DE102018130450B4 - Method for determining object information of at least one object and radar system - Google Patents

Abstract

Method for determining object information of at least one object (18) that is detected by a radar system (12), wherein in the method- transmission signals (24) are sent into a monitoring area (14) of the radar system (12),- echo signals (28) that are reflected on at least one object target (26) of at least one object (18) present in the monitoring area (14) are received by at least one receiving antenna (Rx1, Rx2, Rx3, Rx4) and further processed as received signals assigned to the receiving antenna (Rx1, Rx2, Rx3, Rx4),- at least one received signal is subjected to at least one discrete Fourier transformation,- at least one item of object information of the at least one object (18) is determined from the result of the at least one discrete Fourier transformation, characterized in that the echo signals (28) are received by at least two spatially separated receiving antennas (Rx1, Rx2, Rx3, Rx4),- by at least two of the spatially separate receiving antennas (Rx1, Rx2, Rx3, Rx4), the respective received signals are each subjected to a separate two-dimensional discrete Fourier transformation separately from one another,- from the results of the separate two-dimensional discrete Fourier transformations for each of the spatially separated receiving antennas (Rx1, Rx2, Rx3, Rx4), corresponding signal quantities (30) which characterize properties of the echo signals (28) received by the respective receiving antenna (Rx1, Rx2, Rx3, Rx4) are determined,- each signal quantity (30) is assigned to a corresponding distance-velocity gate (32) of a virtual two-dimensional signal matrix (34), wherein the signal matrix (34) has a distance dimension and a velocity dimension and each distance-velocity gate (32) characterizes a distance (19) and a velocity relative to the radar system (12),- from the Signal matrices (34) of at least two of the spatially separated receiving antennas (Rx1, Rx2, Rx3, Rx4) subject the respective information of at least one distance-velocity gate (32) to a common discrete Fourier transformation,- the maximum of the result of the common discrete Fourier transformation is determined as a direction variable (38) which characterizes a direction relative to the radar system (12), and the direction variable (38) is assigned to the corresponding at least one distance-velocity gate (32) of a virtual two-dimensional direction matrix (44) which has a distance dimension and a speed dimension which are comparable to the signal matrices (34),- from at least one of the signal matrices (34), if present, the distance-velocity gates (32) whose signal variable (30) is above a respective predetermined or predeterminable signal limit value are selected as signal target gates (36) are identified which belong to at least one object target (26),- in the direction matrix (44) for at least one distance-speed gate (32) which corresponds to one of the signal target gates (36) of the at least one signal matrix (34) with respect to the distance-speed dimension, is identified as a direction target gate (46),- an extension (48) of at least one object (18) is determined in the distance-speed dimension of the direction matrix (44) by, starting from at least one direction target gate (46), assigning those distance-speed gates (32) of the extension (48) of the at least one object (18) which have the same direction size (38), optionally plus or minus a predetermined direction size tolerance, as the adjacent at least one direction target gate (46) or, starting from the at least one direction target gate (46) having the adjacent distance-velocity gate (32),- from the extent (48) in the distance-velocity dimension of the at least one object information of the at least one object (18) is determined.

Description

Technical area

• - Sendesignale in einen Überwachungsbereich des Radarsystems gesendet werden,
• - Echosignale, welche an wenigstens einem Objektziel wenigstens eines in dem Überwachungsbereich vorhandenen Objekts reflektiert werden, von wenigstens einer Empfangsantenne empfangen und als der Empfangsantenne zugeordnete Empfangssignale weiterverarbeitet werden,
• - wenigstens ein Empfangssignal wenigstens einer diskreten Fourier-Transformation unterzogen wird,
• - aus dem Ergebnis der wenigstens einen diskreten Fourier-Transformation wenigstens eine Objektinformation des wenigstens einen Objekts ermittelt wird.

The invention relates to a method for determining object information of at least one object that is detected by a radar system, wherein in the method

• - Transmission signals are sent into a surveillance area of the radar system,
• - echo signals which are reflected at at least one object target of at least one object present in the surveillance area are received by at least one receiving antenna and further processed as received signals assigned to the receiving antenna,
• - at least one received signal is subjected to at least one discrete Fourier transformation,
• - at least one item of object information of the at least one object is determined from the result of the at least one discrete Fourier transformation.

• - wenigstens eine Sendeantenne, mit der Sendesignale in einen Überwachungsbereich des Radarsystems gesendet werden können,
• - wenigstens eine Empfangsantenne, mit der Echosignale, welche an wenigstens einem Objektziel wenigstens eines in dem Überwachungsbereich vorhandenen Objekts reflektiert werden, empfangen werden können,
• - wenigstens eine Auswerteeinrichtung, mit der Empfangssignale, welche der wenigstens einen Empfangsantenne zugeordnet sind, weiterverarbeitet werden können, wenigstens ein Empfangssignal wenigstens einer diskreten Fourier-Transformation unterzogen werden kann und aus dem Ergebnis der wenigstens einen diskreten Fourier-Transformation wenigstens eine Objektinformation des wenigstens einen Objekts ermittelt werden kann.

Furthermore, the invention relates to a radar system for determining object information of at least one object, which has

• - at least one transmitting antenna with which transmission signals can be sent into a surveillance area of the radar system,
• - at least one receiving antenna with which echo signals reflected from at least one object target of at least one object present in the surveillance area can be received,
• - at least one evaluation device with which received signals which are assigned to the at least one receiving antenna can be further processed, at least one received signal can be subjected to at least one discrete Fourier transformation and at least one item of object information of the at least one object can be determined from the result of the at least one discrete Fourier transformation.

State of the art

From the EN 10 2007 008 944 A1 A radar system for environmental monitoring is known, whereby the length of surrounding objects is also recorded. The radar system is used to send and receive a high-frequency signal. A sampled signal is subjected to a discrete Fourier transformation; the discrete grid points of the discrete Fourier transformation correspond to the object distance. To determine a Doppler frequency, a second spectral analysis is carried out using values obtained at a respective grid point in several frequency ramps. In this way, spectral power peaks are detected at respective grid points, whereby their frequency corresponds to the relative speed of objects.

From the WO 2010/115418 A2 A radar system for environmental detection is known which is designed in such a way that, in particular for determining the relative speed of objects and for increasing the detection sensitivity, one or more possibly interleaved sequences of identical or similar individual signals are emitted per transmitting antenna in one measuring cycle, wherein the time interval between these individual signals is on average greater than the transit time of received signals corresponding to the maximum distance of interest.

From the EN 102016213007 A1 A system for scanning an object from a motor vehicle is known and comprises a first radar device for scanning first information of the object, which includes a distance, a first object angle and a first relative speed; a periodic continuous wave radar device for scanning second information of the object, which includes a second relative speed and a second object angle; and a processing device for assigning the first and second information relating to the same object, and for classifying the object on the basis of a characteristic of the second relative speed.

The invention is based on the object of designing a method and a radar system of the type mentioned at the outset, in which an effort, in particular a component effort and/or a computation effort and/or a time effort, in determining object information of objects can be reduced.

Disclosure of the invention

• -die Echosignale von wenigstens zwei räumlich getrennten Empfangsantennen empfangen werden,
• - von wenigstens zwei der räumlich getrennten Empfangsantennen die jeweiligen Empfangssignale getrennt voneinander jeweils einer separaten zweidimensionalen diskreten Fourier-Transformation unterzogen werden,
• - aus den Ergebnissen der separaten zweidimensionalen diskreten Fourier-Transformationen für jede der räumlich getrennten Empfangsantennen entsprechende Signal-Grö-ßen, welche Eigenschaften der Echosignale charakterisieren, die von der jeweiligen Empfangsantenne empfangen wurden, ermittelt werden,
• - jede Signal-Größe einem entsprechenden Entfernungs-Geschwindigkeits-Tor einer virtuellen zweidimensionalen Signal-Matrix zugeordnet wird, wobei die Signal-Matrix eine Entfernungsdimension und eine Geschwindigkeitsdimension aufweist und jedes Entfernungs-Geschwindigkeits-Tor eine Entfernung und eine Geschwindigkeit relativ zum Radarsystem charakterisiert,
• - aus den Signal-Matrizen von wenigstens zwei der räumlich getrennten Empfangsantennen die jeweiligen Informationen wenigstens eines Entfernungs-Geschwindigkeits-Tores einer gemeinsamen diskreten Fourier-Transformation unterzogen werden,
• - das Maximum des Ergebnisses der gemeinsamen diskreten Fourier-Transformation als Richtungsgröße ermittelt wird, welche eine Richtung relativ zum Radarsystem charakterisiert, und die Richtungsgröße dem entsprechenden wenigstens einen Entfernungs-Geschwindigkeits-Tor einer virtuellen zweidimensionalen Richtungs-Matrix zugeordnet wird, welche eine Entfernungsdimension und eine Geschwindigkeitsdimension aufweist, die mit den Signal-Matrizen vergleichbar sind,
• - aus wenigstens einer der Signal-Matrizen sofern vorhanden die Entfernungs-Geschwindigkeits-Tore, deren Signal-Größe betragsmäßig über einem jeweiligen vorgegebenen oder vorgebbaren Signal-Grenzwert liegt, als Signal-Zieltore identifiziert werden, welche zu wenigstens einem Objektziel gehören,
• - in der Richtungs-Matrix für wenigstens ein Entfernungs-Geschwindigkeits-Tor, welches bezüglich der Entfernungs-Geschwindigkeits-Dimension einem der Signal-Zieltore der wenigstens einen Signal-Matrix entspricht, als Richtungs-Zieltor identifiziert wird,
• - eine Ausdehnung wenigstens eines Objekts in der Entfernungs-Geschwindigkeits-Dimension der Richtungs-Matrix ermittelt wird, indem ausgehend von wenigstens einem Richtungs-Zieltor diejenigen Entfernungs-Geschwindigkeits-Tore der Ausdehnung des wenigstens einen Objekts zugeordnet werden, die die gleiche Richtungsgröße, gegebenenfalls zuzüglich oder abzüglich einer vorgegebenen Richtungsgrößen-Toleranz, wie das benachbarte wenigstens eine Richtungs-Zieltor oder, ausgehend von dem wenigstens ein Richtungs-Zieltor, das benachbarte Entfernungs-Geschwindigkeits-Tor aufweisen,
• - aus der Ausdehnung in der Entfernungs-Geschwindigkeits-Dimension des wenigstens eine Objektinformation des wenigstens einen Objekts ermittelt wird.

This object is achieved in the method according to the invention in that

• -the echo signals are received by at least two spatially separated receiving antennas,
• - the respective received signals from at least two of the spatially separated receiving antennas are each subjected to a separate two-dimensional discrete Fourier transformation,
• - from the results of the separate two-dimensional discrete Fourier transformation For each of the spatially separated receiving antennas, corresponding signal quantities can be determined which characterize properties of the echo signals received by the respective receiving antenna,
• - each signal quantity is assigned to a corresponding range-velocity gate of a virtual two-dimensional signal matrix, the signal matrix having a distance dimension and a velocity dimension and each range-velocity gate characterizing a distance and a velocity relative to the radar system,
• - from the signal matrices of at least two of the spatially separated receiving antennas, the respective information of at least one range-velocity gate is subjected to a common discrete Fourier transformation,
• - the maximum of the result of the joint discrete Fourier transformation is determined as a direction variable which characterizes a direction relative to the radar system, and the direction variable is assigned to the corresponding at least one distance-velocity gate of a virtual two-dimensional direction matrix which has a distance dimension and a velocity dimension which are comparable to the signal matrices,
• - from at least one of the signal matrices, if present, the distance-velocity gates whose signal magnitude is above a respective predetermined or predeterminable signal limit value are identified as signal target gates which belong to at least one object target,
• - in the direction matrix for at least one distance-speed gate, which corresponds to one of the signal target gates of the at least one signal matrix with respect to the distance-speed dimension, is identified as a direction target gate,
• - an extension of at least one object in the distance-speed dimension of the direction matrix is determined by assigning, starting from at least one direction target gate, those distance-speed gates of the extension of the at least one object which have the same direction size, optionally plus or minus a predetermined direction size tolerance, as the adjacent at least one direction target gate or, starting from the at least one direction target gate, the adjacent distance-speed gate,
• - from the extent in the distance-velocity dimension of the at least one object information of the at least one object is determined.

According to the invention, the echo signals for each receiving antenna are first subjected to a two-dimensional discrete Fourier transformation separately from one another in order to assign any object targets to the signal matrix with the distance-velocity dimension. The result of the two-dimensional discrete Fourier transformation for at least two of the receiving antennas, in particular all receiving antennas, is then subjected to a further common discrete Fourier transformation in order to determine the direction of the object targets. For this purpose, the maxima of the results of the common discrete Fourier transformation are assigned as directional variables to a two-dimensional directional matrix.

From at least one of the signal matrices, the distance-velocity gates, if present, whose signal magnitude is above a respective predetermined or predeterminable signal limit value, are identified as signal target gates which belong to at least one object target.

Advantageously, several, preferably all, signal matrices can be combined, in particular using a Fourier transformation, and from the result, if available, the distance-velocity gates whose signal size is above a respective predetermined or predeterminable signal limit value can be identified as signal target gates which belong to at least one object target.

In the direction matrix, the distance-speed gates are identified as direction-target gates, in which, based on the signal matrix, possible object targets are suspected. It is assumed that there are no large jumps between the corresponding direction sizes between distance-speed gates in the vicinity of a direction-target gate, which are occupied by echo signals at object targets of the same object. As long as an object is detected in the direction matrix, the direction changes continuously, so that no jumps in the direction size can be observed. From the continuity of the direction sizes in the neighboring distance-speed gates, an extension of the object in the direction matrix can be determined. From the extension of the object in the direction matrix, the size and spatial extent of the object in reality. With the method according to the invention, time expenditure and component expenditure, in particular with regard to a memory, can be reduced.

By specifying a directional size tolerance, it can be ensured that echo signals from similar directions to that of at least one directional target gate can also be included in the determination of the extent of the object in the directional matrix. In this way, the assignment of object targets to objects can be improved.

Advantageously, starting from at least one direction target gate, the direction values of the distance-speed gates adjacent to this at least one direction target gate can be compared with the direction value of the at least one direction target gate. If the direction value of a distance-speed gate adjacent to the at least one direction target gate has the same direction value, possibly plus or minus a predetermined or predeterminable direction value tolerance, as the at least one direction target gate, the said distance-speed gate can be assigned to the extent of the object in the direction matrix. Otherwise, the said distance-speed gate is not assigned to the extent of the object in the direction matrix. Starting from at least one distance-speed gate which is assigned to the extent of the at least one direction target gate, the direction variables of the distance-speed gates adjacent to this at least one assigned distance-speed gate can be compared with the direction variable of said at least one assigned distance-speed gate, provided that these adjacent distance-speed gates are not already assigned to the extent of the object in the direction matrix by comparison with the at least one direction target gate. If the directional size of a distance-speed gate adjacent to said at least one associated distance-speed gate has the same directional size, possibly plus or minus a predetermined directional size tolerance, as said at least one associated distance-speed gate, the distance-speed gate to be checked can be assigned to the extent of the object in the directional matrix. Otherwise, the checked distance-speed gate cannot be assigned to the extent of the object in the directional matrix. In this way, the environment of at least one directional target gate in the directional matrix is successively checked for affiliation with the object.

The echo signals are received by at least two spatially separated receiving antennas. In this way, directions in which the detected object targets are located relative to the radar system, or directional variables characterizing the directions, can be determined from the corresponding receiving signals. The directions or directional variables can be determined by considering phase shifts of the receiving signals from the different receiving antennas. Each receiving antenna can have an independent receiving channel. In this way, the corresponding receiving signals can be processed separately from one another.

Advantageously, the echo signals can be received by four or more spatially separated receiving antennas. In this way, the directions or directional quantities can be determined more easily, efficiently and/or more accurately.

The maximum of the result of the common discrete Fourier transformation of at least one distance-velocity gate of at least two receiving antennas is determined as the direction variable. The direction variable is assigned to the corresponding at least one distance-velocity gate of a virtual two-dimensional direction matrix. Since only the maximum of the result is used, the storage space required to store the information in the direction matrix can be reduced. The two dimensions of the direction matrix are sufficient for this.

The received signals are subjected to several discrete Fourier transformations. In this way, coherent results can be achieved. Any noise remains constant during the discrete Fourier transformations. In this way, the signal-to-noise ratio (SNR) can be improved.

Advantageously, the discrete Fourier transforms can be performed as fast Fourier transforms. In this way, the Fourier transforms can be calculated more efficiently.

Object targets within the meaning of the invention are areas of the surface of an object at which transmission signals can be reflected. An object can have one or more such object targets.

Advantageously, the object can be classified on the basis of the object information of the object to which the at least one object target belongs. In particular, it can be determined whether the object is a pedestrian, a vehicle or another type of object.

From at least one of the signal matrices, the distance-speed gates whose signal sizes are above a respective predetermined or predeterminable signal limit value are identified as signal target gates. An individual signal limit value can be set for each distance-speed tour. A corresponding algorithm, in particular a CFAR algorithm, can advantageously be used for this purpose. The signal target gates belong to at least one object target.

The signal quantities can be complex numbers. A signal quantity can in particular indicate an amplitude and a phase of the echo signal and thus characterize properties of the echo signal.

The invention can be used in a vehicle, in particular a motor vehicle. The invention can advantageously be used in a land vehicle, in particular a passenger car, a truck, a bus, a motorcycle or the like, an aircraft and/or a watercraft. The invention can also be used in vehicles that can be operated autonomously or at least partially autonomously.

The radar system can advantageously be connected to or be part of at least one electronic control device of the vehicle, in particular a driver assistance system and/or a chassis control and/or a driver information device and/or a parking assistance system or the like. In this way, the vehicle can be operated autonomously or at least partially autonomously.

The radar system can detect stationary or moving objects, in particular vehicles, persons, animals, obstacles, road bumps, in particular potholes or stones, road markings, open spaces, in particular parking spaces, or the like.

The object information can be distances, speeds, accelerations and/or directions of objects relative to the radar system. In addition, shape, spatial extent, in particular height, width and/or length, or other properties that can be detected with the radar system can be used as object information. The object information can also be movement patterns of objects, which can be used in particular for gesture recognition.

In an advantageous embodiment, if an extension of at least one object in the direction matrix has the same direction size, possibly plus or minus a predetermined direction size tolerance, as the at least one direction target gate, it can be concluded that the object targets of the object detected within the extension lie in approximately the same direction to the radar system. In this way, objects which are relatively narrow in terms of their spatial extension transverse to the direction of propagation of the transmission signals can be identified. In particular, pedestrians can be distinguished from cars.

In a further advantageous embodiment of the method, at least one expansion variable can be determined which characterizes the expansion of the at least one object in the direction matrix, and at least one object information item of at least one object can be determined from the at least one expansion variable. In this way, the expansion can be quantified and thus processed more easily.

In a further advantageous embodiment of the method, at least one expansion size can be determined as the total number of distance-velocity gates of the expansion, as the number of distance-velocity gates or as the number of outer boundaries of the distance-velocity gates at the edge of the expansion of the at least one object in the direction matrix. In this way, the expansion can be quantified more easily with just a few of its expansion sizes.

In a further advantageous embodiment of the method, if a difference between the respective direction sizes is determined in the direction matrix between two adjacent distance-velocity gates which are located between two direction target gates and which is greater than the direction size tolerance, it can be concluded that the two direction target gates originate from two object targets which do not belong to the same object.

In this way, two objects whose targets are detected at a similar distance and speed but in different directions relative to the radar system can be distinguished from each other. Between the targets jumps between the direction values can be detected. In addition, noise in the direction matrix can be detected by sudden changes in the direction values.

In a further advantageous embodiment of the method, an opening angle of the radar system, which defines the surveillance area and whose apex is on the radar system side, can be divided into a predetermined or predeterminable number of defined angle sections, each of which is characterized by an associated direction index, and the direction values from the result of the joint Fourier transformation can be specified as corresponding direction indices. In this way, the amount of data to be stored and processed can be further reduced. In this way, the overall storage space required can be further reduced and/or the processing time required can be shortened. The directions of the detected object targets can be specified with sufficient accuracy using the direction indices. Each direction index corresponds to one of the angle sections.

Advantageously, the surveillance area can be defined by an opening angle in the form of a surface angle or a solid angle, the apex of which is on the radar system side. The direction indices correspond to a corresponding direction angle in which the corresponding object target is detected, with a correspondingly coarse resolution.

Advantageously, the direction variables from the result of the joint Fourier transformation and, if applicable, corresponding direction indices can qualitatively correspond to an azimuth angle and/or an elevation angle, in particular an elevation.

In a further advantageous embodiment of the method, an opening angle of the radar system, which defines the monitoring area and whose apex is on the radar system side, can be divided into a predetermined number of defined angle sections, with at least two adjacent angle sections overlapping. By dividing the opening angle into defined angle sections, the overall amount of data can be reduced. Because at least two adjacent angle sections overlap, blind sections in the monitoring area can be avoided.

In a further advantageous embodiment of the method, the type of object can be determined from at least one extension and/or optionally at least one extension size in the direction matrix by means of pattern recognition. In this way, the object can be efficiently assigned and optionally classified. The radar system can have corresponding means for carrying out pattern recognition for this purpose.

In a further advantageous embodiment of the method, at least one artificial neural network and/or a support vector machine can be used for pattern recognition. In this way, the determination of the type of object can be increasingly improved. In this way, the effort, in particular the time required and/or the computing effort required, can be reduced.

Furthermore, the object is achieved according to the invention in the radar system in that the radar system has at least two spatially separated receiving antennas with which echo signals can be received, and the at least one evaluation device has means for carrying out the method according to the invention.

Furthermore, the features and advantages shown in connection with the method according to the invention and the radar system according to the invention and their respective advantageous embodiments apply to one another accordingly and vice versa. The individual features and advantages can of course be combined with one another, whereby further advantageous effects can arise that go beyond the sum of the individual effects.

Short description of the drawings

• 1 ein Kraftfahrzeug mit einem Radarsystem zu Überwachung eines Überwachungsbereichs in Fahrtrichtung vor dem Kraftfahrzeug und einem Fahrerassistenzsystem;
• 2 eine Funktionsdarstellung des Kraftfahrzeugs mit dem Radarsystem und dem Fahrerassistenzsystem aus der 1;
• 3 eine Draufsicht auf das Kraftfahrzeug aus den 1 und 2 in einer Fahrsituation;
• 4 eine Signal-Matrix, in der Amplituden aus Empfangssignalen von Echosignalen des Radarsystems aus den 1 bis 3 in einer Entfernungs-Geschwindigkeits-Dimension angegeben sind, wobei die Echosignale von einem Fußgänger stammen;
• 5 eine Richtungs-Matrix, welche zu der Signal-Matrix aus 4 gehört, in der Richtungsgrößen aus Empfangssignalen von Echosignalen des Radarsystems aus den 1 bis 3 in einer Entfernungs-Geschwindigkeits-Dimension angegeben sind;
• 6 die Richtungs-Matrix aus der 5, bei der die Ausdehnung des Fußgängers in der Entfernungs-Geschwindigkeits-Dimension mit der gleichen Richtungsgröße markiert ist;
• 7 die Richtungs-Matrix aus der 6, wobei hier die Ausdehnung des Fußgängers in der Entfernungs-Geschwindigkeits-Dimension mit Richtungsgröße zuzüglich einer Richtungsgrößen-Toleranz markiert ist;
• 8 die Richtungs-Matrix aus der 5, wobei hier die Entfernungs-Geschwindigkeits-Tore am Rand der Ausdehnung des Fußgängers in der Entfernungs-Geschwindigkeits-Dimension und die Außengrenzen der Ausdehnung markiert sind;
• 9 eine Signal-Matrix von Echosignalen, die von einer Leitplanke stammen, die mit dem Radarsystem aus den 1 bis 3 erfasst wird;
• 10 eine Richtungs-Matrix, welche zu der Signal-Matrix aus der 9 gehört.

Further advantages, features and details of the invention emerge from the following description, in which embodiments of the invention are explained in more detail with reference to the drawing. The person skilled in the art will expediently also consider the features disclosed in the drawing, the description and the claims in combination individually and combine them to form useful further combinations. They show schematically

• 1 a motor vehicle with a radar system for monitoring a surveillance area in the direction of travel in front of the motor vehicle and a driver assistance system;
• 2 a functional representation of the motor vehicle with the radar system and the driver assistance system from the 1 ;
• 3 a plan view of the motor vehicle from the 1 and 2 in a driving situation;
• 4 a signal matrix in which amplitudes from received signals of echo signals of the radar system from the 1 to 3 in a development distance-speed dimension, where the echo signals originate from a pedestrian;
• 5 a direction matrix, which corresponds to the signal matrix of 4 in which directional quantities from received signals of echo signals of the radar system from the 1 to 3 are given in a distance-velocity dimension;
• 6 the direction matrix from the 5 , where the pedestrian's extension in the distance-speed dimension is marked with the same direction size;
• 7 the direction matrix from the 6 , where the extension of the pedestrian in the distance-speed dimension is marked with directional size plus a directional size tolerance;
• 8th the direction matrix from the 5 , where the distance-velocity gates are marked at the edge of the pedestrian's extension in the distance-velocity dimension and the outer boundaries of the extension;
• 9 a signal matrix of echo signals originating from a guardrail that is connected to the radar system from the 1 to 3 is recorded;
• 10 a direction matrix, which corresponds to the signal matrix from the 9 heard.

In the figures, identical components are provided with identical reference symbols.

Embodiment(s) of the invention

In the 1 a motor vehicle 10 in the form of a passenger car is shown in the front view. The motor vehicle 10 has a radar system 12. The radar system 12 is arranged, for example, in the front bumper of the motor vehicle 10. With the radar system 12, a 2 The radar system 12 can also be arranged at a different location on the motor vehicle 10 and aligned differently. The radar system 12 can also be arranged at the rear of the motor vehicle 10, in particular in the rear bumper. In this way, a monitoring area 14 in the direction of travel 16 behind the motor vehicle 10 can be monitored for objects 18.

The objects 18 can be, for example, other vehicles, people, obstacles, road irregularities, for example potholes or stones, road markings or the like. In the 2 an object 18 is indicated as an example of a pedestrian. The 2 is otherwise merely a functional diagram of some components of the motor vehicle 10 and the radar system 12, which does not serve for spatial orientation.

The radar system 12 is designed as a "frequency modulated continuous wave radar" for example. Frequency modulated continuous wave radars are also referred to in technical circles as FMCW (frequency modulated continuous wave) radars. The radar system 12 can be used to transmit so-called chirp sequences. Such a radar is also referred to in technical circles as a "chirp sequence radar".

The radar system 12 can be used to determine object information, for example a distance 19, a speed and a direction of the object 18 relative to the motor vehicle 10 and spatial extents of the object 18.

The radar system 12 is connected to a driver assistance system 20. The driver assistance system 20 can, for example, support a driver of the motor vehicle 10. For example, the motor vehicle 10 can be operated autonomously or semi-autonomously using the driver assistance system 20, for example driving, parking or reversing. The driver assistance system 20 can be used to influence driving functions of the motor vehicle 10, for example an engine control, a braking function or a steering function, or to issue instructions or warning signals.

The radar system 12 comprises, for example, a transmitter antenna Tx, an electronic control and evaluation device 22 and four spatially separated receiving antennas Rx1, Rx2, Rx3 and Rx4. Each of the receiving antennas Rx1, Rx2, Rx3 and Rx4 belongs to its own independent receiving channel.

The control and evaluation device 22 is connected to the driver assistance system 20 in terms of signals. It is not important for the invention whether electrical/electronic control and/or evaluation devices, such as the control and evaluation device 20, an engine control unit of the motor vehicle 10 or the like, are integrated in one or more components or component groups or are at least partially implemented as decentralized components or component groups.

The transmitter Tx can be used to send transmission signals 24 into the monitoring area 14. The transmission signals 24 are reflected at object targets 26 of the object 18 and sent back as echo signals 28.

The object targets 26 are areas of the surface of the object 18 at which transmission signals 24 can be reflected. An object 18 can have one or more such object targets 26. In the 2 Two object targets 26 of the object 18 are indicated as examples.

The echo signals 28 are received by the receiving antennas Rx1, Rx2, Rx3 and Rx4 and converted for further processing into received signals which are assigned to the receiving antennas Rx1, Rx2, Rx3 and Rx4.

The respective received signals of the four receiving antennas Rx1, Rx2, Rx3 and Rx4 are each subjected to a separate two-dimensional discrete Fourier transformation. From the results of the separate two-dimensional discrete Fourier transformations, corresponding signal quantities 30 are determined for each of the receiving antennas Rx1, Rx2, Rx3 and Rx4. The signal quantities 30 characterize properties of the received signals that belong to the corresponding receiving antenna Rx1, Rx2, Rx3 and Rx4. The signal quantities 30 are, for example, complex numbers that indicate the amplitude and phase of the received signals and thus characterize the properties of the received signals.

The magnitude of each signal quantity 30 is assigned to a corresponding range-velocity gate 32 of a virtual two-dimensional signal matrix 34. A total of four such signal matrices 34 are assigned to the four receiving antennas Rx1, Rx2, Rx3 and Rx4.

In the 4 The signal matrix 34 for one of the receiving antennas Rx1, Rx2, Rx3 or Rx4 is shown as an example for a measurement in which an object 18 in the form of a pedestrian is detected by the radar system 12. The corresponding driving situation of the motor vehicle 10 is shown in the 3 shown.

Each signal matrix 34 has a distance dimension and a speed dimension. The distance dimension corresponds to so-called “range bins” or distance intervals. The speed dimension corresponds to so-called relative speed gates or “Doppler bins”. Each distance-speed gate 32 characterizes a distance 19 and a speed relative to the radar system 12. The 4 For better clarity, shows an example section of the signal matrix 34 with seven distance gates and seven speed gates. In total, the signal matrix 34 has, for example, 128 or 256 distance gates and 128 or 256 speed gates.

For example, from each of the signal matrices 34, the distance-speed gates 32 whose signal size 30 is above the respective predeterminable signal limit values are identified as signal target gates 36. The identified signal target gates 36 belong to the object targets 26. The signal limit values are determined using a CFAR algorithm, for example. An individual signal limit value can be set for each distance-speed gate 32. In the 4 Two signal target gates 36 are shown hatched. These are the distance-speed gates 32 with the coordinates (3, 4) and with the coordinates (5, 2). The signal limit value for the distance-speed gate 32 with the coordinates (3, 4) is determined using the value "30" as an example. The signal limit value for the distance-speed gate 32 with the coordinates (5, 2) is determined using the value "14" as an example.

From the signal matrices 34 of the four receiving antennas Rx1, Rx2, Rx3 and Rx4, the respective information of all range-velocity gates 32 is subjected to a common discrete Fourier transformation.

The respective maximum of the result of the common discrete Fourier transformation of the distance-velocity gates 32 is determined as a direction variable 38. The direction variables 38 each characterize a direction of an object target 26 relative to the radar system with respect to the corresponding distance-velocity gate 32.

The direction variables 38 are specified as corresponding direction indices. The direction variables 38, or the direction indices, characterize predetermined angle sections 40 of an aperture angle 42 of the radar system 12. The aperture angle 42 of the radar system 12 is defined as in the 3 shown, the monitoring area 14. The apex of the opening angle 42 is located at the radar system 12. The opening angle 42 is divided into a predetermined number of defined angle sections 40. Neighboring angle sections 40 overlap.

In the 3 the motor vehicle 10 is shown in a top view in a driving situation. The radar system 12 is designed in such a way that it can determine the direction of an object 18 in azimuth. For example, the opening angle 42 of the radar system 12 is divided into 16 angle sections 40. The limits of the opening angle 42 are indicated by two continuous straight lines. For the sake of better clarity, only three of the angle sections 40 are shown as examples. The limits of the angle sections 40 are indicated by dashed lines. Straight lines are indicated. The object 18 in the form of a pedestrian is, for example, located in a direction which is in the angle section 40 with the direction size 38 with the value “2”.

The direction variables 38 are assigned to the corresponding distance-velocity gates 32 of a virtual two-dimensional direction matrix 44. The direction matrix 44 is in the 5 shown. The direction matrix 44 has a distance dimension and a speed dimension that corresponds to the distance-speed dimension of the signal matrices 34. In the 5 the direction matrix 44 is shown, which is derived from the signal matrix 34 from the 4 For the 3 The driving situation shown results in directional variables 38 with values between 1 and 16.

From the direction matrix 44, the distance-speed gates 32, which correspond with respect to the distance-speed dimension to the signal target gates 36 of the signal matrix 34, namely the distance-speed gates 32 with the coordinates (3, 4) and (5, 2), are identified as direction target gates 46. In the 5 the corresponding directional target gates 46 are marked with hatching. According to the embodiment, the directional target gates 46 each contain the directional variable 38 with the value "2". The associated echo signals 28 therefore come from the same angular section 40 of the monitoring area 14.

An extension 48 of the detected object 18 is then determined in the distance-speed dimension of the direction matrix 44. For this purpose, starting from one of the direction target gates 46, those distance-speed gates 32 are assigned to the extension 48 of the object 18 that have the same direction size 38 plus or minus a predetermined direction size tolerance as the neighboring direction target gate 46 or, starting from the respective direction target gate 46, the neighboring distance-speed gate 32.

For example, starting from the direction target gate 46 with the coordinates (3, 4), the direction variables 38 of the distance-speed gates 32 adjacent to the direction target gate 46 with the coordinates (3, 4) are compared with the direction variable 38 of the direction target gate 46 with the coordinates (3, 4). The value “0” is specified here as an example as the direction variable tolerance. If the direction size 38 of a distance-velocity gate 32 adjacent to the direction target gate 46 with the coordinates (3, 4) has the same direction size 38 plus or minus the direction size tolerance as the direction target gate 46 with the coordinates (3, 4), namely "2", the checked distance-velocity gate 32 is assigned to the extent 48 in the distance-velocity dimension around the direction target gate 46 with the coordinates (3, 4). Otherwise, the said distance-velocity gate 32 is not assigned to the extent 48.

Starting from the distance-speed gates 32 which have been assigned to the extension 48 of the direction target gate 46 with the coordinates (3, 4), the direction variables 38 of the distance-speed gates 32 adjacent to these distance-speed gates 32 assigned to them, provided they have not already been assigned to the extension 48 by comparison with the direction target gate 46 with the coordinates (3, 4), are compared with the direction variable 38 of the assigned distance-speed gates 32. If the direction size 38 of a distance-velocity gate 32 adjacent to an associated distance-velocity gate 32 has the same direction size 38 plus or minus the direction size tolerance as the associated distance-velocity gate 32, the checked distance-velocity gate 32 is assigned to the extent 48. Otherwise, the checked distance-velocity gate 32 is not assigned to the extent 48.

If, as in the present embodiment, the extension 48 of the object 18 in the distance-velocity dimension in the direction matrix 44 around one of the direction target gates 46 has the same direction size 38 as the direction target gate 46 without any gaps, it is concluded that the detected object targets 26 of the object 18 are located in approximately the same direction to the radar system 12.

In this case, “uninterrupted” means that within the corresponding extent 48 there is no distance-speed gate 32 which has a smaller or larger direction size 38 than that of the respective direction target gate 46.

The extent 48 can be characterized by an extent quantity. For example, the number of distance-velocity gates 32 of the extent 48 can be used as an extent quantity.

In the 6 the extension 48 to the direction target gate 46 with the coordinates (3, 4) is indicated with dashed double arrows. The extension 48 of the direction target gate 46 with the coordinates (3, 4) comprises a total of six distance-speed gates 32.

The extension 48 to the direction target gate 46 with the coordinates (5, 2) is indicated with continuous double arrows. The extension 48 to the direction target gate 46 with the coordinates (5, 2) comprises a total of seven distance-speed gates 32.

The range-velocity gates 32 outside the extents 48 of the direction target gates 46 with the coordinates (3, 4) and (5, 2) do not belong to the object 12 to which the object targets 26 of the direction target gates 46 with the coordinates (3, 4) and (5, 2) belong.

If in the direction matrix 44 between two adjacent distance-velocity gates 32, which are located between two direction target gates 46, a difference between the respective direction sizes 38 is determined which is greater than the direction size tolerance, it is concluded that the two direction target gates 46 originate from two object targets 26 which do not belong to the same object 12.

Since in the embodiment shown the direction target gates 46 with the coordinates (3, 4) and (5, 2) each have the direction size 38 with the value "2" and between the two direction target gates 46 no distance-speed gate 32 has a smaller or larger direction size 38 than "2", it is concluded that the object targets 26 belonging to the direction target gates 46 with the coordinates (3, 4) and (5, 2) come from the same object 18, namely the one in the 3 shown pedestrians. Overall, the extension 48 of the two direction target gates 46 with the coordinates (3, 4) and with the coordinates (5, 2) comprises nine distance-speed gates 32.

The type of object 12, namely a pedestrian, is determined from the extent of the direction matrix 44, for example by means of pattern recognition, for example with the aid of an artificial neural network and/or a support vector machine.

From the size of the extension, in this case the number of distance-speed gates 32 of the extension 48 of the direction target gates 46 with the coordinates (3, 4) and (5, 2), in conjunction with the corresponding distances 19, which result from the distance dimensions of the direction target gates 46 with the coordinates (3, 4) and (5, 2), for example, the spatial extension of the object 18 to which the detected object targets 26 belong is deduced. From the spatial extension of the object 18, the type of object can be deduced.

In the 7 is the direction matrix 44 from the 6 shown, where the directional size tolerance is specified with the value “1” for determining the extension 48.

In the 8th is the direction matrix 44 from the 6 shown, where the number of distance-velocity gates 32 at the edge of the extension 48 is determined as an extension size. In the present case, these are eight distance-velocity gates 32.

Alternatively or additionally, the number of outer boundaries of the distance-velocity gates 32 at the edge of the extent 48 is determined here as a further extent quantity. In the present case, these are 14 outer boundaries.

In the 9 is a signal matrix 34 corresponding to the signal matrix 34 from the 4 shown, where the echo signals 28 originate from object targets 26 of an object 18 in the form of a guard rail. The corresponding signal target gates 36, whose signal sizes 30 are greater than 30, are located in the distance-velocity gates 32 with the coordinates (2, 7), (3, 4) and (5, 2). The extension of the object 18 in the signal matrix 34 is shown in the 9 roughly indicated with an ellipse.

10 shows the direction matrix 44, which corresponds to the signal matrix 34 of 9 belongs. The extent 48 of the corresponding direction target gates 46 is roughly indicated with an ellipse. The direction variables 38 in the distance-speed gates 32 within the extent 48 of the direction target gates 46 change continuously from the value "1" to the value "6". The direction variable tolerance, for example "1", between the neighboring distance-speed gates 32 is never exceeded. It can be concluded from this that the object 18, or the guardrail, extends over several angular sections 40.

Claims (10)

Method for determining object information of at least one object (18) which is detected by a radar system (12), wherein in the method - transmission signals (24) are sent into a monitoring area (14) of the radar system (12), - echo signals (28) which are reflected at at least one object target (26) of at least one object (18) present in the monitoring area (14) are received by at least one receiving antenna (Rx1, Rx2, Rx3, Rx4) and further processed as reception signals assigned to the receiving antenna (Rx1, Rx2, Rx3, Rx4), - at least one reception signal of at least one discrete Fourier transformation, - from the result of the at least one discrete Fourier transformation, at least one item of object information of the at least one object (18) is determined, characterized in that the echo signals (28) are received by at least two spatially separated receiving antennas (Rx1, Rx2, Rx3, Rx4), - from at least two of the spatially separated receiving antennas (Rx1, Rx2, Rx3, Rx4), the respective received signals are each subjected to a separate two-dimensional discrete Fourier transformation separately from one another, - from the results of the separate two-dimensional discrete Fourier transformations for each of the spatially separated receiving antennas (Rx1, Rx2, Rx3, Rx4), corresponding signal quantities (30) which characterize properties of the echo signals (28) received by the respective receiving antenna (Rx1, Rx2, Rx3, Rx4) are determined, - each signal quantity (30) is assigned to a corresponding Distance-velocity gate (32) is assigned to a virtual two-dimensional signal matrix (34), wherein the signal matrix (34) has a distance dimension and a speed dimension and each distance-velocity gate (32) characterizes a distance (19) and a speed relative to the radar system (12), - from the signal matrices (34) of at least two of the spatially separated receiving antennas (Rx1, Rx2, Rx3, Rx4), the respective information of at least one distance-velocity gate (32) is subjected to a common discrete Fourier transformation, - the maximum of the result of the common discrete Fourier transformation is determined as a direction variable (38) which characterizes a direction relative to the radar system (12), and the direction variable (38) is assigned to the corresponding at least one distance-velocity gate (32) of a virtual two-dimensional direction matrix (44) which has a Distance dimension and a speed dimension which are comparable with the signal matrices (34), - from at least one of the signal matrices (34), if present, the distance-speed gates (32) whose signal size (30) is above a respective predetermined or predeterminable signal limit value are identified as signal target gates (36) which belong to at least one object target (26), - in the direction matrix (44) for at least one distance-speed gate (32) which corresponds to one of the signal target gates (36) of the at least one signal matrix (34) with respect to the distance-speed dimension is identified as a direction target gate (46), - an extension (48) of at least one object (18) in the distance-speed dimension of the direction matrix (44) is determined by starting from at least one direction target gate (46) those distance-speed gates (32) of the extent (48) of the at least one object (18) are assigned which have the same directional size (38), optionally plus or minus a predetermined directional size tolerance, as the adjacent at least one directional target gate (46) or, starting from the at least one directional target gate (46), the adjacent distance-speed gate (32), - from the extent (48) in the distance-speed dimension of the at least one object information of the at least one object (18) is determined. Procedure according to Claim 1 , characterized in that, if an extension (48) of at least one object (18) in the direction matrix (44) has the same direction size (38) without gaps, optionally plus or minus a predetermined direction size tolerance, as the at least one direction target gate (46), it is concluded that the object targets (26) of the object (18) detected within the extension (48) lie in approximately the same direction to the radar system (12). Procedure according to Claim 1 or 2 , characterized in that at least one extension variable is determined which characterizes the extension (48) of the at least one object (18) in the direction matrix (44), and from the at least one extension variable at least one object information of at least one object (18) is determined. Procedure according to Claim 3 , characterized in that at least one extent size is determined as the total number of distance-velocity gates (32) of the extent (48), as the number of distance-velocity gates (32) or as the number of outer boundaries of the distance-velocity gates (32) at the edge of the extent (48) of the at least one object (18) in the direction matrix (44). Method according to one of the preceding claims, characterized in that, if in the direction matrix (44) between two adjacent distance-speed gates (32) which are located between two direction target gates (46), a difference between the respective direction variables (38) is determined which is greater than the direction variable tolerance, it is concluded that the two direction target gates (46) are from two object targets (26) that do not belong to the same object (18). Method according to one of the preceding claims, characterized in that an opening angle (42) of the radar system (12), which defines the monitoring area (14) and whose apex lies on the side of the radar system (12), is divided into a predetermined or predeterminable number of defined angle sections (40), which are each characterized by an associated direction index, and the direction variables (38) from the result of the common Fourier transformation are specified as corresponding direction indices. Procedure according to Claim 6 , characterized in that an opening angle (42) of the radar system (12), which defines the surveillance area (14) and whose apex lies on the side of the radar system (12), is divided into a predetermined number of defined angular sections (40), wherein at least two adjacent angular sections (40) overlap. Method according to one of the preceding claims, characterized in that the type of object (18) is determined from at least one extension (48) and/or optionally at least one extension size in the direction matrix (44) by means of pattern recognition. Procedure according to Claim 8 , characterized in that at least one artificial neural network and/or a support vector machine is used for pattern recognition. Radar system (12) for determining object information of at least one object (18), which has - at least one transmitting antenna (Tx), with which transmitting signals (24) can be transmitted into a monitoring area (14) of the radar system (12), - at least one receiving antenna (Rx1, Rx2, Rx3, Rx4), with which echo signals (28) which are reflected on at least one object target (26) of at least one object (18) present in the monitoring area (14) can be received, - at least one evaluation device (22) with which receiving signals which are assigned to the at least one receiving antenna (Rx1, Rx2, Rx3, Rx4) can be further processed, at least one receiving signal can be subjected to at least one discrete Fourier transformation and at least one item of object information of the at least one object (18) can be determined from the result of the at least one discrete Fourier transformation, characterized in that the radar system (12) has at least two spatially separated receiving antennas (Rx1, Rx2, Rx3, Rx4) with which echo signals (28) can be received, and the at least one evaluation device (22) has means for carrying out the method according to one of the preceding claims.

Images