pattern adapted to the estimated trajectory and the determined search areas is created.and then to move the measuring light beam along a corresponding search path using the tracking actuator via tracking control signals, wherein the tracking control signals are derived from the estimated trajectory and the determined search pattern, and wherein the search operation is terminated as soon as a tracking control signal can again be generated via the position-sensitive detector.

[0019] Advantageously, the signal processing device is also designed and configured to increase the spatial search areas around the estimated trajectory during the course of the search.

[0020] It is possible that the tracking actuator includes two drives that enable rotation of the measuring light beam around two mutually orthogonal rotation axes, and that each rotation axis is assigned an angle measuring device whose angle measurements are used by the signal processing device for trajectory estimation.

[0021] Furthermore, it may be provided that the position-sensitive detector is designed as a 4-quadrant diode or as a position-sensitive diode.

[0022] Furthermore, the device according to the invention can have an optical assembly comprising at least two beam splitters, a reference reflector, a distance measurement / detection unit, and the position-sensitive detector, wherein a light beam emitted by the light source can be supplied to the optical assembly, so that The supplied light beam in the optical unit strikes a first beam splitter, which splits the incident light beam into the measurement light beam and a reference light beam, and propagates the measurement light beam towards the retroreflector and the reference light beam towards the reference reflector, and the reference light beam reflected back from the reference reflector strikes the first beam splitter again, and the measurement light beam reflected back from the retroreflector strikes a second beam splitter, via which part of the measurement light beam is coupled to the position-sensitive detector and the remaining part of the reflected measurement light beam strikes the first beam splitter again, where it is superimposed with the reference light beam, and the superimposed measurement and reference light beams are propagated towards the distance measurement detection unit.

[0023] A key advantage of the measures according to the invention is that they ensure the retroreflector can be reliably located even after a potentially prolonged beam interruption. This is achieved primarily by taking into account the increasing uncertainty in estimating the retroreflector trajectory during the beam interruption. In this way, the probability of quickly locating the retroreflector is significantly increased. No additional components are generally required for this; furthermore, the respective measurement application does not need to be completely interrupted and potentially reset.

[0024] Further details and advantages of the present invention will be explained with reference to the following description of exemplary embodiments of the device according to the invention and of the method according to the invention in conjunction with the figures. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] It shows Figure 1a is a partial representation of an embodiment of the device according to the invention; Figure 1 is a highly schematic block diagram with various functional components of the device according to the invention; Figure 2 is a schematic representation of a possible beam interruption; Figure 3 is a schematic representation to explain the method according to the invention; Figures 4a-4c each show a first example of a search pattern during the retroreflector search; Figure 5 shows a representation of the path of the measuring light beam resulting from the superposition of the search pattern from the Figures 4a - 4cwith an estimated trajectory of the retroreflector; Figures 6a - 6d each show a second example of a search pattern during the retroreflector search; Figure 7 shows a representation to illustrate a further variant for searching the retroreflector; Figure 8 shows a flowchart to explain the basic procedure within the framework of the method according to the invention. DESCRIPTION OF THE EXECUTION FORMS

[0026] An embodiment of the device according to the invention is described below with reference to the Figure 1a and 1b explained; Figure 1a This shows part of the corresponding device. Figure 1b A highly schematic block diagram showing a number of functional components of this device.

[0027] The device shown in the figures serves to determine the spatial position of a retroreflector 40 that is movable in space and tracked by this device. For this purpose, several distance measurements are taken between a reference point (not shown) and the retroreflector; from these measurements, the spatial position of the retroreflector 40 can then be determined using known methods. The retroreflector 40 is, for example, arranged on a movable component of a machine tool, which is to be calibrated using the device according to the invention. In the present embodiment, an interferometric measuring method is used for the distance measurement.

[0028] With the aid of a tracking actuator 30, a measuring light beam LM emitted by a light source 11 via an optical assembly 10 is tracked towards the retroreflector 40 and aligned as centrally as possible on the retroreflector 40. For this purpose, the tracking actuator 30 in the present embodiment comprises two drives 31.1, 31.2, which enable the measuring light beam LM to be pivoted about two mutually perpendicular rotation axes A1, A2; these are also referred to as the azimuth axis (A1) and elevation axis (A2) of the corresponding device. The tracking actuator 30 also includes two angle measuring devices 32.1, 32.2, which are assigned to the two drives 31.1, 31.2 and by means of which the rotational movements of the measuring light beam LM about the two rotation axes A1, A2 can be measured. The two angle measuring devices 32.1, 32.2 thus generate as measured values ​​MW each angle measurement values ​​that characterize the rotation of the measuring light beam LM around the two rotation axes A1, A2.

[0029] The interferometer used for distance measurement comprises, on the one hand, a measuring arm extending along the tracked measuring light beam LM between a first beam splitter 14 and the movable retroreflector 40. On the other hand, a reference arm of the interferometer is extended between the first beam splitter 14 and a stationary reference reflector 15 in the optical assembly 10 along a reference light beam LR. At the first beam splitter 14, the light beam L emitted by the light source 11, e.g., a suitable laser, and supplied via an optical waveguide 12 and a fiber collimator 13, is split into a measuring light beam LM and a reference light beam LR. The corresponding light beams LM and LR are reflected back from both the movable retroreflector 40 in the measuring arm and the stationary reference reflector 15 in the reference arm, opposite to the direction of incidence.In this process, the measuring beam LM in the optical unit 10 first passes through a second beam splitter 16, via which a portion of the measuring beam LM' is coupled out to a position-sensitive detector 17. The remaining, or non-coupled, portion of the measuring beam LM, as well as the reflected reference beam LR, then strikes the first beam splitter 14, where the two beams LM and LR are superimposed. From the first beam splitter 14, an interfering pair of beams LM and LR propagates towards a distance measurement / detection unit 18. Distance information with respect to the retroreflector 40 can be obtained from the detected interference signal in a known manner. The interferometric distance measurement method used in the illustrated device is not essential to the invention, i.e.,Various known relative or absolute distance measurement methods can be used for this purpose.

[0030] In the device according to the invention, for example triple mirrors, glass triple prisms or spheres with a refractive index n = 2 serve as the retroreflector 40. The stationary reference reflector 15 can also be designed identically.

[0031] To track the retroreflector 40 which is movable in space, a part of the measuring light beam LM reflected back by the retroreflector 40 is coupled out at the second beam splitter 16 and falls according to Figure 1b as a measuring light beam LM ' onto the receiving surface of the downstream position-sensitive detector 17; this can be designed, for example, as a four-quadrant diode or as a position-sensitive diode (PSD).

[0032] The position-sensitive detector 17 is connected to a signal processing unit 20, which generates a tracking control signal RS depending on the position of the measuring light beam LM' on the detector 17. The tracking control signal RS is used by the tracking actuator 30 to track the measuring light beam LM to the retroreflector 40, which is movable in space. For this purpose, a signal acquisition unit 21, to which position signals PS generated by the detector 17 are transmitted, checks whether the coupled measuring light beam LM' is incident on the center of the detector 17 or, if applicable, offset from it. If an offset is present, the offset-dependent tracking control signal RS for the tracking actuator 30 is generated via a control unit 23 in order to control the drives 31.1, 31.2 accordingly and to track the measuring light beam LM to the retroreflector 40, so that the coupled measuring light beam LM ' falls as close as possible back onto the center of the detector 17.For the sake of simplicity, the following description refers to a tracking control signal RS. In practice, the control unit 23 naturally generates separate tracking control signals for the two drives 31.1, 31.2 of the tracking actuator 30, which are used to pivot the measuring light beam LM to track the retroreflector 40 in space; that is, the term tracking control signal RS can of course also be understood to encompass multiple such signals that can be supplied to the tracking actuator 30.

[0033] During measurement operation, the signal processing unit 20 and its associated status monitoring unit 22 continuously check whether the measuring light beam LM of the measuring arm can be tracked to the retroreflector 40 via the tracking actuator 30, or whether the retroreflector 40 has been lost. The latter can occur if—as mentioned earlier—there is a beam interruption, meaning the line of sight between the optical unit 10 and the retroreflector 40 is interrupted for an extended period, for example, by an obstacle in the beam path, while the retroreflector 40 continues to move. In this embodiment, the status monitoring is carried out by continuously determining the intensity of the measuring light beam LM' incident on the detector 17 and comparing it to a predefined intensity threshold.Depending on the comparison result, the status is determined via the status detection unit 22. If the intensity is below the intensity threshold, the retroreflector 40 is considered lost due to a beam interruption, since no tracking control signal RS can then be generated based on a determined offset of the measuring light beam LM incident on the detector 17. Subsequently, a search for the retroreflector 40 must be initiated in a search mode, in the manner according to the invention, which will be described in detail later.

[0034] Furthermore, in the device or method according to the invention, the determination of the condition regarding a possible loss of the retroreflector 40 does not necessarily have to be carried out using the position-sensitive detector 17; alternatively, the intensity of the signal generated by the distance measurement detection unit 18 could be used by the condition detection unit 22 and compared with a suitable intensity threshold.

[0035] If the described procedure reveals that the retroreflector 40 has been lost, a search operation is then initiated to track the measuring light beam LM or to locate the retroreflector 40 using trajectory estimation. This means that the positions of the retroreflector during the beam interruption are estimated, and these positions along an estimated trajectory are used for the necessary search for the retroreflector 40. For this purpose, the trajectory estimation unit 24 uses measurement values ​​MW from the tracking actuator 30, in particular angle measurements from the angle measuring devices 32.1, 32.2, which characterize the rotational movements of the measuring light beam LM around the two rotation axes A1, A2. The trajectory estimation uses corresponding measurement values ​​MW from a period during which tracking the measuring light beam LM in the manner described above was still possible during measurement operation.This means that measured values ​​MW from the angle measuring devices 32.1 and 32.2 of the tracking actuator 30 are used, which were generated during measurement operation until the beam interruption occurred or until the retroreflector 40 was lost. Details of the trajectory estimation will be explained in more detail later in the following description.

[0036] From the trajectory estimate, the control unit 23 generates tracking control signals RS for the tracking actuator 30 during search mode to move the measuring light beam LM within a specific spatial search area Si (i = 1, 2, ...) along the estimated trajectory Ttest of the retroreflector 40 during beam interruption. The search area Si (i = 1, 2, ...) changes during the search, as will be explained below; that is, different search areas Si (i = 1, 2, ...) exist for different time points ti (i = 1, 2, ...). Thus, during the search, the measuring light beam LM does not follow the estimated trajectory Ttest exactly, but rather searches a larger spatial area within a predetermined search area Si (i = 1, 2, ...) around the estimated trajectory Ttest.During the search for the lost retroreflector 40, the direction of the measuring light beam LM emitted by the optical unit 10 deliberately deviates from the estimated trajectory T test. The search area S i (i = 1, 2, ...) swept by the measuring light beam LM corresponds to a specific uncertainty range around the estimated trajectory T test of the retroreflector 40, within which the lost retroreflector 40 could be located. This uncertainty range, or search area S i (i = 1, 2, ...), is determined, like the estimated trajectory T test, using the trajectory estimation unit 24 for a plurality of time points ti (i = 1, 2, ...) during the search. Typically, such an uncertainty range often has the geometric shape of an ellipse, which is why it is referred to as an uncertainty ellipse in the following.

[0037] Within the uncertainty or search range Si (i = 1, 2, ...), the measuring light beam LM moves along a defined search pattern SM around the estimated trajectory Ttest during the search. It is also possible to search a corresponding search range Si (i = 1, 2, ...) multiple times; this approach can increase the probability of locating the retroreflector as quickly as possible. For suitable search routines and search patterns SM, which are generated by a search pattern generation unit 25, please refer to the following description of exemplary implementations.

[0038] In particular, it proves advantageous within the scope of the present invention if the spatial search range Si (i = 1, 2, ...) is increased by the estimated trajectory Ttest of the retroreflector 40 during the search. This takes into account that the retroreflector 40 can increasingly deviate from the estimated trajectory Ttest during the search operation. The increase in the spatial search range Si (i = 1, 2, ...) is based on the consideration that, initially, the movement of the retroreflector 40 can only deviate from the estimated trajectory Ttest to a small extent, which cannot be arbitrarily large. However, the longer the beam interruption lasts, the more the position of the retroreflector 40 can deviate from the estimated trajectory Ttest due to changes in speed and acceleration.

[0039] Based on the Figures 2 and 3The aforementioned considerations will be clearly explained below. Thus, in Figure 2 The diagram illustrates how, at time t0, an obstacle H moves into the line of sight between the optical unit 10 and the retroreflector (not shown) at location P0 (x0, y0, z0). Up to this point, the retroreflector is moving along the actual trajectory T during measurement operation. From time t0 onwards, no tracking control signal can be generated using the detector's position signals; that is, the measurement light beam LM can no longer be tracked by the retroreflector, and the retroreflector has been lost. An estimated trajectory Ttest of the retroreflector is then determined via the trajectory estimation unit, as explained above. Figure 2The figure is shown with dashed lines. As shown in the figure, the actual trajectory, or the current trajectory Tis, of the retroreflector deviates further and further from the estimated trajectory Ttest over time. When the line of sight between the optical assembly 10 and the retroreflector becomes clear again at time t1, the deviation between the estimated trajectory Ttest and the actual position of the retroreflector on the current trajectory Tis can already be considerable, namely, in particular, larger than the so-called capture area FB of the device according to the invention or of the position-sensitive detector. The capture area FB is in Figure 2At time t1, the area around the measuring light beam LM' is defined by the circular disk-shaped region and describes the region within which a deviation of the retroreflector position from the actual trajectory Tist or the estimated trajectory Test can be compensated for, and tracking by the measuring light beam LM' based on the tracking control signals generated by the detector is still possible. At time t1, the retroreflector is therefore located at position P1ist on the actual trajectory Tist and not, as estimated, at position P1est on the estimated trajectory Test. If the measuring light beam LM' were to... Figure 2 If the search operation follows the estimated trajectory T exactly, then in the event of a longer beam interruption the lost retroreflector could not be reliably found again.

[0040] Therefore, according to the present invention, the measuring light beam is moved within a specific spatial search area around the estimated trajectory of the retroreflector during search operation. The corresponding search area is selected to maximize the probability of finding the retroreflector again. Preferably, the search area is further enlarged during the search. This increases the probability of quickly locating the retroreflector and continuing the measurement operation. The corresponding procedure is described in Figure 3This is illustrated in a highly schematic way. Again, at time t0, the retroreflector is lost, having moved along the actual trajectory Tist up to that point. The estimated trajectory Ttest for the subsequent search for the retroreflector is determined via trajectory estimation. For the following times t1, t2, t3 during the search, a spatial search area S1, S2, S3 is also shown in the figure in the form of an uncertainty ellipse. As can be seen, the search area S1, S2, S3 is enlarged during the course of the search; that is, the search area S2 at time t2, or the corresponding uncertainty ellipse, is larger than the search area S1 at the previous time t1, and so on. The reason for this is the fact that the motion of the retroreflector at or after time t0 of the retroreflector loss may be subject to a change in velocity, acceleration, or jerk.A uniform further estimation of the trajectory would not capture these changes and therefore would not reflect the actual trajectory.

[0041] The following outlines a possible approach to trajectory estimation in the event of retroreflector loss. After determining the current state, the actual trajectory of the retroreflector is extrapolated into an estimated trajectory. This trajectory estimation is based at least on the measured values ​​(MW) from the tracking actuators' angle sensors, which record rotations around the axes of rotation—that is, on the corresponding angle measurements. However, it is also possible to use additional information from the period before the beam interruption for the trajectory estimation, such as determined distances to the retroreflector, velocity, acceleration, and jerk information along the axes of rotation, as well as in the measured length, etc.

[0042] Preferably, in the present invention, the trajectory estimation or extrapolation of the actual trajectory is performed using a so-called Kalman filter. A Kalman filter is an iterative state estimator with which the position, location, velocity, and acceleration of the retroreflector can be estimated. The corresponding estimates of position, velocity, and acceleration are continuously generated from the measured values ​​produced by the angle measuring devices during the measurement operation of the device according to the invention. If, at a specific time, a beam interruption occurs, resulting in a loss of the retroreflector, the valid estimates of position available at that time are used to determine the position. ψ̂ , Speed ψ ˙ ^ and acceleration ψ ¨ ^ the extrapolation ψ τ = ψ ^ + ψ ˙ ^ ⋅ τ + 1 2 ψ ¨ ^ ⋅ τ 2

[0043] In Eq. 1, it states τ represents the time elapsed since the beam interruption occurred.

[0044] The following is an example of the possible design of a Kalman filter for the rotation or azimuth axis, which estimates the azimuth angle, azimuth angular velocity and azimuth angular acceleration from measurements of an angle measuring device.

[0045] The Kalman filter estimates a state vector, which is chosen as follows. ψ ψ ˙ ψ ¨ with: ψ := Azimuth angle ψ̇ := Azimuthal angular velocity ψ̈ := Azimuthal angle acceleration.

[0046] The process dynamics of the Kalman filter, i.e., the time behavior of the state vector, can then be modeled as follows: d dt ψ ψ ˙ ψ ¨ = 0 1 0 0 0 1 0 0 0 ψ ψ ˙ ψ ¨ + w

[0047] The measurement equation for the angle measuring device on the azimuth axis is, in this model, y = 1 0 0 ψ ψ ˙ ψ ¨ + v

[0048] In equations 2) and 3), the quantities w and v Noise terms are represented.

[0049] Regarding the equations resulting from this model, which describe how, for example, the generated angle measurements can be used to estimate the state vector, reference should be made to the relevant specialist literature, e.g. the textbook, "Optimal State Estimation - Kalman, H∞, and Nonlinear Approaches", Dan Simon, John Wiley & Sons Inc., 2006.

[0050] In addition to the state vector or estimated trajectory, consisting of position, velocity, acceleration, and optionally other estimated parameters, the Kalman filter also estimates the uncertainty, i.e., the covariance matrix, of its state estimate. The Kalman filter thus also serves to quantify the uncertainty in the trajectory estimate and thereby indicate what deviations of the trajectory estimate from the true location of the retroreflector are to be expected. In particular, the Kalman filter can also be used to quantify the temporal evolution of these uncertainties. It can therefore be used to specify, at any given time, a region around the estimated position of the retroreflector in which it is highly likely to be located. Figure 3As previously explained, the uncertainty regions are shown in the form of three uncertainty ellipses around the estimated trajectory Test at the three time points t1, t2, and t3. During the search operation, these uncertainty regions function as spatial search regions Si (i = 1, 2, ...) within which the lost retroreflector is searched along the estimated trajectory Test. The Kalman filter thus also serves to determine the search regions Si (i = 1, 2, ...) during the search operation.

[0051] The trajectory estimated by the Kalman filter, along with the time-dependent uncertainties determined from it, forms the basis for tracking the measuring light beam in search mode, from the point of beam interruption until the retroreflector is located again. The search mode's intended scanning of a defined spatial search area along the estimated trajectory depends on the uncertainty estimated by the Kalman filter, or rather, on the uncertainty ellipse valid at the respective time.

[0052] The following will be based on the Figures 4a - 7 Several examples are described that illustrate how, during search operation, the measuring light beam can be appropriately moved in space within the spatial search areas during a beam interruption.

[0053] As soon as the loss of the retroreflector and thus the occurrence of a beam interruption is detected, as explained above, the emitted measurement light beam is to be moved along a search path in search mode by the tracking actuators. This involves appropriate rotational movements around the azimuth and elevation axes along a search pattern within a spatial search area, following the estimated trajectory. The purpose of tracking this search path is to ensure that, after the line of sight is cleared, the measurement light beam re-engages with the retroreflector as quickly as possible, so that tracking in the manner described above is once again possible.

[0054] The search patterns described below, which are generated by the search pattern generation unit, merely represent the deviation of the measurement light beam's search path from the estimated trajectory within the respective spatial search area. The final overall movement of the measurement light beam, or search path, results from the superposition of the estimated trajectory and the search patterns described below.

[0055] One first possibility is with the help of the Figures 4a - 4c and 5 explained. Here, the Figures 4a - 4c Each search pattern SM is executed at different times t1 < t2 < t3 during the search operation. Analogous to the example in Figure 3The search area Si (i = 1, 2, 3) is designed to increase over time. This means that the uncertainty ellipse corresponding to search area S2 at time t2 is larger than the uncertainty ellipse, or search area S1, at time t1. Search area S3 at the later time t3 is even larger than search area S2. The respective search pattern SM is spiral-shaped; that is, the measuring light beam repeatedly moves along a spiral path from a center of search area Si (i = 1, 2, 3) to a boundary of search area Si (i = 1, 2, 3) and back again. In the specific example shown, the measuring light beam travels according to... Figure 4aFirst, the movement spirals outwards from the center until, at time t1, it reaches the boundary of the search area S1 valid at that time. The direction of movement then reverses, and the search pattern SM is traversed in the opposite direction back to the center, which at time t2 is defined according to... Figure 4b is achieved. At time t 2, the situation is as follows: Figure 4b An enlarged search area S2 or a correspondingly enlarged uncertainty ellipse is already evident. The measuring light beam then moves spirally outwards along the search pattern SM until time t3, at which time... Figure 4cthe boundary of the then valid search area S 3 is reached, and where the direction of movement changes again, etc. In this example, it is therefore intended that the measuring light beam is repeatedly moved along a spiral search pattern SM from a center of the search area S i (i = 1, 2, ...) to a boundary of the search area S i (i = 1, 2, ...) and back again.

[0056] In Figure 5The superimposed movement or search path B of the measuring light beam, resulting from the estimated trajectory Test and the spiral search pattern SM, is shown in the figure; the estimated trajectory Test is also shown as a dashed line. This figure clearly illustrates how, over time, the resulting search path B of the measuring light beam moves further and further away from the estimated trajectory Test, thus increasing the spatial search area Si (i = 1, 2, ...) by the estimated trajectory Test. Also shown in Figure 5 the capture area FB around the search path B, which ultimately indicates the area that is searched for the lost retroreflector using such a search path B.

[0057] Another example of a suitable search pattern SM is shown using the following: Figures 6a - 6dThe following is explained. Analogous to the previous example, the respective search area S1-S4, or the corresponding uncertainty ellipse, is shown at the various times t1 < t2 < t3 < t4 during the search operation. The search pattern SM within the search areas S1-S4 is serpentine. The measuring light beam moves within the search areas S1-S4 along a horizontal direction between the left and right boundaries of the search areas S1-S4. As can be seen from the figures, the boundaries of the search areas S1-S4 form an envelope for the serpentine search pattern SM. Along the vertical direction shown in the figures, the measuring light beam exhibits an oscillatory movement, while along the perpendicular horizontal direction, the movement is nearly uniform. The reversal points in the search pattern SM result from the boundaries of the search areas S1-S4 changing over time.In this example, the measuring light beam is therefore oscillated along a serpentine search pattern SM in a predetermined direction between opposite boundaries of the search area S i , (i = 1, 2, ...).

[0058] Figure 7 Finally, a third possible procedure for a suitable search operation according to the present invention is shown when the retroreflector has been lost. The figure illustrates this analogously to… Figure 5 already the search path B of the measuring light beam during the search operation, which results from the superposition of a search pattern and the estimated trajectory Test.

[0059] Until time t0 of beam interruption, the retroreflector moves along the actual trajectory Tis. After it has been determined, as explained above, that the retroreflector is lost, the measuring light beam is initially moved along the estimated trajectory Ttest for a predetermined period from time t0 to time t1 at a first velocity. This ensures that the searching measuring light beam is located behind the actual position of the lost retroreflector. The speed of movement of the measuring light beam during this period is chosen to be significantly lower than the speed of movement of the measuring light beam in the previous measurement operation. The duration of this movement of the measuring light beam is chosen to be approximately on the order of the estimated duration of the beam interruption.From time t1, the speed of the measuring light beam along the estimated trajectory T is significantly increased to a second speed. Simultaneously, from this time t1, the measuring light beam is moved in an oscillating motion around the estimated trajectory T, with the oscillation amplitude increasing during the search, as can be seen in the figure. Therefore, an increase in the spatial search area around the estimated trajectory T is also provided for in this search mode.

[0060] During the search operation, the signal processing unit continuously checks whether the retroreflector is within the detection range and whether a tracking control signal can be generated based on a measurement light beam registered again on the position-sensitive detector. As soon as this is possible, the tracking of the light beam is no longer based on the tracking control signal generated via trajectory estimation, but rather on the tracking control signal that, as explained above, is generated from the position of the measurement light beam on the position-sensitive detector.

[0061] The key steps of the inventive method are summarized once again using the flowchart in Figure 8 explained.

[0062] In process step S10, the retroreflector is tracked by the measuring light beam in an undisturbed manner during the measurement operation. A portion of the measuring light beam reflected back from the retroreflector reaches the position-sensitive detector, and the signal processing unit generates the tracking control signals for the tracking actuator based on the detector's position signals. The tracking actuator then follows the moving retroreflector.

[0063] According to procedure step S20, the status check continuously verifies whether the retroreflector can be tracked via the corresponding tracking control signals generated from the position signals of the position-sensitive detector, or whether the retroreflector has been lost, for example, due to a prolonged interruption of the measurement light beam. As already explained, this can be done, for example, by comparing the intensity determined by the detector with a suitable intensity threshold.

[0064] If a corresponding loss of the retroreflector is detected, an estimated trajectory of the retroreflector for future times ti (i = 1, 2, ..) is created from measured values ​​of the tracking actuators in accordance with procedure step S30 using the trajectory estimation unit.

[0065] From the trajectory estimation, uncertainty or search ranges around the estimated trajectory for future times ti (i = 1, 2, ..) are further determined via the signal processing unit according to procedure step S40.

[0066] In process step S50, a suitable search pattern is finally created via the search pattern generation unit, adapted to the estimated trajectory and the specified search area.

[0067] The measuring light beam is then moved along a corresponding search path using the tracking actuator via the tracking control signals derived from the determined search pattern and the estimated trajectory (procedure step S60).

[0068] During the search operation, procedure step S70 continuously checks whether a tracking control signal can again be generated via the position-sensitive detector, i.e., whether the retroreflector is once again within the detector's capture range. If this is the case, the search operation is terminated, and retroreflector tracking, and thus regular measurement operation, can be resumed according to procedure step S10; otherwise, the measurement light beam continues to be moved along the defined search path during the search operation according to procedure step S60.

[0069] In addition to the exemplary embodiments and variants of the device or method according to the invention described above, there are of course further possibilities for its design.

[0070] As already indicated above, it would be possible to use alternative distance measurement methods to the interferometric method described above.

[0071] Furthermore, there are alternative methods for estimating the trajectory of the measurement light beam for search operation. For example, instead of a Kalman filter, trajectory extrapolation using a so-called polynomial fit would also be possible.

[0072] Furthermore, the light source, designed as a semiconductor laser or laser diode, could also be arranged in the optical assembly, etc.

Claims

1. Method for tracking a retroreflector (40) that is movable through space, wherein - a light source (11) emits a measuring light beam (LM) in the direction of the retroreflector (40), off which said light beam is retroreflected oppositely to the direction of incidence and at least some of the retroreflected measuring light beam (LM') is incident on a position-sensitive detector (17), and - a tracking control signal (RS) is generated by way of a signal processing device (20) on the basis of the position of the measuring light beam (LM') on the detector (17), wherein the tracking control signal (RS) is used by a tracking actuator system (30) to track the retroreflector (40) with the measuring light beam (LM), and - a state is continuously determined by way of the signal processing device (20), to the effect of whether the measuring light beam (LM) can track the retroreflector (40) by way of the tracking actuator system (30) or whether the retroreflector (40) has been lost, and - wherein, should a loss of the retroreflector (40) be determined, a trajectory is estimated in a search operation on the basis of measured values (MW) from the tracking actuator system (30), in order to determine an estimated trajectory (Test) of the retroreflector (40) for future times (ti, i = 1, 2...), and - search regions (Si, i = 1, 2..) around the estimated trajectory (Test) are determined for future times ti, (i = 1, 2, ..), with the search regions (Si, i = 1, 2..) differing for different times ti, (i = 1, 2, ..), and - a search pattern (SM) matched to the estimated trajectory and the determined search regions (Si, i = 1, 2..) is created, and - the measuring light beam (LM) is then moved along a corresponding search path (B) with the aid of the tracking actuator system (30) by way of tracking control signals (RS), wherein the tracking control signals (RS) are derived from the estimated trajectory (Test) and the determined search pattern (SM), and wherein - the search operation is terminated as soon as a tracking control signal (RS) can be generated by way of the position-sensitive detector (17) again.

2. Method according to Claim 1, wherein the spatial search regions (Si, i = 1, 2, ..) around the estimated trajectory (Test) of the retroreflector (40) are enlarged over the course of the search.

3. Method according to Claim 1 or 2, wherein the measuring light beam (LM) is repeatedly moved along a spiral search pattern (SM) from a centre of a search region (Si, i = 1, 2, ..) to a boundary of the search region (Si, i = 1, 2, ..) and back again.

4. Method according to Claim 1 or 2, wherein the measuring light beam (LM) is moved along a serpentine search pattern (SM) in a manner oscillating back and forth between opposing boundaries of a search region (Si, i = 1, 2, ..) about a predetermined direction.

5. Method according to Claim 1 or 2, wherein the measuring light beam (LM) is initially moved at a first speed along the estimated trajectory (Test) of the retroreflector (40) over a predetermined period of time, and, after the predetermined period of time, the measuring light beam (LM) is moved at a second speed, which is higher than the first speed, in a manner oscillating around the estimated trajectory (Test), with the oscillation amplitude increasing over the course of the search.

6. Method according to at least one of the preceding claims, wherein the trajectory is estimated with the aid of a Kalman filter, by means of which the actual trajectory (Tist) of the retroreflector (40) traversed until the loss of the retroreflector (40) is extrapolated.

7. Method according to Claim 6, wherein at least measured angle values of the tracking actuator system (30), which are obtained from the rotation of the measuring light beam (LM) about two rotational axes (A1, A2) of the tracking actuator system (30) that are oriented perpendicular to each other, are used as measured values (MW) for the trajectory estimation.

8. Method according to Claim 6, wherein further search regions (Si, i = 1, 2..) around the estimated trajectory (Test) are determined by way of the Kalman filter for a plurality of times (ti, i = 1, 2, ..) over the course of the search.

9. Method according to at least one of the preceding claims, wherein, in order to determine the state, the intensity of the measuring light beam (LM) incident on the detector (17) is determined and compared with at least one predetermined intensity threshold value, and the state is determined on the basis of the comparison result.

10. Apparatus for tracking a retroreflector (40) that is movable through space, having - a light source (11) which emits a measuring light beam (LM) in the direction of the retroreflector (40), which retroreflects said light beam oppositely to the direction of incidence, and - a position-sensitive detector (17) on which at least some of the retroreflected measuring light beam (LM') is incident, and - a signal processing device (20) which is designed and configured to - generate a tracking control signal (RS) on the basis of the position of the measuring light beam (LM') on the detector (17), wherein the tracking control signal (RS) can be used by a tracking actuator system (30) to track the retroreflector (40) with the measuring light beam (LM), and - continuously determine a state, to the effect of whether the measuring light beam (LM) can track the retroreflector (40) by way of the tracking actuator system (30) or whether the retroreflector (40) has been lost, and - should a loss of the retroreflector (40) be determined, estimate a trajectory in a search operation on the basis of measured values (MW) from the tracking actuator system (30), in order to determine an estimated trajectory (Test) of the retroreflector (40) for future times (ti, i = 1, 2...), and - determine search regions (Si, i = 1, 2..) around the estimated trajectory (Test) for future times ti (i = 1, 2, ..), with the search regions (Si, i = 1, 2..) differing for different times ti (i = 1, 2,..), and - create a search pattern (SM) matched to the estimated trajectory and the determined search regions (Si, i = 1, 2..), and - then move the measuring light beam (LM) along a corresponding search path (B) with the aid of the tracking actuator system (30) by way of tracking control signals (RS), wherein the tracking control signals (RS) are derived from the estimated trajectory (Test) and the determined search pattern (SM), and wherein - the search operation is terminated as soon as a tracking control signal (RS) can be generated by way of the position-sensitive detector (17) again.

11. Apparatus according to Claim 10, wherein the signal processing device (20) is designed and configured to enlarge the spatial search regions (Si, i = 1, 2, ..) around the estimated trajectory (Test) over the course of the search.

12. Apparatus according to Claim 10 or 11, wherein the tracking actuator system (30) comprises two drives (31.1, 31.2) which allow a rotation of the measuring light beam (LM) about two rotation axes (A1, A2) which are orthogonal to each other, and each rotation axis (A1, A2) is assigned an angle measuring device (32.1, 32.2), the measured angle values of which are used by the signal processing device (20) to estimate the trajectory.

13. Apparatus according to Claim 10, wherein the position-sensitive detector (17) is in the form of a 4-quadrant diode or a position-sensitive diode.

14. Apparatus according to Claim 10, having an optical assembly (10) comprising at least two beam splitters (14, 16), a reference reflector (15), a distance measurement detection unit (18) and the position-sensitive detector (17), wherein a light beam (L) emitted by the light source (11) can be supplied to the optical assembly (10) such that - the supplied light beam (L) is incident on a first beam splitter (14) in the optical assembly (10), said first beam splitter splitting the incident light beam (L) into the measuring light beam (LM) and a reference light beam (LR), and - the measuring light beam (LM) propagates in the direction of the retroreflector (40), and the reference light beam (LR) propagates in the direction of the reference reflector (15), and - the reference light beam (LR) retroreflected off the reference reflector (15) is incident on the first beam splitter (14) again, and - the measuring light beam (LM) retroreflected off the retroreflector (40) is incident on a second beam splitter (16) by means of which some of the measuring light beam (LM') is output coupled to the position-sensitive detector (17), and the remaining portion of the retroreflected measuring light beam (LM) is incident on the first beam splitter (14) again, where the reference light beam (LR) is superimposed, and - the superimposed measuring and reference light beams (LM, LR) propagate in the direction of the distance measurement detection unit (18).