EP4103962A1

EP4103962A1 - Method for analysing backscattering histogram data in an optical pulse delay method and data processing device

Info

Publication number: EP4103962A1
Application number: EP21702999.0A
Authority: EP
Inventors: Ralf Beuschel; Falko Diebel
Original assignee: Ibeo Automotive Systems GmbH
Current assignee: Microvision Inc
Priority date: 2020-02-11
Filing date: 2021-02-01
Publication date: 2022-12-21
Also published as: CA3167506A1; KR20220119500A; DE102020201637A1; IL295469A; US20230048750A1; JP2023513560A; CN115104038A; WO2021160454A1; JP7471018B2

Abstract

The invention relates to a method (20, 30, 40, 50) for analysing backscattering histogram data in an optical pulse delay method, comprising: receiving (21, 31, 41, 51) backscattering histogram data; and analysing (22, 32, 42, 52) the received backscattering histogram data.

Description

Method for analyzing backscatter histogram data in an optical pulse transit time method and device for data processing

The present invention relates generally to a method for analyzing backscatter histogram data in an optical pulse transit time method and a device for data processing.

In general, various optical pulse transit time methods (z. B. optical transit time measurement) are known that are based on the so-called time-of-flight principle, in which the transit time of a light signal emitted and reflected by an object is measured by the distance to determine the object based on the runtime.

It is known to use sensors in the motor vehicle environment which are based on the so-called LIDAR principle (Light Detection and Ranging), in which pulses are periodically transmitted to scan the environment and the reflected pulses are detected. A corresponding method and a device are known, for example, from WO 2017/081294.

In LIDAR applications in the motor vehicle environment, the amount of ambient light can be relatively large under certain environmental conditions, e.g. when driving during the day, which reduces the signal-to-noise ratio (also known as SNR) can. In such situations, the detection range of the LIDAR application can also be limited.

In general, the type of light signals detected can differ in LIDAR applications, e.g. B. depending on whether the transmitted light signal is reflected on a solid object (object backscattering) or is backscattered by airborne particles (diffuse backscattering), such as in fog or in Abga sen. Conclusions about the ambient conditions can be drawn from the recorded backscatter data. Even if solutions for the analysis of backscatter data in an optical pulse transit time method are known from the prior art, it is an object of the present invention to provide a method for the analysis of backscatter histogram data in an optical pulse transit time method and to provide a device for data processing.

This object is achieved by the method according to claim 1 and the device according to claim 15.

According to a first aspect, the present invention provides a method for analyzing backscatter histogram data in an optical pulse time-of-flight method, comprising:

Receiving backscatter histogram data; and analyzing the received backscatter histogram data.

According to a second aspect, the present invention provides an apparatus for data processing, comprising means for carrying out the method according to the first aspect.

As mentioned, some exemplary embodiments relate to a method for analyzing backscatter histogram data in an optical pulse time-of-flight method, including:

As stated at the beginning, conclusions can be drawn about ambient conditions (e.g. fog or other particles in the air, smoke, spray, etc.) from backscatter data from LIDAR measurements. The detection events of the backscattered light are not related to solid objects. With a more precise knowledge of the ambient conditions, e.g. in the case of autonomous vehicles, the driving style can be adapted to the ambient conditions, thus increasing safety. Furthermore, precise knowledge of the diffuse backscattering in LIDAR measurements also allows (more precise) detection of solid objects in some exemplary embodiments. In this way, for example, traffic situations can be determined more precisely, which also increases the safety and reliability of autonomous vehicles.

A more precise knowledge of the backscatter signal and an amount of ambient light can also contribute to an effective detection range of a LIDAR measurement to determine. This makes it easier to assess whether the detection of solid objects is reliable. This increases the probability of correctly identifying an object from the measurement data, which also increases the safety and reliability of autonomous vehicles.

Therefore, in some exemplary embodiments, the method is used for the analysis in a LIDAR system or the like and is used, for example, in the motor vehicle environment, without the invention being restricted to these cases.

In some exemplary embodiments, LIDAR data typically contain signal contributions from diffuse backscattering, light reflection on objects, ambient light, interfering light signals from other light sources in the vicinity and the like. This data can be presented in a histogram, which is known in principle.

Correspondingly, the analysis of backscatter histogram data in some exemplary embodiments can mean that the signal contributions of the diffuse backscatter, the amount of ambient light and the effective detection area are determined from these data during an optical time of flight measurement (optical pulse time of flight method). The backscatter histogram data can therefore be suitable for such an analysis and determination, since they can in principle contain the signal contributions of the diffuse backscatter and the amount of ambient light.

In some exemplary embodiments, the optical transit time measurement is based on the so-called TCSPC (time correlated single photon counting) measurement principle, in particular in exemplary embodiments which are based on LIDAR. Light pulses are periodically emitted, which are typically a few nanoseconds long and mark a starting point in time for a measurement. During the time until the next light pulse (measurement time), the light reflected from objects or backscattered light is detected by a light-detecting receiving element (e.g. a single photon avelance diode (SPAD)), with light also being used in a short period of time before the light pulse is emitted can be detected. The measurement time is divided into a large number of short time intervals (e.g. 500 ps). A point in time can then be assigned to each time interval, which corresponds to a time interval from the start time (for example, in the case of time intervals of 500 ps, a first time interval can be used a point in time of 250 ps can be assigned and a point in time of 750 ps can be assigned to a second time interval, etc.).

Depending on the distance to the object or to the point of backscattering, the light reaches the light-detecting receiving element at different times. Since it generates an electrical signal in the light-detecting receiving element. With the help of a time-to-digital converter (also known as "TDC"), which is known in principle, the electrical signal can then be assigned to one of the time intervals. By counting the electrical signals ("Events "), which are assigned to a time interval, result in so-called histograms or time-correlated histograms (also called TCSPC histograms), with these histograms, for example, also only being available as pure data and, for example, as value pairs from the time interval and the associated number of entries ( Events) are stored. The time intervals together with the number of events assigned to each time interval accordingly form histogram data that can basically be represented by digital signals (or also analog signals).

With LIDAR measurements based on TCSPC, histogram data of the light reflected or scattered back can therefore be output with a high time resolution. This can correspond to an analog-to-digital conversion of the light power reflected or scattered back as a function of time or distance. In some exemplary embodiments, time-correlated histogram data are data that are generated based on the electrical signals of the light-detecting receiving elements within the (associated) measurement time and shortly before it. These therefore typically contain signal contributions from diffuse backscattering, light reflection on objects, ambient light, interfering light signals from other light sources in the vicinity and the like.

In some embodiments, the backscatter histogram data correspond to an accumulation of time-correlated histogram data from a plurality of light-detecting receiving elements. The accumulation of time-correlated histogram data can be advantageous for the determination of various parameters of the measurement (analysis of the backscatter histogram data), since the signal-to-noise ratio (also called "SNR", signal-to-noise ratio) of the diffuse backscatter contribution and the contribution of the ambient light compared to other signal contributions (e.g. Reflection on objects or interfering light sources) can be increased. As a result, the analysis of the backscatter histogram data can be carried out better in some exemplary embodiments.

In some exemplary embodiments, this follows from the fact that an accumulation of several time-correlated histogram data (backscatter histogram data) can smear the reflection of objects (signal contributions from objects) at different distances and / or in different regions of the field of view of the LIDAR measurement. In such exemplary embodiments, the objects are typically only present in a narrow area of the field of view, the field of view typically describing an area of space that is being detected. In contrast, the contributions of diffuse backscatter and ambient light are typically similar over the entire field of view of the LIDAR system.

The ambient light is also usually constant during the measurement time and thus typically makes a constant contribution in all time intervals. Likewise, the signal contributions from reflections on objects are often sharp peaks, which means that the reflected light is only detected in one or a few time intervals because the light pulse can be received with a weakened amplitude but almost the same pulse duration. With typical pulse durations of z. B. 10 ns can e.g. B. a 250 ps Zeitauf solution are required for an accurate location.

With diffuse backscattering e.g. Fog or particles in the air can cause continuous backscattering during the light propagation with low intensity. The light pulse can be very strongly expanded or smeared over time. At z. B. a 10 ns light pulse with a geometric extent of 1.5 m, diffuse backscattering over a 1.5 m depth range can be generated at any time. Therefore, in some exemplary embodiments, a significantly reduced time resolution is sufficient.

In some exemplary embodiments, the backscatter histogram data are generated by one or more histogram accumulation units and made available for analysis. The histogram accumulation unit has several signal inputs. The histogram accumulation unit receives time-correlated histogram data at the or each signal input. It doesn't always have to be on everyone Signal input the histogram data are received and in some exemplary embodiments there are also further signal inputs at which, for example, no histogram data or only after a corresponding configuration are received. Backscatter histogram data are generated based on the time-correlated histogram data received at the signal inputs.

In some exemplary embodiments, the maximum number of histogram accumulation units is given by the number of light-detecting receiving elements in a system for optical transit time measurement (e.g. LIDAR system). The histogram accumulation unit can in principle be or have an electronic circuit or an electronic circuit that receives (the) digital signals or data, such as the time-correlated histogram data, via the signal inputs and the generation of backscattering described here. Executes histogram data. The electronic circuit may include electronic components, digital storage elements, and the like to perform the functions described herein. The electronic circuit can be implemented by an FPGA (Field Programmable Gate Array), DSP (digital signal processor) or the like. In other exemplary embodiments, the histogram accumulation unit is implemented by a memory and a microprocessor. In further exemplary embodiments, the histogram accumulation unit is implemented by software, the signal inputs in such exemplary embodiments corresponding to the parameters / attributes of a software function / method. The generation of the backscatter histogram data then corresponds to the execution of a sequence of commands for the execution of certain arithmetic operations on a computer, so that backscatter histogram data are available after all commands have been processed. In some exemplary embodiments, the histogram accumulation unit is also implemented by a mixture of hardware- and software-based components, to which the functionalities described herein are appropriately distributed.

In some exemplary embodiments, the histogram accumulation unit generates the backscatter histogram data by adding the received time-correlated histogram data.

The number of events that were detected in a time interval ("bin") can be from all received time-correlated histogram data are added, so that the backscatter histogram data are generated which in each time interval just contain the sum of all events in this time interval. The time-correlated histogram data are preferably accumulated or added as integer numbers in order to make a weak, diffuse backscattering measurable in some exemplary embodiments. This is advantageous because the SNR of the diffuse backscatter contribution can be increased compared to other contributions.

In some exemplary embodiments, the histogram accumulation unit calculates an arithmetic mean from the received time-correlated histogram data in order to generate the backscatter histogram data.

The time-correlated histogram data received are added and divided by the number of signal inputs. This can be advantageous in some exemplary embodiments which have a fixed-point number or floating-point number implementation (in contrast to exemplary embodiments, which accumulate integers).

In some exemplary embodiments, the histogram accumulation unit accumulates the received time-correlated histogram data from a plurality of time intervals in a time interval in order to generate the backscatter histogram data.

In some exemplary embodiments, the histogram accumulation unit is further set up not to take into account received time-correlated histogram data from time intervals that are above a specific time threshold for the generation of the backscatter histogram data.

In some exemplary embodiments, the histogram accumulation unit is further set up to weight the received time-correlated histogram data for generating the backscatter histogram data.

In some exemplary embodiments, the histogram accumulation unit is further set up to output the backscatter histogram data for determining the backscatter. The backscatter histogram data can then, for example, be output to a processor, FPGA or the like to determine the backscatter. A receiving system that is used for optical transit time measurement, such as, for example, a LIDAR system, can in some exemplary embodiments have a receiving matrix with a plurality of light-detecting receiving elements, each of the light-detecting receiving elements being set up to detect light and, in response to it, an electrical light Generate signal.

In some exemplary embodiments, each of the light-detecting receiving elements can be activated and deactivated. In some exemplary embodiments, the light-detecting receiving elements are arranged in columns and rows in the receiving matrix (as is fundamentally known), with some exemplary embodiments being provided in each row with the same number of light-detecting receiving elements without loss of generality.

In some exemplary embodiments, the device comprises several evaluation units, one evaluation unit being connected to the light-detecting receiving elements in a column or one evaluating unit being connected to the light-detecting receiving elements in a row.

In some exemplary embodiments, each of the evaluation units is set up to generate the time-correlated histogram data based on the electrical signals of the light-detecting receiving elements.

In some exemplary embodiments, only those light-detecting receiving elements that are activated are taken into account for generating the time-correlated histogram data.

In some exemplary embodiments, each signal input of a histogram accumulation unit is connected to one of the evaluation units, so that the time-correlated histogram data are transmitted from the evaluation unit to the corresponding histogram accumulation unit.

In the method for analyzing backscatter histogram data in an optical transit time measurement, backscatter histogram data are initially received. These backscatter histogram data are from the histogram or histogram Accumulation units have been generated.

The received backscatter histogram data is analyzed. Analyzing can be a calculation or a sequence of calculations in order to determine various parameters (backscatter signal, amount of ambient light, effective detection range, etc.) of the optical transit time measurement. The calculation takes the backscatter histogram data as input values for mathematical operations such as arithmetic averaging or applying a predetermined function or the like.

The analysis of the received backscatter histogram data can in principle be carried out by a processor, an FPGA, DSP or the like. In such exemplary embodiments, the analysis is implemented by software. The analysis of the backscatter histogram data then corresponds to an execution of a sequence of commands for the execution of specific arithmetic operations on a computer, so that the backscatter histogram data are analyzed after all commands have been processed. In other exemplary embodiments, a specific electronic circuit with corresponding electronic components can be used for analyzing the backscatter histogram data. In some exemplary embodiments, the analysis of the backscatter histogram data is implemented by a mixture of hardware- and software-based components, over which the method described herein is appropriately distributed. The above-mentioned exemplary embodiments can be exemplary embodiments of a device for data processing which can additionally contain storage elements for data storage.

The amount of light that is detected due to diffuse backscattering is often small compared to the amount of ambient light, e.g. in daylight, and the amount of light reflected from objects, so that a determination of the backscattering can be difficult and imprecise. Therefore, the method for analyzing backscatter histogram data, which correspond to accumulated time-correlated histogram data, is used in some exemplary embodiments for the determination of a backscatter signal in an optical transit time measurement.

Typically, the diffuse backscattering is higher at short distances (e.g. 5 m) than with long distances (e.g. 200 m) and can drop continuously. The diffuse backscattering in an optical transit time measurement can therefore have a typical signal shape in some exemplary embodiments, which can have a maximum of the backscattered light at short distances and quickly drops for longer distances. In such exemplary embodiments, the backscatter signal then correlates in the backscatter histogram data with the typical signal shape. This is advantageous because the backscatter signal can be identified on the basis of the signal shape over short distances. However, in such exemplary embodiments, when an object is present at short distances, the backscatter signal often cannot be determined either.

Consequently, in some exemplary embodiments, a similarity measure, such as e.g. B. a correlation is calculated between the received backscatter histogram data and a predetermined reference backscatter signal in order to determine a backscatter signal.

The correlation can be a measure of the similarity between two or more temporal or spatial signal curves or for a statistical relationship between the signal curves. In some exemplary embodiments, the correlation is calculated using a correlation integral, which is known in principle.

The predefined reference backscatter signal can correspond to a typical signal shape of the backscatter, wherein the typical signal shape of the backscatter can differ in the various exemplary embodiments. In some exemplary embodiments, the predefined reference backscatter signal can be present as histogram data in a memory which is accessible for the analysis of the backscatter histogram data, for example by a processor. In other exemplary embodiments, the predefined reference backscatter signal can be calculated dynamically (for example at the required time during the analysis) from a predefined function.

The typical signal shape of the backscatter can be determined, for example, by searching for a characteristic peak or a sequence of peaks at short distances, the position, signal shape and intensity being evaluated. For example, in embodiments with a LIDAR system for optical transit time measurement Solution, which has a parallax between a transmitter (from which light pulses are emitted) and a receiver (e.g. a receiving matrix with several light-detecting receiving elements), the backscattering has a system-dependent signal form and the position of the peak. In such exemplary embodiments, it can be checked whether the intensity of the peak corresponds to an object or not in order to establish the typical signal shape. Typical reference backscatter signals are illustrated, for example, in FIGS. 1, 2 and 3, which are described in more detail below. For example, a distinction can be made between single-beam and multi-beam LIDAR systems (single-beam systems only emit one light pulse (beam) at the same time and multi-beam systems can emit several light pulses from different positions at the same time, such as the one in DE 10 2017 222 971 A1 described Ll-DAR system).

In other exemplary embodiments, the reference backscatter signal can be determined experimentally by simulating various environmental conditions and measuring the typical signal shape, position and intensity of the backscatter.

In further exemplary embodiments, the reference backscatter signal can be determined in that a correlation integral (cross-correlation) is calculated between an expected backscatter signal and the received backscatter histogram data. The level of correlation can be used in such execution examples to assess whether the expected backscatter signal can be used as a reference backscatter signal. Several expected backscatter signals can be tested and the level of correlation can be compared to determine a reference backscatter signal for the specific system.

In exemplary embodiments in which the correlation between the received backscatter histogram data and the specified reference backscatter signal is calculated by a correlation integral, a backscatter signal that varies over the measurement time (and shortly before it) results, with the amplitude of the calculated backscatter signal being the backscattered signal Light output corresponds to (also abbreviated to AB in the following). Therefore, in such embodiments, the backscatter can be determined from the backscatter histogram data.

In some exemplary embodiments, the method for analyzing backscatter ungs histogram data is used for determining an amount of ambient light from the received backscatter histogram data.

The amount of ambient light (also abbreviated as AL in the following) basically corresponds to the signal contribution, for example in a LIDAR system, which is present and detected, for example, by sunlight or street lamps, regardless of the emitted light pulse. Furthermore, the amount of ambient light can also contain components of electronic noise from the receiver, which is temperature-dependent. However, in such exemplary embodiments, the effect of contributions to the amount of ambient light by ambient light and by noise is the same and is therefore not differentiated.

During the measurement time (and shortly before it) the ambient light is usually constant and thus typically makes a constant contribution in all time intervals. Likewise, the signal contributions from reflections on objects are often sharp peaks, which means that the reflected light is only detected in one or a few time intervals. Therefore, in some exemplary embodiments, the amount of ambient light can be determined from the received backscatter histogram data from several time intervals that are shortly before the emission of the light pulses (start time). In other exemplary embodiments, the amount of ambient light can be determined from the received backscatter histogram data from several time intervals which correspond to a long distance (as long as no objects are present in long distances). In further exemplary embodiments, the amount of ambient light can be determined from a combination of the above methods. In some exemplary embodiments, the method for analyzing the received backscatter histogram data is used to determine an effective detection range of the optical transit time measurement based on the backscatter signal and the amount of ambient light.

An effective detection area (also abbreviated to EDR in the following) of an optical time of flight measurement can correspond to a distance at which signals that arise from reflection on solid objects can still be clearly detected and assigned. The effective detection area is basically referenced to an object with a given reflectivity with a given amount of ambient light and given constant probability of light detection. In some exemplary embodiments, the effective detection range can correspond to an absolute value (for example 100 m). In other exemplary embodiments, the effective detection range can correspond to a relative value, wherein in such exemplary embodiments the effective detection range is related to a nominal detection range which was determined for the above reference values.

The effective detection range can be reduced by the fact that the backscatter is high, since in such exemplary embodiments the light power of the emitted light pulse is attenuated by the backscatter as the distance increases. Consequently, the light output that is available for reflection on the solid objects is lower than in embodiments with less backscattering, so that the reflected light output is lower and this is further reduced on the way to the receiver.

In exemplary embodiments with a high amount of ambient light, the effective detection area can be reduced by reducing the SNR, since the amount of ambient light basically makes a contribution to the level of noise. In other exemplary embodiments, the effective detection area is increased when the amount of ambient light is low, since this increases the SNR. The effective detection range of the optical transit time measurement can therefore be determined from the received backscatter histogram data based on the backscatter signal and the amount of ambient light.

In some exemplary embodiments, a transformation function is applied to the backscatter signal in order to determine a signal attenuation factor.

As stated above, a high backscatter can reduce the effective detection range by attenuating the light power of the emitted light pulse with increasing distance. Therefore, the backscatter signal can be used to determine the amount of such attenuation.

Consequently, a transformation function is applied to the backscatter signal, it being possible for the transformation function in some exemplary embodiments to be a predetermined (mathematical) function that results from the backscatter signal Signal damping factor calculated. In other exemplary embodiments, the transformation function can be a sequence of calculations. The transformation function can be determined experimentally or gained from experience. In some exemplary embodiments, the experimental determination can take place beforehand in the development and key figures and / or characteristic curves etc. corresponding to the transformation function can be stored in software, for example.

In some exemplary embodiments, the signal attenuation factor can then be a distance-dependent percentage reduction in the light output caused by the backscatter.

The transformation function was therefore determined experimentally in some exemplary embodiments. The signal attenuation can, for example, have been measured under different ambient conditions, whereby a transformation function can be found that determines the signal attenuation factor from the backscatter signal (which, as explained above, can also be determined experimentally), which corresponds well to the measured values of the signal attenuation.

In some exemplary embodiments, an arithmetic mean is calculated from the received backscatter histogram data from a plurality of time intervals that lie before a start time in order to determine the amount of ambient light.

The start time is the time at which the light pulse for determining the distance from solid objects is emitted. Typically, only the ambient light can be detected shortly before the light pulse is emitted (e.g. 20 ns), so it is advantageous for determining the amount of ambient light in some exemplary embodiments to take into account those time intervals of the received backscatter histogram data that are before the start time. Furthermore, the calculation of an arithmetic mean from several time intervals is advantageous in order to compensate for statistical fluctuations in the ambient light and thus to obtain a more precise value for the amount of ambient light. However, the amount of ambient light determined in such exemplary embodiments can be falsified by the reflection on objects that are very distant. In some exemplary embodiments, an arithmetic mean is calculated from the received backscatter histogram data from several time intervals which lie above a certain time threshold in order to determine the amount of ambient light.

The diffuse backscattering in an optical time of flight measurement is typically no longer detectable over long distances because the amount of light is too small. In addition, after the light pulse hits the road surface or a solid object, for example in a LIDAR system, a constant amount of ambient light can be detected because the light energy or light power has been absorbed or reflected.

It is therefore advantageous for the determination of the amount of ambient light in some exemplary embodiments to take into account those time intervals of the received backscatter histogram data that are above a certain time threshold (e.g. from a time threshold corresponding to a distance of 20 m). Furthermore, the calculation of an arithmetic mean from several time intervals is advantageous in order to compensate for statistical fluctuations in the ambient light and thus to obtain a more precise value for the amount of ambient light. However, the amount of ambient light determined in such exemplary embodiments can be falsified by the reflection on objects that are far away.

In some exemplary embodiments, an arithmetic mean is calculated from the received backscatter histogram data from several time intervals that lie before a start time, and from the received backscatter histogram data from several time intervals that are above a certain time threshold, an arithmetic mean of the amount of ambient light to determine.

In such exemplary embodiments, the two methods described above are combined in order to determine the amount of ambient light. This is advantageous because it reduces the possible influence of distant objects on the determination of the amount of ambient light. Furthermore, the calculation of an arithmetic mean is advantageous, since this allows the statistical fluctuations to be reduced even further.

In some exemplary embodiments, the received backscatter histo- gram data that are above a certain time threshold, a local minimum is determined which meets a certain or predetermined criterion in order to determine the amount of ambient light. As mentioned above, in some embodiments a constant amount of ambient light can be determined for greater distances (corresponding to the specific time threshold value) if no reflections on objects in this time range have contributed to the received backscatter histogram data. In such embodiments, the amount of ambient light may correspond to a local minimum in the received backscatter histogram data that is above a certain time threshold.

Typically, signal contributions due to reflection on objects have a certain signal shape that can extend over several time intervals. If several objects are present for larger distances, in some exemplary embodiments the two signal forms can overlap in such a way that a local minimum arises which does not correspond to the amount of ambient light. Therefore, in some exemplary embodiments, only those local minima are taken into account for determining the amount of ambient light that meet a certain or predetermined criterion, wel ches the various possibilities for the presence of a local minimum in the received backscatter histogram data that is above a certain time threshold lie. The local minima can be classified based on this criterion.

In some exemplary embodiments, the amount of ambient light can be determined by first determining and classifying a first local minimum (from the time threshold value). If the local minimum does not meet the requirements for determining the amount of ambient light according to the classification, a further local minimum can be searched for in the received backscatter histogram data, this local minimum in such exemplary embodiments being behind and behind the time threshold the first local minimum. The search can accordingly be continued until the end of the measurement period. If the local minimum meets the requirements for determining the amount of ambient light according to the classification, the amount of ambient light is determined as the local minimum. In some exemplary embodiments, the determination of the amount of ambient light has the following steps: Calculating an arithmetic mean from the received Backscatter histogram data of a plurality of time intervals prior to a start time to obtain a first amount of ambient light;

Determining a local minimum which fulfills a certain criterion from the received backscatter histogram data which is above a certain time threshold in order to obtain a second amount of ambient light; and determining the amount of ambient light from a comparison between the first amount of ambient light and the second amount of ambient light, the amount of ambient light being determined as the smaller of the two amounts of ambient light.

First, for the first amount of ambient light, an arithmetic mean is calculated from the received backscatter histogram data from a number of time intervals which are before a start time. This can be expressed algorithmically as min_ambient = first amount of ambient light, where min_ambient corresponds to a minimum to be determined for the amount of ambient light. Then a local minimum in the received backscatter histogram data is determined which is above a certain time threshold value which corresponds to the second amount of ambient light. This can be written as current_far_ambient = second amount of ambient light. According to the above statements, this local minimum is classified according to a specific criterion. If the local minimum meets the requirements for determining the amount of ambient light according to the classification, then the smaller of the two amounts of ambient light is set as min_ambient. This can be expressed algorithmically as follows: min_ambient = min (min_ambient, current_far_ambient). If the local minimum does not meet the requirements for determining the amount of ambient light according to the classification, the next local minimum is determined as the second amount of ambient light and classified again, etc. If the search has reached the end of the measurement time, the amount of ambient light is determined set as AL = min_ambient.

The method described can thus correspond to a combination of two of the methods outlined above. This can be advantageous since it reduces statistical fluctuations and influences from objects that are far away, so that the amount of ambient light can be determined more precisely.

In some embodiments, the effective detection area of the optimal see transit time measurement determined using a predetermined function.

As mentioned above, the effective detection area can be determined based on the backscatter signal (AB) and the amount of ambient light (AL). The given function can be a given (mathematical) function in some exemplary embodiments, which is calculated from the backscatter signal and the amount of ambient light, the effective detection area. In other exemplary embodiments, the predetermined function can be a sequence of calculations. The given function can be determined experimentally or obtained from experience. Formally this can be expressed as: DER = f (AL, AB), where at f is the given function.

In some exemplary embodiments, the effective detection range of the optical transit time measurement is determined from a characteristic diagram.

A characteristic diagram can be a table-like, simple and, with regard to the required computing capacity, not very demanding type of mapping of a model that maps the relationship between input and output variables of a system. Almost any mathematical relationship or formula can be represented with characteristic diagrams, whereby the number of input variables is limited. In some exemplary embodiments, the characteristic diagram can therefore be a mapping of the function f (AL, AB), which has correspondingly stored the values of the effective detection range for a large number of values of AL and AB. This is advantageous since the determination of the effective detection area does not have to be calculated in such exemplary embodiments and thus saves computing capacity.

As mentioned above, the effective detection range of the optical transit time measurement is determined in some exemplary embodiments from a comparison with predetermined reference values.

Some exemplary embodiments relate to a device for data processing, comprising means for carrying out the steps of the method as set out herein. The device can be installed in a motor vehicle or implemented in a component of the motor vehicle, e.g. B. in an on-board computer, a controller or the like. In addition, the means can be one or more (mic- ro) processors, storage means, and other electronic components typically required to implement the functions described herein.

Embodiments of the invention will now be described by way of example and with reference to the accompanying drawing, in which:

Fig. 1 illustrates a first embodiment of a reference backscatter signal.

Fig. 2 illustrates a second embodiment of a reference backscatter signal;

Fig. 3 illustrates a third embodiment of a reference backscatter signal.

Fig. 4 illustrates a diagram of an embodiment of a receiving system for an optical cal distance measurement;

5 illustrates a flow chart of a first embodiment of a method for analyzing backscatter histogram data in an optical time of flight measurement;

6 illustrates a flow chart of a second embodiment of a method for analyzing backscatter histogram data in an optical time of flight measurement;

7 illustrates a flow diagram of a third embodiment of a method for analyzing backscatter histogram data in an optical time of flight measurement; and

8 illustrates a flow chart of a fourth exemplary embodiment of a method for analyzing backscatter histogram data in an optical time of flight measurement. Fig. 1 illustrates the first embodiment of a reference backscatter signal.

The reference backscatter signal illustrated in Fig. 1 corresponds to a typical signal form as it occurs in a coaxial LIDAR system, i.e. in an LLAR system in which there is no parallax between transmitter and receiver. The reference backscatter signal (in other words the intensity of the backscattered light) falls monotonically over time.

Fig. 2 illustrates the second embodiment of a reference backscatter signal.

The reference backscatter signal illustrated in Fig. 2 corresponds to a typical waveform as it occurs in a bi-axial single-beam LIDAR system. In biaxial systems (i.e. transmitting and receiving systems are at a defined distance, e.g. 10 cm and have a defined beam divergence), an overlap only occurs from a minimum distance (start of the signal increase of the reference backscatter signal). The reference backscatter signal then decreases in accordance with the curve in FIG. 1. The reference backscatter signal has a low intensity at very short distances and rises to a maximum, which then decreases monotonically over time.

Fig. 3 illustrates the third embodiment of a reference backscatter signal.

The reference backscatter signal illustrated in Fig. 3 corresponds to a typical signal form as it typically occurs in a bi-axial multi-beam LIDAR system (e.g. according to DE 10 2017 222 971 A1). The reference backscatter signal is similar to a bi-axial single-beam LIDAR system, but has several maxima at distances at which different beams cross the field of view of the light-detecting receiving elements. The intensity falls monotonously over time for greater distances.

FIG. 4 illustrates a diagram of an exemplary embodiment of a receiving system 1 for an optical distance measurement. The receiving system 1 has a receiving matrix 2 on which several light-detecting receiving elements (ENxM, in this exemplary embodiment E0.0 to £ 127.255) are arranged in rows (ZO to Z127) and columns (SO to S255). In each of the N = 128 lines (ZO to Z127), M = 256 light-detecting receiving elements (E0,0 to E127,255) are arranged (corresponding to the M = 256 columns (SO to S255)). The light-detecting receiving elements (E0,0 to E127,255) are SPADs in this exemplary embodiment.

The receiving system 1 also has several evaluation units (A0 to A127), one evaluation unit (AO to A127) with the light-detecting receiving elements (E0,0 to E127,255) of a line (ZO to Z127) via a multiplexer (not shown) ) connected is. In each line (ZO to Z127) only the two light-detecting receiving elements (E0,0 and E0,1 to E127,0 and E127, l) in the columns SO and S1 are activated at a given point in time (illustrated by the second circle within the light-detecting receiving elements (E0,0 and E0,1 to E127,0 and E127, l)). The activated light-detecting receiving elements (E0,0 and E0,1 to E127,0 and E127, l) generate electrical signals when light is detected, from which a time-to-digital converter (not shown) is time-correlated in each of the evaluation units (AO to A127) Histogram data are generated. In this embodiment, the time-correlated histogram data of the two activated light-detecting receiving elements (E0,0 and E0,1 to E127,0 and E127, l) are added in the evaluation units (AO to A127) in order to generate and output time-correlated histogram data . In other exemplary embodiments, any number of M = 256 light-detecting receiving elements (E0,0 to E127,255) can be activated in each line, for example E0,0 to EO, IO, E1,0 to E1, 10, E2,0 to E2,10, ..., E127,0 to E127,10.

The receiving system 1 also has several histogram Akkumulationseinhei th (HAO to HAX). Each histogram accumulation unit (HAO to HAX) has P = 16 signal inputs (not explicitly shown), each signal input being connected to an evaluation unit (AO to A127). Therefore, with N = 128 lines (ZO to Z127) in this exemplary embodiment, X = N / P = 8 histogram accumulation units are required, which accordingly accumulate the time-correlated histogram data from P = 16 evaluation units (AO to A127). The Time-correlated histogram data outputted to value units (AO to A127) are transmitted to the histogram accumulation units (HAO to HAX) so that they are received at the signal inputs. The histogram accumulation units (HAO to HAX) generate backscatter histogram data based on the received time-correlated histogram data. In this embodiment, the time-correlated histogram data received at each signal input are added to generate the backscatter histogram data.

The receiving system 1 also has a device 3 for data processing, which has a processor and memory elements (not shown). The histogram accumulation units (HAO to HAX) output the generated backscatter histogram data which are received by the device 3 for data processing. The data processing device 3 analyzes the received backscatter histogram data. In this embodiment, the device 3 for data processing calculates a correlation between the received backscatter histogram data and the reference backscatter signal from FIG. 3 in order to determine a backscatter signal, which is determined, for example, as a backscatter indicator or backscatter signal strength.

Fig. 5 illustrates a flowchart of the first embodiment of a method 20 for analyzing backscatter histogram data in an optical transit time measurement.

At 21, backscatter histogram data is received as set forth herein.

At 22, the received backscatter histogram data is analyzed as set out herein.

At 23, a measure of similarity between the received backscatter histogram data and a predetermined reference backscatter signal is calculated to determine a backscatter signal as set forth herein.

At 24, a transform function is applied to the backscatter signal to obtain a signal attenuation factor as set forth herein. Where, at 25, the transformation function from step 24 was (previously) determined experimentally, as set out herein.

FIG. 6 illustrates a flow chart of the second exemplary embodiment of a method 30 for analyzing backscatter histogram data in an optical transit time measurement.

At 31, backscatter histogram data is received as set forth herein.

At 32, the received backscatter histogram data is analyzed as set out herein.

At 33, an amount of ambient light is determined from the received backscatter histogram data, as set forth herein.

Steps 34 to 36 are options that are carried out individually.

At 34, an arithmetic mean is calculated from the received backscatter histogram data from a plurality of time intervals that precede a start time to determine the amount of ambient light, as set forth herein.

At 35, an arithmetic mean is calculated from the received backscatter histogram data of multiple time intervals that are above a certain time threshold to determine the amount of ambient light, as set forth herein.

At 36, an arithmetic mean is calculated from the received backscatter histogram data from several time intervals that are in time before a start time, and from the received backscatter histogram data from several time intervals that are above a certain time threshold in order to determine the amount of ambient light as set forth herein.

Fig. 7 illustrates a flowchart of the third embodiment of a method 40 for analyzing backscatter histogram data in an optical transit time measurement. At 41, backscatter histogram data is received as set forth herein.

At 42, the received backscatter histogram data is analyzed as set out herein.

At 43, an amount of ambient light is determined from the received backscatter histogram data, as set forth herein. Steps 44 and 45 are options that are each carried out individually.

At 44, a local minimum that meets a specific criterion is determined from the received backscatter histogram data that is above a specific time threshold in order to determine the amount of ambient light, as set forth herein.

At 45, an arithmetic mean is calculated from the received backscattering histogram data from several time intervals that lie before a start time in order to obtain a first amount of ambient light, and a local minimum is determined, which meets a certain criterion, from the received back scatter histogram data that is above a certain time threshold to determine a second amount of ambient light, and an amount of ambient light is determined from a comparison between the first amount of ambient light and the second amount of ambient light, the amount of ambient light being determined as the smaller of the two amounts of ambient light, such as detailed herein.

FIG. 8 illustrates a flowchart of the fourth exemplary embodiment of a method 50 for analyzing backscatter histogram data in an optical transit time measurement.

At 51, backscatter histogram data is received as set forth herein.

At 52, the received backscatter histogram data is analyzed as set out herein.

At 53 a similarity measure between the received backscatter histo- gram data and a predetermined reference backscatter signal are calculated to determine a backscatter signal as set forth herein.

At 54, an amount of ambient light is determined from the received backscatter histogram data, as set forth herein.

At 55, an effective detection range of the optical transit time measurement is determined based on the backscatter signal, which is determined, for example, as a backscatter indicator or backscatter signal strength, and the amount of ambient light, as set out herein.

Steps 56 to 58 are options that are each carried out individually.

At 56, the effective detection range of the optical transit time measurement is determined with the aid of a predetermined function, as set out herein.

At 57, the effective detection range of the optical transit time measurement is determined from a characteristic diagram, as set out herein.

At 58, the effective detection range of the optical transit time measurement is determined from a comparison with predetermined reference values, as set out herein.

Reference number

1 receiving system

2 reception matrix

3 device

20, 30, 40, 50 procedures

21, 31, 41, 51 Receive backscatter histogram data

22, 32, 42, 52 Analyze the received backscatter histogram data. 23, 53 calculating a measure of similarity between the received backscatter histogram data and a predetermined reference backscatter signal in order to determine a backscatter signal.

24 Apply a transform function to the backscatter signal to obtain a signal attenuation factor

25 Experimental determination of the transformation function

33, 43, 54 determine an amount of ambient light from the received backscatter histogram data

34 Calculation of an arithmetic mean from the received backscatter histogram data of several time intervals, which are temporally before a start time, in order to determine the amount of ambient light

35 Calculating an arithmetic mean from the received backscatter histogram data of several time intervals which are above a certain time threshold in order to determine the amount of ambient light

36 Calculation of an arithmetic mean from the received backscatter histogram data from several time intervals that lie before a start time and from the received backscatter histogram data from several time intervals that lie above a certain time threshold in order to determine the amount of ambient light

44 Determination of a local minimum which fulfills a certain criterion from the received backscatter histogram data which is above a certain time threshold in order to determine the amount of ambient light

45 Calculation of an arithmetic mean from the received backscatter histogram data of several time intervals that occur before a start Point in time, and from the received backscatter histogram data from a plurality of time intervals which are above a certain time threshold, in order to determine a first amount of ambient light; Determining a lo cal minimum which meets a certain criterion from the received backscattering histogram data that are above a certain time threshold in order to determine a second amount of ambient light; and determining the amount of ambient light from a comparison between the first amount of ambient light and the second amount of ambient light, the amount of ambient light being determined as the smaller of the two amounts of ambient light

55 Determination of an effective detection range of the optical transit time measurement based on the backscatter signal and the amount of ambient light

56 Determining the effective detection range of the optical transit time measurement with the aid of a predefined function

57 Determining the effective detection range of the optical transit time measurement from a characteristic map

58 Determination of the effective detection range of the optical transit time measurement from a comparison with specified reference values

A0 to A127 evaluation units

ENcM, EO, O to E127,255 light-detecting receiving elements HAO to HAX His togram accumulation units SO to S255 columns

Z0 to Z127 lines

Claims

1 claims

1 . A method (20, 30, 40, 50) for analyzing backscatter histogram data in an optical pulse time-of-flight method, comprising:

Receiving (21, 31, 41, 51) backscatter histogram data; and analyzing (22, 32, 42, 52) the received backscatter histogram data.

The method (20, 30, 40, 50) of claim 1, wherein analyzing the received backscatter histogram data comprises:

Calculating (23, 53) a degree of similarity between the received backscatter histogram data and a predetermined reference backscatter signal in order to determine a backscatter signal.

The method (20, 30, 40, 50) of claim 1 or 2, wherein analyzing the received backscatter histogram data comprises:

Determining (33, 43, 54) an amount of ambient light from the received backscatter histogram data.

4. The method (20, 30, 40, 50) according to claim 3 insofar as it depends on claim 2, wherein the analyzing of the received backscatter histogram data comprises:

Determining (55) an effective detection range of the optical transit time measurement based on the backscatter signal and the amount of ambient light.

5. The method (20, 30, 40, 50) according to claim 2, wherein a transformation function is applied (24) to the backscatter signal in order to determine a signal damping factor.

6. The method (20, 30, 40, 50) according to claim 5, wherein the transformation function was determined experimentally (25).

7. The method (20, 30, 40, 50) according to claim 3, wherein from the received backscatter histogram data from a plurality of time intervals, the time before a 2 nem starting time, an arithmetic mean is calculated (34) to determine the amount of ambient light.

8. The method (20, 30, 40, 50) according to claim 3, wherein an arithmetic mean is calculated (35) from the received backscatter histogram data from a plurality of time intervals which are above a certain time threshold value in order to match the amount of ambient light determine.

9. The method (20, 30, 40, 50) according to claim 3, wherein from the received backscatter histogram data from a plurality of time intervals that are in time before a start time, and from the received backscatter histogram data from a plurality of time intervals that are over a certain time threshold, an arithmetic mean is calculated (36) to determine the amount of ambient light.

10. The method (20, 30, 40, 50) according to claim 3, wherein from the received backscatter histogram data, which are above a certain time threshold, a local minimum is determined (44) which fulfills a certain criterion in order to determine the amount of ambient light.

11. The method (20, 30, 40, 50) of claim 3, wherein determining the amount of ambient light comprises:

Calculating an arithmetic mean from the received backscatter histogram data of a plurality of time intervals which precede a start time in order to obtain a first amount of ambient light; Determining a local minimum which fulfills a certain criterion from the received backscatter histogram data which are above a certain time threshold in order to obtain a second amount of ambient light; and

Determining the amount of ambient light from a comparison between the first amount of ambient light and the second amount of ambient light, the amount of ambient light being determined as the smaller of the two amounts of ambient light (45). 3

12. The method (20, 30, 40, 50) according to claim 4, wherein the effective detection area of the optical time-of-flight measurement is determined using a predetermined function (56).

13. The method (20, 30, 40, 50) according to claim 4, wherein the effective detection area of the optical transit time measurement is determined from a characteristic map (57).

14. The method (20, 30, 40, 50) according to claim 4, wherein the effective detection area of the optical transit time measurement from a comparison with predetermined reference values is determined (58).

15. Device (3) for data processing, comprising means for carrying out the method according to one of the preceding claims.