DE102021111013A1 - Calibration object and method for determining a position of an individual object in a scan plane of a LiDAR-based environmental sensor - Google Patents

Abstract

The invention relates to a calibration object (32) with at least three elongated individual objects (28, 30), the individual objects (28, 30) being arranged in one plane and two first individual objects (28) being arranged parallel to one another, and a second individual object (30 ) extends between the two first individual objects (28), the individual objects (28, 30) being arranged within the plane in such a way that, when the calibration object (32) is set up, they are in relation to an expected scan plane (24), in particular a horizontal scan plane (24 ), have a shallow intersection angle (α) with the expected scan plane (24). The invention also relates to a method for determining a position (38) of an individual object (28, 30) in a scanning plane (24) of a LiDAR-based environmental sensor (14), in particular for locating a calibration object (3) with a number of individual objects (28, 30 ) for calibrating the LiDAR-based environmental sensor (14), in particular with an above calibration object (32), comprising the steps of emitting a plurality of laser pulses in the scan plane (24) of the LiDAR-based environmental sensor (14), receiving reflections of the emitted laser pulses from the scanning plane (24), determining intensities (26) of the received reflections, determining reflections belonging to the individual object (28, 30), and determining the position (38) of the individual object (28, 30) based on the intensities (26 ) of the received reflections belonging to the individual object (28, 30). The invention also relates to a driving support system (12) for carrying out the above method.

Description

The present invention relates to a calibration object with at least three elongated individual objects, the individual objects being arranged in one plane and two first individual objects being arranged parallel to one another, and a second individual object extending between the two first individual objects.

The present invention also relates to a method for determining a position of an individual object in a scan plane of a LiDAR-based surroundings sensor, in particular for locating a calibration object with several individual objects for calibrating the LiDAR-based surroundings sensor, in particular with a calibration object mentioned above.

Furthermore, the present invention relates to a driving support system with a LiDAR-based environment sensor and a control unit, which are connected to one another via a data connection, the LiDAR-based environment sensor being designed to receive reflections of emitted laser pulses from at least one scan plane and intensities of the received reflections to determine, wherein the driving support system is designed to carry out the above method for determining a position of an individual object, in particular for locating a calibration object with a plurality of individual objects for calibrating the LiDAR-based environmental sensor.

LiDAR-based environmental sensors (Light Detection and Ranging) are becoming more and more important in current vehicles in order to monitor their surroundings and implement driving support systems for various support functions up to and including autonomous driving. An extrinsic calibration of LiDAR-based environmental sensors is a prerequisite for their use for many purposes. The aim is to determine the position and orientation of the LiDAR-based environmental sensors as precisely as possible.

The calibration is particularly important when other environmental sensors are fitted to the vehicle, for example cameras, so that it can be ensured that the information from the various environmental sensors is consistent in relation to the vehicle's surroundings. The various environmental sensors can then be calibrated sequentially or simultaneously.

The calibration can take place both in the moving vehicle and on a calibration stand. Calibration in the moving vehicle is mostly based on a feature tracking approach based on detected objects (“features”). For example, special calibration objects can be placed specifically for the calibration. Alternatively, "everyday" objects such as traffic signs, crash barriers, etc. can be used for calibration, which preferably have a known orientation (horizontal or vertical). The movement of the vehicle must be determined with the greatest possible accuracy. This requires additional sensors, which makes the calibration very complex. In addition, the identification of static objects and their tracking contain further sources of error.

With static calibration, calibration objects are placed in the field of view of the LiDAR-based environmental sensor. By knowing the positions of the calibration objects in a reference coordinate system and their representation in the point cloud of the LiDAR-based environmental sensor in the LiDAR coordinate system, the position and alignment of the LiDAR-based environmental sensor in the reference coordinate system can be determined.

In both cases, special calibration objects are preferably used for calibrating the environmental sensors, which, due to their design, facilitate detection by the environmental sensors and an exact determination of their position. The calibration objects are preferably suitable both for detection with LiDAR-based environmental sensors and with cameras, in order to be able to calibrate the various environmental sensors with the same calibration objects.

With the different approaches, the position of the calibration objects and/or unique points on the calibration object can be uniquely determined in both coordinate systems. A transformation between the two coordinate systems and thus the extrinsic calibration between the coordinate systems is determined by a so-called point-matching algorithm.

For the detection of calibration objects with the LiDAR-based environmental sensor, the underlying measurement principle of the LiDAR-based environmental sensors is important. LiDAR-based environmental sensors used in vehicles are typically designed to capture a large horizontal area around the vehicle. Less information is required in the vertical direction. Typical LiDAR-based environmental sensors therefore capture the environment in scan planes (layers). A series of scan points is generated for each scan plane, which contains runtime information from received reflections and, in the case of newer LiDAR-based environmental sensors, also an intensity of the received reflections included. The number of scan planes determines the vertical polar angle resolution. The azimuthal resolution in the individual scan planes is typically significantly higher.

For multilayer LiDAR sensors with high polar resolution, i.e. with multiple scan planes, there is a large number of known calibration objects that can be used, for example, for static calibration. For example, squares, triangles or circular sections are used to clearly determine points in space with the LiDAR-based environment sensor.

In the field of autonomous driving and industrial applications, however, LiDAR-based environmental sensors with few scan planes are often used for cost reasons. Particularly cost-effective LiDAR-based environmental sensors work with just one scan plane. Fewer scan lines have the advantage that the environment can be scanned at a higher frequency. However, these LiDAR-based environmental sensors cannot be extrinsically calibrated with sufficient accuracy using such known calibration objects. The lower the vertical resolution of the LiDAR-based environmental sensors, the more difficult it is to determine all six degrees of freedom during calibration. In the case of LiDAR-based environmental sensors with only one scan plane, calibration with such known calibration objects is significantly more difficult or even impossible. In order to be able to estimate the shapes and orientations of the calibration objects, typically at least two scan planes have to intersect the calibration objects.

There is also the problem that conventional LiDAR-based environmental sensors provide inaccurate runtime information at the edges of individual objects, making assignment to the calibration object more difficult. A binary output of the discrete scanning of the scan plane with the scan points means that the position of the edges and above them the position of the calibration object can only be estimated discretely. Particularly in the context of a calibration, however, particularly high demands are made on determining the position of the calibration objects.

In this context, from the DE 10 2015 118 874 A1 a method for operating a sensor system of a motor vehicle is known, in which a first piece of information from an area surrounding the motor vehicle is recorded with a first sensor of the sensor system arranged locally at a first position on the motor vehicle. A second piece of information from the area surrounding the motor vehicle is recorded with a second sensor of the sensor system, which is arranged locally at a second position on the motor vehicle that differs from the first position. The first information and the second information are recorded and evaluated in relation to a common reference point of the first sensor and the second sensor on the motor vehicle and external to the sensor.

Based on the prior art mentioned above, the invention is therefore based on the object of providing a calibration object with at least three elongated individual objects, a method for determining a position of an individual object in a scan plane of a LiDAR-based environmental sensor, in particular for locating a calibration object with several individual objects Calibration of the LiDAR-based environment sensor, in particular with a calibration object above, and a driving support system with a LiDAR-based environment sensor and a control unit for performing the method, which at least partially overcome the disadvantages of the prior art and in particular a simple calibration of environment sensors for only allow one scan plane.

The object is achieved according to the invention by the features of the independent claims. Advantageous refinements of the invention are specified in the dependent claims.

According to the invention, a calibration object with at least three elongated individual objects is specified, the individual objects being arranged in one plane and two first individual objects being arranged parallel to one another, and a second individual object extending between the two first individual objects. The individual objects are arranged within the plane in such a way that when the calibration object is set up, they have a flat angle of intersection with the expected scan plane, in particular a horizontal scan plane, in particular an angle of intersection of less than 45°, preferably an angle of intersection of less than 30 °, particularly preferably a cutting angle of about 20 °. The minimum intersection angle depends on the resolution of the LiDAR and the size of the calibration object and its individual objects.

According to the invention, a method for determining a position of an individual object in a scan plane of a LiDAR-based environmental sensor is also specified, in particular for locating a calibration object with a plurality of individual objects for calibrating the LiDAR-based environmental sensor, in particular with an above calibration object, comprising the steps of sending of a plurality of laser pulses in the scan plane of the LiDAR-based environmental sensor, receiving reflections of the emitted laser pulses from the scan plane, determining intensities of the received reflections, determining reflections belonging to the individual object, and determining the position of the individual object based on the intensities of the received reflections belonging to the single object.

Furthermore, according to the invention, a driving support system with a LiDAR-based environment sensor and a control unit, which are connected to one another via a data connection, is specified, the LiDAR-based environment sensor being designed to receive reflections of transmitted laser pulses from at least one scan plane and to intensities of the received reflections determine, wherein the driving support system is designed to carry out the above method for determining a position of an individual object, in particular for locating a calibration object with a plurality of individual objects for calibrating the LiDAR-based environmental sensor.

The basic idea of the present invention is therefore to improve the ability to determine the positions of individual objects not only to improve the performance of LiDAR-based environmental sensors as such for the detection of individual objects, but in particular to enable an improved determination of individual objects from calibration objects in order to improve the performance of LiDAR-based environmental sensors based on an improved calibration that can be performed easily and reliably. The improved determination of the position of individual objects can be achieved individually or in combination by various factors.

Individual reflections of the emitted laser pulses from the scan plane, which are received with the LiDAR-based environmental sensor, are also referred to as scan points in the following.

A first aspect of the present invention relates to the provision of an improved calibration object. Calibration objects that have two vertical individual objects for calibrating LiDAR-based environmental sensors with horizontal scanning planes are widespread in the prior art. However, their position can only be determined imprecisely, since only a few reflections, in the worst case only one reflection, are received from the individual object. This can result in an uncertainty with regard to the position of the individual object in the range of an angular distance between adjacent scan points. The proposed alignment of the individual objects with a flat angle of intersection can ensure that an increased number of reflections from the individual object is detected, so that its position can be better determined based on these reflections. A further improvement can be achieved if the intensities of the received reflections are also considered, since reflections from laser pulses, which only partially detect the individual object, allow an improved determination of the position of the individual object due to their different intensities. There is therefore not only a binary distinction between scan points that capture the individual object and scan points that do not capture the individual object, but a degree of coverage of the individual object can also be determined.

A second aspect of the present invention concerns the consideration of intensities of the received reflections. As a result, scan points that only partially cover the individual object can contribute to determining the position of the individual object by determining their intensities. There is therefore not only a binary distinction between scan points that capture the individual object and scan points that do not capture the individual object, but a degree of coverage of the individual object can also be determined. The result is an increased number of scan points, which contribute to determining the position of the individual object and enable the position of the individual object to be determined with improved accuracy. A further improvement can be achieved if an improved calibration object described above is also used. The proposed alignment of the individual objects with a flat angle of intersection can ensure that an increased number of reflections of the laser pulses on the individual object is detected, so that its position can be better determined based on these reflections. In addition, the reflections with their different intensities in the edge area of the individual objects can contribute to determining the position of the respective individual object.

In both aspects, an increased number of reflections can therefore be received from the corresponding individual object and assigned to it. If the intensities of the received reflections are taken into account, those reflections that cannot be used in the prior art can also be used to determine the position of the respective individual object, i.e. reflections of laser pulses that only partially hit the individual object due to their real extent can also be used are taken into account and their intensity provide additional information for determining the position of the individual object.

In addition, an increased number of reflections from the corresponding individual object makes it easier to identify the respective individual object and, via this, to determine reflections belonging to the individual object. In particular, automatic recognition of the respective individual object is facilitated. In principle, special prior knowledge with regard to the position of the individual objects can be dispensed with.

A flat cutting angle within the meaning of the present invention is assumed if there is a deviation from a right angle, so that a cutting length with the scan plane is increased. A larger deviation results in a larger cut length, so that more reflections are received from the individual object and the determination of the position is improved

However, the intersection angle of the individual objects in the calibration object must be selected in such a way that it can be distinguished from the expected scan plane, i.e. it is larger than the tolerance range for an expected horizontal scan plane with a tolerance range of the scan plane. For example, the cutting angle is twice as large as the tolerance range.

The first individual objects and the second individual object can have independent angles of intersection.

Each of the individual objects preferably has a uniform extension transversely to its longitudinal direction, i.e. the individual objects each have a width that does not change over their length. The width of the individual objects can be chosen differently.

In the case of the calibration object, the individual objects can be at least partially connected to one another. Alternatively, the individual objects can be arranged separately and together form the calibration object. At least in the imaginary extension of the individual objects, there is an intersection of the second individual object with the two first individual objects.

When determining the position of the individual object in the scan plane and of the calibration object, it is necessary for the scan plane to intersect the individual object or the calibration object, with the scan plane having to intersect all individual objects in the calibration object. The scan object must therefore be aligned and positioned accordingly.

The control unit includes a processor and a memory to execute a program for performing an assistance function of the driving assistance system. The control unit processes sensor information provided by the LiDAR-based environmental sensor.

The data connection is designed, for example, in the manner of a bus system that is customary in the automotive sector. Various bus systems such as CAN, FlexRay, LON or others are known in this context.

LiDAR-based environmental sensors are known as such. The method can be performed with LiDAR-based environment sensors with one or more scan planes, with one scan plane being sufficient to determine the position of the individual object. When considering multiple scan planes, the position of the individual object can typically be determined with increased accuracy. However, only those scan planes that intersect the individual object or the calibration object can be considered.

The majority of laser pulses are emitted in the scan plane of the LiDAR-based environmental sensor, i.e. the scan plane is defined by the propagation of the laser pulses. The emitted laser pulses preferably have the same intensity. For LiDAR-based environmental sensors with multiple scan planes, scan points of each scan plane can be acquired sequentially or in any order. One laser pulse can be emitted at a time so that a reflection is received based on it. Alternatively, several laser pulses can be emitted simultaneously and, based on this, several reflections can be received. For this purpose, for example, a single laser pulse can be split into several laser pulses.

The scan plane is typically a substantially horizontally oriented plane, however, deviations in the scan plane from horizontal are possible based on misalignment of the LiDAR-based environmental sensor on the vehicle and/or variations in manufacturing and assembly of the LiDAR-based environmental sensor. In principle, vertical or diagonal scanning planes are also conceivable.

Receiving reflections of the emitted laser pulses from the scan plane typically includes a runtime detection from the emission of the laser pulses to the receipt of the corresponding reflections in order to determine a distance of an object at which the laser pulse was reflected based thereon.

Intensities of the received reflections are preferably determined as integral step in receiving the reflections of the emitted laser pulses from the scan plane.

Reflections belonging to the individual object can be determined in various ways based on intensities and/or propagation times of the received reflections.

The position of the individual object can be determined based on the intensities of the received reflections belonging to the individual object in different ways, as is explained below by way of example.

In an advantageous embodiment of the invention, the calibration object has a surface, in particular as a coating, with a high reflectivity for emitted laser pulses, at least on the side of the plane for the individual objects. Due to the high reflectivity for emitted laser pulses, individual objects can be easily detected if they have a particularly high reflectivity compared to surfaces with "natural" reflectivities. Reflections with particularly high intensities can thus be identified as belonging to the individual object. In addition or as an alternative, it is possible to generate a large dynamic range of the intensities of the reflections and to evaluate them accordingly, so that the position of the individual object can be reliably determined. Such high reflectivities relate in particular to reflectivities that are practically unachievable in “natural” environments.

In an advantageous embodiment of the invention, the calibration object is designed as a calibration panel, and the individual objects are formed on the calibration panel by the surface with a high reflectivity for emitted laser pulses. The individual objects can thus be formed, for example, simply by applying a coating with a high level of reflectivity to the calibration panel. Alternatively or additionally, the calibration table can be composed of the individual objects with a surface with a high reflectivity and parts with a surface with a low reflectivity. By using the calibration panel, the intensities of the reflections can be focused. The intensities of the reflections can thus be reliably determined for the individual object if partial reflections on the individual object can still be attributed to the individual object due to a matching propagation time or can be used to determine its position. The reflections on the calibration panel all have the same transit time and therefore differ in their intensity, with reflections on the individual object having a high intensity, some reflections on the individual object having a lower intensity, and reflections on the calibration panel that are not from the individual object originate, have a lowest intensity relative to the calibration panel.

In an advantageous embodiment of the invention, the calibration object is designed with a three-dimensional structure and the individual objects form the three-dimensional structure. The three-dimensional structure is thus generated by the elongated individual objects, which are designed, for example, as beams or strips. The elongated individual objects can, for example, be connected to one another at their ends to form the calibration object. Additionally or alternatively, the calibration object can have a frame structure or a support structure, for example, on which the individual objects are held to form the calibration object. Alternative arrangements of the individual objects are also possible. Due to the three-dimensional structure, reflections on the individual objects can be distinguished not only based on the intensity of the reflections, but also taking into account different propagation times of the reflections that do not originate from the individual objects. The frame structure or support structure is preferably designed with small dimensions compared to the individual objects in order to avoid reflections on it.

In an advantageous embodiment of the invention, the determination of reflections belonging to the individual object includes a comparison of propagation times of the received reflections. For example, a reflection can be determined as belonging to the individual object if this reflection has a different propagation time from the emission of the laser pulse to the reception of the reflection compared to neighboring reflections. Correspondingly, a group of adjacent reflections from the scan plane can be determined as belonging to the individual object if these reflections have the same propagation time and/or if these reflections have a deviating propagation time compared to neighboring reflections. A rough comparison of the position with an expected position of the individual object can help in determining the reflections associated with the individual object. In particular during the calibration, the individual object can be positioned in an area in which there are no other individual objects within a certain distance range, which can help to identify the individual object and determine reflections belonging to the individual object, in particular to automatically identify the individual object to carry out a determination of reflections belonging to the individual object.

In an advantageous embodiment of the invention, the determination of the individual object includes associated reflections comparing the intensities of the received reflections. Reflections with similar intensities can be attributed together to the individual object. For example, a group of adjacent reflections from the scan plane can be determined as belonging to the individual object if these reflections have approximately the same intensity. Alternatively or additionally, a group of adjacent reflections from the scan plane can be determined as belonging to the individual object if these reflections have an increased intensity compared to adjacent reflections. As a further alternative or in addition, a profile of the intensities can be considered in order to determine reflections belonging to the individual object. In this way, in particular, scan points on edges or edges of the individual object can be recognized and reliably assigned to the corresponding individual object. A rough comparison of an expected position of the individual object with positions of the reflections in the scan plane can help in determining the reflections that belong to the individual object. In particular, this can enable automatic determination of the reflections that belong to the individual object. In particular during calibration, a number of individual objects can be positioned in an area in which there are no other individual objects with a similar reflectivity, which can help to identify the individual object and to determine reflections belonging to the individual object. In addition, the intensity can help to recognize the individual object as such. For example, the individual object can have a particularly high reflectivity for emitted laser pulses, so that reflections on the individual object have a particularly high intensity. As a result, reflections on the individual object can be distinguished from reflections on "natural" objects, which usually have a lower reflectivity.

In an advantageous embodiment of the invention, determining the position of the individual object based on the intensities of the received reflections belonging to the individual object includes determining the received reflection belonging to the individual object with the highest intensity as the position of the individual object. The reflection with the highest intensity is taken as a middle position of the single object. The position of the individual object can thus be determined in a simple manner, in particular if only a few reflections are received from the individual object. Laser pulses which, due to a real expansion of the laser pulses, partially hit a border or edge area of the individual object, lead to reflections with a reduced intensity compared to reflections of laser pulses which hit the individual object in their entirety, so that a maximum intensity can indicate the position of the individual object .

In an advantageous embodiment of the invention, determining the position of the individual object based on the intensities of the received reflections belonging to the individual object includes determining an intensity profile over the received reflections belonging to the individual object and a maximum value of the determined intensity profile as the position of the individual object. The course of the intensity can be determined, for example, as an interpolation of the intensities for the reflections belonging to the individual object. Various mathematical methods for interpolation are known per se to those skilled in the art and can be used. Preferably, the intensity profile has a peak whose apex indicates the maximum value.

As an alternative or in addition, the position of the individual object can be determined based on the intensities of the received reflections belonging to the individual object as the center point between edges or borders of the individual object. As a result, the position of the individual object can be reliably determined when a plurality of laser pulses completely hit the individual object, so that no peak of the intensity profile with a sharp maximum arises.

The invention is explained in more detail below with reference to the attached drawing based on preferred embodiments. The features shown can represent an aspect of the invention both individually and in combination. Features of different exemplary embodiments can be transferred from one exemplary embodiment to another.

• 1 eine schematische Ansicht eines Fahrzeugs mit einem Fahrunterstützungssystem mit einem LiDAR-basierten Umgebungssensor und einer Steuereinheit, die über eine Datenverbindung miteinander verbunden sind, gemäß einer ersten, bevorzugten Ausführungsform,
• 2 eine schematische Darstellung eines Kalibierobjekts gemäß einer zweiten Ausführungsform mit drei länglichen Einzelobjekten, wobei die Einzelobjekte in einer Ebene angeordnet sind und zwei erste Einzelobjekte parallel zueinander angeordnet sind, und ein zweites Einzelobjekt sich zwischen den zwei ersten Einzelobjekten erstreckt,
• 3 eine schematische Darstellung eines Kalibierobjekts gemäß einer dritten Ausführungsform mit drei länglichen Einzelobjekten, wobei die Einzelobjekte in einer Ebene angeordnet sind und zwei erste Einzelobjekte parallel zueinander angeordnet sind, und ein zweites Einzelobjekt sich zwischen den zwei ersten Einzelobjekten erstreckt, und wobei die Einzelobjekte an ihren Enden miteinander verbunden sind,
• 4 eine schematische Darstellung von zwei unterschiedlich ausgerichteten Einzelobjekten in Bezug auf eine Scanebene des LiDAR-basierten Umgebungssensors der ersten Ausführungsform, mit jeweils einer Kurve eines Verlaufs einer Intensität basierend auf empfangenen Reflektionen in der Scanebene,
• 5 eine schematische Darstellung des Kalibierobjekts aus Figur zwei mit einer zusätzlichen Kurve eines Verlaufs einer Intensität basierend auf empfangenen Reflektionen in der Scanebene,
• 6 eine schematische Darstellung eines Kalibierobjekts gemäß einer vierten Ausführungsform mit drei länglichen Einzelobjekten, wobei die Einzelobjekte in einer Ebene angeordnet sind und zwei erste Einzelobjekte parallel zueinander angeordnet sind, und ein zweites Einzelobjekt sich zwischen den zwei ersten Einzelobjekten erstreckt, mit einer Kurve eines Verlaufs einer Intensität basierend auf empfangenen Reflektionen in einer Scanebene, und
• 7 ein Ablaufdiagramm eines Verfahrens zum Bestimmen einer Position von Einzelobjekten in einer Scanebene des LiDAR-basierten Umgebungssensors des ersten Ausführungsbeispiels.

It shows

• 1 a schematic view of a vehicle with a driving support system with a LiDAR-based environment sensor and a control unit, which are connected to one another via a data connection, according to a first preferred embodiment,
• 2 a schematic representation of a calibration object according to a second embodiment with three elongated individual objects, the individual objects being arranged in one plane and two first individual objects being arranged parallel to one another, and a second individual object extending between the two first individual objects,
• 3 a schematic representation of a calibration object according to a third embodiment with three elongated individual objects, wherein the individual objects are arranged in one plane are net and two first individual objects are arranged parallel to one another, and a second individual object extends between the two first individual objects, and the individual objects are connected to one another at their ends,
• 4 a schematic representation of two differently aligned individual objects in relation to a scan plane of the LiDAR-based environmental sensor of the first embodiment, each with a curve of a course of an intensity based on received reflections in the scan plane,
• 5 a schematic representation of the calibration object from FIG. 2 with an additional curve of a course of an intensity based on received reflections in the scan plane,
• 6 a schematic representation of a calibration object according to a fourth embodiment with three elongated individual objects, wherein the individual objects are arranged in one plane and two first individual objects are arranged parallel to one another, and a second individual object extends between the two first individual objects, with a curve of a course of an intensity based on received reflections in a scan plane, and
• 7 a flow chart of a method for determining a position of individual objects in a scan plane of the LiDAR-based surroundings sensor of the first exemplary embodiment.

the 1 shows a vehicle 10 with a driving support system 12 according to a first preferred embodiment.

The driving support system 12 includes a LiDAR-based environment sensor 14 and a control unit 16 which are connected to one another via a data connection 18 .

The LiDAR-based environment sensor 14 is designed to capture an environment 20 of the vehicle 10 . The detection takes place with scan points 22 in a scan plane 24 in that laser pulses are emitted in the scan plane 24 and reflections of the emitted laser pulses are received in the scan plane 24 . The scan points 22 are spanned by the received reflections of the emitted laser pulses in the scan plane 24. The scan plane 24 is spanned by the emitted laser pulses and the received reflections. In the scanning plane 24, the laser pulses are emitted with a uniform angular spacing. In addition, intensities 26 of the received reflections can be determined with the LiDAR-based environment sensor 14 . Corresponding LiDAR-based environmental sensors 14 are known as such.

The control unit 16 includes a processor and a memory to execute a program for performing an assistance function of the driving assistance system 12 . The control unit 16 controls the LiDAR-based surroundings sensor 14 and receives and processes sensor information provided by the LiDAR-based surroundings sensor 14 .

The data connection 18 is designed, for example, in the manner of a bus system that is customary in the automotive sector. Various bus systems such as CAN, FlexRay, LON or others are known in this context.

The driving support system 12 is implemented, the method described below for determining a position of individual objects 28, 30 in the scan plane 24 of the LiDAR-based surroundings sensor 14 and for locating a calibration object 32 with the individual objects 28, 30 for calibrating the LiDAR-based surroundings sensor 14, to perform.

The method is carried out by way of example with the calibration object 32 that is 2 and 5 and partially additionally in 4 is shown. The calibration object 32 includes three elongated individual objects 28, 30 which are arranged in one plane. The individual objects 28, 30 include two first individual objects 28, which are each arranged perpendicularly and thus parallel to one another, and a second individual object 30, which extends diagonally between the two first individual objects 28. The calibration object 32 is designed as a calibration table, on which the individual objects 28, 30 are formed by applying a coating with a high level of reflectivity.

At the in the 2 and 5 Calibration object 32 shown, the individual objects 28, 30 are not in contact. With an alternative calibration object 32, which is 3 As shown, the second individual object 30 is connected at its ends to one end of each of the first individual objects 28 . In addition, the alignment of the second individual object 30 with the calibration object 32 is in 3 compared to the calibration object 32 in the 2 and 5 vice versa. Incidentally, the calibration objects are 32 of 2 and 3 identical and can be used alternatively to carry out the method described below.

The calibration object 32 is positioned in such a way that the scan plane 24 intersects all of the individual objects 28 , 30 of the calibration object 32 .

The method begins with step S100, which includes emitting a plurality of laser pulses 22 in the scan plane 24 of the LiDAR-based environmental sensor 14. The emitted laser pulses each have the same pulse intensity, with one laser pulse being emitted at a time in this exemplary embodiment.

Step S110 relates to receiving reflections of the emitted laser pulses from the scan plane 24. The reflections are received by the LiDAR-based surroundings sensor 14. Receiving the reflections of the emitted laser pulses from the scan plane 24 includes a runtime detection from the emission of the laser pulses to the receipt of the corresponding reflections, so that for each scan point 22 distance information for the reflection of the emitted laser pulse is determined.

The laser pulses are sent out one after the other, so that the reflections are also received one after the other, so that in detail a sequence of repetitions of steps S100 and S110 results in order to record all scan points 22 in the scan plane 24 . In principle, however, it may be sufficient to detect scanning points 22 in the area of the respective individual object 28, 30, for example if its position is approximately known in advance or a rough detection is first carried out to approximately determine the position.

Step S120 relates to determining the intensities 26 of the received reflections. In this exemplary embodiment, the intensities 26 of the received reflections are determined as an integral step when the reflections of the emitted laser pulses are received from the scan plane 26 in step S110.

Step S130 relates to determining reflections belonging to the individual object 28, 30. Scanning points 22 can thus be assigned to the individual object 28, 30 for further processing.

Reflections belonging to the individual object 28, 30 are initially determined based on the propagation times of the received reflections. As a result, reflections from the calibration object 32 can be identified on the basis of their identical propagation times. A group of adjacent reflections from the scan plane 24 is thus identified as belonging to the individual object 28, 30 if these reflections have the same transit time and if these reflections have a different transit time compared to neighboring reflections.

In an additional step, the intensities 26 of the received reflections in the area of the calibration object 32 are compared. Due to the high reflectivity, reflections on the individual objects 28, 30 have an increased intensity 26 compared to reflections from other areas of the calibration object 32. Adjacent reflections of increased intensities 26 are therefore assigned to the respective individual object 28, 32 together. This includes scan points 26 at the edges of the individual objects 28, 30, where the laser pulses only partially hit the respective individual object 28, 30 and their intensity is therefore increased compared to reflections that do not originate from the individual objects 28, 30.

Step S140 relates to determining the position 38 of the individual object 28, 30 based on the intensities 26 of the received reflections belonging to the individual object 28, 30.

For this purpose, based on the intensities 26 of the received reflections belonging to the individual object 28, 30, a profile 34 of the intensity 26 over the received reflections belonging to the individual object 28, 30 is determined. The profile 34 of the intensity 26 is interpolated based on the reflections belonging to the individual object 28, 30. Various mathematical methods of interpolation are known per se to those skilled in the art and can be used. For the calibration object 32 used here, the profiles 34 of the intensity 26 are in FIG 4 and 5 shown. The curves 24 of the intensity 26 each have a maximum 36 (peak). The maximum 36 of the determined profile 34 of the intensity 26 indicates the position 38 of the respective individual object 28, 30.

Based on the positions 38 of the individual objects 28, 30, a position determination for the calibration object 32 can take place, as exemplified in FIG 2 is shown. There is a relationship between a height h of an intersection 40 of the scan plane 24 with the second individual object 30 and the distances a, b between the positions 38 of the individual objects 28, 30. If the method is carried out for a number of calibration objects 32, the LiDAR-based Environmental sensor 14 are calibrated in all directions.

Alternatively, the procedure with the in 3 shown calibration object 32 are carried out, which is not designed as a calibration panel. The calibration object 32 is designed with a three-dimensional structure, which is formed by the individual objects 28, 30. As a result, the reflections belonging to the individual object 28, 30 are determined under joint and simultaneous consideration of the propagation times of the reflections and their intensities 26. A prior determination of the reflections belonging to the calibration panel can be omitted.

The method described above is preferably carried out with a calibration object 32 that 6 is shown. This in 6 Calibration object 32 shown also includes three elongated individual objects 28, 30, which are arranged in one plane. Two first individual objects 28 are arranged parallel to one another, and a second individual object 30 extends between the two first individual objects 28. The calibration object 32 from 6 differs from the previously described calibration objects 32 in that all the individual objects 28, 32 have flat angles of intersection α with the scanning plane 24. The intersection angles α are each set to about 45° here, so that they differ from the expected scan plane 24 . In addition, the individual objects 28, 30 are connected to one another at their ends.

Also the calibration object 32 off 6 is designed as a calibration panel. The individual objects 28, 30 are formed on the calibration table by a surface with a high reflectivity for emitted laser pulses.

reference list

10

vehicle

12

driving support system

14

LiDAR-based environmental sensor

16

control unit

18

Data Connection

20

vicinity

22

scan point

24

scan plane

26

intensity

28

first single object

30

second single object

32

calibration object

34

course of the intensity

36

maximum

38

position

40

intersection

a

cutting angle

H

Height

a

Distance

b

Distance

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

• DE 102015118874 A1 [0014]

Claims (10)

Calibration object (32) with at least three elongated individual objects (28, 30), wherein the individual objects (28, 30) are arranged in one plane and two first individual objects (28) are arranged parallel to one another, and a second individual object (30) is between the two first individual objects (28), characterized in that the individual objects (28, 30) are arranged within the plane in such a way that, when the calibration object (32) is set up, they are in relation to an expected scan plane (24), in particular a horizontal scan plane (24) , have a flat intersection angle (a) with the expected scan plane (24), in particular an intersection angle (a) of less than 45°, preferably an intersection angle (a) of less than 30°, particularly preferably an intersection angle (α) of about 20 °. Calibration object (32) after claim 1 , characterized in that the calibration object (32) has a surface, in particular as a coating, with a high reflectivity for emitted laser pulses, at least on the side of the plane for the individual objects (28, 30). Calibration object (32) after claim 2 , characterized in that the calibration object (32) is designed as a calibration panel, and the individual objects (28, 30) are formed by the surface with a high reflectivity for emitted laser pulses on the calibration panel. Calibration object (32) according to one of Claims 1 or 2 , characterized in that the calibration object (32) is designed with a three-dimensional structure and the individual objects (28, 30) form the three-dimensional structure. Method for determining a position (38) of an individual object (28, 30) in a scanning plane (24) of a LiDAR-based environmental sensor (14), in particular for locating a calibration object (3) with a plurality of individual objects (28, 30) for calibrating the LiDAR -based environmental sensor (14), in particular with a calibration object (32) according to one of Claims 1 until 4 , comprising the steps of emitting a plurality of laser pulses in the scan plane (24) of the LiDAR-based environmental sensor (14), receiving reflections of the emitted laser pulses from the scan plane (24), determining intensities (26) of the received reflections, determining to reflections belonging to the individual object (28, 30), and determining the position (38) of the individual object (28, 30) based on the intensities (26) of the received reflections belonging to the individual object (28, 30). procedure after claim 5 , characterized in that the determination of the individual object (28, 30) belonging to the reflections includes a comparison of transit times of the received reflections. Procedure according to one of Claims 5 or 6 , characterized in that the determination of the individual object (28, 30) associated reflections comprises a comparison of the intensities (26) of the received reflections. Method according to any of the preceding Claims 5 until 7 , characterized in that determining the position (38) of the individual object (28, 30) based on the intensities (26) of the received reflections belonging to the individual object (28, 30) involves determining the belonging to the individual object (28, 30). includes received reflection with the highest intensity (26) as the position (38) of the individual object (28, 30). Method according to any of the preceding Claims 5 until 8th , characterized in that the determination of the position (38) of the individual object (28, 30) based on the intensities (26) of the individual object (28, 30) belonging to the received reflections a determination of a profile (34) of the intensity (26) about the received reflections belonging to the individual object (28, 30) and a maximum value of the determined profile (34) of the intensity (26) as the position (38) of the individual object (28, 30). Driving support system (12) with a LiDAR-based environment sensor (14) and a control unit (16), which are connected to one another via a data connection (18), the LiDAR-based environment sensor (14) being implemented in at least one scanning plane (24) Receiving reflections of emitted laser pulses and determining intensities (26) of the received reflections, the driving support system (12) being implemented, the method for determining a position (38) of an individual object (28, 30), in particular for locating a calibration object (32 ) with several individual objects (28, 30) for calibrating the LiDAR-based environmental sensor (14), according to one of Claims 5 until 9 to perform.

Images