Method and arrangement for detecting an object using a movable sensor

Abstract

Method for detecting an object using a movable sensor (215), comprising: Moving the sensor (215) relative to the object and repeatedly determining an instantaneous position of the sensor (215) by a position measuring system, wherein the determined position is output to an evaluation device (220) at a predetermined or determinable time, wherein the sensor (215) repeatedly detects the object during the duration of a detection time interval (18), which is an exposure time interval, and outputs a sensor measurement signal to the evaluation device (220) according to the information acquired during the detection time interval (18), at a predetermined or determinable time, whereby a distance value is determined on the basis of the sensor measurement signal in each case, where the duration of the exposure time interval is continuously adjusted according to the reflectance of the object, and wherein for each of the recording time intervals (18) the following is done: Determining a recording time that lies within the recording time interval (18); Determining position values ​​that were issued at times that include the recording time in between; Determining a principal position value based on the determined position values, which approximates the position assumed by the sensor (215) at the time of detection, and Calculating the positional main value and the distance value to determine a coordinate value of the object.

Description

[0001] The invention relates to a method and an arrangement for detecting an object by means of a movable optical sensor or by means of a movable sensor of another type, which repeatedly detects the object during a detection time interval. In particular, the invention relates to the preferably non-contact detection of an object, e.g., with a coordinate measuring machine or an industrial robot on which a sensor, in particular an optical, capacitive, or inductive sensor, is attached in order to determine the coordinates and / or dimensions of the object.

[0002] It is known to use coordinate measuring machines (hereinafter referred to as CMMs) to measure the coordinates of an object. The object can be, for example, a workpiece, an industrial product, and / or an assembly. The object is scanned by at least one sensor of the CMM, either by tactile probing with a probe and / or non-contact. Non-contact sensors include optical sensors. Optical sensors are known, for example, in the form of laser triangulation sensors or cameras. Another type of optical sensor is confocal white light sensors. Their use as sensors in coordinate measuring machines is known, for example, from DE 103 40 803 A1.

[0003] A coordinate measuring machine is understood to be a device that can measure the coordinates of an object, or more precisely, an object's surface, using at least one sensor. The present invention relates specifically to coordinate measuring machines that can measure the coordinates of surfaces and / or material interfaces of objects. The present invention further specifically relates to coordinate measuring machines that allow relative movement of the sensor and the object. However, the invention also extends to other motion devices, for example, in the form of an industrial robot, for moving the sensor relative to the object. Furthermore, the invention also includes solutions in which a sensor is moved manually relative to an object, and the sensor's position can be detected, for example, by means of camera monitoring or so-called camera tracking.

[0004] One way to move the sensor and the object relative to each other is offered by CMMs with one or more sensors that can be moved relative to a stationary base.

[0005] Examples include coordinate measuring machines in portal or gantry designs. The object to be measured is typically placed directly on the stationary base, e.g., a measuring table, or on the base via an object holder (e.g., a rotary table).

[0006] A confocal white light sensor, a type of optical sensor for use in a coordinate measuring machine (CMM) according to the present invention, is understood to be a sensor that utilizes the principle of chromatic confocal distance measurement. White light (i.e., electromagnetic radiation, not necessarily visible, with radiation components of several wavelengths) is directed from a light source onto a focusing optic. The focusing optic causes a dispersion of the radiation, i.e., chromatic aberration occurs. As a result, the radiation components of the different wavelengths are focused at different distances from the focusing optic. If an object is located at the respective focus (focus point or focus line) that reflects the radiation back towards the sensor, the sensor detects radiation of the wavelength with maximum intensity that was reflected at the focus.

[0007] It is also possible that radiation of different wavelengths is reflected simultaneously at their respective foci. In this case, the sensor detects a (local) intensity maximum at each of these wavelengths. From the wavelength of the single intensity maximum, or from the wavelengths of the intensity maxima, the distance between the sensor and the object can be determined if the distance of the focus to the sensor (e.g., to the focusing optics) is known. However, this information is not initially available and is usually obtained through a reference measurement in which the distance between the sensor and the reflection point is also measured by another method, e.g., using a laser interferometer.

[0008] Confocal white light sensors are high-resolution and precise distance sensors compared to other distance sensors (such as capacitive sensors). White light sensors with maximum resolutions of, for example, one hundredth of a micrometer and measuring ranges from several tenths of a millimeter to several tens of millimeters are available on the market. One example is the confocal white light sensor with the type designation "confocal DT IFS 2405" from Micro-Epsilon Messtechnik GmbH & Co. KG, Ortenburg, Germany.

[0009] To determine the object's coordinates, the sensor generates sensor measurement signals or values ​​that, for example, specify the distance between the sensor and the object's surface, or from which this distance can be calculated. When measuring the thickness of preferably at least partially transparent workpieces, the distance can also refer to a material interface from which radiation is reflected. In this case, at least two distance values ​​can be obtained, one of which can refer to the object's surface and the other to a material interface in a depth dimension of the object. The thickness of the object, or of its measured material layer, can be determined from the difference between these distance values, corrected for the refractive indices of the measured material.

[0010] Furthermore, a position value from the CMM is obtained, which typically consists of the position or setpoint of each CMM axis and can therefore also be referred to as an axis value. The position values ​​can be determined and / or read by a position measuring system of the coordinate measuring machine (CMM), whereby the position measuring system, for example, detects the individual axis positions of the CMM and determines the position of a sensor mounted on the CMM from this data. If no CMM is used, the position value can be determined by camera tracking or by a position measuring system of the motion device.

[0011] A single position value contains information about the spatial position of the CMM and / or sensor at a given time. Specifically, position values ​​can be output at predefined, predetermined, subsequently determined, and / or uniquely identifiable times, where these times can define a position cycle. In particular, the position value can specify the position of a sensor or sensor interface positioned by the CMM. Furthermore, the position value can comprise multiple components, such as an X, Y, or Z component.

[0012] The sensor is moved by the CMM or by other means relative to and along the object to generate sensor measurement signals (hereinafter also referred to simply as sensor signal or signal) for or to multiple measurement or sampling points. The generation and output of the signals can occur at defined intervals or according to a defined sensor clock. The intervals or the sensor clock can be defined by predetermined, predeterminable, subsequently determinable, and / or uniquely identifiable time points at which the signals acquired by the sensor can be output to an evaluation unit, which may, for example, be provided in a CMM.

[0013] Furthermore, the sensor's movement is accompanied by changing position values ​​of the CMM and / or the sensor, particularly when the sensor is moved essentially continuously relative to and along the object during a so-called scan operation. As a result, each sensor measurement signal can be uniquely assigned a position value, where the position value corresponds to the axis values ​​or positions of the CMM or other motion device that were assumed by the sensor during object detection to generate that sensor measurement signal, and / or where the position value was determined directly from the sensor and, for example, by means of camera tracking.

[0014] As mentioned, the sensor measurement signals and position values ​​can be output to an evaluation unit according to a sensor cycle or position cycle, which ultimately performs an evaluation with regard to the object coordinates. A value can be output after or at the end of a predetermined cycle interval. The cycle-by-cycle reading of sensor measurement signals and position values ​​is also referred to as cycle-by-cycle triggering of this information by the evaluation unit. In particular, if the sensor measurement signals relate directly or indirectly to a distance to the object, the evaluation unit can add the sensor measurement signals and their corresponding position values ​​in a known manner. This yields a total value from which the object's coordinates can be derived. The determined coordinates can therefore be referenced to a coordinate system that includes the base of the CMM.Similarly, a thickness measurement of the object can be carried out in the manner described above.

[0015] Regarding the sensor measurement signals, it should be noted that the sensor's measurement signals generally need to be processed to obtain the final sensor measurement signals or values ​​(for example, by converting the measurement signals into a suitable format, applying calibration factors, etc.). This can take place within a sensor controller, which, due to the described processing steps, only outputs the sensor measurement signals to the evaluation unit with a time delay, or in other words, with a certain latency.

[0016] Additionally or alternatively, delays can occur because the sensor measurement signals are transmitted to the evaluation unit via a communication bus. The transmission time required for this and / or the waiting time until a sensor measurement signal can be transmitted from the communication bus according to its communication clock can also contribute to time delays or latencies.

[0017] As a result, a certain dead time can occur between the point at which the object is detected, or in other words, the point at which measurement signals are detected by the sensor, and the point at which the evaluation unit finally receives a corresponding sensor measurement signal. This dead time can, for example, last for several sensor cycles.

[0018] To ensure sufficient accuracy of the measurement results, it is generally advantageous to assign the appropriate position value to each measured sensor signal during the evaluation process. In particular, the sensor signal and the position value should refer to the same point in time (i.e., be considered and / or assigned synchronously). Preferably, the position value used should be the one that was present when the sensor signal was being measured (i.e., at the time the object was detected by the sensor) or that was assumed by the CMM and / or the sensor at that time.

[0019] Therefore, approaches exist in which, assuming a constant dead time between object detection and the receipt of the sensor signal by the evaluation unit, time-synchronous value pairs of a sensor signal and a position value are at least approximated. For example, it is assumed that a sensor signal is always received by the evaluation unit with a delay of two sensor cycles compared to its detection time. If the received position values ​​are stored and, in particular, continuously recorded, a position value shifted by two sensor cycles relative to the current evaluation time can be read out for a currently received sensor signal in order to approximate a time-synchronous value pair. More precisely, a position value two sensor cycles prior to the receipt of the sensor signal is considered.

[0020] However, it has been shown that such a mapping of position values ​​and sensor measurement signals does not always guarantee the desired measurement accuracy. Assuming a constant dead time is generally already an imprecise simplification. Even if this is accepted, a suitable value for the assumed constant dead time can often only be determined empirically. This leads to further losses in accuracy.

[0021] DE 10 2016 212 650 A1 discloses a solution for generating geolocated sensor data by assigning position values ​​that are as synchronous as possible with time to data acquired by a sensor. In particular, individual pixels of a sensor data set can have individual acquisition times.

[0022] DD 208 858 A1 concerns a coordinate measurement of moving objects using photoreceiver arrays, in which an integration time during image acquisition is used to determine a time of acquisition.

[0023] DE 100 20 842 A1 concerns the coordination of higher sensor measurement frequencies with the control frequencies of the positioning system of a coordinate measuring machine (CMM). In particular, the sensor measurements and the position measurements are brought to a common time scale.

[0024] WO 2005 / 119 173 A1 also reveals a solution for reconciling a higher sensor sampling frequency with a lower controller clock frequency.

[0025] US 2018 / 080766 A1 concerns a non-contact coordinate measuring machine with an electro-optical sensor that uses a structured light code.

[0026] One object of the present invention is therefore to improve the accuracy of the detection of an object, in particular by means of optical sensors.

[0027] This problem is solved by an arrangement and a method according to the attached independent claims. Advantageous further developments are specified in the dependent claims. Furthermore, it is understood that the features mentioned in the introductory description may also be provided individually or in any combination in the solution disclosed herein, unless otherwise specified or apparent.

[0028] The inventors recognized, among other things, that known approaches, due to their inherent disadvantages, cannot accurately capture cases where the sensor's acquisition time and the associated dead time are unpredictable and / or variable. This can result, in particular, from the sensor requiring a certain acquisition time during object detection, which is difficult to predict and / or can vary when capturing multiple sampling points. An acquisition time is generally necessary when the sensor must integrate and / or accumulate incident radiation or comparable measured quantities over a certain period to determine an actual measurement signal. In the case of an optical sensor, this can involve integrating incident radiation using a photosensitive acquisition or detector unit.A typical example of acquisition time is therefore a required exposure time during which a photosensitive acquisition or detector unit of the sensor is exposed.

[0029] The invention is also applicable to sensors other than optical sensors that detect the object within a detection period in order to generate sensor information usable by the evaluation device. Examples of such sensors are capacitive sensors that integrate over time intervals and inductive sensors.

[0030] More precisely, the exposure time of an optical sensor serves to ensure a sufficient overall intensity of the incident radiation (i.e., a sufficient amount of radiation) so that measurement signals with the desired accuracy can be generated. This is particularly relevant when the intensity of the radiation emitted by the sensor cannot or should not be increased further. To nevertheless achieve a sufficient amount of usable radiation, the exposure time (or, more generally, the acquisition time) can be appropriately increased. In the case of CCD or CMOS sensors, for example, this allows the incident radiation to be integrated over a longer period to increase the usable radiation amount and generate a sufficiently strong measurement signal.The exposure time can be varied, for example, by closing and opening a so-called shutter, which is positioned in front of the actual detection unit of the sensor (for example, in front of a photosensitive detector surface) when viewed from the perspective of the object.

[0031] When measuring objects, the intensity of the incident radiation depends particularly on the object's reflectance, which can be difficult to predict and / or vary locally. Consequently, even when measuring along a continuous object surface, fluctuating intensities of the radiation detected by the sensor can occur. This may require continuous adjustment of the exposure time (generally, the lower the reflectance, the longer the required exposure time). Setting a suitable exposure time can be done automatically via sensor control in a known manner and is explained in more detail below.

[0032] If the sensor's acquisition time varies, the point in time at which the object is actually detected also varies. Consequently, the dead time between the actual object detection and the final output / receipt of a sensor measurement signal also varies. Even assuming a constant acquisition time, the difficult-to-predict reflective properties of the object make it nearly impossible to predict the required acquisition time, and thus the actual detection time and the resulting dead time. In any case, this would require at least time-consuming test measurements on the object to be measured. Therefore, known solutions based on the assumption of a predetermined and, above all, constant dead time are only applicable at the cost of high measurement inaccuracies.

[0033] For example, a sensor reading acquired midway through a sensor cycle interval due to the required acquisition time, and subsequently received by the evaluation unit with an additional two sensor cycles of processing and transmission dead time, would have a total dead time of 2.5 sensor cycles. If, as in the previous example, a constant dead time of two sensor cycles were assumed, this sensor reading would be incorrectly assigned a position value that occurred at a later time, shifted by 0.5 sensor cycles. This would result in a temporal discrepancy between the assigned sensor reading and the position value. This is particularly critical when the sensor is moved relative to the object, especially if the movement is essentially continuous or uninterrupted (for example, in a so-called scan mode).Within the aforementioned discrepancy of 0.5 sensor cycles, the CMM or the sensor may already be in a position that differs from its position during the actual object and sensor measurement acquisition. The coordinate of the scanning point on the object, determined based on the position value and the sensor measurement signal (and especially a distance value derived from it), would therefore have only low accuracy.

[0034] One aspect of the invention is therefore to consider the actual time of object detection by the sensor, taking into account the applied detection time. This is used to determine a primary position value, which also relates to the actual detection time. As a result, the position value that was actually present at the time of object detection can be considered with greater accuracy. Since this can be done separately for each sensor measurement signal, varying exposure times and the associated varying dead times during object detection can also be compensated for.

[0035] In particular, the invention proposes a method for detecting an object by means of a movable, in particular optical, sensor, comprising: Moving the sensor relative to the object and repeatedly determining the instantaneous position of the sensor by a position measuring system, wherein the determined position is output to an evaluation device at a predetermined or determinable time. wherein the sensor repeatedly detects the object during the duration of a detection time interval and, in each instance, outputs at least one signal to the evaluation device according to the information acquired during a detection time interval, specifically at a predetermined or determinable time, and wherein, for at least one of the recording time intervals, the following is performed: Determining a mean recording time that lies within the recording time interval; Determining position values ​​that were issued at times that include the recording time in between; and Determining a primary position value based on the determined position values, which approximates the position assumed by the sensor at the time of detection.

[0036] The invention further relates in particular to an arrangement for detecting an object by means of a movable, in particular optical, sensor, comprising: a sensor that is movable relative to the object; a position measuring system designed to repeatedly determine the instantaneous position of the sensor, and an evaluation unit to which the determined position is output by the position measuring system at a predetermined or determinable time, wherein the sensor is configured to repeatedly detect the object during the duration of a detection time interval and to output at least one signal to the evaluation device according to the information detected during a detection time interval, at a predetermined or determinable time, and wherein the evaluation device is configured to perform the following for at least one of the detection time intervals: Determining a recording time that lies within the recording time interval; Determining position values ​​that were issued at times that include the time of recording between them; and Determining a primary position value based on the determined position values, which approximates the position assumed by the sensor at the time of detection.

[0037] The arrangement may include any further step, any further development, and any further feature to provide all of the foregoing or following interactions, operating states, and functions. In particular, the device may be configured to perform a procedure according to any of the foregoing or following aspects.

[0038] The arrangement can in particular be part of a coordinate measuring machine, i.e. the coordinate measuring machine incorporates the arrangement.

[0039] The sensor can be configured according to any of the aforementioned variants. In particular, an optical sensor can be configured as a white light sensor. The detection time can be in the form of an exposure time. The detection time can be automatically determined and / or set by a sensor controller or other control device of the arrangement, for example, using an exposure meter. The exposure meter can include a light sensor, which, for example, comprises a photosensitive electrical resistor (LDR - Light Dependant Resistor).

[0040] The sensor can be mounted on a coordinate measuring machine (CMM) and moved by it. The CMM can also include the position measuring system and / or the evaluation unit. Alternatively, the sensor can be mounted on another moving device, for example, an industrial robot. The sensor can also be moved manually. In both of the above cases, the position measuring system can include a camera system for tracking the sensor's position, with the sensor, for example, including suitable markers for camera detection. The evaluation unit can be a conventional PC.

[0041] As described, the points in time at which position values ​​and sensor measurement signals are output can define a position clock or a sensor clock, respectively. These clocks can have a frequency of several Hertz, several hundred Hz, or at least 1 kHz. Examples include frequencies of 1 kHz, 500 Hz, and 100 Hertz. The frequency of the position clock can generally be the same as that of the sensor clock or different from it.

[0042] The sensor clock and / or the position clock can be divided into individual clock intervals, each defined by a start time and an end time. The start time can coincide with the end time of a preceding clock interval, and the end time can coincide with the start time of a subsequent clock interval. The start and end times can, in turn, be defined by points in time at which position values ​​or sensor measurement signals are output.

[0043] After each sensor cycle, at least one sensor measurement signal can be output. This signal may be based on an object detection by the sensor within that cycle. However, considering the described dead times, the sensor measurement signal received by an evaluation unit after a cycle may also be based on an object detection within a previous cycle. The sensor measurement signal for the currently elapsed cycle would then also be received with a corresponding delay.

[0044] After each cycle of the position measurement, at least one position value can be output. This position value can refer to the current position (or at least a position within the current cycle) of the CMM and / or the connected sensor. However, considering potential dead times, the position value can also refer to a position during a previous cycle. A dead time can result, for example, from the fact that the output position value is determined by averaging various individual values ​​to increase measurement accuracy.

[0045] The position value can be defined in such a way that the position of the sensor can be calculated from it, at least indirectly. For example, the position value can relate to the position of a sensor interface of the CMM or another motion device, whereby, given the dimensions of the sensor interface and / or the sensor, the position of a predetermined area of ​​the sensor can be deduced (for example, the position of the origin of a sensor coordinate system).

[0046] The sensor clock and / or the position clock can generally be regular (i.e., have a constant frequency) and output at least one sensor measurement signal or position value at predetermined time intervals. This regularity can be achieved by ensuring that the predetermined, predetermined, subsequently determined, and / or uniquely identifiable times at which sensor measurement signals or position values ​​are output are regularly spaced apart. In general terms, the sensor measurement values ​​or position values ​​can be output at regular intervals, preferably according to the sensor clock or position clock.

[0047] The sensor can be moved essentially continuously, or in other words, without interruption, while detecting the object. This can include movement at an essentially constant or varying speed. Since a certain detection time is required for the sensor to detect the object, it can be moved along a surface area during exposure, i.e., along several possible measurement or sampling points. The final sensor measurement value and / or signal, which, as described above, can generally refer to at least one distance of the object (for example, from its surface and / or an internal material interface) from the sensor, can thus correspond to an average distance value along the moving surface area during the exposure time.

[0048] The acquisition time can refer to the point in time that coincides with the midpoint of the acquisition time duration (or the duration of the acquisition time interval) (i.e., a point in time at which half of the exposure time has elapsed). It can be determined, for example, by subtracting half of the acquisition time duration from the end time of a sensor clock interval or by adding it to the start time of a sensor clock interval.

[0049] Unless otherwise stated or apparent, all of the times specified above or below may be defined as absolute times. Alternatively, they may be defined relative to a given time, and in particular relative to a current time (for example, relative to an evaluation time). For instance, the times may refer to a negative time axis extending from the evaluation time. In this case, the times may indicate the time interval from the evaluation time (for example, as a value of -4 ms).

[0050] The time points that include the acquisition time can, in particular, directly include the latter between them. For example, they can be the time points immediately adjacent to the acquisition time at which a position value is output. In particular, they can be time points that delimit a position clock interval that contains the acquisition time. The interval of the position clock that contains the exposure time can be determined based on the usually known position clock frequency and / or with knowledge of a relative ratio of the acquisition time to a given time (for example, to an evaluation time). If the mean exposure time (i.e., the one in the middle of the exposure time interval) is...If the acquisition time interval (i.e., the time frame within the acquisition time interval) is, for example, 3.5 ms before the evaluation time, and the position clock frequency is 1000 Hz, then the relevant interval of the position clock (i.e., the acquisition time interval) is the one that begins 4 ms before the evaluation time and ends 3 ms before the evaluation time. The position values ​​output at these times, 4 ms and 3 ms before the evaluation time, which were stored and / or recorded according to one of the following variants, can then be determined as the position values ​​encompassing the acquisition time.

[0051] As described, the acquisition time can be a mean point within the acquisition time interval. In particular, if the sensor and / or position measurements change due to accelerated movement within an acquisition interval, the position and sensor acquisition should refer to the same (acquisition) point in time. Preferably, in this case, the duration of each acquisition interval should also be the same, which, for example, defines an integration time over which the respective values ​​are integrated. In the case of accelerated movement, the mean point in time can be defined differently than being in the middle of the acquisition time interval: for example, the mean point in time can be the point in time at which the sensor has covered half of the total distance traveled during the acquisition time interval.

[0052] The main position value is preferably a value calculated purely computationally based on other position values ​​and not a position value actually measured and / or output according to the position clock. According to the invention, the main position value can be assigned to the sensor measurement signal for which the acquisition time was determined. According to the invention, the main position value is combined with this sensor measurement signal (and in particular a sensor measurement value or distance value derived therefrom) to determine a coordinate value of the object. This can be done by adding these values ​​as already described. Alternatively, a thickness measurement can be performed by determining two distance values ​​at different material interfaces of the object (for example, at a surface and a back side and / or at an internal material interface) and calculating the difference between them.In this case, the acquisition time can be assumed to be identical for both distance values. An exact correlation with the position value (or primary position value) recorded during the thickness measurement may be necessary, for example, to determine the precise thickness profile along or within the object. In other words, this allows the obtained thickness value to be assigned a precise position value in space and / or on the object. The position value in space can also be determined from the primary position value, a second sensor measurement signal or distance value, and the orientation in space, the latter being determined, for example, from the device moving the sensor and / or its joint angles.

[0053] As mentioned, the acquisition time can correspond to the mean time of the acquisition time interval, i.e., coincide with this time. In particular, the acquisition time, which can generally be a specific point in time along a defined time axis, can be calculated or determined using this mean, where the mean can coincide with the point in the acquisition or exposure time interval that divides the time interval into two equal halves.

[0054] Furthermore, the acquisition time interval can correspond to an exposure interval in which the sensor, designed as an optical sensor, is exposed. Alternatively, the acquisition time interval can correspond to an interval in which the system waits for the receipt of back-reflected optical signals, for example, in the case of an optical sensor in the form of a time-of-flight camera. When this description refers to one time interval corresponding to another, it specifically means that the beginning and end times of the time intervals are identical.

[0055] In a further development of the method and the arrangement, it is provided that the acquisition time interval has a predetermined relative relationship to an output time of the sensor measurement signal and / or a position value to the evaluation device and / or to an at least partially parallel interval of the sensor clock. The relative relationship can refer to a temporal relative relationship or, in other words, to a relative position along a common time axis of the acquisition time interval and the sensor clock.

[0056] In general, the relative ratio can be chosen such that the acquisition time interval ends at an end time of the sensor clock interval and / or begins at a start time of the sensor clock interval. In particular, the acquisition time interval can end at a time that coincides with at least one of the output times. Alternatively, an output time can divide the acquisition time interval into two sub-intervals according to the predetermined relative ratio, and / or the acquisition time interval can have a predetermined time interval from the end, start, and / or output times.

[0057] Knowing the relative ratio, the mean exposure time can be determined in particular, for example by subtracting half of the exposure time from the end time or output time, or by adding it to the start time.

[0058] According to another embodiment of the arrangement and method, a plurality of the output position values ​​are stored. This storage can be accomplished using a memory unit in which the position values ​​are sorted according to their input time. The memory unit can, for example, be configured as a rolling memory or as a FIFO (First-In-First-Out) memory. The rolling memory can also be referred to as a ring buffer or ring buffer. In general, in this case, a memory address can be incremented clock cycles, and when the end of the memory area is reached, data is written again from the beginning of the memory area, overwriting the data stored there. Consequently, the memory area can be written to virtually indefinitely.

[0059] In particular, the position values ​​can be saved taking into account their respective corresponding position cycle intervals and / or output times. In other words, the position values ​​can be saved in such a way that their output times, corresponding position cycles, and / or a chronological sequence of the position values, determined by the output time and / or the position cycle, can be read or reconstructed. For this purpose, the position values ​​can be digitally timestamped and saved.

[0060] In a further development of the procedure and the arrangement, the main position value is calculated by averaging the determined position values; that is, an average of the determined position values ​​is calculated. Specifically, the arithmetic mean of the determined position values ​​assigned to the interval containing the mean exposure time can be calculated to determine the main position value. Alternatively, interpolation can be performed between the determined position values, and in particular, linear interpolation can be used. All of the described calculation methods can also be applied when more than two determined position values ​​are available. This can particularly involve considering additional adjacent position intervals, as explained below, whereby the main position value can be calculated by averaging all position values ​​of the individual intervals or by interpolating based on these values.

[0061] According to a further variant of the method and arrangement, an additional position value is determined for the purpose of determining the main position value. This further position value is output at a time that precedes or follows the position value output times that include the acquisition time between them. In particular, this position value may have been output at a time that limits an interval of the position clock that precedes or follows the interval containing the mean exposure time. Preferably, this further interval immediately precedes or follows the position clock interval containing the mean exposure time, such that no other intervals exist between them.In other words, additional position values ​​of a position clock interval can be taken into account, which is adjacent to the position clock interval containing the mean exposure time.

[0062] This allows, for example, the compensation of different position clock and sensor clock frequencies. In particular, at least enough position clock intervals and associated position values ​​can be considered such that the total duration of these position clock intervals is at least equal to the duration of one sensor clock interval.

[0063] As mentioned, the arrangement and the procedure can further provide that the determined principal position value and the sensor signal for which the acquisition time was determined are correlated. Alternatively or additionally, at least one property (for example, a coordinate value or a thickness value) of the object can be determined based on the principal position value and the sensor signal. For example, a coordinate of at least one measurement point, or in other words, sampling point on the object, which was detected by the optical sensor during the exposure time, can be determined.

[0064] Another embodiment of the method and arrangement provides that the determination of the main position value is carried out by an evaluation unit operated according to a system clock as follows: Defining a determination interval containing the time of acquisition, which has the duration of a single system clock interval or an integer multiple thereof; Determining position values ​​that were output at times which limit at least a time interval (and in particular in the form of an interval of the position clock) that at least partially overlaps with the determination interval; Determining the main position value based on the determined position values.

[0065] By determining position values ​​within the acquisition interval or at the aforementioned time points, the evaluation for determining 3D object coordinates (or general positional key values) is simplified, and in particular, the required computing power is reduced. To achieve high accuracy, all position values ​​output within the acquisition interval or at the aforementioned time points can be determined. For example, all intervals of the position measurement cycle, along with their corresponding position values ​​that overlap with the acquisition interval, can be considered. Determining the positional key value can then be performed using any of the methods discussed above, such as calculating an average or by interpolation.

[0066] The measurement interval can refer to or define a time interval. Preferably, it is defined such that it includes the acquisition time at its midpoint, or, in other words, that the acquisition time forms a temporal midpoint of the measurement interval. The system clock can be, for example, a CPU clock, a processor clock, or a general processing clock of the evaluation unit. The system clock can, in turn, be several hundred Hz or at least 1 kHz and, for example, be of the same type as the sensor clock and / or the position clock, or be an integer multiple thereof.

[0067] Finally, it can generally be provided that coordinate determination and / or thickness measurement is carried out based on the sensor signal(s).

[0068] One embodiment of the invention is explained below with reference to the accompanying schematic figures. Features that are identical in type and / or function may be designated with the same reference numerals across different embodiments. These represent: Fig. 1 a schematic representation of an arrangement according to the invention, which carries out a method according to the invention; Fig. 2 a representation to explain the method according to the invention in a first embodiment; Fig. 3 a representation to explain the method according to the invention in a second embodiment; and Fig. 4 A representation to explain the method according to the invention in a third embodiment.

[0069] In Fig. Figure 1 shows an arrangement 100 according to an embodiment of the invention, wherein the arrangement 100 can perform all of the variants of the methods of the invention described below.

[0070] The arrangement 100 comprises a coordinate measuring machine (CMM) 211, which is designed as a portal and has a measuring table 201 above which columns 202, 203 are movably arranged in the Y-direction of a Cartesian coordinate system. The columns 202, 203, together with a crossbeam 204, form a portal for the CMM 211. The crossbeam 204 is connected to the columns 202 and 203 at their opposite ends. Electric motors (not shown) cause the linear movement of the columns 202, 203 in the Y-direction, along the axis of movement, which runs in the Y-direction. For example, each of the two columns 202, 203, or only one of them (e.g., column 202), is assigned an electric motor.

[0071] The crossbeam 204 is combined with a cross slide 207, which is movable along the crossbeam 204 in the X-direction of the Cartesian coordinate system, for example, via air bearings. The movement of the cross slide 207 along the axis of movement in the X-direction is driven by at least one further electric motor (not shown). A vertically movable quill 208 is mounted on the cross slide 207. At its lower end, this quill is connected via a mounting device 210 and a rotary device 205 to an interchangeable interface 209, to which a white light sensor 215 is coupled via an angled bracket. The interchangeable interface 209 can also be referred to as a sensor interface.

[0072] Due to its angular position, the measuring direction of the white light sensor 215 is approximately parallel to the XY plane. The interchangeable interface 209, driven by another electric motor, can be moved relative to the cross slide 207 in the Z direction, along the Z axis of motion of the Cartesian coordinate system. The electric motors of the CMM allow the white light sensor 215, coupled to the interchangeable interface 209, to be moved to almost any position in the area below the crossbeam 204. Furthermore, the rotary device 205 can rotate the white light sensor 215 about the Z axis, allowing it to be oriented in different directions. Alternatively, a rotary swivel device can be used instead of the rotary device 205, enabling other degrees of freedom of movement, such as additional rotational mobility about an axis perpendicular to the vertical (Z direction).

[0073] Also shown is an evaluation unit 220, which receives the measurement signals from the white light sensor 215 via a schematically depicted connection 230. The connection 230 can, for example, be an optical fiber such as a fiber optic cable. The connection 230 can be different from the one shown in the diagram. Fig. 1 shown is directly connected to the white light sensor 215.

[0074] Furthermore, in Fig. Figure 1 schematically shows a control unit 222 of the CMM 211, which in particular controls the drives (e.g. the aforementioned electric motors). In particular, the control unit 222 is able to move the white light sensor 215 to a desired position by controlling the drives and also to control the rotary device 205 in order to align the white light sensor 215 in a desired measuring direction.

[0075] The controller 222 is combined with a data storage device 221, in which information about the relationship between measurement signals from the confocal white light sensor and the actual distance of the white light sensor to the surface of a measurement object 235 is stored. To determine such relationships, a reference body 231 is also shown, which is connected to the measuring table 201 via a support 232. The purpose and use of the reference body 231 are known, for example, from DE 10 2015 217 637 A1, which is why a more detailed explanation is omitted here. The data storage device 221 can also store position values ​​and / or sensor measurement values ​​explained below and can generally be designed as a rolling memory. The spatial position of the individual axes relative to each other is also stored in the data storage device 221, in particular the position of the white light sensor 215 relative to the CMM axes and / or the angles relative to that axis (e.g.,in the form of a robot arm or limb), to which the sensor 215 is attached.

[0076] During operation, the CMM 211 moves the white light sensor 215 relative to the object 235 being measured via its individual axes and electric motors. More precisely, the white light sensor 215 is moved in such a way that it scans the surface of the object 235 without contact and determines distance values ​​between itself and the object surface for several individual scanning or, in other words, measuring points. However, since the white light sensor 215 can also be moved relative to the object 235 during the measurement process, and since a certain acquisition time is required for the measurement due to the exposure time explained below, the final distance value output may also be an average distance value of points on the object surface that move through the detection range of the white light sensor 215 during the measurement.

[0077] The distance values ​​ultimately determined are sensor readings from the white light sensor 215 (one distance value corresponds to one sensor reading). They are determined based on sensor measurement signals, which are output via connection 230 to the evaluation unit 220 and converted there into corresponding measured values. It should be noted that the sensor measurement signals received by the evaluation unit 220 may also represent or contain already completed distance values, thickness values, and / or sensor readings. For this purpose, signals acquired by sensor 215 can be directly evaluated and / or converted in a sensor controller.

[0078] In addition to the sensor measurement signals, the evaluation unit 220 also receives position values ​​from the CMM 211, which describe the position of the CMM 211 or its individual axes. In particular, the position values ​​can be values ​​that describe the position of the white light sensor 215 in a base coordinate system of the CMM 211, or from which this position can be determined. For example, the position values ​​can indicate the position of the interface 209. The output of the sensor measurement signals and the position values ​​occurs in cycles according to a sensor or position cycle, as explained below with reference to the additional figures.

[0079] The position values ​​are stored in the data storage device 221 or another storage device, which may, for example, be integrated into the evaluation device 220. In particular, they are stored in such a way that their chronological sequence and preferably also their specific output and / or input times can be reconstructed. This can be achieved by using a rolling memory or a FIFO memory and / or by providing the position values ​​with a digital timestamp.

[0080] A pair of values ​​can generally be formed from a position value and a sensor measurement signal (which, in the example shown, can generally represent n distance and / or thickness values, where n is a natural number of 1 or more). If this pair refers to a common point in time, and in particular the point in time when the white light sensor 215 detects the object 235, the values ​​of this pair can be calculated and / or assigned to each other. In particular, the values ​​can be added to determine a coordinate of the surface of the object 235 in the base coordinate system of the CMM 211 (i.e., a coordinate of the scanning point or the scanning area in the case of continuous motion measurement). Alternatively, a thickness value can be assigned the exact measured object location or another spatial coordinate using the corresponding position value.

[0081] A variety of corresponding coordinates can be used to check properties of the measured object 235. These properties can be, for example, quality parameters such as dimensional accuracy or surface finish.

[0082] As described, obtaining meaningful measurement results requires that the position values ​​of the CMM 211 and the sensor measurement signals or sensor readings of the white light sensor 215 are meaningfully correlated. In particular, time-synchronized pairs of values ​​should be generated for this purpose, which refer to a common detection time of the object 235 by the white light sensor 215.

[0083] The following refers to Fig. 2. First, possible error potentials identified by the inventors are discussed, which have so far made a corresponding time-synchronous assignment of the aforementioned values ​​difficult.

[0084] In Fig. Figure 2 shows a single evaluation cycle of the evaluation unit 220, in which the coordinates for a single sampling point or sampling range are determined. In other words, the evaluation cycle shown refers to a single measurement cycle and can be repeated for each of the total measured sampling points or sampling ranges.

[0085] In detail, within an evaluation cycle, coordinates for a single sampling point or sampling area on the object surface are determined by considering at least one position clock interval and one sensor clock interval 16, 14, which are assigned to each other and occur simultaneously or at least overlapping in time. This is illustrated below using the following examples: Fig. The procedure described above can therefore be repeated for each individual sampling point or sampling area, whereby individual sensor clock intervals or position clock intervals are considered in each case, which relate to the corresponding measurement process of the specific sampling point.

[0086] In Fig. Figure 2 shows two time axes, 10 and 12. The upper time axis, 10, represents a position cycle, according to which position values ​​from the CMM 211 can be output to the evaluation unit 220. The position cycle is defined by specific, predefined points in time at which position values ​​are, or could be, output. These points in time are marked by points along time axis 10.

[0087] The lower time axis 12 represents a sensor cycle, according to which sensor measurement signals are output from the white light sensor 215 to the evaluation unit 220. The sensor cycle is defined by individual predefined time points at which sensor measurement signals are, or could be, output. These time points are marked by points along the time axis 12.

[0088] Both time axes 10 and 12 are negative time axes, starting from a current evaluation time of 0 ms and extending in a negative direction. Each represents a time duration from 0 ms to -4 ms. In other words, they represent individual clock intervals that lie before the current evaluation time of 0 ms and thus refer to an already completed object acquisition, for which the evaluation (i.e., coordinate determination) still needs to take place.

[0089] It can be seen that the position clock and the sensor clock have an identical frequency of 1,000 Hz. Accordingly, the duration of a single position clock interval is 16 ms, and a single sensor clock interval is 14 ms, as shown in Fig. 2. Examples of sensor clock and position clock intervals 14 and 16 are shown. For the exemplary highlighted intervals 14 and 16 in Fig. In case 2, the start time is -4 ms and the end time is -3 ms.

[0090] The following section will first examine the lower time axis 12 of the sensor clock in more detail to explain the associated dead times. When the white light sensor 215 determines a distance value or distance signal, this information must first be processed within the sensor control unit. This can include, for example, evaluating the measurement signals taking into account any calibration factors and / or converting them into a suitable digital format.

[0091] Furthermore, a time delay arises because the sensor measurement signal must be transmitted to the evaluation unit 220 via connection 230. If connection 230 is configured as a bus connection, it may be necessary, for example, to wait for predefined time windows, which are defined according to a communication clock of bus connection 230. In the case shown, this contributes to a delay of at least 1 ms before a currently acquired sensor measurement reaches the evaluation unit 220. To achieve additional reliability and to be able to compensate for potentially longer dead times, the evaluation unit 220 assumes an inherent dead time of 2 ms.In other words, for a given evaluation time, the sensor measurement signal that is shifted by two sensor measurement cycles relative to the evaluation time and occurs before it is always considered current and used for evaluation. In the example shown, this corresponds to a shift of -2 ms.

[0092] In Fig. 2 means that, from the perspective of the evaluation time of 0 ms, the preceding sensor clock interval 14, which lasts from -3 ms to -2 ms, is considered. However, the exact acquisition time within this sensor clock interval 14 remains unknown. This is primarily because the white light sensor 215 requires a previously mentioned exposure time to actually measure the object surface at a given sampling point or sampling area.

[0093] More precisely, a sensor measurement signal is output at each individual clock cycle or after each sensor clock interval 14, according to the sensor clock, i.e., also at -2 ms. However, this sensor measurement signal was not acquired at this specific time of -2 ms, but rather within a acquisition or exposure time interval 18 that lasts from -2.6 ms to -2 ms (i.e., has an exposure time of 0.6 ms). Therefore, if one were to assume a constant dead time of, for example, -2 ms, as is common in the prior art described above, one would be working with an incorrect acquisition time for the sensor measurement signal.

[0094] This is problematic because it would result in an incorrect, non-time-synchronized value pair being formed from the sensor measurement signal (or the distance value derived from it) and the position value. More precisely, the position values ​​are calculated according to the... Fig. The upper time axis 10 is also output in a clock-like fashion and according to the position clock with the same frequency as the sensor clock. That is, at each of the displayed time points, and especially at -2 ms, a position value is output that also describes the actual position of the CMM 211 at that time with sufficient accuracy. As described, this position value can first be stored in the memory unit 221 and read from it as needed.

[0095] Assuming constant dead times according to the prior art, one would therefore add the position value, which refers to the output time of -2 ms, and the sensor measurement value, which was supposedly acquired at time -2 ms in the form of the sensor measurement signal, together to determine a coordinate of the object surface. However, since, as the inventors recognized, this assignment is only possible within the framework of an imprecise simplification due to the exposure time interval 18, no accurate measurement result can be obtained in this way. This applies in particular to cases in which the white light sensor 215 is moved continuously in the sense of a scan operation. In this case, a continuous change in the position values ​​can occur even within a single sensor clock interval 14. That is to say,, the position value for the actual detection time within a sensor clock interval 14 can deviate significantly from the position value at the end time of -2 ms.

[0096] In summary, the assumption of constant dead times between object detection and the final receipt of the sensor measurement signal or value by the evaluation unit 220, which has been made so far according to the state of the art, leads to the possibility that position values ​​are considered which refer to times when object detection by the white light sensor 215 did not take place or at least not primarily.

[0097] Therefore, it is planned to estimate the actual detection time of the white light sensor 215, which in the example shown is chosen as an average detection time. For this purpose, the exposure time duration of 0.6 ms, which was determined by the sensor control based on a known exposure measurement and is therefore known, is averaged or halved. This corresponds to a value of 0.3 ms. This value is subtracted from an end time of the detection interval 18 or added to an start time (here -2.6 ms) of the detection interval 18.

[0098] In the case shown, the system takes advantage of the fact that the detection interval 18 is chosen such that it always has a predetermined relative relationship to the sensor clock interval 14. More precisely, the exposure is carried out by a sensor control of the white light sensor 215 such that the exposure time interval 18 ends at the end point of a sensor clock interval 14, i.e., in the case shown, at -2 ms. This only requires that the necessary exposure time is first determined, whereupon the sensor control appropriately sets the trigger time of the exposure so that the exposure interval 18 ends at the corresponding end point of the sensor clock interval.

[0099] As a result, the mean exposure time in the example shown is -2.3 ms. This time is considered the actual detection time of the object 235 by the white light sensor 215 for the sensor measurement signal that was subsequently output at -2 ms.

[0100] Since the actual acquisition time is thus known or at least approximated, the exemplary embodiment further provides for the determination of a position value that existed at this actual acquisition time. More precisely, the position value that existed at the mean exposure time of -2.3 ms is to be determined or at least approximated. For this purpose, the position measurements output at times of -3 ms and -2 ms are considered, which consequently define the position interval in which the mean acquisition time of -2.3 ms lies. In particular, these times include the acquisition time between them and, in the case shown, also directly between them (i.e., without any further output times positioned in between). To illustrate this, the upper time axis 10 is also shown. Fig. 2 (i.e., at the position clock) the corresponding exposure interval 18 is entered.

[0101] To determine the actually relevant position value present at -2.3 ms (i.e., at the mean exposure time), which is also referred to as the principal position value within the scope of this disclosure, the position values ​​output at -3 ms and -2 ms are calculated together by the evaluation unit 220. This can, in particular, include averaging the corresponding position values.

[0102] For example, in the case shown, an averaging can be performed by multiplying the position value output at -2 ms by 0.7 and the position value output at -3 ms by 0.3, in order to then calculate the sum of the correspondingly weighted position values. The result represents the main position value for the time of -2.3 ms.

[0103] As a result, the evaluation unit 220 receives a sensor measurement signal or a distance value derived therefrom, which refers to the time of -2 ms, but was actually acquired at the actual acquisition time of -2.3 ms, as calculated or at least estimated according to the invention. Furthermore, the main position value is available, which refers to the time of -2.3 ms and was calculated in the manner described above (e.g., by the evaluation unit 220 itself). Thus, a position value in the form of the main position value can be assigned to the distance value, which represents the position value at the actual acquisition time of -2.3 ms much more precisely. By adding the main position value to the distance value, the accuracy in determining the coordinate of the sampling point or sampling area measured in this measurement cycle can be significantly improved. The addition can, for example,in the form that the scaled direction vector of the (known) beam direction of the sensor is added to the main position value, where the length of the vector is defined by the distance value of the sensor.

[0104] Analog time axes, as in [reference to relevant document], can be used to determine a subsequent sampling point or sampling range. Fig. Figure 2 is shown and considered. In particular, the evaluation time can again be set to 0 ms. A sensor measurement value obtained in this evaluation cycle can be assigned to a sensor clock interval 14, which ranges from -2 ms to -3 ms, given the minimum dead time of -2 ms, which is assumed to be inherent to the system due to the described evaluation and communication processes. Accordingly, the position values ​​that were available at these times (i.e., at -2 ms and -3 ms) can also be read from the storage unit 221.

[0105] Despite the identical time indications (see -2 ms and -3 ms), it is understood that the position values ​​and sensor readings in this subsequent evaluation cycle are different from those in the preceding cycle. More precisely, a currently available sensor reading (or, in other words, the most recently output sensor reading) is considered, and not the one that was already used to determine a coordinate in the previous evaluation cycle. Similarly, differing position values ​​are considered. In the latter case, in particular, considering current values ​​(or position values) can be achieved by reading the position values ​​sequentially (for example, from memory unit 221). If these have already been used for a coordinate evaluation, they are therefore not read again and can even be deleted.

[0106] Knowing the exposure time used to acquire the sensor measurement signal considered in the subsequent cycle, the mean acquisition time can be determined. This mean acquisition time can then be used, analogously to the calculation methods described above, to calculate a principal position value from the corresponding position values ​​(in particular, from the position values ​​output at times -3 ms and -2 ms). The principal position value and the sensor measurement signal, or a distance value derived from it, can then be added to determine the coordinate of the sampling point or sampling area measured in this measurement cycle.

[0107] In Fig. Figure 3 shows another example of the method according to the invention, which is shown with the arrangement 10 from Fig. 1 is feasible. Since this embodiment is essentially based on the above in connection with Fig. Based on the principles explained in section 2, only the essential differences to the variant from will be highlighted below. Fig. 2 received.

[0108] One can again see an upper time axis 10, which represents the position clock, according to which individual position values ​​are output (one position value after each position clock interval). Furthermore, a lower time axis 12 is again shown, which represents the sensor clock, according to which sensor measurement signals are output (one sensor measurement signal after each sensor clock interval). Both time axes start from a current evaluation time, which again corresponds to a time of 0 ms. Analogous to the previous example, a detection or exposure interval 18 is again considered, and based on this, a mean detection time is determined that corresponds to (or at least approximates) the actual detection time of the sensor measurement signal considered in this cycle. In the case shown, the mean exposure time is again -2.3 ms.

[0109] To determine the main position value, in this case, not only the position clock interval 16 containing the corresponding mean exposure time is considered. Instead, a system clock of the evaluation unit 220 is also taken into account. To limit the required computing power and, in particular, computing time, it has proven advantageous to use whole system clocks or integer multiples thereof for determining the main position value based on the mean acquisition time.

[0110] In the example shown, the clock duration of a single system clock interval 20 is 1 ms (i.e., the system clock has a frequency of 1,000 Hz). To determine the main position value, a determination interval 22 is defined in the case shown, which corresponds to a single system clock interval 20 and thus also has a clock duration of 1 ms. Furthermore, the determination interval 22 is defined such that it contains the mean acquisition time of -2.3 ms in the middle. Consequently, it lasts from -2.8 ms to -1.8 ms. As can be seen from Fig. As can be seen in Figure 3, it overlaps with two adjacent position intervals 16, with the first position interval 16 lasting from -3 ms to -2 ms and the second from -2 ms to -1 ms. Thus, there are three points in time that define the respective position intervals 16, which overlap with the determination interval 22. The position values ​​output at the corresponding times (-1 ms, -2 ms, and -3 ms) are read from the storage unit 221 by the evaluation unit 220 and used as the basis for determining the main position value. More precisely, the individual position values ​​are weighted again by averaging, taking into account the temporal relative relationship of the determination interval 22 to the corresponding times.

[0111] In the example shown, the following weighting is used: The position value output at -1 ms is multiplied by 0.2, the position value output at -2 ms is multiplied by 1, and the position value output at -3 ms is multiplied by 0.8. The sum of these weighted values ​​is then calculated and divided by two. The result is the primary position value, which can be added to the sensor measurement signal and / or the derived distance value, as described above, to determine a coordinate value of the object's surface or the object itself.

[0112] In Fig. Figure 4 shows a further embodiment of a method according to the invention, which again builds on the principles explained above and uses an arrangement 100 as shown in Figure 4. Fig. As shown in Figure 1, this is executable. The negative time axes originating from a current evaluation time of 0 ms are considered again, with the upper time axis 10 representing the position clock and the lower time axis 12 the sensor clock. It is immediately apparent that in this case the sensor clock differs from the position clock. More precisely, the frequency of the sensor clock is 333 Hz, while that of the position clock is 1,000 Hz. This allows for longer exposure times within a sensor clock interval 14, for example, to capture weakly reflective surfaces with sufficient accuracy.

[0113] Again, a constant dead time of 2 sensor cycles is assumed, which is why a sensor measurement signal is considered that was output at -6 ms and acquired in the interval between -9 ms and -6 ms. The exact acquisition time is again determined identically to the previous method from a given exposure time. In the case shown, this is 2.4 ms. The mean acquisition time is therefore -7.2 ms.

[0114] To calculate or at least estimate a position value at time -7.2 ms, a primary position value is generated again. This is done analogously to the example from Fig. 3. A determination interval 22 is defined, which in this case is an integer multiple of the system clock of 1 ms and therefore comprises two system clock intervals 20. More precisely, the duration of the determination interval 22 is 2 ms.

[0115] Analogous to the situation in connection with Fig.As explained in section 3, all position increment intervals 16 that overlap with the determination interval 22 are considered for determining the main position value. This includes the interval lasting from -9 ms to -8 ms, the interval lasting from -8 ms to -7 ms, and the interval lasting from -7 ms to -6 ms. Therefore, to determine the main position value, the position values ​​output at -9 ms, -8 ms, -7 ms, and -6 ms are considered. These can then be averaged, particularly by means of a weighting based on the temporal relative relationship of the determination interval 22 to the individual position increment intervals.

[0116] For the example shown, the position values ​​output at -8 ms and -7 ms can be multiplied by a factor of 1. The position value output at -6 ms can be multiplied by a factor of 0.8, and the position value output at -9 ms by a factor of 0.2. The weighted values ​​can be added together and then divided by three. From the primary position value determined in this way, a coordinate value of the object's surface or the object itself can be calculated, identical to the methods described above.

Claims

[1] Method for detecting an object using a movable sensor (215), comprising: Moving the sensor (215) relative to the object and repeatedly determining an instantaneous position of the sensor (215) by a position measuring system, wherein the determined position is output to an evaluation device (220) at a predetermined or determinable time, wherein the sensor (215) repeatedly detects the object during the duration of a detection time interval (18), which is an exposure time interval, and outputs a sensor measurement signal to the evaluation device (220) according to the information acquired during the detection time interval (18), at a predetermined or determinable time, whereby a distance value is determined on the basis of the sensor measurement signal in each case, where the duration of the exposure time interval is continuously adjusted according to the reflectance of the object, and wherein for each of the recording time intervals (18) the following is done: Determining a recording time that lies within the recording time interval (18); Determining position values ​​that were issued at times that include the recording time in between; Determining a principal position value based on the determined position values, which approximates the position assumed by the sensor (215) at the time of detection, and Calculating the positional main value and the distance value to determine a coordinate value of the object. [2] Method according to claim 1, wherein the detection time corresponds to the mean time of the detection time interval (18). [3] Method according to claim 1 or 2, wherein the detection time interval (18) corresponds to an exposure interval in which the sensor (215) designed as an optical sensor is exposed. [4] Method according to one of the preceding claims, wherein the detection time interval (18) has a predetermined relative ratio to an output time of the sensor measurement signal and / or a position value to the evaluation device (220). [5] Method according to any of the preceding claims, wherein a plurality of the output position values ​​are stored. [6] Method according to any of the preceding claims, wherein the main position value is formed by averaging the determined position values. [7] Method according to any of the preceding claims, wherein, for the purpose of determining the main position value, an additional position value is determined which was issued at a time that precedes or follows those position value issue times which include the recording time between them. [8] Method according to any of the preceding claims, further comprising: Assigning the positional main value to the sensor measurement signal for which the acquisition time was determined; and / or Determine at least one property of the object based on the main position value and the sensor measurement signal for which the detection time was determined. [9] Method according to one of the preceding claims, wherein the determination of the position principal value is carried out by an evaluation unit (220) operated according to a system clock as follows: Defining a determination interval (22) containing the time of acquisition, which has the duration of a single system clock interval or an integer multiple thereof; Determining position values ​​that were output at times which limit at least a time interval (16) that overlaps at least partially with the determination interval (22); Determining the main position value based on the determined position values. [10] Method according to one of the preceding claims, wherein a thickness measurement is carried out on the basis of the sensor measurement signal(s). [11] Arrangement (100) for detecting an object by means of a movable sensor (215), comprising: a sensor (215) that is movable relative to the object; a position measuring system configured to repeatedly determine an instantaneous position of the sensor (215), and an evaluation unit (220) to which the determined position is output by the position measuring system at a predetermined or determinable time, wherein the sensor (215) is configured to repeatedly detect the object during the duration of a detection time interval (18), which is an exposure time interval, and to output a sensor measurement signal to the evaluation device (220) in accordance with the information acquired during the detection time interval (18), at a predetermined or determinable time, whereby a distance value is determined on the basis of the sensor measurement signal in each case, where the duration of the exposure time interval is continuously adjusted according to the reflectance of the object, and wherein the evaluation unit (220) is configured to perform the following for each of the recording time intervals (18): Determining a recording time that lies within the recording time interval (18); Determining position values ​​that were issued at times that include the recording time in between; Determining a principal position value based on the determined position values, which approximates the position assumed by the sensor (215) at the time of detection, and Calculating the positional main value and the distance value to determine a coordinate value of the object.

Images