ACTIVE SENSOR SYSTEM AND OBJECT DETECTION

Description

[0001] The present invention relates to a method for operating an active sensor system, wherein, by means of the sensor system, electromagnetic radiation is emitted into an environment of the sensor system for generating a first point cloud during a first measurement period and for generating a second point cloud during a second measurement period, and reflected portions of the emitted radiation are detected. The first and second point clouds each contain a plurality of points, each of which is described by a spatial position and an energy characteristic. Furthermore, the invention relates to a method for automatic object recognition, a method for at least partially automatically driving a motor vehicle, an active sensor system, an electronic vehicle guidance system, and a computer program product.

[0002] Active sensor systems comprise a transmitter unit for transmitting electromagnetic signals and at least one detector for detecting reflected portions of the electromagnetic radiation. Based on the reflected and detected portions, the detector generates corresponding sensor signals. Examples of active sensor systems include lidar systems and radar systems.

[0003] As a beam of electromagnetic radiation propagates through its environment, a small portion of the electromagnetic radiation's power is generally absorbed, which can lead to fluctuations in the refractive index and, consequently, to beam distortion. This effect, also known as blooming, is particularly noticeable in highly reflective objects and correspondingly high reflected radiation powers. This effect can ultimately cause the objects detected by the active sensor system to appear larger than they actually are.

[0004] Document US 10,593,029 B2 describes a method for reducing pixel blooming. First, an image of the surroundings is captured without illuminating the surroundings. This image is subtracted from another image captured while the surroundings are illuminated with infrared light. Saturated areas are identified and removed in the subtraction image to reduce the effects of blooming.

[0005] A disadvantage of such methods is that, especially in the case of highly reflective objects, it is not possible to clearly distinguish between artifacts resulting from blooming effects and actual objects or parts thereof. This results in either unreliable identification of blooming artifacts or incorrectly treating actual objects as blooming artifacts. A sensor system for detecting highly reflective objects is known from US Pat. No. 9,080,866 B1.

[0006] Against this background, it is an object of the present invention to provide an improved concept for an active sensor system by which artifacts due to blooming effects can be reliably identified and at the same time the risk of false-positive identification of artifacts is reduced.

[0007] This object is achieved by the respective subject matter of the independent claims. Advantageous further developments and preferred embodiments are the subject matter of the dependent claims.

[0008] The improved concept is based on the idea of ​​comparing the spatial extent of areas that potentially correspond to artifacts due to blooming effects for several measurement periods and identifying artifacts depending on the result of the comparison.

[0009] According to the improved concept, a method for operating an active sensor system is specified, wherein, by means of the sensor system, electromagnetic radiation is emitted into an environment of the sensor system to generate a first point cloud during a first measurement period and to generate a second point cloud during a second measurement period, which in particular lies after the first measurement period, and respective reflected portions of the emitted radiation are detected. The first point cloud and the second point cloud each contain a plurality of points, each of which is described by a spatial position and an energy characteristic.A computing unit of the sensor system identifies a first subset of the first point cloud and a second subset of the second point cloud, wherein the respective energy characteristic of each point of the first subset and each point of the second subset is greater than or equal to a predetermined first energy threshold. A third subset of the first point cloud and a fourth subset of the second point cloud are identified by the computing unit, wherein the respective position of each point of the third subset lies within a predefined spatial environment of the points of the first subset, and the respective position of each point of the fourth subset lies within a predefined spatial environment of the points of the second subset.By means of the computing unit, a spatial extent of the fourth subset is compared with a spatial extent of the third subset and the points of the third subset and / or the points of the fourth subset are marked as artifacts depending on a result of the comparison.

[0010] The electromagnetic waves can be emitted, for example, as radio waves, in particular if the sensor system is designed as a radar system, or as light, for example as infrared light, in particular if the active sensor system is designed as a lidar system.

[0011] To generate a point cloud, the active sensor system, in particular a transmitting unit of the active sensor system, emits the electromagnetic radiation into the environment of the sensor system, where it is at least partially reflected by one or more objects in the environment. The reflected components are detected by the sensor system, in particular by a detector unit of the sensor system, and one or more detector signals are generated based on the detected components. The computing unit generates the points of the point cloud based on the detector signals. In doing so, the computing unit can determine the respective spatial position of the points, in particular three-dimensional coordinates of the respective points. For example, the radial distance of the respective point from the sensor system can be determined based on a signal propagation time measurement, also referred to as a time-of-flight measurement.Furthermore, the computing unit determines the respective energy characteristics based on the detector signals, which quantify an energy of the reflected and detected components that lead to the respective detector signal.

[0012] Apart from the radial distance, the coordinates of the respective points can be determined by a position of the respective detector in a detector array of the detector unit, such as in a flash lidar system, and / or based on a position of a deflection unit that directs the reflected portions of the electromagnetic radiation onto the detector unit, in combination with the position of the detector with respect to the deflection unit, such as in a laser scanner.

[0013] The spatial environment of the points in the first point cloud or the second point cloud can be defined differently depending on the coordinate system used or the technology employed in the sensor system. For example, if a spatial position of a point is defined by a two-dimensional position on a pixel array and a radial distance, the spatial environment can be given by a corresponding radial distance range and a corresponding two-dimensional region on the pixel array adjacent to the respective subset. If the spatial positions of the points are defined by a radial distance and two angles, as is the case with laser scanners, for example, the environments can be given by a radial distance range and corresponding angular ranges.

[0014] Since the points in the first and second subsets have energy values ​​greater than the specified first energy threshold, the points in the first and second subsets can be assigned to an object in the vicinity of the sensor system with high reflectivity. The probability of blooming effects being generated is therefore potentially increased. The points in the third and fourth subsets are spatially adjacent to the object with high reflectivity and are therefore potential artifacts.

[0015] If the sensor system moves relative to the object during or between measurement periods, the actual extent of the object does not change, so the extent of the first and second subsets should remain essentially constant. However, this is not the case for blooming artifacts. Here, the extent of the artifacts changes with the distance of the sensor system from the highly reflective object; in particular, the spatial extent of the corresponding regions increases with decreasing distance.

[0016] The situation is analogous if, according to the invention, the sensitivity of the sensor system during the first measurement period differs from the sensitivity of the sensor system during the second measurement period. Even then, the extent of the first and second subsets remains essentially constant, whereas the extent of the regions corresponding to blooming artifacts changes.

[0017] The spatial extent of the third and fourth subsets can be understood as the respective spatial area occupied by the corresponding points, whereby the spatial area can be one-dimensional, two-dimensional or three-dimensional.

[0018] The first and second measurement periods do not necessarily follow one another immediately, so there may be an additional period between the first and second measurement periods. However, in various embodiments, the first and second measurement periods may also follow one another immediately.

[0019] By comparing the spatial extent of the fourth subset with the spatial extent of the third subset, it is possible to estimate whether the apparent extent of the corresponding regions changes, which indicates that they are most likely blooming artifacts. If, however, the extent of the fourth subset is essentially the same as the extent of the third subset, they are most likely not artifacts but actual objects.

[0020] By marking the points of the third or fourth subset as artifacts depending on the result of the comparison, the quality of the corresponding point clouds is increased for their further use. Marking them as artifacts, or the process according to the improved concept, can therefore be viewed as filtering the point clouds. Depending on whether the points of the third or fourth subset are marked as artifacts or not, these points can be considered in subsequent processing steps, ignored, or considered with a correspondingly reduced confidence value. For example, in object detection, points marked as artifacts can be omitted, allowing object detection to be carried out more reliably.As a consequence, more reliable electronic map materials can be created based on object recognition and / or motor vehicles can be guided automatically or partially automatically based on the point clouds with greater reliability or safety.

[0021] According to at least one embodiment, the sensor system is moved between the first measurement period and the second measurement period and / or during the first measurement period and / or during the second measurement period, so that a distance of the sensor system to an object in the environment, which is represented by the first subset and by the second subset, is changed.

[0022] As explained above, this can be achieved by changing the spatial extent of points corresponding to artifacts due to blooming, so that a corresponding difference between the spatial extent of the third and the fourth subset becomes visible.

[0023] In particular, the sensor system can be mounted on a motor vehicle which moves with the sensor system.

[0024] According to the invention, the computing unit sets a first sensitivity of the sensor system to generate the first point cloud during the first measurement period, and a second sensitivity of the sensor system to generate the second point cloud during the second measurement period. The first sensitivity is different from the second sensitivity.

[0025] In particular, the first sensitivity is set first, and then the first point cloud is generated with the first sensitivity set. Next, the second sensitivity is set, and then the second point cloud is generated with the second sensitivity set.

[0026] The sensitivity of the sensor system corresponds, in particular, to the sensitivity of the detector unit. For example, the detector unit can comprise one or more optical detectors, such as photodiodes, in particular avalanche photodiodes, also known as avalanche photodiodes. The sensitivity is then determined, for example, by an operating voltage of the optical detectors, for example, by a reverse voltage with which the photodiodes or avalanche photodiodes are operated.

[0027] As explained above, this can also be achieved by changing the spatial extent of points corresponding to artifacts due to blooming, so that a corresponding difference between the spatial extent of the third and the fourth subset becomes visible.

[0028] The change in sensitivity can replace the movement of the sensor system between the measurement periods or be combined with it in order to achieve an even more pronounced effect and thus to be able to detect the artifacts even more reliably.

[0029] According to at least one embodiment, only points of the first point cloud are identified as part of the third subset whose energy characteristic value is less than or equal to a predetermined second energy limit, wherein the second energy limit is less than the first energy limit.

[0030] According to at least one embodiment, only points of the second point cloud are identified as part of the fourth subset whose energy characteristic is less than or equal to the second energy threshold.

[0031] In other words, only those points whose energy value is less than or equal to the second energy threshold are marked as artifacts. This takes into account the fact that the energy of reflected portions of the electromagnetic radiation that lead to artifacts due to blooming is usually significantly smaller than the energy reflected by the neighboring highly reflective objects that ultimately cause the blooming. By limiting the detection based on the second energy threshold, the risk of false-positive detection of artifacts can be further reduced.

[0032] For example, the second energy limit may be less than or equal to 50% of the first energy limit, for example less than or equal to 30% or less than or equal to 20% of the first energy limit.

[0033] By modifying the second energy threshold, a balance can also be made between the respective risks for false-negative and false-positive detection of artifacts, depending on the specific requirements for further use of the point clouds.

[0034] According to at least one embodiment, the points of the third subset and / or the points of the fourth subset are marked as artifacts only if the spatial extent of the fourth subset differs from the spatial extent of the third subset by at least a predetermined threshold value.

[0035] This can further reduce the risk of false-positive detection of artifacts. In a specific implementation, the threshold value can also depend on the length of each measurement period or the time elapsed between the first and second measurement periods.

[0036] According to at least one embodiment, at least one further point cloud is generated by means of the sensor system during at least one corresponding further measurement period after the first measurement period, wherein points of the at least one further point cloud are each described by a spatial position and an energy characteristic. For each of the at least one further point clouds, a fifth subset is identified by means of the computing unit, wherein the respective energy characteristic of each point of the fifth subset is greater than or equal to the first energy limit value. For each of the at least one further point clouds, a sixth subset is also identified, wherein the respective position of each point of the sixth subset lies within a predefined spatial environment of the points of the fifth subset of the respective further point cloud.For each of the at least one further point cloud, the points of the fourth point cloud are marked as artifacts depending on a spatial extent of the sixth subset.

[0037] The at least one further measurement period lies, for example, between the first measurement period and the second measurement period. By comparing the spatial extent of the third subset, the sixth subsets, and the fourth subsets with one another, it can be determined whether the space occupied by the points potentially corresponding to artifacts continuously increases or decreases as the distance of the sensor system from the corresponding object continuously decreases or increases. In particular, a more precise assessment can be made from the temporal progression of the spatial extent of the third subset, the sixth subsets, and the fourth subsets, so that the detection of artifacts can be carried out with greater reliability and / or with a lower risk of false-positive detection of artifacts.

[0038] According to at least one embodiment, the spatial position of the points of the respective point clouds is determined depending on a respective radial distance from the sensor system. Only points of the first point cloud are identified as part of the third subset whose radial distance lies within a predetermined radial environment of the first subset, and / or only points of the second point cloud are identified as part of the fourth subset whose radial distance lies within a predetermined radial environment of the second subset.

[0039] This takes into account the fact that the points corresponding to artifacts due to blooming effects belong to illusory objects that appear to be located at a distance from the sensor system that is at least approximately equal to the distance of the actual highly reflective object. By taking the radial surrounding area into account, the risk of false-positive detection of artifacts can be further reduced.

[0040] According to at least one embodiment, the spatial position of the points of the respective point clouds is determined depending on a horizontal or azimuthal angle of incidence of the corresponding reflective portions. Only points of the first point cloud are identified as part of the third subset whose horizontal angle of incidence lies within a predefined horizontal angular neighborhood of the first subset or of the points of the first subset, and / or only points of the second point cloud are identified as part of the fourth subset whose horizontal angle of incidence lies within a predefined horizontal angular neighborhood of the second subset or of the points of the second subset.

[0041] Such embodiments are particularly relevant when the active sensor system is a scanning sensor system, for example, a lidar system configured as a laser scanner. Such systems incorporate deflection units, such as deflection mirrors mounted for rotation or pivoting, so that at specific times, only reflected portions of the electromagnetic radiation from specific horizontal angles of incidence can be directed onto the detector unit. This allows spatial resolution to be achieved in the horizontal direction.

[0042] Since artifacts due to blooming effects primarily occur in areas in the immediate vicinity of the object with high reflectivity, the risk of false-positive detection of artifacts is further reduced by limiting the horizontal angular environment.

[0043] According to at least one embodiment, the spatial position is determined depending on a vertical angle of incidence or polar angle of the corresponding reflected portions. Only points of the first point cloud are identified as part of the third subset whose vertical angle of incidence lies within a predefined vertical angular neighborhood of the first subset, and / or only points of the second point cloud are identified as part of the fourth subset whose vertical angle of incidence lies within a predefined vertical angular neighborhood of the second subset.

[0044] The vertical angle of incidence can also be defined by a pivotable or rotatable deflection unit, as described for the horizontal angle of incidence. Alternatively, several detectors can be arranged next to one another in the vertical direction, so that reflected portions of the electromagnetic radiation from different vertical angular ranges are detected by different detectors. This also makes it possible to achieve vertical resolution. In this case, points that are all detected by the same detector are also referred to as a layer. The vertical angular environment of the second subset can, for example, be defined by a layer, so that only points of the first point cloud that belong to the same layer as the points of the first subset are identified as part of the third subset. In further embodiments, the vertical angular environment can, for example, be defined by two or more consecutive layers.The same applies to the points of the fourth subset of the second point cloud.

[0045] According to the improved concept, a method for automatic object detection is also specified. For this purpose, a method for operating an active optical sensor system is implemented according to the improved concept; thus, an active sensor system is operated according to a method according to the improved concept. By means of the computing unit, object detection is carried out depending on the second point cloud and / or depending on the first point cloud, whereby the points of the third subset are only taken into account in object detection if they are not marked as artifacts and / or the points of the fourth subset are only taken into account in object detection if they are not marked as artifacts.

[0046] In other words, the points of the third or fourth subset are not taken into account in object detection if and to the extent that they are marked as artifacts.

[0047] For object detection itself, the computing unit can implement known algorithms for processing point clouds. In particular, the computing unit can determine a position and / or extent and / or type of the object based on the first and / or second point cloud. Object detection can also include object classification, determining a bounding box for the object, and / or object tracking.

[0048] According to the improved concept, a method for at least partially automatically driving a motor vehicle is also specified. For this purpose, a method for automatic object recognition according to the improved concept is carried out, and the motor vehicle is at least partially automatically guided, in particular by means of an electronic vehicle guidance system of the motor vehicle, depending on a result of the object recognition.

[0049] An electronic vehicle guidance system can be understood here and below as an electronic system that is designed to guide or control the motor vehicle fully automatically or fully autonomously, in particular without requiring intervention by a driver. The motor vehicle or the electronic vehicle guidance system performs all necessary functions, such as any necessary steering, braking and / or acceleration maneuvers, the observation and detection of road traffic, and the associated required reactions, independently and fully automatically. In particular, the electronic vehicle guidance system can be used to implement a fully automatic or fully autonomous driving mode of the motor vehicle according to Level 5 of the SAE J3016 classification.An electronic vehicle guidance system can also be understood as a driver assistance system (ADAS), which supports the driver during partially automated or semi-autonomous driving of the motor vehicle. In particular, the electronic vehicle guidance system can be used to implement a partially automated or semi-autonomous driving mode of the motor vehicle according to one of the levels 1 to 4 according to the SAE J3016 classification. Here and below, "SAE J3016" refers to the corresponding standard in the June 2018 version.

[0050] The at least partially automated vehicle guidance may therefore involve driving the motor vehicle according to a fully automated or fully autonomous driving mode of level 5 according to SAE J3016. The at least partially automated vehicle guidance may also involve driving the motor vehicle according to a partially automated or semi-autonomous driving mode according to one of levels 1 to 4 according to SAE J3016.

[0051] According to the improved concept, an active sensor system is also specified, which has an emitter unit configured to emit electromagnetic radiation into an environment of the sensor system during a first measurement period and during a second measurement period. The sensor system has a detector unit configured, in particular controlled by a control unit of the sensor system, to detect first reflected portions of the electromagnetic radiation emitted during the first measurement period and, based thereon, to generate at least one first detector signal. The detector unit is configured to detect second reflected portions of the electromagnetic radiation emitted during the second measurement period and, based thereon, to generate at least one second detector signal.

[0052] The sensor system comprises a computing unit configured to generate a first point cloud based on the at least one first detector signal and to generate a second point cloud based on the at least one second detector signal, wherein the first and second point clouds each contain a plurality of points, each described by a spatial position and an energy characteristic. The computing unit is configured to identify a first subset of the first point cloud and to identify a second subset of the second point cloud, wherein the respective energy characteristic of each point of the first subset and the respective energy characteristic of each point of the second subset are greater than or equal to a predetermined first energy limit value.The computing unit is configured to identify a third subset of the first point cloud and a fourth subset of the second point cloud, wherein the respective position of each point of the third subset lies within a predefined spatial environment of the points of the first subset and the respective position of each point of the fourth subset lies within a predefined spatial environment of the points of the second subset. The computing unit is configured to compare a spatial extent of the fourth subset with a spatial extent of the third subset and to mark the points of the third subset and / or the fourth subset as artifacts depending on a result of the comparison.

[0053] According to at least one embodiment of the active sensor system according to the improved concept, the sensor system is designed as a radar sensor system or as a lidar sensor system, in particular as a laser scanner or as a flash lidar sensor system.

[0054] According to at least one embodiment, the sensor system is designed as a laser scanner and includes a deflection unit configured to deflect the emitted radiation in order to establish a horizontal emission angle of the emitted radiation and to deflect the reflected portions in order to establish a horizontal angle of incidence of the reflected portions.

[0055] According to at least one embodiment, the computing unit is configured to determine the energy characteristic value for each point of the first point cloud as a function of a pulse width of a signal pulse of the at least one detector signal and / or the computing unit is configured to determine the energy characteristic value for each point of the second point cloud as a function of a pulse width of a signal pulse of the at least one second detector signal.

[0056] In the context of laser scanners, these pulse widths are sometimes also referred to as echo pulse widths.

[0057] Further embodiments of the active sensor system according to the improved concept follow directly from the various embodiments of the method for operating an active sensor system according to the improved concept, the method for automatic object recognition according to the improved concept, and the method for at least partially automatically driving a motor vehicle according to the improved concept, and vice versa. In particular, an active sensor system according to the improved concept can be configured to carry out a method according to the improved concept or it carries out such a method.

[0058] According to the improved concept, an electronic vehicle guidance system is also specified which has an active sensor system according to the improved concept.

[0059] According to the improved concept, a computer program with instructions is specified, wherein the instructions, when the computer program or the instructions are executed by an active sensor system according to the improved concept, cause the active sensor system to carry out a method for operating an active sensor system according to the improved concept.

[0060] According to the improved concept, a further computer program with further instructions is also specified, wherein the further instructions, when the further instructions or the further computer program are executed by an electronic vehicle guidance system according to the improved concept, cause the vehicle guidance system to carry out a method for at least partially automatically driving a motor vehicle according to the improved concept.

[0061] According to the improved concept, a computer-readable storage medium is also provided which stores a computer program and / or another computer program according to the improved concept.

[0062] The computer program, the further computer program and the computer-readable storage medium can be referred to as respective computer program products with the instructions or the further instructions.

[0063] Further features of the invention emerge from the claims, the figures and the description of the figures. The features and combinations of features mentioned above in the description as well as the features and combinations of features mentioned below in the description of the figures and / or shown alone in the figures can be encompassed and disclosed by the improved concept not only in the respectively specified combination, but also in other combinations. Thus, even those embodiments of the improved concept are encompassed and disclosed which are not explicitly shown and / or explained in the figures, but which emerge and can be generated from the explained embodiments by means of separate combinations of features. Thus, in particular, embodiments and combinations of features are also encompassed and disclosed which do not have all the features of an originally formulated claim.Furthermore, embodiments and combinations of features are included and disclosed which go beyond or deviate from the combinations of features set out in the references to the claims.

[0064] The figures show: Fig. 1 shows a schematic representation of a motor vehicle with an exemplary embodiment of an electronic vehicle guidance system according to the improved concept; Fig. 2 shows a flowchart of an exemplary embodiment of a method for automatic object recognition according to the improved concept; Fig. 3 shows a schematic representation of a lidar point cloud; and Fig. 4 shows a schematic representation of another lidar point cloud.

[0065] In Fig. 1 A motor vehicle 1 is schematically shown, which includes an active sensor system 2 according to the improved concept. The sensor system 2 is configured, for example, as a lidar system.

[0066] The sensor system 2 includes a computing unit 3, an emitter unit 5, and a detector unit 4, each of which is connected to the computing unit 3. The detector unit 4 and the emitter unit 5 can be controlled, for example, by the computing unit 3, which can then also assume the function of a control unit.

[0067] The emitter unit 5 can emit electromagnetic radiation 6, for example infrared light, into an environment of the sensor system 2, where it can be at least partially reflected by an object 7, so that reflected portions 6' can be detected by the detector unit 4. Based on the detected portions 6', the detector unit 4 can generate at least one detector signal and transmit it to the computing unit 3.

[0068] The functioning of the sensor system 2 is explained below with reference to exemplary embodiments of a method for automatic object recognition according to the improved concept, in particular with reference to the figures Fig. 2 bis Fig. 4 , explained in more detail.

[0069] In Fig. 2 is a flowchart of an exemplary embodiment of a method for automatic object recognition according to the improved concept.

[0070] In step S1, the emitter unit 5 transmits the electromagnetic radiation 6 into the environment of the sensor system 2 during a first measurement period. The detector unit 4 detects the corresponding reflected portions 6 and generates at least one first detector signal based thereon. The computing unit 3 generates a first point cloud 8 based on the at least one first detector signal, as shown schematically in Fig. 3 The right-pointing arrow in Fig. 3 represents a direction of movement of motor vehicle 1.

[0071] In the presentation of the Fig. 3 The points of the first point cloud 8 are shown as two-dimensional projections into the road plane, with each point being assigned a straight line that corresponds to an echo pulse width of the corresponding detector signal. As shown in Fig. 3 As can be seen, an object with relatively high reflectivity is located in the vicinity of the sensor system 2, which leads to high values ​​of the respective energy characteristics. The corresponding points of the first point cloud 8 are identified by the computing unit 3 in step S2 as a subset 8a if their energy characteristic is greater than or equal to a predetermined first energy limit value.

[0072] In addition, in step S1, the emitter unit 5 again transmits electromagnetic radiation 6 into the environment of the sensor system 2 during a second measurement period, which occurs after the first measurement period. The detector unit 4 again detects the corresponding reflected portions 6 and, based thereon, generates at least one second detector signal. The computing unit 3 generates a second point cloud 9 based on the at least one second detector signal, as shown schematically in Fig. 4 is shown.

[0073] In step S2, the points of the second point cloud 9 with an energy characteristic value greater than or equal to the first energy limit value are identified by the computing unit 3 as a subset 9a.

[0074] In step S3, the computing unit 3 identifies, for example, all remaining points of the point clouds 8, 9 as points that do not belong to a highly reflective object.

[0075] In step S4, the computing unit 3 performs a check to determine whether points immediately adjacent to subset 8a are potential artifacts. For this purpose, those points of the first point cloud 8 are identified that lie within a predefined spatial area around the points of subset 8a and whose energy characteristic is less than or equal to a predefined second energy threshold that is less than the first energy threshold. Thus, points are identified that appear to be close to the highly reflective object but are due to less energetic reflections. Such points are identified as subsets 8b, 8c. An analogous check is also performed with respect to the second point cloud 9, whereby corresponding subsets 9b, 9c are identified.

[0076] In step S5, the computing unit 3 can, for example, mark all points of the subsets 8b, 8c, 9b, 9c as potential artifacts, for example by setting corresponding flags. Optionally, in step S6, the remaining points of the point clouds 8, 9 are marked as non-artifacts, for example by setting a corresponding flag.

[0077] In step S7, the computing unit 3 compares the spatial extent of the points in subset 8b with the spatial extent of the points in subset 9b or the spatial extent of the points in subset 8c with the spatial extent of subset 9c. For artifacts resulting from blooming effects, the corresponding spatial areas increase in size as the distance between the highly reflective object and sensor system 2 decreases. In step S7, the computing unit 3 therefore checks, for example, whether the spatial extent of subset 9b is at least a threshold value greater than the spatial extent of subset 8b or whether the spatial extent of subset 9c is at least the threshold value greater than the spatial extent of subset 8c.For example, a threshold value can be chosen that corresponds to an increase in the spatial extent of the relevant points by at least 10% or at least 20% every 40 ms or the like.

[0078] If this is the case, the respective points of the subsets 8b, 8c, 9b, 9c are marked as artifacts, for example, using a corresponding flag. Otherwise, the corresponding points can be identified as points representing real objects. In step S8, the computing unit 3 discards, for example, all points marked as artifacts or does not further consider these points, in particular for a subsequent object recognition algorithm.

[0079] In step S9, the computing unit 3 performs the object detection algorithm based on the points that are not marked as artifacts.

[0080] According to the improved concept, as described in particular with regard to the figures, it can be achieved that artifacts due to blooming effects can be identified with greater reliability without increasing the probability of false-positive detection of artifacts or whereby the probability of false-positive detection of artifacts can be reduced.

[0081] According to the improved concept, for example, physical constraints are defined that allow artifacts due to blooming effects to be distinguished from points representing actual objects. In various embodiments, potential artifacts near highly reflective objects are first identified. These potential artifact points can then be clustered and treated specifically. To avoid false-positive artifact detection, the spatial extent of the artifact clusters can be observed over time. Since an increase in the spatial extent is implausible for real objects, the corresponding cluster can be identified as an artifact if necessary.

[0082] In various embodiments, points corresponding to highly reflective objects are therefore first identified and clustered. Neighboring points, for example, those located within a similar radial distance range, are identified and clustered; in various embodiments, a restriction on the horizontal and / or vertical angle of incidence can be taken into account. It can be checked whether the energy characteristics of the neighboring points are small enough to be considered artifacts. Clusters that meet these criteria can be monitored over time to determine whether their spatial extent exceeds a defined maximum increase in size over time. If this is the case, these points can be treated as artifacts and not used further, for example, for object detection or object tracking algorithms.

Claims

1. Method for operating an active sensor system (2), wherein electromagnetic radiation (6) is emitted into an environment of the sensor system (2) by means of the sensor system (2) in each case in order to generate a first point cloud (8) during a first measurement period of time and to generate a second point cloud (9) during a second measurement period of time and reflected components (6') of the emitted radiation (6) are detected and the first and the second point cloud (9) each contain a plurality of points, which are each described by a spatial position and an energy characteristic value, wherein, by means of a computing unit (3) of the sensor system (2), - a first subset (8a) of the first point cloud (8) and a second subset (9a) of the second point cloud (9) are identified, wherein the respective energy characteristic value of each point in the first subset (8a) and the second subset (9a) is greater than or equal to a predetermined first energy limiting value; - a third subset (8b, 8c) of the first point cloud (8) and a fourth subset (9b, 9c) of the second point cloud (9) are identified, wherein the respective position of each point in the third subset (8b, 8c) is within a predefined spatial environment of the points in the first subset (8a) and the respective position of each point in the fourth subset (9b, 9c) is within a predefined spatial environment of the points in the second subset (9a); - a spatial extent of the fourth subset (9b, 9c) is compared with a spatial extent of the third subset (8b, 8c), and the points in the third subset (8b, 8c) and / or the fourth subset (9b, 9c) are marked as artefacts depending on a result of the comparison, characterized in that a first sensitivity of the sensor system (2) is set in order to generate the first point cloud (8) during the first measurement period of time; and a second sensitivity of the sensor system (2), which is different from the first sensitivity, is set in order to generate the second point cloud (9) during the second measurement period of time.

2. Method according to Claim 1, characterized in that the sensor system (2) is moved between the first measurement period of time and the second measurement period of time and / or during the first measurement period of time and / or during the second measurement period of time, so that a distance of the sensor system (2) to an object (7) in the environment which is represented by the first subset (8a) and the second subset (9a) is changed.

3. Method according to one of the preceding claims, characterized in that - only points in the first point cloud (8) whose energy characteristic value is less than or equal to a predetermined second energy limiting value are identified as part of the third subset (8b, 8c), wherein the second energy limiting value is less than the first energy limiting value; and / or - only points in the second point cloud (9) whose energy characteristic value is less than or equal to the second energy limiting value are identified as part of the fourth subset (9b, 9c).

4. Method according to one of the preceding claims, characterized in that the points in the third subset (8b, 8c) and / or the fourth subset (9b, 9c) are only marked as artefacts if the spatial extent of the fourth subset (9b, 9c) differs from the spatial extent of the third subset (8b, 8c) at least by a predetermined threshold value.

5. Method according to one of the preceding claims, characterized in that - by means of the sensor system (2), at least one further point cloud is generated during at least one corresponding further measurement period of time, wherein points in the at least one further point cloud are each described by a spatial position and an energy characteristic value; - for each of the at least one further point cloud, by means of the computing unit (3), - a fifth subset is identified, wherein the respective energy characteristic value of each point in the fifth subset is greater than or equal to the first energy limiting value; - a sixth subset is identified, wherein the respective position of each point in the sixth subset is within a predefined spatial environment of the points in the fifth subset of the respective further point cloud; and - the points in the fourth subset (9b, 9c) are marked as artefacts in dependence on a spatial extent of the sixth subset.

6. Method according to one of the preceding claims, characterized in that - the spatial position is determined in dependence on a respective radial distance from the sensor system (2); and - only points in the first point cloud (8) whose radial distance is in a predetermined radial environment of the first subset (8a) are identified as part of the third subset (8b, 8c) and / or only points in the second point cloud (9) whose radial distance is in a predetermined radial environment of the second subset (9a) are identified as part of the fourth subset (9b, 9c).

7. Method according to one of the preceding claims, characterized in that - the spatial position is determined in dependence on a horizontal angle of incidence of the corresponding reflected components (6'); and - only points in the first point cloud (8) whose horizontal angle of incidence is in a predetermined horizontal angle environment of the first subset (8a) are identified as part of the third subset (8b, 8c) and / or only points in the second point cloud (9) whose horizontal angle of incidence is in a predetermined horizontal angle environment of the second subset (9a) are identified as part of the fourth subset (9b, 9c).

8. Method for automatic object detection, characterized in that - an active sensor system (2) is operated according to a method according to one of the preceding claims; and - by means of the computing unit (3), the object detection is carried out in dependence on the second point cloud (9), wherein the points in the third subset (8b, 8c) and / or the points in the fourth subset (9b, 9c) are only taken into consideration if they are not marked as artefacts.

9. Active sensor system including - an emitter unit (5), which is configured to emit electromagnetic radiation (6) into an environment of the sensor system (2) in each case during a first measurement period of time and during a second measurement period of time; - a detector unit (4), which is configured, during the first measurement period of time, to detect first reflected components (6') of the radiation (6) and to generate at least one first detector signal based thereon and, during the second measurement period of time, to detect second reflected components (6') of the radiation (6) and to generate at least one second detector signal based thereon; and - a computing unit (3), which is configured to generate a first point cloud (8) based on the at least one first detector signal and to generate a second point cloud (9) based on the at least one second detector signal, wherein the first and the second point cloud (8, 9) each contain a plurality of points, which are each described by a spatial position and an energy characteristic value; wherein the computing unit (3) is configured - to identify a first subset (8a) of the first point cloud (8) and a second subset (9a) of the second point cloud (9), wherein the respective energy characteristic value of each point in the first subset (8a) and the second subset (9a) is greater than or equal to a predetermined first energy limiting value; - to identify a third subset (8b, 8c) of the first point cloud (8) and a fourth subset (9b, 9c) of the second point cloud (9), wherein the respective position of each point in the third subset (8b, 8c) is within a predefined spatial environment of the points in the first subset (8a) and the respective position of each point in the fourth subset (9b, 9c) is within a predefined spatial environment of the points in the second subset (9a); - to compare a spatial extent of the fourth subset (9b, 9c) with a spatial extent of the third subset (8b, 8c), and - to mark the points in the third subset (8b, 8c) and / or the points in the fourth subset (9b, 9c) as artefacts depending on a result of the comparison, - characterized in that the computing unit (3) is configured to set a first sensitivity of the sensor system (2) in order to generate the first point cloud (8) during the first measurement period of time; and - to set a second sensitivity of the sensor system (2), which is different from the first sensitivity, in order to generate the second point cloud (9) during the second measurement period of time.

10. Active sensor system according to Claim 9, characterized in that the sensor system (2) is designed as a radar sensor system or as a lidar sensor system.

11. Active sensor system according to Claim 9, characterized in that the sensor system (2) is designed as a laser scanner and includes a deflection unit, which is configured to deflect the emitted radiation (6), in order to define a horizontal emission angle of the emitted radiation (6), and to deflect the reflected components (6'), in order to define a horizontal angle of incidence of the reflected components (6').

12. Active sensor system according to Claim 11, characterized in that - the computing unit (3) is configured, for each point in the first point cloud (8), to determine the energy characteristic value in dependence on a pulse width of a signal pulse of the at least one first detector signal; and / or - the computing unit (3) is configured, for each point in the second point cloud (9), to determine the energy characteristic value in dependence on a pulse width of a signal pulse of the at least one second detector signal.

13. Electronic vehicle control system for a motor vehicle (1), the vehicle control system including an active sensor system (2) according to one of Claims 9 to 12.

14. Computer program product with instructions which, when executed by an active sensor system (2) according to one of Claims 9 to 12, prompt the active sensor system (2) to carry out a method according to one of Claims 1 to 8.