METHOD FOR DETERMINING OPTICAL OVERLAY OF A LIDAR SENSOR AND LIDAR SENSOR

Description

State of the art

[0001] The present invention relates to a method for determining optical crosstalk of a lidar sensor, in particular a spatially resolving lidar sensor, and to such a lidar sensor.

[0002] Lidar sensors of various designs are known from the state of the art and are used particularly in semi-automated or highly automated vehicles for environmental perception. Depending on the technology used, crosstalk of varying degrees can occur in the receiver path of the lidar sensors (e.g., due to scattering of received light by a protective glass of the lidar sensors), which can have a more or less pronounced effect depending on the reflectivity of objects in the environment.

[0003] CN111366941 A discloses a time-of-flight (TOF) measuring device which emits light in the form of a predefined dot pattern to scan an environment, wherein a portion of the dot pattern is used in a receiving module of the measuring device to detect direct reflections of the emitted light, and wherein another portion of the dot pattern is used to detect emitted light that has been indirectly reflected to the measuring device, particularly via multipath propagation. Based on such separate detection of the respective light components, it is possible to separate indirectly reflected light components from the directly reflected light components.

[0004] US2013293684 A1 discloses an optical measuring device comprising a first projector, a first camera, a second projector and a second camera, on the basis of which a processor enables three-dimensional detection of an environment of the measuring device.

[0005] DE 10 2015 101 902 A1 describes a detector for a lidar system with a series of adjacent radiation-sensitive pixels. Due to the design, radiation-insensitive intermediate areas are located between the pixels to reduce crosstalk between them.

[0006] EP 2 998 700 B1 describes an electro-optical distance measuring device and a distance measuring method. According to one embodiment, a transmitting light beam and a receiving light beam of the electro-optical distance measuring device are arranged biaxially to each other. Because each receiving segment can be assigned a separate signal processing path, electronic crosstalk between signals from different receiving segments can at least be reduced. Disclosure of the invention

[0007] According to a first aspect of the present invention, a method for determining optical crosstalk of a lidar sensor, in particular a spatially resolved lidar sensor, is proposed. The lidar sensor is configured, for example, as a point scanner, flash lidar sensor, and preferably as a line scanner. Preferably, the lidar sensor is a lidar sensor of a means of transportation, which is used for scanning the environment of the means of transportation. In a first step of the method according to the invention, laser light from the lidar sensor is emitted into the environment of the lidar sensor. In a second step of the method according to the invention, a signal from a light detector of the lidar sensor, representing components of the emitted laser light reflected or scattered in the environment of the lidar sensor, is received. The light detector is preferably configured as an area detector, but is not limited to such a configuration.The light detector also has a first receiving area, the extent and position of which on the detector correspond to the extent and position of the laser light focused onto the detector when the laser light scattering is equal to or less than a predefined threshold. Furthermore, the light detector has a second receiving area, distinct from the first, which is directly adjacent to the first and is configured to detect portions of the laser light focused onto the detector when the laser light scattering exceeds the predefined threshold. Such laser light scattering can be caused, for example, by the protective glass of the lidar sensor and / or by other optical elements (e.g., lenses) in the transmit and / or receive path.Other causes of such scattering include, for example, raindrops and / or dirt on the protective glass of the lidar sensor. The extent of laser light scattering in the area of ​​the light detector also depends on the reflectivity of objects in the vicinity of the lidar sensor. In particular, highly reflective objects such as retroreflectors (e.g., from traffic signs, traffic control devices, etc.) can thus lead to a high degree of laser light scattering on the light detector. It should be noted that the method according to the invention is based on the assumption that the scattered laser light received at the light detector essentially corresponds to isotropic scattering, i.e., scattering that is uniformly scattered in all directions.In a third step of the method according to the invention, the extent of the optical crosstalk of the lidar sensor is determined based on the portions of the laser light received in the second receiving area. At least the steps of receiving the signal and determining the information about the extent of the optical crosstalk are preferably performed using an evaluation unit according to the invention. According to the method described above, a particularly simple and cost-effective way of determining the optical crosstalk of the lidar sensor can therefore be realized.

[0008] The dependent claims describe preferred embodiments of the invention.

[0009] Preferably, information about the extent of optical crosstalk in environmental detection is taken into account based on the signal from the lidar sensor, and in particular based on the signal components representing the first reception area. In other words, the method according to the invention makes it possible to evaluate the reliability of environmental detection, since a high degree of scattering can potentially lead to false detections of objects in the vicinity of the lidar sensor.

[0010] In an advantageous embodiment of the present method, the portions of the signal representing the second reception area are used to at least partially compensate for crosstalk in the first reception area. This offers the advantage of improving the reliability of the result of subsequent signal processing (e.g., in the environmental detection described above), since interference introduced into the useful signal by the scattering can be at least partially eliminated. While determining the extent of optical crosstalk from the lidar sensor is fundamentally feasible with any lidar system (point scanner, line scanner, flash lidar) of the aforementioned lidar sensors, crosstalk compensation is generally most advantageously applicable in conjunction with line scanners.However, it should not be explicitly excluded that such compensation may also be carried out in connection with other lidar systems based on the method according to the invention.

[0011] The lidar sensor is particularly advantageous when configured as a spatially resolving line scanner. In other words, the lidar scanner is designed to emit laser light in a line pattern into the area surrounding the lidar sensor and to spatially resolve the echoes generated in this area along the direction of the scan line in the receiving path of the lidar sensor. The first receiving area, in the context of such a line scanner, comprises at least one row of pixels aligned with an image of the lidar sensor's scan line on the light detector.To at least partially compensate for crosstalk in the first receiving area, for each pixel under consideration in at least one pixel row of the first receiving area, a number of pixels in the second receiving area adjacent to the respective pixel under consideration are determined. These pixels are arranged on an imaginary line that runs orthogonally to the pixel row of the first receiving area and intersects the pixel under consideration. Subsequently, the respective brightness values ​​of the respective determined pixels in the second receiving area are subtracted from the respective brightness values ​​of the pixel row of the first receiving area such that the respective brightness values ​​of those pixels that are equidistant from the pixel under consideration are subtracted from each other.The compensation described above is preferably performed by successively executing the aforementioned processing steps for all pixels or a suitable subset of pixels in the pixel array of the first receiving area (i.e., the pixels to be considered in each case), with the respective results of previous compensation steps serving as the basis for calculation for each subsequent compensation step. Preferably, after completion of all compensation iterations, those portions of the light detector signal in the signal that represent the first receiving area are replaced by the respective compensated brightness values. Alternatively or additionally, it is conceivable to generate a new signal based on the compensated brightness values, which is then subjected to downstream processing.In a particularly simple and therefore cost-effective embodiment of a light detector applicable in this context, the detector has, for example, a single row of pixels in the first reception area and another row of pixels arranged parallel to this row, representing the second reception area. If the scatter in the pixel row of the second reception area exceeds the predefined threshold, at least the pixels immediately adjacent to the pixel under consideration in the first area (e.g., above and / or below) can be compensated for by the brightness value present in the second area.By using a number of additional parallel rows of pixels in the second area, additional pixels located further away from the pixel being viewed in the first reception area can be compensated for by means of the additional brightness values ​​recorded in the second reception area.

[0012] Advantageously, the brightness values ​​of the second reception area used for compensation in the first reception area are extrapolated, at least partially, using a predefined dispersion characteristic of the lidar sensor. This is particularly advantageous when the second reception area, as described above, has only one row of pixels or a small number of parallel pixel rows. In such a case, for example, when using a single pixel row in the second reception area, it is possible to calculate further brightness values ​​along a virtual (since it does not physically exist) pixel row in the second reception area based on the brightness value of the single pixel used for a given compensation run.For this purpose, a function describing the dispersion characteristic of the lidar sensor and / or a lookup table can be used, allowing further brightness values ​​to be extrapolated by inserting or comparing the brightness value of the pixel in the second area. Furthermore, extrapolation is also possible based on multiple pixels in the second area if it has more than one pixel in width. The dispersion characteristic can, for example, be an average dispersion characteristic for multiple lidar sensors with similar or identical designs, or a dispersion characteristic determined individually for each lidar sensor.

[0013] Furthermore, it is conceivable that the first receiving area comprises a plurality of parallel pixel rows, from which a representative pixel row is determined, to which the crosstalk compensation steps are applied. The representative pixel row can be defined, for example, based on average values ​​of parallel pixels or based on maximum brightness values ​​in the parallel pixel rows.

[0014] In an advantageous embodiment, crosstalk compensation is applied only to those pixels of the at least one pixel row that exhibit a predefined minimum dispersion in their respective corresponding pixels in the second reception area. The predefined minimum dispersion can be determined, for example, based on a minimum brightness of a pixel in the second reception area corresponding to the pixel under consideration in the first reception area, and / or based on an average minimum brightness of a plurality of pixels in the corresponding second reception area, and / or based on a minimum number of illuminated pixels in the respective corresponding second reception area.

[0015] Preferably, the surface of the light detector is essentially square, so that in the case of maximum dispersion of the scanning line imaged on the light detector, optimal or nearly complete compensation of the brightness values ​​of the first pixel rows is enabled.

[0016] The information about the extent of optical crosstalk is advantageously used to detect contamination and / or moisture on the protective glass of the lidar sensor. Alternatively or additionally, the information about the extent of optical crosstalk is used to detect a highly reflective object (e.g., a retroreflector) in the vicinity of the lidar sensor whose reflectance exceeds a predefined value.

[0017] Furthermore, it is advantageous to determine the position and / or extent of a highly reflective object in the vicinity of the lidar sensor based on a distribution of the scattering in the second reception area.

[0018] According to a second aspect of the present invention, a lidar sensor, in particular a spatially resolved lidar sensor, is proposed. The lidar sensor comprises an evaluation unit, a light emitter, and a light detector. The evaluation unit is configured, for example, as an ASIC, FPGA, processor, digital signal processor, microcontroller, or similar, and is connected to at least the light detector, and preferably also to the light emitter, via an information technology connection. The lidar sensor is configured to emit laser light into the vicinity of the lidar sensor by means of the light emitter, while the evaluation unit is configured to receive a signal from the light detector representing the reflected or scattered portions of the laser light in the vicinity of the lidar sensor.The light detector has a first receiving area whose extent and position on the detector correspond to the extent and position of the laser light projected onto the detector when the laser light scattering is equal to or less than a predefined threshold. Additionally, the light detector has a second receiving area, distinct from the first, which is directly adjacent to the first and is configured to detect portions of the laser light projected onto the detector when the laser light scattering exceeds the predefined threshold. Furthermore, the evaluation unit is configured to determine the extent of optical crosstalk from the lidar sensor based on the portions of the laser light received in the second receiving area. Brief description of the drawings

[0019] Exemplary embodiments of the invention are described in detail below with reference to the accompanying drawing. The drawings show: Figure 1 is a schematic overview of components of a lidar sensor according to the invention; Figure 2 is a top view of a light detector of a lidar sensor according to the invention in a first receiving state; and Figure 3 is a top view of a light detector of a lidar sensor according to the invention in a second receiving state. Embodiments of the invention

[0020] Figure 1Figure 1 shows a schematic overview of components of a lidar sensor 10 according to the invention. The lidar sensor 10 comprises a light emitter 70, which, in conjunction with a transmitting optic 90, is configured to emit laser light in the form of a scanning line through a protective glass 15 of the lidar sensor 10 into the vicinity of the lidar sensor 10. Components of the emitted laser light reflected or scattered by an object 80 in the vicinity of the lidar sensor 10 re-enter the lidar sensor 10 through the protective window 15 and are focused by a receiving optic 95 of the lidar sensor 10 onto a planar light detector 20 of the lidar sensor 10. Due to a scattering property of the protective glass 15, the received laser light, especially with a high reflectance of the object 80, includes scattered light components 100, which can lead to crosstalk and thus to a reduction in the accuracy of the spatial resolution of the lidar sensor 10.An evaluation unit 60 according to the invention, which is designed here as an ASIC, is connected to the light emitter 70 and the light detector 20 via information technology. Based on a computer program executed by the evaluation unit 60, which implements the method steps described above according to the invention, the evaluation unit 60 is configured to determine the respective extent of crosstalk and, furthermore, to perform at least partial compensation of the crosstalk.

[0021] Figure 2Figure 1 shows a top view of a light detector 20 of a lidar sensor according to the invention in a first reception state. The light detector 20 is configured here as a square light detector 20, which has a first reception area 30 and a second reception area 35. The light detector 20 is configured to receive a scan line projected onto the light detector 20 completely via the pixel array 40, which forms the first reception area 30, provided that the dispersion of the projected scan line is equal to or less than a predefined threshold value. The second reception area 35 consists of a plurality of pixels 55 which, in the first reception state described here, are not illuminated or are only illuminated to a negligible extent due to a very low dispersion of the scan line.

[0022] Figure 3Figure 1 shows a top view of a light detector 20 of a lidar sensor according to the invention in a second receiving state. Due to the similarities between Figure 2 and Figure 3 To avoid repetition, only the differences between the two figures will be described below. Figure 3Figure 1 shows a second reception state in which a portion of a scan line projected onto the light detector 20 is scattered to such an extent in the region of the uppermost left pixel of the first reception area 30 that stray light components are detected in the second reception area 35. It should be noted that this is a simplified representation which does not show the actual radial scattering around the uppermost left pixel in the second reception area 35. Using the method described above according to the invention, the brightness information present horizontally to the uppermost left pixel in the second reception area 35 is algorithmically rotated, as it were, towards the pixel row 40 of the first reception area 30 (indicated by the arrow shown) and then subtracted pixel by pixel from the first reception area 30 to compensate for the scattering of the scan line.

Claims

1. Method for determining an optical crosstalk of a lidar sensor (20) comprising: • emitting (100) a laser light of the lidar sensor (10), in particular of a spatially resolving lidar sensor (10), into an environment of the lidar sensor (10), • receiving (200) a signal of a light detector (20) of the lidar sensor (10) representing portions of the laser light reflected or scattered in the environment of the lidar sensor (10), wherein ∘ the light detector (20) has a first reception region (30), the extent and position of which on the light detector (20) correspond to an extent and position of the laser light imaged onto the light detector (20) if a scattering of the laser light is equal to or less than a predefined threshold value, and ∘ the light detector (20) has a second reception region (35), which differs from the first reception region (30) and which is directly adjacent to the first reception region (30) and which is configured to detect portions of the laser light imaged onto the light detector (20) if the scattering of the laser light is greater than the predefined threshold value, and • determining (300) information about an extent of the optical crosstalk of the lidar sensor (10) on the basis of the portions of the laser light received in the second reception region (35).

2. Method according to Claim 1, further comprising: • taking into account the information about the extent of the optical crosstalk in the course of environment recognition on the basis of the signal of the lidar sensor (10).

3. Method according to either of the preceding claims, further comprising: • using the portions of the signal that represent the second reception region (35) for the at least partial compensation of the crosstalk in the first reception region (30).

4. Method according to Claim 3, wherein • the lidar sensor (10) is a spatially resolving line scanner, • the first reception region (30) comprises at least one pixel row (40) oriented in the direction of an image of a scanning line of the lidar sensor (10) on the light detector (20), and • for the at least partial compensation of the crosstalk in the first reception region (30) ∘ for each pixel (50) to be considered of the at least one pixel row (40) of the first reception region (30), a number of pixels (55) of the second reception region (35) adjacent to the respective pixel (50) to be considered is determined, said number of pixels being arranged on an imaginary line which runs orthogonally with respect to the pixel row (40) of the first reception region (30) and which intersects the pixel (50) to be considered, and ∘ respective brightness values of the respective determined pixels (55) of the second reception region (35) are subtracted from respective brightness values of the pixel row (40) of the first reception region (30) in such a way that respective brightnesses of those pixels which are each at the same distance from the considered pixel (50) are subtracted from one another.

5. Method according to Claim 4, wherein respective brightness values of the second reception region (35) that are to be used for the compensation in the first reception region (30) are at least partially extrapolated by means of a predefined scattering characteristic of the lidar sensor (10).

6. Method according to Claim 4 or 5, wherein • the first reception region (30) comprises a plurality of pixel rows (40) arranged parallel, • from the plurality of pixel rows (40) arranged parallel, a representative pixel row (40) is determined, to which the steps for the compensation of the crosstalk are applied.

7. Method according to any of Claims 4 to 6, wherein the compensation of the crosstalk is applied only to those pixels (50) of the at least one pixel row (40) which have a predefined minimum scattering in their respectively corresponding pixels (55) in the second reception region (35).

8. Method according to any of the preceding claims, wherein a surface of the light detector (20) is substantially square.

9. Method according to any of the preceding claims, wherein the information about the extent of the optical crosstalk is used for determining • contamination and / or wetness on a protective glass of the lidar sensor (10), and / or • a highly reflective object in the environment of the lidar sensor (10), the reflectance of which exceeds a predefined reflectance.

10. Method according to Claim 9, wherein a position and / or an extent of a highly reflective object are / is determined on the basis of a distribution of the scattering in the second reception region (35).

11. Lidar sensor (10), in particular spatially resolving lidar sensor (10), comprising: • an evaluation unit (60), • a light emitter (70), and • a light detector (20), wherein the evaluation unit (60) is configured • in conjunction with the light emitter (70) to emit a laser light into an environment of the lidar sensor (10), • to receive a signal of the light detector (20) representing portions of the laser light reflected or scattered in the environment of the lidar sensor, wherein ∘ the light detector (20) has a first reception region (30), the extent and position of which on the light detector (20) correspond to an extent and position of the laser light imaged onto the light detector (20) if a scattering of the laser light is equal to or less than a predefined threshold value, and ∘the light detector (20) has a second reception region (35), which differs from the first reception region (30) and which is directly adjacent to the first reception region (30) and which is configured to detect portions of the laser light imaged onto the light detector (20) if the scattering of the laser light is greater than the predefined threshold value, and • to determine information about an extent of the optical crosstalk of the lidar sensor (10) on the basis of the portions of the laser light received in the second reception region (35).