Method for distance measurement using a time-of-flight camera

Abstract

Method for measuring distance by means of a time-of-flight camera having an illumination and a time-of-flight sensor, where a distance is determined from a phase shift of a modulated emitted and received light, whereby at least two individual measurements are carried out to measure the distance, wherein a modulation signal is formed by replacing bit values ​​of a maximum sequence by subbit sequences according to a substitution rule, wherein the modulation signal for the illumination differs from the modulation signal of the time-of-flight sensor in at least one individual measurement in the subbit sequence, where a uniqueness range is determined via the length of the maximum sequence and the modulation frequency, and wherein a measuring range within the uniqueness range is defined via the sub-bit sequence provided for the time-of-flight sensor and / or illumination, and wherein a shift of the measuring range can be adjusted via a bitwise and / or analog cyclic shift of the illumination and / or sensor signal, where in successive individual measurements a different subbit sequence.

Description

The invention relates to a method for distance measurement and to a time-of-flight camera suitable for carrying out the method according to the preamble of the independent claims.The time-of-flight camera according to the invention and the method according to the invention relate to time-of-flight cameras which determine distances from a phase shift of emitted and received modulated radiation. In particular, PMD cameras with a photomix detector are suitable for this purpose, as disclosed, for example, in DE 197 04 496 A1.As set forth in DE 10 2018 131 580 A1, for example, maximum length sequences (MLS) or maximum sequences are particularly well suited for the modulation of an iToF camera because of their special correlation properties. Due to its odd length (different numbers of "0" and "1", however, a suitable substitution or a suitable spreading code for the bits ("0" and "1") of the base maximum sequence is required for the modulation of an iToF camera in order to achieve a uniform distribution of charge carriers in the time of flight pixels during the demodulation.For a phase measurement with an iToF camera, at least two independent individual measurements are required. As a rule, even four, less frequently three, individual measurements are used in order to compensate for asymmetry caused by hardware as well as possible. This is particularly successful in the case of four individual measurements, since the individual measurements are respectively inverted in pairs, with the result that asymmetry is compensated directly. In principle, however, this is also possible with three individual measurements.Furthermore, DE 10 2018 131 580 A1 describes how a phase or distance measurement can be carried out on the basis of maximum sequences with the aid of suitable substitution rules or with the aid of specific spreading codes, and how the length of the measurement range can be influenced and the measurement range can be shifted.The modulation of iToF cameras is usually carried out with simple, rectangular or sinusoidal pulse patterns. Due to the resulting continuous periodic correlation signal, this is generally referred to as continuous wave (CW) modulation.U.S. Pat. No. 9,778,363 B2 discloses the use of maximum sequences for modulating iToF cameras. In this case, the measurement range is successively shifted in order to selectively record different regions of a scene. For modulation, however, the maximum sequence is used in its original form, i.e. without the use of substitutions or without code spreading.Also in U.S. Pat. No. 11 393 115 B2 and in U.S. Pat. No. 11 668 829 B2, maximum sequences are used, among other things, for modulating a PMD camera. However, the maximum sequences only assist in the validation of the measurement range. Here, too, no substitution is used. The use of modulation schemes for creating sequences for sensor and illumination modulation with specific measurement properties will be explained below.In order to measure a distance using a PMD camera, two to four individual measurements are carried out in each case. In the modulation with periodic sine or square wave signals, a continuous, periodic correlation signal results. This uniformity leads to objects always producing a measurement signal regardless of their distance, the amplitude of which only varies slightly and decreases with increasing distance, while the phase position continuously repeats itself (>2π periodicity) in proportion to the period duration of the modulation signal. Thus, unique distance measurements can only be carried out within a closely limited distance range which is defined as a uniqueness range (engl. Unambigual Range) (see below).They show schematically FIG. 1 shows a measurement and uniqueness range for a continuous wave (CW) modulation with a modulation frequency of 120 MHz FIG. 2 shows a reference between the substitution codes according to Table 1 for a bit with a value "1" for the sensor modulation of the four individual measurements 0-3 by successive left-hand shift by one bit each. FIG. 3 shows a measurement and uniqueness range for a substituted maximum sequence modulation with a modulation frequency of 120 MHz. FIG. 4 shows a measurement range validation by checking the ratio between the signal strength of the maximum sequence-modulated measurement signal and the signal strength of a static measurement signal.FIG. 1 shows, by way of example, a continuous wave (CW) modulation with a modulation of 120 MHz. In the case of CW modulation, measurement range MB and uniqueness range EB are identical. Distances that exceed the measurement range MB or uniqueness range EB do indeed provide a distance value but an incorrect distance value.Furthermore, four correlation functions of a CW range finding are shown. Assuming a modulation frequency of 120 MHz, the measurement range MB extends from 0 to 1.25 m. For all longer distances, the CW signal repeats and does not allow a clear measurement. For a later comparison with the procedure according to the invention, an ambiguity range MDB to 38.75 m was shown. <row><cell>Dist.ber.< / cell><cell>Ratio:< / cell><cell>Measurement Characteristic< / cell>< / row><row><cell>0 -1.25 m< / cell><cell>EB_CW< / cell><cell>MB=EB, noise such as CW< / cell>< / row><row><cell>1,15 - 38.75 m< / cell><cell>Remainder: Remainder:< / cell><cell>MDB - Invalid Measurements< / cell>< / row><p xml:id="_6b758c0039" n="0013">If, on the other hand, a PMD camera is modulated, as shown in FIG. 3, with a sequence which is based on a maximum sequence, the periodicity of the correlation signal can be greatly reduced and thus a significantly larger uniqueness range EB can be achieved. The length of the uniqueness range EB increases in this case in proportion to the length of the maximum sequence.<p xml:id="_6b758c0040" n="0014">Each additional bit of the maximum sequence increases the uniqueness range EB by c / 2f <hi rend="subscript">mod< / hi>, where c denotes the speed of light and f <hi rend="subscript">mod< / hi> denotes the modulation frequency. This is the distance which light covers during half the period duration for an unsubstituted, i.e. non-spread, maximum sequence bits, that is to say one bit of the maximum sequence base sequence. However, measurements are not possible in the entire uniqueness range EB, but only in a limited distance range, which is referred to as measurement range MB, while objects outside the measurement range MB do not produce a signal. This region is referred to as a blanking or also blind region BB. The regions in which the blind region BB merges into the measurement region MB and the measurement region MB merges into the blind region BB can be referred to as transition regions ÜB. Thus, the range of uniqueness is divided into the following three ranges: 1) Measurement Range (MB) in which valid phase measurements can be made. This generally comprises a similar distance range as the uniqueness range in CW modulation, i.e. c / 2f <hi rend="subscript">mod< / hi>, but can also be smaller or larger-depending on the chosen substitution scheme. 2) Masking or blind area (BB) in which no signal is present and objects are thus masked out. This makes up the majority of the uniqueness range EB. 3) transition regions (ÜB) between blind BB and measurement region MB or vice versa, in which signal is present, but no measurement is possible in these regions on account of missing or coherent correlation signals of the individual measurements. These regions cannot be completely eliminated because of limited edge steepness of the correlation signals, but as a rule should be as short as possible.<p xml:id="_6b758c0044" n="0015">With the aid of different substitution rules or suitable spreading codes, the correlation properties in the measurement range MB and in the transition ranges ÜB can be controlled. This relates substantially to the length of the measurement and transition regions MB, UMand to the shape of the correlation signal (in particular signal strength, edge slope) within these regions.<p xml:id="_6b758c0045" n="0016">A suitable form of a substitution rule is explicitly shown in Tables 1 and 2 for a distance measurement consisting of four individual measurements:<list xml:id="_6b758c0046" type="bulleted"><item>Table 1 below shows a substitution of a maximum sequence with 4 bits.< / item>< / list>Table 1<title desc="title">Table 10000001001011011000011010011000000011110011000 111000011200000011010011100010010110300000110000111100000111100As shown, for the individual measurements, the subbit strings for PMD gate B are the inverse of the subbit strings for PMD gate A. The PMD gates are modulation gates via which the modulation signal is introduced as potential into the PMD or time of flight sensor.For all individual measurements, for both gates, the subbit sequence for MLS bit "0" is the inverse of the subbit sequence for MLS bit "1.".Further, all sub-bit sequences of the modulation gates have the same number of "0" and "1" bits.While the substitution for the illumination signal is identical for each individual measurement, the substitution for the sensor code differs for each individual measurement. Analogously to CW modulation, the sensor code or the substitution for a specific individual measurement can be derived from that of the respective preceding individual measurement simply by a cyclic left shift by one bit in each case. Thus, as in CW modulation, the sensor codes for the individual measurements #0 and #2 and #1 and #3 are respectively inverted with respect to one another.FIG. 2 shows a reference between the substitution codes according to Table 1 for a bit with a value "1" for the sensor modulation of the four individual measurements 0-3 by successive left-hand shift by one bit each.The complete modulation sequence is generated by arranging a plurality of these substitution codes in succession in accordance with the basic maximum sequence on which it is based. This will be explained below with reference to a simple exemplary basic maximum sequence: "1 1 0" consisting of three bits. Using the substitution rule mentioned above, the modulation sequences shown in Table 2 for the illumination and for the sensor modulation for the four individual measurements in each case are produced therefrom: Table 2: Table 2:01000 1000 00000110 0110 10011001 1001 011011000 1000 00001100 1100 00110011 0011 110021000 1000 00001001 1001 01100110 0110 100131000 1000 00000011 0011 11001100 1100 0011For a modulation frequency of 120 MHz, the arrangement of the various regions EB, BB, ÜB, MB shown in FIG. 3 results when using this substitution rule. The measurement range MB together with the transition ranges ÜB can be shifted as desired within the uniqueness range EB, and the blind range BB is shifted accordingly. The step size of the shift in the case of the 4-bit substitution specified here corresponds to a 1⁄4 of the CW uniqueness range EB CW, which is given by the underlying modulation frequency. <row><cell>0 -0.31 m< / cell><cell>1 / 4EBcw< / cell><cell>BB - no signal< / cell>< / row><row><cell>0,31 - 0.63 m< / cell><cell>1 / 4EBcw< / cell><cell>SU - No valid distance values< / cell>< / row><row><cell>0,63 - 1.56 m< / cell><cell>3 / 4EBcw< / cell><cell>MB - Measurement range< / cell>< / row><row><cell>1,56 - 1.88 m< / cell><cell>1 / 4EBcw< / cell><cell>SU - No valid distance values< / cell>< / row><row><cell>1.88 - 38,75< / cell><cell>Remainder EB< / cell><cell>BB - no signal< / cell>< / row><p xml:id="_6b758c0171" n="0024">We will now consider the displacement of the measurement range MB within the uniqueness range EB. As has already been described in the preceding section, the measurement range in the case of maximum sequence modulation generally comprises only a small part of the uniqueness range EB, namely ≈c / 2f <hi rend="subscript">mod< / hi>, i.e. that distance range which the light travels within half the period duration (>in and return path of the light) of an unsubstituted maximum sequence bit, i.e. of a bit of the maximum sequence base sequence. However, the measurement range MB can be freely shifted within the uniqueness range EB in a simple manner. This is done by a relative (circular) shift (rotation) of the modulation codes (as shown above) for illumination and / or image sensor. The granularity of the measurement range shift depends on the length of the substitution code. The greater the bit length of the substitution code, the smaller the step size of the shift and the more finely granular the measurement range MB can thus be shifted within the uniqueness range EB. Generally, the granularity of the measurement range shift is 1 / n <hi rend="subscript">Subbit< / hi>.<p xml:id="_6b758c0174" n="0025">By using an analog phase shift between sensor and illumination, the restriction of this granularity can also be bypassed, provided the module provides this capability.<head xml:id="_6b758c0175">Measurement Area Check< / head><p xml:id="_6b758c0176" n="0026">Within the transition ranges ÜB already described, there is generally the problem that objects, although causing a signal, cannot be valid phase measurement there because of missing or coherent signals of the individual measurements. In these ranges, false measurements occur. In order to avoid this, pixels within these transition regions ÜB must be reliably detected. In addition to restricting the permitted phase range to valid values, a further criterion for this purpose is to check the ratio between the amplitude of the measurement signal modulated with a substituted maximum sequence and a reference amplitude, wherein the reference amplitude is obtained, for example, from the amplitude of a continuously modulated measurement signal (CW) or of an unmodulated measurement signal (intensity measurement). The amplitude is determined in a known manner from unmodulated illumination, it is advantageous if a portion of extraneous light is subtracted by a further measurement.<p xml:id="_6b758c0177" n="0027">If a pixel is within the measurement range MB, this ratio exceeds a specific threshold value. This is suitable for measurement range validation.<p xml:id="_6b758c0178" n="0028"> FIG. 4 shows a measurement range validation by checking the ratio between the signal strength of the measurement signal modulated with a substituted maximum sequence and the signal strength of a static measurement signal. Table 3 shows a more complex substitution rule, in which, in contrast to the above, 8 sub-bits are substituted instead of 4 sub-bits.<title desc="title">Table 3 shows a more complex substitution rule, in which, in contrast to the above, 8 sub-bits are substituted instead of 4 sub-bits.1[0][00000000][11000011][00111100]0[1][01110000][00111100][11000011]2[0][00000000][00001111][11110000]0[1][01100000][11110000][00001111]3[0][00000000][00111100][11000011]0[1][11100000][11000011][00111100]4[0][00000000][11110000][00001111]0[1][01100000][00001111][11110000]Such substitution results in a measurement range MBcorresponding to 5 / 4 of the CW uniqueness range EB CW.References included in the specificationThis list of documents cited by the applicant has been produced in an automated manner and is only included for the better information of the reader. The list is not part of the German patent application or utility model application. The DPMA does not take any adhesion for any faults or omissions.Patent Literature citedDE 197 04 496 A1

[0002] DE 10 2018 131 580 A1 [0003, 0005]U.S. Pat. No. 9,778,363 B2

[0007] U.S. Pat. No. 11 393 115 B2

[0008] U.S. Pat. No. 11 668 829 B2

[0008]

Claims

Method for distance measurement by means of a time-of-flight camera having an illumination and a time-of-flight sensor, wherein a distance is determined from a phase shift of a light emitted and received in a modulated manner, wherein at least two individual measurements are carried out for the distance measurement, wherein a modulation signal is formed in that bit values of a maximum sequence are replaced by subbit sequences according to a substitution rule, wherein the modulation signal for the illumination differs from the modulation signal of the time-of-flight sensor in at least one individual measurement in the subbit sequence, wherein a uniqueness range is determined over the length of the maximum sequence and the modulation frequency, and wherein a measurement range within the uniqueness range is defined via the subbit sequence provided for the time-of-flight sensor and / or illumination, and wherein a shift of the measurement range can be adjusted via a bit-wise and / or analog cyclical shift of the illumination signal and / or sensor signal, wherein a different subbit sequence has in successive individual measurements.Method according to Claim 1, in which, for the validation of a distance measurement, an amplitude of a distance measurement is compared with a reference amplitude, and wherein a distance measurement is evaluated as valid if a comparison preferably exceeds a limit value by a quotient of the amplitudes.Method according to one of the preceding claims, in which the modulation signals are conducted to a first and second modulation gate of the time of flight sensor, wherein for all individual measurements the subbit sequences for the first and second modulation gates are inverse to one another.A method according to any preceding claim, wherein all subbit sequences for the modulation gates have the same number of "0" and "1" bits.Time-of-flight camera, having illumination for emitting a modulated light and having a time-of-flight sensor for receiving the modulated light emitted and reflected by a scene, wherein distances are determined from a phase difference between the emitted and received light, wherein the time-of-flight camera is designed to carry out one of the aforementioned methods.

Images