Computer-implemented method, electronic vehicle system and computer program product

Abstract

The application relates to a computer-implemented method (30) for determining a degree of reflection (RG) in a point cloud (28) of a captured environment (22), said point cloud having been generated by means of an active optical sensor system (10), wherein the point cloud (28) has point information (28.P) relating to respective reflection points, wherein the point information (28.P) has information relating to a respective echo signal (40), the method comprising: determining a degree of reflection (RG) for a respective echo signal (40) of a respective reflection point using the point information (28.P) of the reflection point. The invention further relates to a method for the at least semi-autonomous guidance of a vehicle (20), to an electronic vehicle system, to a vehicle (20), to a computer program product, to a computing unit (24) and to a data carrier.

Description

[0001] 2023PF01570 1

[0002] COMPUTER-IMPLEMENTED PROCESS, ELECTRONIC VEHICLE SYSTEM AND COMPUTER PROGRAM PRODUCT

[0003] Technical field

[0004] The application relates to a computer-implemented method for performing an evaluation of a point cloud of a detected environment generated by an active optical sensor system, e.g., a lidar system. The application further relates to an electronic vehicle system comprising a processing unit and an active optical sensor system, e.g., a lidar system, as well as a computer program product. The active optical sensor system is configured to generate a point cloud.

[0005] background

[0006] Modern vehicles (cars, vans, trucks, motorcycles, etc.) are equipped with a multitude of sensor systems whose data serves to inform the driver and / or provide data to driver assistance systems. These sensor systems detect the vehicle's surroundings, including other road users. Based on the collected data, a model of the vehicle's environment can be created, and the system can react to changes in this environment.

[0007] Sensor systems are constantly being developed for various functions, such as environmental sensing in the near and far range of vehicles, including people, cars, and commercial vehicles. Sensor systems can also be used for driver assistance systems, particularly those for autonomous or semi-autonomous vehicle control. They can be used specifically for detecting obstacles and / or other road users in the front, rear, or blind spot areas of a vehicle. Sensor systems can be based on various sensor principles, such as radar, ultrasound, optics, etc.

[0008] Lidar technology (Lidar stands for Light Detection and Ranging) is an important sensor principle for environmental perception, based on optical electromagnetic 2023PF01570 2

[0009] This technology is wave-based and used in active optical sensor systems. An active optical sensor system, such as a lidar system, has an optical transmitter and an optical receiver. The transmitter emits light in the form of an optical signal. This optical signal can be pulsed and / or modulated. In a lidar system, laser beams in the ultraviolet, visible, or infrared range can be used as the light source. The receiver picks up the optical signal after reflection from the lidar system's surroundings. Using the transmitted optical signal, the lidar system's evaluation unit can analyze the received optical signal, for example, using a time-of-flight (TOF) method, to determine the spatial location and distance of the objects from which the reflection occurred.Furthermore, it is possible to determine a relative velocity. In this context, reflection or reflected light is understood to mean any light that is thrown back and is specifically intended to include light thrown back by scattering or absorption-emission.

[0010] Light reflected from the surroundings can be detected in the receiving device by a receiving sensor. Receiving sensors in lidar systems can have multiple receiving elements, called pixels, for opto-electrical conversion. The pixels can be configured to receive optical signals from different angles.

[0011] US20240061091A1 describes methods and apparatus for measuring the reflectivity of an object using a lidar system. A receiver of the lidar system receives an echo signal of a transmitted optical signal reflected by the object and converts the echo signal into an electrical signal to obtain a photocurrent integral of the echo signal. The reflectivity of the object is determined based on a predefined calibration curve and the photocurrent integral of the echo signal.

[0012] US20230194666A1 describes methods, devices, systems, and computer program products for estimating the reflectivity of objects in a lidar system. Echo signals from reflections at a lidar are described.

[0013] The object's reflectivity is estimated by accumulating data over the multitude of sampling cycles and using features of the accumulated echo signal.

[0014] US20230243919A1 describes a vehicle assistance system for classifying objects in the vehicle's environment. The system includes at least one memory that stores classification information for classifying a multitude of objects; and at least one processor that receives a multitude of acquisition results from a lidar system. The acquisition results pertain to specific objects and include location information of the object and the reflectivity of the object's surface.

[0015] Overview

[0016] A point cloud of a detected environment, generated by an active optical sensor system, contains point information for each reflection point, with the point information including information for a corresponding echo signal. A computer-implemented method for determining the reflectance in the point cloud of a detected environment generated by the active optical sensor system exhibits:

[0017] Determining the reflectance value for a given echo signal from a given reflection point using the point information of the reflection point.

[0018] Reflectance refers to the ratio of the reflected energy of an optical signal to the incident energy of the optical signal at a given wavelength. It is a measurable quantity that depends on how much light was reflected from a surface at the reflection point under the conditions during point cloud recording by the active optical sensor system.

[0019] The point cloud is generated by the active optical sensor system, e.g., a lidar system. The points in the point cloud contain point information. This point information depends on the position of the reflection point in the field of view, the distance of each reflection point, and the echo signal of the received reflection. The reflection points are those points where the reflection of the signal emitted by the sensor system and 2023PF01570 4

[0020] The point cloud is generated from an optical signal received as an echo signal. It can be understood as a set of points, each with point information including coordinates in a three-dimensional coordinate system. In the case of a three-dimensional point cloud, the three-dimensional coordinates can be determined, for example, by the direction of incidence (solid angle) of a light ray reflected at the respective reflection point and the corresponding travel time or radial distance measured for that point. In addition to spatial information, such as the three-dimensional coordinates which depend on the distance to the reflection point, and point information which depends on the echo signal, the point cloud can also contain additional information for each point, which may depend on further measurement data of the received echo signal.The point cloud may contain information relating to the echo signal, such as information concerning the signal width, peak, and / or area. In particular, it may include information from which the signal width, peak, and / or area of ​​the echo signal can be determined or approximately determined.

[0021] The three-dimensional coordinate system of the point cloud can be a three-dimensional polar coordinate system, which is based, for example, on solid angles. However, the spatial information can also be specified for the points in other coordinate systems, such as Cartesian coordinates. A conversion can take place.

[0022] To receive the optical signal, the sensor system includes an optical receiver. The optical receiver comprises an optical receiver sensor and a readout device. The receiver sensor can have multiple receiver pixels, each configured to convert an optical signal into an electrical receiver signal. Individual pixels can be configured to receive the echo signal from specific spatial directions, so that the pixel array of the receiver sensor covers the environment detected by the sensor system. The electrical receiver signal can, for example, represent the shape of the echo signal. 2023PF01570 5

[0023] The pixels of the optical receiver sensor contain light-sensitive components that capture light and convert it into an electrical signal, which can then be read out by the readout device and made available for further processing. These light-sensitive components can be photodiodes, charge-coupled devices (CCDs), CMOS sensors, photomultiplier tubes, SPADs, or similar devices.

[0024] The computer-implemented method for determining the reflectance uses the point information from the point cloud of the detected environment, generated by the active optical sensor system, to determine the reflectance of a given echo signal from a given reflection point. The reflectance is determined by calculating the ratio of the echo signal's energy to the energy of the emitted optical signal. In one embodiment of the method, the reflectivity at the reflection point is determined using the reflectance. This determination of the reflectivity at the reflection point is performed, in particular, using artificial intelligence.

[0025] Reflectivity refers to the inherent or maximum potential of a material to reflect light. It is a material property and describes the ability of a surface to reflect energy. Reflectivity is a property of the surface at the point of reflection.

[0026] Reflectivity thus refers to the reflection potential of a surface and depends in particular on the material of the surface. Reflectance refers to the proportion of incident light that is reflected at the reflection point under the conditions at the time of reflection.

[0027] This method offers the advantage that it is possible to determine reflectivity as a material property from the reflectance value, which can be derived from measured values. The reflectance values ​​were measured by the active optical sensor system during the generation of the point cloud. Given a known emitted optical signal, the energy of the corresponding echo signal can be compared to it, and the reflectance at the reflection point can be determined. This information is contained within the point information of the point cloud. 2023PF01570 6

[0028] Determining reflectivity using the reflectance value can be achieved, for example, with artificial intelligence. The artificial intelligence can, for instance, consist of one or more neural networks trained specifically for this task. The reflectance value, or a dependent variable, can serve as the input for the artificial intelligence. The output can include the reflectivity value and / or a dependent variable.

[0029] In one embodiment, the method includes extracting the reflectance as a feature from the point information using artificial intelligence feature extraction.

[0030] Feature extraction allows one or more features to be extracted as patterns from point information. This can include automated feature extraction using artificial intelligence. The extracted feature(s) can then be used for further processing, such as classification. Feature extraction reduces data complexity and allows the determination of reflectance, an essential property of surfaces and reflection points in the environment. Various feature extraction methods can be used, such as convolutional neural networks (CNNs) and / or deep learning techniques.

[0031] In one embodiment of the method, the reflectance is used for object detection in the point cloud. The reflectance, which, for example, was extracted as a feature from the point cloud's point information, can thus be used for object detection in a subsequent step. The reflectance can be one of several extracted features used for object detection.

[0032] In one embodiment of the method, the reflectivity of the detected object is determined using artificial intelligence and its reflectance value. Thus, for a detected object, the reflectivity can be determined from the reflectance value calculated for the object and / or its reflection points using artificial intelligence. The reflectivity of the detected object can then serve as additional information in the further processing of the point cloud, e.g., in 2023PF01570 7

[0033] its use in an algorithm for the automatic or semi-automatic driving of a vehicle.

[0034] In one embodiment of the method, the reflection points and / or objects are classified according to their respective determined reflectivity. This classification can be used in further processing or analysis of the point cloud to identify, for example, optical artifacts such as blooming or crosstalk, which may depend on reflectivity.

[0035] In one embodiment of the method, the point information used to determine the reflectance depends on the peak signal of the echo signal, the signal width of the echo signal, the signal area of ​​the echo signal, and / or the distance to the reflection point. This point information is particularly advantageous for pulsed optical signals, i.e., for a point cloud generated by an active optical sensor system that operates with pulsed signals. The peak signal of the echo signal corresponds to the maximum amplitude of the received reflection of the optical signal emitted by the active optical sensor system. The signal width of the echo signal corresponds to the duration for which the received reflection of the optical signal emitted by the active optical sensor system remains above a predefinable threshold.The signal area of ​​the echo signal relates to the area corresponding to a surface integral that lies under the received reflection of the optical signal emitted by the active optical sensor system.

[0036] To transmit pulsed optical signals, the optical sensor system, e.g., a lidar system, can have an optical transmitter that emits the optical signal in pulses. The pulsed optical signal has short periods during which the optical signal is transmitted. This can be referred to as a pulse. Between the pulses, no light is transmitted by the optical transmitter. The reflections of the optical signal pulses L at reflection points in the environment are then received by an optical receiver of the sensor system as an echo signal.

[0037] In a method for at least partially autonomous driving of a vehicle, an active optical sensor system of the vehicle is used to determine a 2023PF01570 8

[0038] A point cloud is generated. The point cloud contains point information for each reflection point. This point information includes information about an echo signal. The described computer-implemented procedure is carried out using at least one processing unit in the vehicle. The vehicle is guided at least partially autonomously based on the result of the reflection coefficient determination.

[0039] An electronic vehicle system comprises a processing unit and an active optical sensor system. The active optical sensor system is configured to generate a point cloud of the detected environment. The point cloud contains point information for each reflection point. This point information includes information about an echo signal. The processing unit is configured to determine the reflectance of the echo signal, calculating the reflectance for each echo signal from each reflection point using the point information from that reflection point.

[0040] In one embodiment of the vehicle system, the computing unit is configured to determine the reflectivity at the reflection point using artificial intelligence.

[0041] In one embodiment of the vehicle system, the active optical sensor system includes the processing unit. Alternatively, the processing unit can be separate from the active optical sensor system and, for example, have a communication link to it. The processing unit can, for example, be located in a central vehicle computer.

[0042] In one embodiment, the vehicle system is configured to guide the vehicle at least partially autonomously, depending on the result of the determination of the reflectance.

[0043] The vehicle may have the described vehicle system.

[0044] One embodiment of a computer program product includes instructions which, when executed by a processing unit, cause the processing unit to perform the described computer-implemented method for determining a reflectance. 2023PF01570 9

[0045] One embodiment of the computer program product includes commands which, when executed by the described electronic vehicle system, cause the electronic vehicle system to perform the described method for at least partially autonomous vehicle control. The processing unit includes means to perform the described method for determining a reflectance.

[0046] The described computer program product can be stored on a computer-readable non-volatile data carrier.

[0047] Tour list

[0048] The following section provides further explanation and description of exemplary implementations of this application with reference to the figures. They show

[0049] Fig. 1 schematically shows a vehicle with an active optical sensor system and a computing unit,

[0050] Fig. 2 schematically shows a classification of points in a point cloud; Fig. 3 schematically shows a processing of the point cloud using artificial intelligence.

[0051] Fig. 4 schematically illustrates an echo signal.

[0052] Fig. 5 schematically illustrates echo signals.

[0053] Fig. 6 schematically illustrates exemplary training data.

[0054] The same reference symbols are used in the figures for identical or similar elements. Representations in the figures may not be to scale.

[0055] Tour description

[0056] Figure 1 schematically depicts a vehicle 20, for example a passenger car. The vehicle 20 has an active optical sensor system 10, e.g. a lidar system, and a processing unit 24. The optical sensor system 10 is arranged in a front area of ​​the vehicle 20, and the environment 22 it detects is located in front of the vehicle 20 in the direction of travel.

[0057] The active optical sensor system 10 comprises an optical transmitter 12, an optical receiver 14, an optical deflector 16, and an evaluation unit 18. The evaluation unit 18 can include a processor, an FPGA, or similar device for processing data.

[0058] The optical transmitter 12 emits light in the form of an optical signal L. It has a light source for emitting, for example, laser light. The optical receiver 14 receives the optical signal L reflected at reflection points in the environment 22 as an echo signal (40).

[0059] Optionally, the optical transmitter 12 can transmit the optical signal L in pulses. The pulsed optical signal L has short periods during which the optical signal L is transmitted. This can be referred to as a pulse. Between the pulses, no light is transmitted by the optical transmitter 12. The reflections of the pulses of the optical signal L in the environment 22 are then received by the optical receiver 14 as an echo signal (40).

[0060] Preferably, the optical receiving device 14 comprises a receiving sensor that serves as an optoelectronic detector. The receiving sensor can, for example, be a point sensor, line sensor, or area sensor, in particular one or more avalanche photodiodes, photodiode cells, CCD sensors, active pixel sensors, for example, CMOS sensors, or the like. The optical signal L can be received by the receiving sensor as an echo signal 40 and converted into electrical receiving signals. The electrical receiving signals can be processed by the evaluation device 18.

[0061] The optical deflection device 16 is configured to deflect the optical signal L transmitted by the optical transmitter 12 into the environment 22 and / or to deflect the optical signal L reflected at reflection points in the environment 22 as an echo signal 40 to the optical receiver 14. The deflection device 16 can be controlled such that the optical signal L performs a scanning motion 26 over the environment 22. For example, the deflection device 16 can include a rotating mirror device that performs a rotational movement to deflect the optical signal L such that the scanning motion 26 is carried out by the optical signal L. 2023PF01570 11

[0062] During the rotational movement, the angular position of the deflection device 16 is changed.

[0063] The evaluation unit 18 is configured to control the transmission of the optical signal L, particularly as a function of the angular position of the deflection device 16. The evaluation unit 18 is further configured to evaluate the transmitted optical signal L and the received echo signal 40. Using the evaluation data generated in this way, a point cloud 28 can be created. The point cloud 28 can be transmitted to a processing unit 24 of the vehicle 20. The processing unit 24 can optionally be part of the optical sensor system 10. In the processing unit 24, the point cloud 28 can be further evaluated and / or processed. The processed point cloud 28 can be output by the processing unit 24 as a processed point cloud 28'.

[0064] Point cloud 28 can be generated, for example, in the optical sensor system 10 from the evaluation data, or point cloud 28 can be generated, for example, in the processing unit 24 from raw data of the optical sensor system 10. In particular, point cloud 28 can be generated using the evaluation data generated by the evaluation unit 18.

[0065] The point cloud 28 contains the reflection points of the optical signal L in the environment 22. Point information 28.P is specified for each point in the point cloud 28, which depends on the evaluation data, such as the distance DIST and further information about the echo signal 40. The distance DIST depends on the distance of the sensor system 10 to the reflection point. The further information about the echo signal 40 includes, for example, the width of a received pulse of the echo signal 40 and / or a peak value of the received pulse of the echo signal 40. The point cloud 28 can be used, for example, to detect objects O, to determine the distance to objects O, and / or to perform further evaluations. In the embodiment shown in Figure 1, the point cloud 28 is generated in the evaluation unit 18 of the optical sensor system 10.In the computing unit 24, a procedure 30 is carried out to determine a reflectance RG, in which the point cloud 28 is further evaluated and optionally post-processed-2023PF01570 12.

[0066] The processing unit 24 can then optionally output the point cloud 28 processed in this way as a processed point cloud 28'. The point cloud 28' processed in this way can then be further processed, for example, by other control units of the vehicle 20, e.g., for the execution of driving functions.

[0067] The computing unit 24 can, for example, be configured as the central vehicle computer of the vehicle 20, in which data from several sensor systems of the vehicle 20 can be received, evaluated, and / or further processed. The computing unit 24 can, for example, be used to implement autonomous or semi-autonomous driving functions. The computing unit 24 can further process the processed point cloud 28' and use it, for example, to execute the driving functions.

[0068] The optical sensor system 10 can, for example, be mounted or integrated at the front of the vehicle 20. Optical sensor systems 10 are also possible for other parts of the vehicle 20, e.g., for surround-view functions, such as on the sides and / or rear of the vehicle 20. It is also possible to arrange further optical sensor systems 10, such as lidar sensors, and / or other sensor systems such as radar, ultrasound, etc., on the vehicle 20, particularly in corner areas of the vehicle 20.

[0069] The optical sensor system 10 can be used to detect stationary or moving objects O in the environment 22. Such objects O can include items such as vehicles, people, animals, plants, obstacles, road surface irregularities, in particular potholes or stones, road boundaries, traffic signs, open areas, in particular parking lots, precipitation, or the like.

[0070] Figure 2 schematically shows a classification of points in a point cloud 28 based on the point information 28. P.

[0071] The point information 28.P of the points in point cloud 28 is processed by method 30 to determine the reflectance RG. Method 30 determines the reflectance RG for the respective echo signal 40 of the respective reflection point using the point information 28.P of the reflection point in point cloud 28. Method 30 can, for example, include artificial intelligence such as at least a neural network. 2023PF01570 13

[0072] The reflectance RG is extracted as a feature from the point information 28. P by means of a feature extraction 32 of an artificial intelligence using method 30.

[0073] Method 30 then determines the reflectivity RFL at the respective reflection point using the reflectance RG. This determination can also include an estimate. For this purpose, the method includes artificial intelligence 34 for determining the reflectivity RFL, e.g., a neural network, which has already been trained with measurement data of reflectance RG together with environmental factors to determine the reflectivity RFL from the reflectance RG. The training of the artificial intelligence 34 for determining the reflectivity RFL can, in particular, have been carried out using training data TD.

[0074] Depending on the determined reflectivity RFL, the points of the point cloud 28 are then classified using a classification 36 and assigned to the classes CL1, CL2, CL3, and CL4. The processed point cloud 28' then comprises the point information classified in this way: 28.PI, 28.P2, 28.P3, and 28.P4. The classification 36 can also incorporate artificial intelligence. The reflectance RG of a surface is the amount of light reflected by the surface. It is a ratio between the radiance of the surface and the irradiance of the surface and, as such, is dimensionless, with values ​​between 0 and 1. This property is particularly useful for surfaces with predominantly Lambertian diffuse reflection. For other surfaces, it may otherwise be too dependent on the angle of incidence. The reflectance RG is sometimes also referred to as (relative) reflectivity.The reflectance RG is a property of the surface at the reflection point and refers to the proportion of the incident optical power that is reflected at the reflection point at a specific wavelength.

[0075] Reflectivity, as a material property of the surface at the reflection point, is given in decibels (dB).

[0076] For a determined or estimated reflectivity RFL, Oest, the following four classes are conceivable, for example. 2023PF01570 14

[0077] • CL1 (Low) for ( <P est < c k)

[0078] • CL2 (High) (cp

[0079]

[0080] < est < <P k+n)

[0081] • CL3 (Very High) ( <P k+n)

[0082] • CL4 (Retro) ( > <P k+n oder Sättigung des Empfangssensors)

[0083] Class CL4 includes retroreflective points, such as those found on objects 0 like traffic signs. Retroreflective points on objects 0 reflect a very large portion of the optical signal L emitted by the active optical sensor system. Thus, the receiving sensor receives an echo signal 40 with high optical intensity. Such echo signals 40, and the associated high optical intensity of the received signal, can cause problems with the dynamics of the detected optical signal in the receiving sensor. Pixels of the optical receiving sensor that receive optical signals of particularly high intensity become saturated. Saturation means that more light reaches the pixels than they are capable of converting into an electrical received signal.Optical signals reflected by such retroreflective objects can cause crosstalk to other pixels of the receiving sensor. This can impair the evaluation of the echo signal.

[0084] Method 30 optionally allows the determined reflectance RG to be used for object detection in the point cloud 28. For this purpose, features extracted via feature extraction 32 can be used. The extracted features include, in particular, the reflectance, which can serve as a feature for object detection.

[0085] For objects O detected in this way, the reflectivity RFL of the respective object O can then be determined using artificial intelligence 34 to determine the reflectivity RFL using the reflectance RG of the respective object O and / or the respective reflection points.

[0086] Optionally, the identified objects O can be classified according to their determined reflectivity RFL. Here, too, classification into the four classes mentioned above—CL1, CL2, CL3, CL4—is possible. 2023PF01570 15

[0087] Figure 3 schematically illustrates an example of the processing of the point cloud by method 30 using artificial intelligence.

[0088] The point information 28. P includes, for example, the distance DIST to the reflection point, the width BR of the received echo signal 40 and the value of the peak SP of the echo signal 40.

[0089] The point information 28.P is used by a feature extraction 32 of the procedure 30 to extract the reflectance RG as a feature to points of the point cloud 28.

[0090] The artificial intelligence 34 determines the reflectivity RFL at the respective reflection point using the reflectance RG. This determination can also include an estimate. For this purpose, the artificial intelligence 34, e.g., a neural network, has already been trained with measurement data of reflectance RG together with environmental factors, e.g., a threshold TH, to determine the reflectivity RFL from the reflectance RG. The training can, for example, have been carried out using training data TD. One embodiment of possible training data TD is shown in Figure 6.

[0091] Depending on the determined reflectivity RFL, the points of the point cloud 28 are then classified using the classification 36 and divided into the classes CL1, CL2, CL3, CL4.

[0092] Optionally, the reflectance RG and / or the reflectivity RFL related to the reflection points and / or detected objects O can be stored in the point information 28. P and the processed point cloud 28' with this point information can be output by method 30.

[0093] Figure 4 schematically shows an exemplary echo signal 40 with its amplitude A over time t.

[0094] The received power of the echo signal 40, the distance DIST of the reflection point, and the solid angle at which the echo signal 40 strikes the receiving sensor are interdependent. The reflectance RG can be estimated using the ratio between the area FL of the echo signal 40 and the maximum area. The maximum area, as a reference quantity, refers to the surface of a white, flat object O in the same 2023PF01570 16

[0095] Area that is orthogonal to the beam axis of the optical signal L and whose size is larger than the illumination of the surroundings 22 by the optical signal L.

[0096] The peak SP represents the highest value of the amplitude A of the echo signal 40. The distance DIST is determined from the midpoint of the width BR of the echo signal 40. The width BR of the echo signal 40 is determined between the two points where the echo signal intersects a predefined threshold TH.

[0097] The relative area i R, which can represent an estimate of the reflectance RG, can be determined as: p R = (BR* SP) / Max, with relative area i R, width BR of the echo signal 40, peak SP of the echo signal 40 and the maximum area i Max.

[0098] Figure 5 schematically illustrates exemplary echo signals 40 under different environmental conditions.

[0099] The measurement and thus the estimation of the reflectance RG is influenced by ambient radiation, weather conditions, and similar factors, and therefore depends on environmental conditions and, in particular, on ambient light 38. The area FL and the estimated area i R are, for example, dependent on the time of day and night. During the day, ambient light 38 is present in the form of constant ambient radiation. This is illustrated in the right part of Figure 5. The left part of Figure 5, for example, shows the situation at night. The ambient light 38 influences the estimation of the peak SP and the determination of the threshold TH.

[0100] Figure 6 schematically illustrates exemplary training data TD, which can be used, for example, for training the artificial intelligence 34 to determine reflectivity RFL. The artificial intelligence 34 for determining reflectivity RFL can, for example, consist of a neural network, in particular a multilayer neural network.

[0101] The artificial intelligence neural network 34 for determining reflectivity RFL can be trained with measurement data obtained from measurements. The measurements can be taken with the optical sensor system 10, e.g., Li-2023PF01570 17

[0102] Measurements may have been performed using a DRA sensor with respect to various reflectances (e.g., from 0% to 100% reflectance). These measurements may also include measurements on retroreflective materials.

[0103] The measurement data can also be acquired for selected angular positions, i.e., solid angles, in the horizontal field of view (H_FOV) and vertical field of view (V_FOV). Horizontal and vertical refer to the field of view of the sensor system when installed in the vehicle. Measurements can be performed both during the day and at night using non-reflective objects, weakly to strongly reflective objects, and retroreflective objects at various distances, under different weather conditions, and / or with different solid angles. Examples of such acquired measurement data are shown in Figure 6.

[0104] These measurement data can then be used as training data (TD) for the artificial intelligence 34, e.g., the neural network of the artificial intelligence 34. Various possible combinations can be trained to classify the point information 28. P into the four described classes CL1, CL2, CL3, CL4.

[0105] The columns of the training data TD in Figure 6 can also be referred to as channels, which can be used as inputs for training. The channels can include, for example, the peak SP, the width BR, the area FL, the distance DIST, or similar parameters. The features generated from these channels, particularly those relating to the reflectance RG, can then be used as described to determine the reflectivity RFL of an object 0. 2023PF01570 18

[0106] Reference mark

[0107] 10 optical sensor system

[0108] 12 Optical transmitting device 14 Optical receiving device 16 Optical deflection device 18 Evaluation device

[0109] 20 vehicles

[0110] 22 surroundings

[0111] 24 computing units

[0112] 26 scanning movement

[0113] 28 point cloud

[0114] 28' processed point cloud

[0115] 28. P Point Information

[0116] 28. PI, 28. P2, 28. P3, 28. P4 classified point information 30 artificial intelligence

[0117] 32 Feature Extraction

[0118] 34 Determining Reflectivity

[0119] 36 Classification

[0120] 38 ambient lights

[0121] 40 Echo signal

[0122] A Amplitude

[0123] CL1, CL2, CL3, CL4 classes

[0124] DIST distance

[0125] BR width

[0126] SP Peak

[0127] FL area

[0128] L optical signal

[0129] RG Reflectance

[0130] RFL reflectivity

[0131] TH threshold

[0132] TD Training Data

Claims

2023PF01570 19 REQUIREMENTS 1. Computer-implemented method (30) for determining a reflection coefficient (RG) in a point cloud (28) of a detected environment (22) generated by means of an active optical sensor system (10), wherein the point cloud (28) has respective point information (28.P) for each reflection point, wherein the point information (28.P) has information about a respective echo signal (40), wherein the method comprises: Determining a reflection coefficient (RG) for a given echo signal (40) of a given reflection point using the point information (28. P) of the reflection point.

2. The method according to claim 1, further comprising: Determining the reflectivity (RFL) at the reflection point using artificial intelligence and the reflectance coefficient (RG).

3. Method according to claim 1 or 2. Extracting the reflectance (RG) as a feature from the point information (28. P) using artificial intelligence feature extraction.

4. Method according to any one of the preceding claims, further comprising: Using the reflectance (RG) for object detection in the point cloud (28).

5. The method according to claim 4, further comprising: Determining the reflectivity (RFL) of the selected object (0) using artificial intelligence and the degree of reflection (RG).

6. A method according to any one of claims 2 to 5, wherein the reflection points and / or the objects (O) are classified depending on the determined reflectivity (RFL). 2023PF01570 20 7. Method according to one of the preceding claims, wherein the point information (28. P) used to determine the reflectance (RG) depends on the signal peak (SP) of the echo signal, the signal width (BR.) of the echo signal, the signal area (FL) of the echo signal and / or the distance (DIST) to the reflection point.

8. Method for at least partially autonomous driving of a vehicle (20), wherein a point cloud (28) is generated by means of an active optical sensor system (10) of the vehicle (20), wherein the point cloud (28) has respective point information (28 :P) for respective reflection points, wherein the point information (28. P) has information about an echo signal (40), wherein a computer-implemented method (200) for determining a reflectance (RG) according to one of the preceding claims is carried out by means of at least one computing unit (24) of the vehicle (20), and the vehicle (20) is driven at least partially autonomously depending on a result of the investigation.

9. Electronic vehicle system comprising a computing unit (24) and an active optical sensor system (10), wherein the active optical sensor system (10) is configured to generate a point cloud (28) of a detected environment (22), wherein the point cloud (28) contains point information (28.P) for each reflection point, wherein the point information (28.P) contains information about an echo signal (40), wherein the computing unit (24) is configured to determine a reflectance (RG) of the echo signal (40), wherein the reflectance (RG) for each echo signal (40) of each reflection point is determined using the point information (28.P) of the reflection point.

10. Vehicle system according to claim 9, wherein the computing unit (24) is configured to determine the reflectivity (RFL) at the reflection point using artificial intelligence. 2023PF01570 21 11. Vehicle system according to claim 9 or 10, wherein the active optical sensor system (10) comprises the computing unit (24).

12. Vehicle system according to one of claims 9 to 11, wherein the vehicle system is configured to guide the vehicle (20) at least partially autonomously depending on a result of the determination of the reflectance (RG).

13. Vehicle (20) comprising a vehicle system according to any one of claims 9 to 12.

14. Computer program product with commands that when executed by a computing unit (24), cause the computing unit (24) to carry out a computer-implemented method (30) according to one of claims 1 to 7, or When implemented by an electronic vehicle system according to one of claims 9 to 12, cause the electronic vehicle system to perform a method according to claim 8.

15. Computing unit (24) comprising means for carrying out the method (30) according to any one of claims 1 to 7.

16. Computer-readable non-volatile data carrier on which the computer program product according to claim 14 is stored.

Images