DE102020213161A1 - LiDAR system and control method with surface roughness determination - Google Patents

Abstract

A LiDAR system (1) is described, in particular for a vehicle, comprising an emitter arrangement (2) which comprises at least one laser (6). A detector arrangement (3) comprises at least one detector unit (4), each detector unit (4) comprising a matrix arrangement of detector elements. Each detector unit (4) is aligned to the same solid angle range as one of the lasers (6), so that the detector unit (4) can at least detect the laser light of an associated laser (6) reflected by surrounding objects. A control unit is set up to evaluate the signals detected by the detector elements in order to detect the distance from surrounding objects. According to the invention, the control unit is set up to evaluate the individual signals of the detector elements in order to record a spatial speckle pattern generated by the reflection on a surface of a surrounding object. In addition, the control unit is set up to analyze the spatial speckle pattern in order to calculate a surface roughness of the surface of the surrounding object. Furthermore, a control method for a LiDAR system (1) is described.

Description

The present invention relates to a LIDAR system, in particular for a vehicle, comprising an emitter arrangement which comprises at least one laser, a detector arrangement which comprises at least one detector unit, each detector unit comprising a matrix arrangement of detector elements, each detector unit being aligned to the same solid angle range such as one of the lasers, so that the detector unit can at least detect the laser light of an associated laser reflected by surrounding objects, and a control unit that is set up to evaluate the signals detected by the detector elements in order to detect the distance from surrounding objects.

State of the art

Highly and fully automated vehicles will account for an ever larger proportion of road traffic in the coming years. All known concepts of automated vehicles use sensors such. B. cameras, radar and LiDAR (Light Detection and Ranging). The latter are laser scanners that emit a short pulse of laser light and capture the light reflected from an object. LiDAR systems then calculate the distance to the object from the measured transit time in the case of time-of-flight (ToF) LiDAR or by detecting beat frequencies in the case of frequency-modulated continuous-wave LiDAR (FMCW-LiDAR).

LiDAR systems typically include a laser that emits short pulses of light and a detector that detects the reflected laser light. In most cases, LiDAR systems work with multiple lasers and multiple detector units (also called macropixels), with each detector unit aligned to the same solid angle section as one of the lasers.

Lasers emit coherent light, i.e. light waves with a fixed phase relationship over a certain time and distance (coherence time and coherence length). This property allows light to interfere, creating interference patterns. For example, when laser light is shined on rough surfaces, a dotted interference pattern called a speckle pattern forms in the detector plane. The granular interference phenomena that can be observed with sufficiently coherent illumination of optically rough object surfaces (with unevenness in the order of magnitude of the wavelength) are generally referred to as speckle patterns, light granulation or laser granulation, or speckle for short. Speckle patterns can also be seen by the human eye when looking at a laser spot on a rough wall - the laser spot then does not appear homogeneous but rather patterned with light and dark areas.

In the prior art for LiDAR systems in particular, speckle effects are viewed more as an effect that disrupts the signal-to-noise ratio and that needs to be avoided.

Out of WO2012019871A1 a velocity measuring device is known from another technical field, for example for determining the velocities of particles suspended in a gas or fluid flow or for determining the velocity, displacement, rotation or vibration of a solid surface. The speed measuring device includes a laser arrangement for emitting a measuring beam to illuminate an object. A signal beam emanating from the object is formed as a reaction to the illumination of the object by the measuring beam. A detector system includes a detector array arranged such that the signal beam and a reference beam are incident on the first detector array, with detector elements converting the intensity of the interfering signal beam and the reference beam incident thereon into a corresponding electronic detector element signal. In particular, a speckle pattern/fringe pattern is analyzed in order to determine the speed. The detector elements can be arranged in a matrix.

Disclosure of Invention

According to the invention, a LiDAR system of the type mentioned is provided, characterized in that the control unit is set up to evaluate the individual signals of the detector elements in order to record a spatial speckle pattern generated by the reflection on a surface of a surrounding object, the control unit being set up to do so to analyze the spatial speckle pattern to calculate a surface roughness of the surface of the surrounding object.

Advantages of the Invention

This invention provides a LiDAR sensor that determines the surface roughness of an object by analyzing the speckle pattern picked up by the detector. Previous LiDAR systems can often only determine the distance and speed of a surrounding object and in this respect have so far provided significantly less information than, for example, camera-based sensor systems (albeit with more computing effort) through image analysis. The inventive The LiDAR system now provides additional information about surrounding objects without requiring a computationally intensive image analysis, as is the case with camera-based sensor systems, for example.

The LiDAR sensor presented here comprises at least one laser with a sufficiently long coherence time and at least one detector unit/macropixel and preferably a lens that images the laser light reflected from surrounding objects onto the macropixel. In addition, each detector unit/each macro-pixel comprises several detector elements/micro-pixels arranged in a matrix structure. The detector elements can be, for example, single-photon avalanche diodes (SPAD), but avalanche photodiodes (APD) or charge-coupled devices (CCD) are also possible. The micro-pixels can then act like an image sensor to recognize patterns within a macro-pixel.

In time-of-flight LiDAR systems, for example, the signal from all micro-pixels is combined within a macro-pixel to measure the time of flight in order to obtain a sufficiently high signal-to-noise ratio. In the case of SPAD detectors in particular, the combination of several micropixels is usually necessary due to the binary detection character of the diode. However, to estimate the surface roughness, the LiDAR system according to the invention now uses the individual signals of each micropixel to measure the spatial speckle pattern. So it is used information that has been included but in the evaluation z. B. is discarded again by averaging.

• - Texturanalyse-Verfahren,
• - Speckle-Kontrastverfahren,
• - Speckle-Korrelationsverfahren.

Die Schätzung der Oberflächenrauheit kann also mit für Speckle-Effekte aus der Laser-Grundlagenforschung bekannten Methoden wie Texturanalyse, Speckle-Kontrastverfahren und Speckle-Korrelationsverfahren durchgeführt werden, die beispielsweise im Artikel „Measure of roughness of paper using speckle“, Abdiel Pinoa, Josep Pladellorensa, CD6 Optics and Optometry Department, Polytechnical University of Catalonia, Proc. SPIE 7432, Optical Inspection and Metrology for Non-Optics Industries, 74320E, genauer erläutert werden.The control unit is preferably set up to carry out one or more of the following methods to calculate the surface roughness:

• - texture analysis method,
• - speckle contrast method,
• - Speckle correlation method.

The surface roughness can therefore be estimated using methods known for speckle effects from basic laser research, such as texture analysis, speckle contrast methods and speckle correlation methods, which are described, for example, in the article "Measure of roughness of paper using speckle", Abdiel Pinoa, Josep Pladellorensa , CD6 Optics and Optometry Department, Polytechnical University of Catalonia, Proc. SPIE 7432, Optical Inspection and Metrology for Non-Optics Industries, 74320E.

In one embodiment, at least one laser is a vertical-cavity surface-emitting (VCSEL) laser or a distributed feedback (DBR) laser. These types of lasers can be arranged particularly effectively in a compact matrix arrangement.

It is preferred if the emitter arrangement comprises a multiplicity of lasers in a matrix arrangement and the detector arrangement comprises a multiplicity of detector units in a matrix arrangement, with each laser being assigned at least one detector unit for the same solid angle range. The LiDAR system thus includes a laser matrix and a detector matrix. Each laser can be aimed at a different solid angle range, with some overlap being possible. The emitter array/laser matrix includes multiple laser diodes (e.g. VCSEL lasers, DBR lasers, etc.) arranged in a matrix array. An objective/lens is preferably arranged in front of the emitter arrangement and directs the laser light of the individual lasers in different directions. A second objective is preferably arranged in front of the detector arrangement and images the light reflected from surrounding objects onto the detector units. The detector units of the detector arrangement are arranged in a matrix structure similar to the lasers in the emitter arrangement. The lens is advantageously chosen so that a detector unit in each case measures the light emitted by one of the lasers and then reflected back, ie a pair of a laser and a detector unit/macropixel covers a corresponding solid angle section.

Each detector unit/macro-pixel includes multiple detector elements/micro-pixels, allowing image-like data to be captured within a macro-pixel to measure speckle patterns for surface roughness estimation. The surface roughness method according to the invention is not limited to the static LiDAR system described, but can also be used in other LiDAR concepts, such as rotating scanners or micromirror scanners.

In one embodiment, the control unit is set up to determine the type of the surrounding object from the determined surface roughness. In a vehicle application, for example, other vehicles (painted surfaces) have a comparatively low surface roughness, while building walls (stone, plaster, etc.), floor coverings (stone, asphalt, etc.), pedestrians and vegetation have a higher but distinguishable surface roughness.

• - einen bestimmten Abstand des Umgebungsobjekts,
• - eine bestimmte Raumwinkelgröße des Umgebungsobjekts durch Vergleich der Messungen von benachbarten Detektoreinheiten,
• - eine bestimmte Relativgeschwindigkeit zwischen dem LiDAR-System und dem Umgebungsobjekt. Diese Informationen erlauben es beispielweise eine absolute Größe und Absolutgeschwindigkeit (mit der bekannten Geschwindigkeit des eigenen Fahrzeugs) eines Umgebungsobjekts abzuschätzen und diese Information zusammen mit der bestimmten Oberflächenrauheit zur Unterscheidung des Umgebungsobjekts heranzuziehen.

The control unit is preferably set up to be involved in the determination of the type of environment exercise object to include one or more of the following information:

• - a certain distance of the surrounding object,
• - a certain solid angle size of the surrounding object by comparing the measurements of neighboring detector units,
• - a certain relative speed between the LiDAR system and the surrounding object. This information makes it possible, for example, to estimate an absolute size and absolute speed (with the known speed of one's own vehicle) of a surrounding object and to use this information together with the determined surface roughness to differentiate between the surrounding object.

• - Emittieren von Laserlicht durch den Laser in einen Raumwinkelbereich,
• - Auswerten der individuellen Signale der dem Laser zugeordneten Detektorelemente um ein durch die Reflektion an einer Oberfläche eines Umgebungsobjekts erzeugtes räumliches Speckle-Muster aufzunehmen,
• - Analysieren des räumlichen Speckle-Musters um eine Oberflächenrauheit der Oberfläche des Umgebungsobjekts zu berechnen. Auch das erfindungsgemäße Steuerverfahren erlaubt das Bestimmen der Oberflächenrauheit eines Objekts durch Analyse des vom Detektor aufgenommenen Speckle-Musters. Bisherige Steuerverfahren für ein LiDAR-System können oft nur die Entfernung und Geschwindigkeit eines Umgebungsobjekts bestimmen und liefern insofern bisher deutlich weniger Informationen als dies beispielsweise kamerabasierte Sensorsysteme grundsätzlich könnten. Das erfindungsgemäße Steuerverfahren für ein LiDAR-System stellt nun zusätzliche Informationen über Umgebungsobjekte bereit, ohne dass dies eine rechenintensive Bildanalyse wie beispielsweise bei kamerabasierten Sensorsystemen erfordert.

According to the invention, a control method for a LiDAR system according to one of the above embodiments is also provided, comprising the following steps:

• - Emission of laser light by the laser in a solid angle range,
• - Evaluation of the individual signals of the detector elements assigned to the laser in order to record a spatial speckle pattern generated by the reflection on a surface of a surrounding object,
• - Analyzing the spatial speckle pattern to calculate a surface roughness of the surface of the surrounding object. The control method according to the invention also allows the surface roughness of an object to be determined by analyzing the speckle pattern recorded by the detector. Previous control methods for a LiDAR system can often only determine the distance and speed of a surrounding object and in this respect have so far provided significantly less information than camera-based sensor systems, for example, could in principle do. The control method according to the invention for a LiDAR system now provides additional information about surrounding objects without requiring a computationally intensive image analysis, as is the case, for example, with camera-based sensor systems.

• - Texturanalyse-Verfahren,
• - Speckle-Kontrastverfahren,
• - Speckle-Korrelationsverfahren.

Die Schätzung der Oberflächenrauheit kann also mit für Speckle-Effekte aus der Laser-Grundlagenforschung bekannten Methoden wie Texturanalyse, Speckle-Kontrastverfahren und Speckle-Korrelationsverfahren durchgeführt werden, die beispielsweise im Artikel „Measure of roughness of paper using speckle“, Abdiel Pinoa, Josep Pladellorensa, CD6 Optics and Optometry Department, Polytechnical University of Catalonia, Proc. SPIE 7432, Optical Inspection and Metrology for Non-Optics Industries, 74320E, genauer erläutert werden.In one embodiment, the step of analyzing the spatial speckle pattern to calculate the surface roughness includes at least one of the following methods:

• - texture analysis method,
• - speckle contrast method,
• - Speckle correlation method.

The surface roughness can therefore be estimated using methods known for speckle effects from basic laser research, such as texture analysis, speckle contrast methods and speckle correlation methods, which are described, for example, in the article "Measure of roughness of paper using speckle", Abdiel Pinoa, Josep Pladellorensa , CD6 Optics and Optometry Department, Polytechnical University of Catalonia, Proc. SPIE 7432, Optical Inspection and Metrology for Non-Optics Industries, 74320E.

• - Bestimmen des Typs des Umgebungsobjekts aus der berechneten Oberflächenrauheit,

wobei vorzugsweise eine oder mehrere der folgenden Informationen miteinbezogen werden:

• - ein bestimmter Abstand des Umgebungsobjekts,
• - eine bestimmte Raumwinkelgröße des Umgebungsobjekts durch Vergleich der Messungen von benachbarten Detektoreinheiten,
• - eine bestimmte Form des Umgebungsobjekts durch Vergleich der Messungen von benachbarten Detektoreinheiten,
• - eine bestimmte Relativgeschwindigkeit zwischen dem LiDAR-System und dem Umgebungsobjekt. Diese Informationen erlauben es beispielweise eine absolute Größe und Absolutgeschwindigkeit (z. B. mit der bekannten Geschwindigkeit eines mit dem LiDAR-System verbundenen Fahrzeugs) eines Umgebungsobjekts abzuschätzen und diese Information zusammen mit der bestimmten Oberflächenrauheit zur Unterscheidung des Umgebungsobjekts heranzuziehen.

Preferably, the control method comprises the following step after the surface roughness calculation:

• - determining the type of surrounding object from the calculated surface roughness,

preferably including one or more of the following information:

• - a certain distance of the surrounding object,
• - a certain solid angle size of the surrounding object by comparing the measurements of neighboring detector units,
• - a specific shape of the surrounding object by comparing measurements from neighboring detector units,
• - a certain relative speed between the LiDAR system and the surrounding object. This information makes it possible, for example, to estimate an absolute size and absolute speed (e.g. with the known speed of a vehicle connected to the LiDAR system) of a surrounding object and to use this information together with the determined surface roughness to distinguish between the surrounding object.

It is preferred if the steps of emitting laser light and evaluating the individual signals to record a spatial speckle pattern are performed at least twice with different power settings of the laser before the spatial speckle patterns recorded at different powers are analyzed for surface roughness of the surface of the surrounding object. Especially when the detector elements are SPAD micropixels, it can be advantageous to perform repeated measurements of the same detector unit with different laser power outputs. This makes it possible to obtain more accurate speckle pattern intensity information for each detector element instead of binary information, which facilitates speckle pattern analysis. For example the laser power could be reduced logarithmically starting from the highest possible laser power until no more detector element of the detector unit detects a signal. The lowest laser power at which the respective detector element still detects a signal is then a measure of the intensity of the signal and an intensity distribution of the spatial speckle pattern can be determined for all detector elements. For other detector element types such. B. APDs or CCDs, this power variation of the laser may not be necessary, although these are also less sensitive than, for example, SPADs and therefore have other disadvantages for the basic function of the LiDAR system, the distance determination. The power variation should take place in a sufficiently short period of time so that the speckle pattern does not change due to a relative movement between the surrounding object and the LiDAR system.

Features disclosed for the LiDAR system can also be combined with features disclosed for the control method according to the invention and vice versa.

Advantageous developments of the invention are specified in the dependent claims and described in the description.

character list

• 1 eine Ausführungsform eines erfindungsgemäßen LiDAR-Systems,
• 2 eine Detektoreinheit / Makropixel eines erfindungsgemäßen LiDAR-Systems und
• 3 eine Emitteranordnung und eine Detektoreinheit / Makropixel einer Ausführungsform eines erfindungsgemäßen LiDAR-Systems.

Exemplary embodiments of the invention are explained in more detail with reference to the drawings and the following description. Show it:

• 1 an embodiment of a LiDAR system according to the invention,
• 2 a detector unit/macropixel of a LiDAR system according to the invention and
• 3 an emitter array and a detector unit/macropixel of an embodiment of a LiDAR system according to the invention.

Embodiments of the invention

In the 1 an embodiment of a LiDAR system 1 according to the invention is shown. The LiDAR system 1 comprises an emitter arrangement 2, which comprises at least one laser 6 (see also 3 ), and a detector arrangement 3, the at least one detector unit 4 (see also 2 ) includes. Each detector unit 4 is aligned to the same solid angle range as one of the lasers 6, so that the detector unit 4 can at least detect the laser light of an associated laser 6 reflected by surrounding objects.

In 1 a simplified layout of a LiDAR system 1 is shown in side view. The laser 6 emits light in a specific direction (a specific solid angle section) determined by the optics (not shown) in the laser light path (solid line). The detector unit 4 is oriented to detect reflected light from the same direction as shown by the broken lines. In the far field, the illuminated area completely overlaps the area imaged on the detector unit, as can be seen by the increasing overlap between the area enclosed by the solid lines and the dashed lines in the figure.

2 now shows the structure of a detector unit 4 or a macropixel in front view. The detector unit 4 comprises a uniform matrix arrangement of detector elements 5 (micropixels), with the number of lines and rows of detector elements 5 preferably being the same, but can also be different.

3 10 now shows an embodiment of a LiDAR system 1 according to the invention with an emitter arrangement 2, with a multiplicity of lasers 6 in a matrix arrangement, adjacent to the detector arrangement 3, also with a matrix arrangement of detector units 4/macropixels. The detector units 4 each comprise a matrix arrangement of detector elements 5/micropixels as in FIG 2 shown. One laser 6 is assigned to one detector unit 4 each.

Although the invention has been illustrated and described in detail by means of preferred exemplary embodiments, the invention is not limited by the disclosed examples and other variations can be derived therefrom by a person skilled in the art without departing from the protective scope of the invention.

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• WO 2012019871 A1 [0006]

Claims (10)

LiDAR system (1), in particular for a vehicle, comprising - an emitter arrangement (2) which comprises at least one laser (6), - a detector arrangement (3) which comprises at least one detector unit (4), each detector unit (4 ) comprises a matrix arrangement of detector elements (5), - each detector unit (4) being aligned to the same solid angle range as one of the lasers (6), so that the detector unit (4) can at least detect the laser light of an associated laser (6) reflected by surrounding objects , - a control unit that is set up to evaluate the signals detected by the detector elements (5) in order to detect the distance from surrounding objects , characterized in that the control unit is set up to evaluate the individual signals of the detector elements (5) in order to Recording spatial speckle pattern generated by reflection on a surface of a surrounding object, the control unit being used to do this The goal is to analyze the spatial speckle pattern to calculate a surface roughness of the surface of the surrounding object. LiDAR system (1) after claim 1 , wherein the control unit is set up to carry out one or more of the following methods for calculating the surface roughness: - texture analysis method, - speckle contrast method, - speckle correlation method. LiDAR system (1) after claim 1 or 2 wherein at least one laser (6) is a vertical cavity surface emitting laser or a distributed feedback laser. LiDAR system (1) according to one of the preceding claims, wherein the emitter arrangement (2) comprises a plurality of lasers (6) in a matrix arrangement and the detector arrangement (3) comprises a plurality of detector units (4) in a matrix arrangement, each laser (6) at least one detector unit (4) is assigned for the same solid angle range. LiDAR system (1) according to one of the preceding claims, wherein the control unit is set up to determine the type of the surrounding object from the determined surface roughness. LiDAR system (1) after claim 5 , wherein the control unit is set up to also include one or more of the following information in the determination of the type of surrounding object: - a specific distance of the surrounding object, - a specific solid angle size of the surrounding object by comparing the measurements of neighboring detector units (4), - a certain relative speed between the LiDAR system (1) and the surrounding object. Control method for a LiDAR system (1) according to any one of the preceding claims, comprising the following steps: - Emission of laser light by the laser (6) in a solid angle range, - evaluating the individual signals of the detector elements (5) assigned to the laser (6) in order to record a spatial speckle pattern generated by the reflection on a surface of a surrounding object, - Analyzing the spatial speckle pattern to calculate a surface roughness of the surface of the surrounding object. tax procedure claim 7 wherein the step of analyzing the spatial speckle pattern to calculate the surface roughness comprises one of the following methods: - texture analysis method, - speckle contrast method, - speckle correlation method. tax procedure claim 7 or 8th , comprising the following step after the calculation of the surface roughness: - determining the type of the surrounding object from the calculated surface roughness, one or more of the following information preferably being included: - a specific distance from the surrounding object, - a specific solid angle size of the surrounding object by comparing the measurements from neighboring detector units (4), - a specific shape of the surrounding object by comparing the measurements from neighboring detector units (4), - a specific relative speed between the LiDAR system (1) and the surrounding object. Tax procedure according to one of the Claims 7 until 9 , wherein the steps of emitting laser light and evaluating the individual signals for recording a spatial speckle pattern are carried out at least twice with different power settings of the laser (6) before the spatial speckle patterns recorded at different powers are analyzed for surface roughness of the surface of the surrounding object.

Images