DE102019216536A1 - Method and control device for vehicle navigation - Google Patents

Abstract

The invention describes a method for navigating a vehicle (10), with the following steps: reading in a predetermined ground radar map, reading in a predetermined semantic map which has at least one piece of traffic system information, recording a ground radar gram with a ground radar device attached to the vehicle (10) , Reading in the recorded ground radargram, determining a spatial position (9) of the vehicle (10) based on a correlation of the recorded and read in ground radargram with the predetermined and read in ground radar map, and establishing a travel trajectory (11) based on the determined spatial Location (9) of the vehicle (10) and depending on the semantic map that has been read. The invention also describes a control device (100) for carrying out such a method. The invention also describes a vehicle (10) with a ground radar device and such a control device (100).

Description

The invention relates to a method for navigating a vehicle. The invention also relates to a control device for navigating a vehicle and a vehicle with such a control device.

It is known to attach a radar device to a vehicle and to detect objects in the vicinity of the vehicle with the radar device. In addition, it is known to move a radar device over a ground in order to record ground properties.

For example is off DE102015208364A1 a method for determining properties of a surface on which a vehicle that can be operated away from paved roads is moved. Signals are sent in the direction of the ground on which the vehicle is moving.

In one aspect, the invention relates to a method for navigating a vehicle. The navigation can include controlling the vehicle along a driving trajectory, wherein driving dynamics of the vehicle can be controlled. For this purpose, the method has a determination of a spatial position of the vehicle. The vehicle can be designed as an automated or autonomous vehicle. The vehicle can be a non-rail-bound vehicle or a rail-bound vehicle. The non-rail vehicle can be a road vehicle, for example a passenger car or a truck, or an off-highway vehicle, for example a construction site vehicle or a tractor.

The spatial position of the vehicle can have a spatial position of the vehicle, wherein the position can have at least one coordinate in a spatial coordinate system. As an alternative or in addition to the position, the spatial position of the vehicle can have a spatial orientation of the vehicle, wherein the orientation can have at least one orientation of the vehicle with respect to a coordinate axis of the spatial coordinate system.

The method has as a step reading in from a predetermined ground radar map. The ground penetrating radar map can have a variety of ground penetrating radar programs. The multiplicity of ground radar programs can have been prerecorded with a ground radar device attached to a vehicle. The ground penetrating radar device can be a multifrequency ground penetrating radar device. The pre-recorded ground radar programs can be combined in a coordinate system of the ground radar map, that is to say they can be mapped together with respect to the coordinate system. The ground radar map can be a three-dimensional ground radar map. The ground radar map can therefore also be a radar mapping of a ground. The ground penetrating radar can contain raw data from the radar measurement or be based on this. Alternatively, the ground penetrating radar map can have preprocessed or filtered raw data from the radar measurement. For example, the ground radar map can have a voxel-based 3D interpolation of measured reflections.

As a further step, the method includes reading in a predetermined semantic map. The semantic map has at least one piece of traffic system information. The semantic map can have information about a traffic system which the vehicle uses or to which the vehicle belongs. The traffic system can have at least one traffic infrastructure object. Alternatively or additionally, the traffic system can have at least one means of transport. Alternatively or additionally, the traffic system can have at least one traffic control system. The traffic control system can function to coordinate the means of transport.

The semantic map can also have at least one piece of street equipment information. The semantic map can thus have at least one piece of information about road equipment. The street equipment can have a traffic light system. Alternatively or additionally, the road equipment can have a traffic sign.

The semantic map can also be a dynamic semantic map which has at least one piece of dynamic road equipment information or traffic information. The dynamic road equipment information or traffic information can have at least one corresponding change in the state of the traffic infrastructure, of the means of transport, of the traffic sign, of the traffic control system or of the road equipment. The dynamic semantic map can also have at least two of the state changes described.

As a further step, the method has the acquisition of a ground radar gram with a ground radar device attached to the vehicle. The ground penetrating radar program can be a result of a ground penetrating radar measurement with the ground penetrating radar device. The term “ground radar” used in connection with the invention can also be understood as “ground penetrating radar” or as “ground penetrating radar (GPR)”.

The ground penetrating radar program can have a ground penetrating radar image. The ground radar program or the ground radar image can be a corresponding two-dimensional or three-dimensional result of the ground radar measurement with the ground radar device. The ground penetrating radar can have reflections or transmissions of electromagnetic waves transmitted and received by the ground penetrating radar device. The reflections can be caused by reflected objects or boundary surfaces in the ground. An interface can be formed between soil layers with different soil properties or between material layers with different material properties. The ground radargram can therefore have depth values of ground objects or ground boundaries. The depth values can be based on a detected speed of propagation of the emitted, reflected and re-received electromagnetic waves.

The ground penetrating radar device can have a radar transceiver. The transmission and reception of the electromagnetic waves can be carried out with the radar transceiver. The ground penetrating radar device can have a large number of transmitting and receiving antennas for this purpose. The ground penetrating radar (GPR) device can be a so-called ground penetrating radar device. The ground radar device can be attached to an underside of the vehicle or to a vehicle underbody. Alternatively, the ground penetrating radar device can be attached to one side of the vehicle. The ground penetrating radar device can furthermore have at least one high-frequency antenna and / or at least one low-frequency antenna. In this way, different depths of penetration of the electromagnetic waves into the ground can be achieved in an advantageous manner.

As a further step, the method includes reading in the recorded ground penetrating radar gram. This and all further steps of the method can be carried out by a control device provided on the vehicle.

As a further step, the method has a determination of a spatial position of the vehicle based on a correlation of a recorded and read-in ground radar gram with the predetermined and read-in ground radar map. The ground penetrating radar can be recorded in a vehicle-related coordinate system. If the vehicle moves along a trajectory, a large number of ground penetrating radargrams can be recorded, each of which can relate to the vehicle-related coordinate system. An attachment position of the ground penetrating radar device in the vehicle-related coordinate system can be predefined in order to establish a relationship between a detected ground radar program and the trajectory driven by the vehicle.

The correlation can be a mathematical or geometric correlation of the ground penetrating radar gram with the ground penetrating radar map. The reflections in the ground recorded with the ground penetrating radar program can be compared and correlated with the predetermined or pre-recorded reflections contained in the ground penetrating radar map. This can be done using methods of pattern recognition or using matching algorithms. As a result of the correlation, a position of the detected ground radar gram in the coordinate system of the predetermined ground radar map can be derived.

The spatial position of the vehicle can be deduced from the position derived from the ground penetrating radar gram in the ground penetrating radar map. The basis for this can be that the position of the ground penetrating radar device is predetermined in the vehicle-related coordinate system. The spatial position of the vehicle can thus be determined in the coordinate system of the predetermined ground radar map. The determination of the spatial position can be based on corresponding coordinate transformations. The coordinate system of the ground radar map can be a global coordinate system, for example a GNSS coordinate system, or a local coordinate system, for example a street coordinate system or a rail coordinate system, or any other known geodetic coordinate system.

As a further step, the method has the determination of a driving trajectory based on the determined spatial position of the vehicle and in Depending on the read semantic map. The driving trajectory can have at least one polygon which can describe the trajectory of the vehicle to be driven.

The travel trajectory can define a travel path of the vehicle. The travel trajectory can also have a course of travel of the vehicle. To navigate the vehicle, the vehicle can be steered along the defined driving trajectory. The transverse dynamics of the vehicle or a transverse dynamics component of the vehicle can be controlled, with the vehicle being able to be steered along the trajectory. As an alternative or in addition, the longitudinal dynamics of the vehicle or a longitudinal dynamics component of the vehicle can also be activated, with the vehicle being able to be braked or accelerated when following the driving trajectory. Navigating or driving the vehicle can be carried out as a function of the at least one traffic system information item and thus as a function of the semantic map that has been read.

In the context of the invention, the spatial position of a moving vehicle can thus be determined solely on the basis of ground radar measurements or reflections in the ground for navigating the vehicle. The ground radar device can therefore be designed as a dynamic ground radar for navigating a vehicle. The vehicle can therefore be localized and navigated on-the-fly using the method. The method can thus also be carried out for radar-based localization and navigation of the vehicle. The method can therefore be carried out independently of satellite-based navigation of the vehicle. Above-ground environmental influences and disruptive influences above the ground can thus advantageously not have an effect on the underground spatial position determination according to the method, which can be carried out more robustly and reliably than other above-ground navigation methods.

According to one embodiment of the method, this has as a further step reading in a predetermined semantic soil map. The semantic ground map does not have any ground radargrams necessary for carrying out the method. Rather, the semantic soil map can contain information about soil objects. As an alternative or in addition, the semantic soil map can contain information about soil boundary surfaces. The semantic soil map can also have information about a type of object or a type of soil or soil moisture. The semantic soil map can thus be a map which, in addition to mere coordinate-based mapping of soil objects, has at least one semantic of the mapped soil objects in the form of additional information about the mapped soil objects. The semantic soil map can also be a three-dimensional soil map.

A coordinate system of the semantic ground map can coincide with the coordinate system of the ground radar map or a spatial reference between the coordinate systems of the two maps can be predetermined. The semantic ground map can also have been derived from the predetermined ground radar map.

According to this embodiment, the step of acquiring from the ground radargram can be carried out as a function of the predetermined and read-in semantic ground map. A ground penetrating radar measurement carried out with the ground penetrating radar device to generate the ground penetrating radar gram can depend on the semantic ground map that has been read. A map section of the semantic soil map can be taken into account, which has information about soil objects or soil boundary surfaces which are currently located under the vehicle. In this way, at least one measurement parameter of the ground penetrating radar device can be set as a function of the ground objects or boundary surfaces currently located below the vehicle.

The predetermined and read-in semantic soil map can have information about at least one object installed on or in the soil. The object can be a structure, for example a road, a tunnel, a bridge or a foundation, or at least one structure part, for example at least one structure part of the structures mentioned above by way of example. The objects can also be pipes, lines or channels. The objects can lie on or under the surface of the ground. As an alternative to the object, the semantic soil map can also have information about a natural object present on or in the soil. The natural object can be a soil layer or vegetation, for example a root. The information on the semantic soil map is not, however, reflections from a ground penetrating radar measurement, but rather information about ground objects or ground boundary surfaces derived, modeled or simulated from such reflections. The information can, for example, have been derived from a radar program, a construction plan or a construction drawing.

The predetermined and read-in semantic soil map can contain information about a road structure. The road structure can have been derived from a construction plan of the road on which the vehicle is traveling. As an alternative to this, the road structure can also have been derived from pre-recorded radar programs of the road currently being traveled on by the vehicle. The information about the road structure can include at least one piece of information about a road superstructure or a road surface. Alternatively or additionally, the information about the road structure can have at least one piece of information about a road substructure. As an alternative or in addition, the information about the road structure can have at least one item of information about a road surface.

The road structure or the road superstructure or the road substructure can have at least one building layer. The at least one building layer can have an asphalt layer. The at least one building layer can also have a concrete layer. The least a build-up layer can also have a composite layer. In a further example, the at least one build-up layer can also have a gravel layer.

In the context of the invention, the floor can be a floor outside of buildings or a floor inside or below a building. The method can thus be carried out when the vehicle is driving outside of buildings or outdoors or inside a building or indoors. The ground penetrating radar can thus also have reflections in a building ceiling or in a building foundation, for example. The ground penetrating radar can also have reflections in a bridge structure that the vehicle is traveling on.

As a further step, the method can have a stipulation of at least one frequency of electromagnetic waves emitted by the ground penetrating radar device during the detection of the ground penetrating radar gram as a function of the semantic ground map read in. If the radar device is a multi-frequency radar device, a large number of frequencies can be defined. The frequency can therefore be a selective or dynamic frequency. By defining the at least one frequency, a penetration depth of the electromagnetic waves emitted by the ground penetrating radar device can be changed or adjusted. The depth of penetration of the electromagnetic waves of the ground penetrating radar device can thus be adapted to a depth of objects built on or in the ground or to the road structure. The depth of the objects can be derived from the semantic soil map. Thus, depending on the depth of the objects built on or in the ground or of the boundary surfaces of the road structure, the frequency and thus also the depth of penetration of the ground penetrating radar measurement with the ground penetrating radar device can be determined. While the vehicle is in motion, a frequency of the ground radar device can therefore be dynamically adapted to different depths of different objects or a changing road structure below the vehicle. The penetration depth of the electromagnetic waves can thus be adapted to existing potential reflection objects, which means that the ground penetrating radar measurement can be more reliable and more accurate.

The method can include, as a further step, defining at least one frequency band of electromagnetic waves emitted by the ground penetrating radar device when the ground penetrating radar gram is detected. The frequency band can therefore be a selective or dynamic frequency band. The ground penetrating radar device can be a multifrequency ground penetrating radar device. The frequency band can have a large number of emitted electromagnetic waves of different frequencies. In detection areas with objects or boundary surfaces of a road structure at different depths, a large number of frequency bands can also be defined depending on the different depths. Reliable radar-based detection of the reflections necessary for correlating can thus be ensured even in a transition between different traffic routes or in a transition area between different road structures. The transition can also be a transition from an outdoor environment to an indoor environment or vice versa.

As a further step, the method can include the acquisition of a meteorological parameter in the surroundings of the vehicle with a meteorological sensor attached to the vehicle. The meteorology sensor can be a precipitation sensor, an air humidity sensor, a soil moisture sensor or a solar radiation sensor. The meteorological parameter can therefore be at least one of precipitation, air humidity, soil moisture or solar radiation. In addition to the known meteorology sensors, the meteorology sensor can also be a radar sensor or an infrared camera. This means that the meteorological parameters in the vicinity of the vehicle can be recorded directly with a radar sensor or with an infrared camera. In a further embodiment, the radar sensor can be the ground radar device. The ground penetrating radar can therefore also function as the meteorological sensor.

As a further step, the method can include correcting the recorded ground penetrating radar program based on the recorded at least one meteorological parameter. The measured reflections of the ground penetrating radar gram and the associated depth values of objects or interfaces can thus be corrected based on the at least one meteorological parameter. For this purpose, soil moisture or soil water content, for example, can be derived from the recorded meteorological parameters, which can have an effect on the propagation speed of the electromagnetic waves emitted by the ground penetrating radar device. The derived soil moisture can thus be used to correct the recorded ground radargram.

As a further step, the method can include detecting at least one ground property of the ground in the vicinity of the vehicle with the ground penetrating radar device attached to the vehicle, based on electromagnetic waves transmitted and received by the ground penetrating radar device. A soil type or soil moisture can thus be derived from the soil radargram, especially taking into account corresponding reference values. According to this embodiment, the method can have, as a further step, correcting the recorded ground radargram based on the at least one recorded ground property.

The soil property can have a physical property of the soil. The soil property can for example have a soil water content, a soil water movement or a soil water flow or a soil moisture content. Alternatively or additionally, the at least one property of the soil can have a clay content of the soil, a grain size composition of the soil, a storage density of the soil, a pore content of the soil, a degree of compaction of the soil or a soil resistance. Each of the at least one property of the soil can be derived directly or indirectly from the received electromagnetic waves.

The step of determining the spatial position of the vehicle can include deriving at least one spatial position parameter of the vehicle and at least one spatial orientation parameter of the vehicle. If the coordinate system of the ground penetrating radar map is a three-dimensional coordinate system, three spatial position parameters and at least two spatial orientation parameters can be derived. The position parameter can be a coordinate difference or a translation parameter.

The orientation parameter can be a rotation about a coordinate axis or a rotation parameter. The determination of the spatial position of the vehicle can also include deriving transformation parameters between the vehicle-related coordinate system and the coordinate system of the ground radar map. This can be two-dimensional or three-dimensional transformation parameters. The transformation parameters can include the position parameter and the orientation parameter.

According to a further embodiment of the method, the at least one traffic system information item has information about a traffic infrastructure object. The traffic infrastructure object can be a traffic route. In a further example, the traffic infrastructure object is a traffic structure. The traffic system information can include information on lane boundaries, lane boundaries, lane crossings, pedestrian crossings or parking spaces, for example.

Alternatively or additionally, the at least one traffic system information item has information about a means of transport. The means of transport can be a vehicle. The vehicle can be any vehicle for moving people or goods. The vehicle can also be a non-rail-bound vehicle or a rail-bound vehicle.

Alternatively or additionally, the at least one traffic system information item can have information about a traffic sign. The traffic sign can be a traffic sign. The traffic system information can include information on a stop sign or a right of way sign, for example. The traffic sign can also be a permanent traffic sign or a variable message sign.

Alternatively or additionally, the at least one traffic system information item can have information about a traffic control system. The traffic control system can be a system for the automated control of traffic. The traffic guidance system can also be a parking guidance system. The traffic control system can also be a traffic light system. The information described can be relevant states or control commands for car-to-car communication (C2C). The established trajectory can thus be established as a function of at least one of the items of information described.

A further advantage of the invention can thus be seen in the fact that the spatial position of the vehicle can be determined independently of disruptive influences within a traffic system. Traffic infrastructure objects, means of transport, traffic signs and systems of a traffic control system cannot influence the ground radar-based determination of the spatial position of the vehicle, since measurements with the ground radar device can only be carried out in the direction of the ground, starting from the vehicle. In contrast, the described components of the traffic system are not located in the ground penetrating measurement path of the ground penetrating radar device. Measurement failures due to above-ground shadowing can therefore can be avoided with the method in an advantageous manner. Based on such a robust and reliable underground position determination with the ground penetrating radar device, the traffic system information can be used efficiently to navigate the vehicle.

According to a further embodiment of the method, the at least one traffic system information item has a predefined travel trajectory for the vehicle at a traffic junction. The traffic junction can be an intersection lying ahead in the direction of travel of the vehicle. The predefined travel trajectory can be a travel trajectory which, at least in sections, does not coincide with a center line of a lane. The predefined driving trajectory can thus take into account a turning behavior of the vehicle. The predefined driving trajectory can therefore be a so-called “virtual” driving trajectory, which can be predefined independently of lane markings or lane boundaries.

If the vehicle is in a spatial position which has a lateral distance from the predefined driving trajectory, a driving trajectory which guides the vehicle onto the predefined driving trajectory can be defined in the step of defining. The specified travel trajectory can therefore be a corrective travel of the vehicle in order to travel on the predefined travel trajectory again at a later point in time. If, on the other hand, the vehicle is already on the predefined driving trajectory, the predefined driving trajectory itself can also be defined in the step of defining in order to continue navigating the vehicle on it.

According to a further embodiment of the method, the at least one traffic system information item has a predefined driving dynamics profile for the vehicle. The predefined driving dynamics course can relate to the predefined driving trajectory. The driving dynamics profile can have a longitudinal dynamics profile. Alternatively or additionally, the driving dynamics course can have a transverse dynamics course. The traffic system information can thus, for example, provide a turning behavior for the vehicle. As an alternative or in addition, the traffic system information can provide a speed specification, a lower speed limit or an upper speed limit for the vehicle speed. The predefined driving dynamics course can be information about the traffic control system described.

According to a further embodiment of the method, the at least one traffic system information item has a predefined driving operation setting for the vehicle. The predefined driving operation specification can relate to the predefined driving trajectory. The driving operation specification can be, for example, a local driving ban. In further examples, the driving operation specification can be a prohibition of overtaking, a priority control over another vehicle, or a platooning specification. The driving mode specification can also be a maneuver profile for the vehicle, which takes into account a turning circle or a drag curve when cornering, for example, taking into account a vehicle model.

According to a further embodiment of the method, this has, as a further step, a merging of the defined driving trajectory with a further driving trajectory which has been derived based on a SLAM algorithm for navigating the vehicle. The vehicle can have further sensors. For example, the vehicle can have a laser scanner for navigating the vehicle based on the SLAM algorithm. Furthermore, the vehicle can also have image-based sensors for image-based navigation of the vehicle. Furthermore, the vehicle can also have a satellite-based navigation system or an inertial measuring system for navigating the vehicle. Driving trajectories derived from various sensors can thus be merged to increase the reliability and accuracy of navigation.

According to a further embodiment of the method, this has as a further step a detection of a position of the vehicle with a position detection sensor attached to the vehicle. According to this embodiment, the correlation can be carried out as a function of the detected position of the vehicle. The position detection sensor can be, for example, a GNSS antenna attached to the vehicle or a WLAN antenna. The detection of the position of the vehicle can function for the predetermination or rough determination of the spatial position of the vehicle. The predetermination or rough determination can be improved or refined based on the step of correlating, wherein the spatial position determined based on the step of correlating can be more reliable and more precise than the spatial position of the vehicle determined with the position detection sensor. Thus, the spatial position of the vehicle determined based on the step of correlating can be more robust with respect to the spatial position predetermined with the position detection sensor. By detecting the position, ambiguities in the correlation step can also be resolved.

According to a further embodiment of the method, the method has, as a further step, communication of the recorded ground radar program with a map information system. The map information system can be a geographic information system (GIS). The map information system can have the predetermined ground radar map. The predetermined ground radar map can thus be updated with the detected ground radar program. The communication can also be car-to-X communication (C2X).

According to a further embodiment of the method, the vehicle is designed as an automatically operated vehicle. If the vehicle is an autonomous vehicle, the vehicle can thus advantageously be operated more reliably with one degree of autonomy or the vehicle can be operated with a higher degree of autonomy. According to this embodiment, the steps of the method for automated operation can be carried out automatically by the vehicle.

In a further aspect, the invention relates to a control device for navigating a vehicle. The control device can be arranged on the vehicle. The control device can be set up to carry out the method according to the preceding aspect.

The control device has an interface for reading in a predetermined semantic map which has at least one piece of traffic system information. The control device also has an interface for reading in a ground radar program recorded by a ground radar device attached to the vehicle. The control device can also have an interface for reading in a predetermined semantic soil map.

Each of these interfaces can be designed as a wired interface or a radio-based interface. If the corresponding interface is a wired interface, the ground radar map or the semantic ground map can be called up from a memory provided on the vehicle, which stores the ground radar map or the semantic ground map. If the interface in question is a radio-based interface, the ground radar map or the semantic ground map can also be called up in a cloud-based or mobile radio-based manner.

The control device has a correlation unit for correlating the read in ground radar gram with the read ground radar map. The correlation unit of the ground penetrating radar gram can automatically perform the correlation of the ground penetrating radar gram with the ground penetrating radar map.

The control device also has a determination unit for determining a spatial position of the vehicle based on a correlation result of the correlation unit. The correlation result can have the spatial position of the ground radargram in the ground radar map or corresponding spatial location parameters. The spatial position of the vehicle can be derived as described based on this relative ground radargram position.

The control device also has a definition unit for defining a travel trajectory based on the determined spatial position of the vehicle and as a function of the semantic map that has been read.

In a further aspect, the invention relates to a vehicle. The vehicle can be designed as described for the preceding aspects. A ground radar device is attached to the vehicle. The vehicle has a control device according to the preceding aspect for controlling the ground radar device and for navigating the vehicle.

• 1 zeigt ein Fahrzeug gemäß einer Ausführungsform der Erfindung in einer Verkehrssituation mit einer in Fahrtrichtung des Fahrzeugs vorausliegender Kreuzung in einer Draufsicht.
• 2 zeigt das Fahrzeug von 1 mit einer Steuereinrichtung zum Bestimmen von einer räumlichen Lage des Fahrzeugs gemäß einer Ausführungsform der Erfindung.
• 3 zeigt die Steuereinrichtung von 2 mit Einrichtungsschnittstellen und Einrichtungseinheiten.
• 4 zeigt ein Flussdiagramm zur Erläuterung eines Verfahrens zum Bestimmen von einer räumlichen Lage des Fahrzeugs gemäß einer Ausführungsform der Erfindung.

The method can also be carried out based on the determined spatial position and the established trajectory for controlling the driving dynamics of the vehicle along the trajectory.

• 1 shows a vehicle according to an embodiment of the invention in a traffic situation with an intersection lying ahead in the direction of travel of the vehicle in a top view.
• 2 shows the vehicle from 1 with a control device for determining a spatial position of the vehicle according to an embodiment of the invention.
• 3 shows the control device of 2 with equipment interfaces and equipment units.
• 4th FIG. 3 shows a flow chart to explain a method for determining a spatial position of the vehicle according to an embodiment of the invention.

1 shows a vehicle 10 in a traffic situation when the vehicle is retracted 10 in one in the direction of travel of the vehicle 10 crossing ahead 2 . The vehicle 10 has a control device 100 on.

The control device 100 is set up in one embodiment to a spatial location 9 of the vehicle 10 to determine. The control device 100 is set up in a further embodiment to the vehicle 10 along a driving trajectory 11 to steer or navigate. The control device 100 is set up in yet another embodiment to a driving dynamics of the vehicle 10 along the driving trajectory 11 to control. The driving dynamics can be at least one of longitudinal dynamics or transverse dynamics of the vehicle 10 exhibit.

The one from the control device 100 pre-determined driving trajectory 11 becomes independent of road markings 4th the intersection 2 set. The driving trajectory 11 is based on that from the controller 100 specific spatial location 9 of the vehicle 10 set. The driving trajectory 11 is also independent of the edges of the road 6th the intersection 2 set.

In the 1 The traffic situation shown also has a road equipment facility 3 on. In the case of the 1 street equipment shown 3 it is a traffic light system 5 . The condition of the traffic light system 5 is operated by an in 2 shown and on the vehicle 10 attached imaging sensor 14th detected. The recorded status of the traffic light system 5 is controlled by the control device 100 read in to the vehicle 10 along the defined driving trajectory 11 to control, for example to brake or accelerate.

2 shows the vehicle 10 of 1 with the control device arranged on the vehicle 100 . The vehicle 10 instructs the imaging sensor 14th on which one in the direction of travel of the vehicle 10 aligned image capture area 15th in which the in 1 traffic light system shown 5 lies. The imaging sensor 14th is with the control device 100 connected.

The vehicle 10 has a ground penetrating radar 12th on, which is on an underside of the vehicle 10 is appropriate. The ground penetrating radar 12th has a radar detection area 13th which in the direction of the ground on which the vehicle 10 drives, is aligned. The floor is in the in 1 The embodiment shown is formed by the road structure of the intersection or the road forming the intersection and the soil located below the road structure. The ground penetrating radar 12th records a ground penetrating radar. While the vehicle is in motion 10 detected by the ground penetrating radar 12th dynamically a variety of ground penetrating radar programs. The ground penetrating radar 12th is with the control device 100 connected.

The vehicle 10 also has a position detection sensor 18th on, which is on an upper side of the vehicle 10 is arranged. In one embodiment, it is the position detection sensor 18th a GNSS receiver. The position detection sensor 18th is with the control device 100 connected.

The vehicle also has a meteorology sensor 16 on, which is on an outside of the vehicle 10 is arranged. With the meteorology sensor 16 in one embodiment it is at least one of a temperature sensor, an air humidity sensor, a soil moisture sensor, an infrared camera, a solar radiation sensor and a precipitation sensor. The meteorology sensor 16 is designed in a further embodiment as a radar sensor to determine at least one of a humidity, a soil moisture and a precipitation. The meteorology sensor 16 is with the control device 100 connected.

The vehicle 10 has a driving dynamics device 19th , for example at least one of a vehicle steering and a vehicle drive. The control device 100 is with the driving dynamics device 19th via a driving dynamics control device 150 for controlling the driving dynamics device 19th connected.

3 shows the control device 100 the 1 and 2 , which with the ground penetrating radar 12th , the imaging sensor 14th , the in 3 meteorology sensor not shown 16 , the driving dynamics control device 150 and a card system 50 connected is. The control device 100 is with the ground penetrating radar 12th via a radar interface 104 connected. The control device 100 is with the card system 50 via a card interface 102 connected. The control device 100 is with the imaging sensor 14th via a sensor interface 106 connected. The control device 100 is with the driving dynamics control device 150 via an output interface 108 connected.

The control device 100 also has a correlation unit 110 , a determination unit 120 , a definition unit 130 and an evaluation unit 140 which are connected to each other.

The card system 50 has a ground penetrating radar map 20th on. The ground penetrating radar 12th records a ground penetrating radar 22nd . The ground radar map 20th is via the card interface 102 from the control device 100 read in and the ground penetrating radar 22nd is via the radar interface 104 from the control device 100 read in. The read Ground penetration radar 22nd is scanned with the ground radar map 20th from the correlation unit 110 read in and correlated. The correlation unit 110 gives as a correlation result 112 a location of the ground penetrating radar gram 22nd in the ground radar map 20th based on correlating or matching from the ground penetrating radar gram 22nd with the ground radar map 20th out. The correlation unit 110 thus determines where this is from the ground penetrating radar 12th recorded ground radar gram 22nd in the ground radar map 20th is located. The correlation unit determines 110 the location of the ground penetrating radar 22nd in the soil map 20th .

The card system 50 also has a semantic floor map 40 which information about objects built on or in the ground is contained. The control device 100 reads information from the semantic soil map 40 on, evaluates them, for example with the correlation unit 110 , and issues a control command via the radar interface 104 to the ground penetrating radar 12th depending on the evaluated semantic soil map 40 out. The control device determines 100 as a control command a frequency or a frequency band with which the ground penetrating radar 12th has to emit electromagnetic waves into the ground. The control device 100 thus controls the current frequencies of the one with the ground penetrating radar 12th currently emitted electromagnetic waves depending on the current, that is, located below the vehicle 10 located, information from the semantic soil map 40 .

The card system 50 also has a semantic map in one embodiment 30th on. The semantic map 30th has at least one of information on a traffic system and information on road equipment.

The unit of determination 120 is set up for the in 1 spatial location shown 9 of the vehicle 10 based on the correlation result 112 the correlation unit 100 to determine. The unit of determination 120 determines the vehicle's position parameters 10 with respect to its surroundings in a coordinate system of the ground penetrating radar map or in another global coordinate system. So determines the determining unit 120 where the vehicle is 10 on the in 1 shown intersection 2 is located. In addition, the determining unit determines 120 like the vehicle 10 relative to the intersection 2 , the lane markings 4th or the edge of the road 6th is aligned. The unit of determination 120 is with the definition unit 130 connected and gives the spatial location 9 of the vehicle 10 to this out. The definition unit 130 puts the in 1 shown driving trajectory 11 based on the spatial location 9 of the vehicle 10 firmly. The definition unit takes effect 130 also on the read semantic card 30th to.

The evaluation unit 140 is via the sensor interface 106 with the imaging sensor 14th connected. One from the sensor 14th detected state 35 from the road equipment facility 3 or from the traffic light system 5 is from the evaluation unit 140 via the sensor interface 106 read in and evaluated.

The control device 100 gives via the output interface 108 a driving dynamics control command to the driving dynamics control device 150 out. The driving dynamics control command is based on that from the determining unit 120 specific spatial location 9 of the vehicle 10 and that of the definition unit 130 defined driving trajectory 11 .

In one embodiment, the driving dynamics control command is alternatively or additionally based on that with the evaluation unit 140 evaluated state 35 . As an alternative to the output of the control command, the specific spatial position 9 or the defined driving trajectory 11 directly to the driving dynamics control device 150 issued.

4th shows a schematic flow diagram to explain a method for determining the spatial position 9 of the vehicle 10 . In one embodiment, the method is a method for navigating the vehicle 10 along the driving trajectory 11 . In a further embodiment, the method is a method for controlling the vehicle 10 based on a state 35 from the road equipment facility 3 .

Position detection takes place in an optional step S0. In step S0, the position detection sensor 18th a position 8th of the vehicle 10 detected. The recorded position 8th serves as a rough positioning for selecting data segments from the map system 50 and to reduce the amount of data that is read out and read in.

The card system 50 points the ground radar map 20th , the semantic soil map 40 and the semantic map 30th on. The maps are so-called "layers" of the map system 50 . In a further step S1 the ground radar map is read in 20th by the control device 100 . Based on the captured position 8th becomes an optional ground radar map segment 21 read in. In a further step S2 the semantic soil map is read in 40 by the control device 100 . Based on the captured position 8th optionally becomes a semantic ground map segment 41 read in.

In a further step S3 a ground penetrating radar gram is recorded with the ground penetrating radar device 12th . The step S3 will depend on the step S2 carried out. Assigns the read-in semantic soil map 40 Information about the one below the vehicle 10 a frequency or a frequency band is determined by the ground penetrating radar device 12th emitted electromagnetic waves to this information of the read semantic soil map 40 customized. So becomes a penetration depth of the ground penetrating radar device 12th in the ground below the vehicle 10 adapted to the road structure in such a way that objects and boundary surfaces in the ground according to the information on the semantic soil map 40 are recorded. A generated ground radar gram 22nd thus has a penetration depth and therefore also measurement data with regard to the detected objects and interfaces. In a further step S4 the recorded ground penetrating radar program is read in 22nd by the control device 100 .

In a further step S5 there is a correlation from the recorded ground radar gram 22nd with the scanned ground radar map 20th or the scanned ground radar map segment 21 . The recorded ground radar gram 22nd is with the ground penetrating radar map 20th or the ground radar map segment 21 correlated to the spatial location of in the ground penetrating radar gram 22nd detected objects and interfaces with those in the ground penetrating radar map 20th correlate pre-detected objects and interfaces. Correlating involves adapting and superimposing the objects and interfaces. Is the ground radar map 20th Georeferenced are the reflections of the ground penetrating radar that relate to the spatial position of the objects and interfaces 22nd in that of the georeferenced ground radar map 20th underlying coordinate system is determined.

In a further step S6 the position of the vehicle is determined 10 . Based on the correlation result 112 becomes the spatial location 9 of the vehicle 10 certainly. The vehicle 10 is then based on the determined spatial location 9 controlled or localized and navigated.

In a further step S7 there is a determination of the driving trajectory. Based on the specific spatial location 9 will the in 1 shown driving trajectory 11 for the vehicle 10 set. When driving at the intersection 2 becomes the vehicle 10 so based on the set driving trajectory 11 controlled.

Optionally takes place in a further step S3a a meteorological survey. Meteorological data are recorded with the meteorological sensor 16 recorded and in a further optional step S3b to correct the in step S3 recorded ground penetrating radar 22nd used. In one embodiment, soil moisture is radar-based with the radar device 12th measured and to correct the recorded ground penetrating radar gram 22nd used.

In one embodiment, the card system has a semantic card 30th on. The semantic map 30th has at least one of traffic system information 32 and road facilities information 34 on. In a further step S2a the semantic card is read in 30th . Based on the captured position 8th becomes a map segment 31 from the control device 100 read in. The step S7 According to this embodiment, the determination of the driving trajectory is based on at least one of the imported traffic system information 32 and the read-in road equipment information 34 . The traffic system information 32 and the road equipment information 34 can each be designed as described.

Based on the step S2a In one embodiment, a status detection takes place S8 from the road equipment facility 3 , which is in the in 1 embodiment shown to a traffic light system 5 acts. The condition of the traffic light system 5 is with the imaging sensor 14th in step S8 detected. A currently sent out light signal of the light signal system 5 is used as the status of the traffic light system 5 detected. In a further step S9 becomes the detected state 35 from the control device 100 read in.

In a further step S10 a vehicle control takes place. Driving the vehicle 10 is based on the specific spatial location 9 . Driving the vehicle 10 in one embodiment is also based on the defined driving trajectory 11 . Driving the vehicle 10 in a further embodiment is also based on the detected state 35 from the road equipment facility 3 . When driving the vehicle 10 its driving dynamics is controlled.

List of reference symbols

2

crossing

3

Road equipment facility

4th

Road marking

5

Traffic light system

6th

Edge of the road

8th

position

9

spatial location

10

vehicle

11

Driving trajectory

12th

Ground penetrating radar

13th

Radar detection area

14th

imaging sensor

15th

Image capture area

16

Meteorology sensor

18th

Position detection sensor

19th

Driving dynamics device

20th

Ground radar map

21

Ground radar map segment

22nd

Ground penetration radar

30th

semantic card

31

Card segment

32

Traffic system information

34

Road equipment information

35

State of the road equipment

40

semantic soil map

41

semantic ground map segment

50

Card system

100

Control device

102

Card interface

104

Radar device interface

106

Sensor interface

108

Output interface

110

Correlation unit

112

Correlation result

120

Determination unit

130

Definition unit

140

Evaluation unit

150

Driving dynamics control device

50

Position detection

S1

Reading in ground radar map

S2

Reading in semantic soil map

S2a

Reading in semantic map

S3

Ground penetrating radar capture

S3a

Meteorology recording

S3b

Radar gram correction

S4

Reading in ground radar gram

S5

correlation

S6

Orientation

S7

Determination of driving trajectories

S8

Condition assessment

S9

Reading in state

S10

Vehicle control

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed 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 102015208364 A1 [0003]

Claims (12)

Method for navigating a vehicle (10), with the steps: - Reading (S1) from a predetermined ground radar map (20), - Reading (S2a) from a predetermined semantic map (30) which has at least one piece of traffic system information (32), - Acquisition (S3) of a ground radar program (22) with a ground radar device (12) attached to the vehicle (10), - Reading (S4) of the recorded ground radar program (22), - Determination (S6) of a spatial position (9) of the vehicle (10) based on a correlation (S5) of the recorded and read-in ground radargram (22) with the predetermined and read-in ground radar map (20), and - Determination (S7) of a driving trajectory (11) based on the determined spatial position (9) of the vehicle (10) and as a function of the semantic map (30) that has been read. Procedure according to Claim 1 , with the further step of reading in (S2) a predetermined semantic soil map (40), the step of capturing (S3) from the ground radargram (22) being carried out as a function of the predetermined and imported semantic soil map (40). Procedure according to Claim 1 or 2 wherein the at least one traffic system information item (32) has at least one item of information about a traffic infrastructure object, information about a means of transport, information about a traffic sign and information about a traffic management system. Method according to one of the preceding claims, wherein the at least one traffic system information item (32) has a predefined travel trajectory for the vehicle (10) at a traffic junction. Method according to one of the preceding claims, wherein the at least one traffic system information item (32) has a predefined driving dynamics profile for the vehicle (10). Method according to one of the preceding claims, wherein the at least one traffic system information item (32) has a predefined driving operation setting for the vehicle (10). Method according to one of the preceding claims, with the further step of merging the established driving trajectory (11) with a further driving trajectory which has been derived based on a SLAM algorithm for navigating the vehicle (10). Method according to one of the preceding claims, with the further step of detecting (S0) a position (8) of the vehicle (10) with a position detection sensor (18) attached to the vehicle (10), wherein the correlating (S5) is dependent on the detected position (8) of the vehicle (10) is carried out. Method according to one of the preceding claims, with the further step of communicating the recorded ground radar program (22) with a map information system. Method according to one of the preceding claims, wherein the vehicle (10) is designed as an automatically operated vehicle (10), and wherein the steps of the method for the automated operation of the vehicle (10) are carried out automatically. Control device (100) for navigating a vehicle (10), with an interface for reading in a predetermined ground radar map (20), an interface (102) for reading in a predetermined semantic map (30) which has at least one piece of traffic system information (32), an interface (104) for reading in a ground radargram (22) recorded by a ground penetrating radar device (12) attached to the vehicle (10), a correlation unit (110) for correlating the ground radargram (22) read in with the ground radar map (20) read in, a determination unit (120) for determining a spatial position (9) of the vehicle (10) based on a correlation result (112) of the correlation unit (110), and a determination unit (130) for determining a travel trajectory (11) based on the determined spatial position (9) of the vehicle (10) and as a function of the read-in semantic map (30). Vehicle (10) on which a ground radar device (12) is attached and which has a control device (100) according to Claim 11 having.

Images