DE102016207181B4 - Method and system for determining a position of a vehicle on a roadway - Google Patents

Abstract

Method for determining a position of a vehicle (1) on a roadway (4), wherein
a reference vehicle (2) is guided on the roadway (4), reference vehicle sensors (21) determining a reference position (x ₁ , x ₂ , x ₃ , x ₄ ) and a reference height profile (20a, 20b, 20c , 20d, 20e) a local environment of the reference vehicle (2) are detected, wherein
the reference height profile (20a, 20b, 20c, 20d, 20e) includes local unevenness in the local environment of the reference vehicle (2),
reference map data are generated based on the reference position (x ₁ , x ₂ , x ₃ , x ₄ ) and the reference height profile (20a, 20b, 20c, 20d, 20e) and transmitted to the vehicle (1),
the vehicle (1) is guided on the roadway (4), an actual height profile (10e) of a local area surrounding the vehicle (1) being recorded by vehicle sensors (11),
an actual position of the vehicle (1) is determined by comparing the actual height profile (10e) with the reference map data,
characterized in that
the reference height profile (20a, 20b, 20c, 20d, 20e) and the actual height profile (10e) are recorded along a transverse extension (y) of the roadway (4) and
the actual position of the vehicle (1) relative to the transverse extent (y) of the roadway (4) is determined.

Description

The invention relates to a method and a system for determining a position of a vehicle on a roadway.

For the increasing automation of driving functions and extended comfort functions in modern vehicles, precise knowledge of one's own vehicle position in relation to the environment or a high-precision map is required. For example, navigation systems can be significantly supported by information about which lane of a roadway a vehicle is in. The vehicle should have its own lane-precise localization in real time, which should be as accurate as possible.

The known methods for this localization are often susceptible to interference, for example due to poor conditions for a satellite-based positioning system or interference from individual sensors.

the DE 10 2010 055 371 A1 proposes a method for determining a position of a vehicle, in which three-dimensional information from the surroundings of the vehicle is recorded and compared with a three-dimensional geographic map. In particular, the course of the roadway is recorded.

the DE 10 2013 016 974 A1 describes a method for detecting a road height profile ahead of a vehicle. Data from the vehicle's sensors is supplemented by further data about the altitude profile ahead, which is retrieved from a storage unit. This allows, for example, the anticipatory adjustment of an active chassis to the condition of the road. In addition, localization can be carried out by comparing the recorded height profile with stored data that is linked to position data. The stored profiles can be updated based on the measured data.

It is the object of the present invention to provide a method and a system for determining the position of a vehicle on a roadway, which allow a particularly high resolution and are also robust against disturbances.

According to the invention, this object is achieved by a method having the features of claim 1 and a system having the features of claim 14. Advantageous refinements and developments result from the dependent claims.

In the method according to the invention, a reference vehicle is guided on the roadway, reference vehicle sensors detecting a reference position and a reference height profile of a local area surrounding the reference vehicle. The reference height profile includes local unevenness in the local environment of the reference vehicle.

Based on the reference position and the reference height profile, reference map data is generated and transmitted to the vehicle. The vehicle is guided on the roadway, with an actual height profile of a local area surrounding the vehicle being recorded by vehicle sensors. An actual position of the vehicle is then determined by comparing the actual height profile with the reference map data.

As a result, properties of the local surroundings of the vehicle are advantageously detected using a height profile and used to determine the position of the vehicle. In particular, the waviness and comparable properties of the roadway can be used. Since the presence of an imperfect, ie not completely flat, roadway can typically be assumed, its properties also allow the vehicle to be positioned in surroundings with few features, so that particularly good coverage for determining the position can be achieved.

According to the invention, “local bumps” in the local environment designate variations in the height of the road surface that do not relate to the global course of the road relative to a geographical space, but are important for driving in the area on the road itself. The local environment of the reference vehicle essentially corresponds to the local environment of the vehicle when the actual position is a short distance from the reference position, in particular when the reference vehicle and the vehicle are essentially in the same position along the longitudinal extent of the roadway . In this case, the local environment can be determined by the detection ranges of the sensors for detecting the height profiles.

The local bumps affect in particular the local surroundings of the vehicle, i.e. their dimensions are in the same order of magnitude as the dimensions of the vehicle itself small area on a geographical scale and are typically not covered by map data, for example for a navigation system.

In particular, deviations from a completely flat height profile of the local surroundings of the vehicle are considered. Such deviations can be caused by wear and tear. For example, the local bumps can be ruts, the width and height of which are typically significantly smaller than the width and height of the vehicle. Furthermore, bumps, tar scars or road damage such as potholes can be detected as local bumps whose dimensions are also smaller than those of the vehicle or which have a small effect on the height of the road surface relative to the dimensions of the vehicle.

In an analogous manner, roadway boundaries, such as curbs, noise barriers, delineators or crash barriers and their local course can be detected as local bumps within the meaning of the invention. These and similar elements of a typical road infrastructure can be detected using the height profile, for example as steep edges, steps or lateral boundaries of the detection area. In this way, the position of the structures within the height profile can also be determined, in particular also relative to the reference position. For example, a determination of the lateral position of the structure relative to the vehicle can be determined. Unevenness can also occur and be detected in these structures. For example, crash barriers typically run parallel to the roadway in a global view, but variations in the course can be detected when viewing a local environment, for example due to an arrangement that is not completely parallel relative to the course of the road. Furthermore, the bumps can be detected, for example at the position of a deformation after a vehicle impact or other damage.

A height profile can be used to record various pieces of information about local bumps, for example their shape, height and position, in particular relative to one another or relative to the vehicle by which the height profile is recorded.

The method according to the invention uses these local unevennesses to determine the position, while with known methods the problem typically has to be solved that the local irregularities disturb the identification of global properties of the route and roadway and have to be compensated for. The invention, on the other hand, makes use of the fact that many local bumps in the roadway, such as bumps, are characteristic of a section of road and typically remain unchanged over longer periods of time. Deviations from an ideal-typical profile of the surface in the area surrounding the vehicle are therefore used to determine the position.

In particular, the invention takes into account that local bumps can occur in the entire local environment of the vehicle, for example on the entire surface of the road, in particular with a characteristic distribution across the width of the road, and can use this for precise position determination on the road.

On the other hand, global properties of the roadway typically relate to its course, ie only to its longitudinal extent, so that only one position in this direction can be determined in a conventional manner.

The method uses reference map data that is generated using data about the course of the roadway and its elevation profile recorded by a reference vehicle. This reference map data is then used by other vehicles that drive on the roadway and can determine their own position in the process. One's own position can be determined on different scales of the environment, for example in the longitudinal direction of the roadway as the location of a road network to which the roadway belongs, or as a lateral position on the roadway.

According to the invention, the reference map data includes information that was obtained using the reference height profile and assigns the respective reference position to this. The reference position can be assigned to a geographic position using data about a road network, but it can also be determined relative to a reference point, for example relative to a specific section of motorway.

When comparing the reference map data with the actual height profile, a correlation is determined in particular, which indicates a quantitative value for the similarity of the recorded values and allows an assessment of which data sets are associated with a specific position on the road. In particular, this can be used to determine which reference map data are relevant for the respective position.

In one embodiment of the invention, the reference height profile and/or the actual height profile also includes an incline of the roadway. As a result, relevant properties of the roadway can be detected in an advantageous manner.

The detection of the reference height profile and the actual height profile can be known per se ter kind, for example by means of a laser scanner, in particular a light detection and ranging (LiDAR) sensor, a radio detection and ranging (RADAR) sensor, an ultrasonic sensor or a camera, in particular a stereo camera. Furthermore, several sensors, also of different types, can be used, for example to achieve greater robustness through redundancy. In particular, sensors can be used that are already provided for other systems of the vehicle, such as distance control sensors.

During the detection, the height profile can be detected in an area of the roadway in front of the reference vehicle, in particular in the direction of a transverse extent of the roadway. By moving the reference vehicle along the length of the roadway, the data of the reference height profile can be recorded over a specific distance. In particular, height profiles that follow one another spatially close together can be put together in order to determine a height profile of the local environment, in particular the roadway. In this way, for example, variations in the course of a lane boundary, such as crash barriers, can be detected with a high level of accuracy.

In a further embodiment, the actual position of the vehicle is also determined using an inertial navigation system and/or a satellite-based position determination system. As a result, a combination of different data about the position of the vehicle can advantageously be used in order to obtain values that are as precise as possible on different scales.

Information of different scales and levels of accuracy can be combined by merging data from different sources, which are recorded in particular using methods known per se. For example, a position of the vehicle can be recorded using a global positioning system (GPS), with an accuracy of several meters being able to be achieved under good conditions. This position can then be specified using the method according to the invention.

In a further development, characteristic features are determined using the reference height profile and the reference map data include the determined characteristic features. This advantageously allows a simple comparison of the height profiles to take place.

Furthermore, the characteristic features can include a lateral delimitation and/or a groove in the roadway. This advantageously allows the detection of particularly relevant data about the roadway and the actual position of the vehicle.

For example, the parameters and positions of individual bumps, ruts, tar scars, road damage or roadway boundaries and possibly detectable traffic signs on the roadway using the height profile can be recorded. Furthermore, curves, folds or furrows in the roadway can be detected. In addition to determining the actual position, the information about characteristic features can also be used for other tasks, such as locating lanes, since ruts are typically to be assumed along these areas of the road.

When determining the characteristic features, the features are parameterized in particular in order to facilitate a comparison of the reference map data with the actual height profile. In particular, the position of the identified characteristic features is determined. For example, the recorded reference height profile can be processed to generate the reference map data in such a way that smaller amounts of data have to be taken into account for comparison with the actual height profile, for example by only taking into account the occurrence and the relative positions of the characteristic features to one another.

In one embodiment, the reference vehicle sensors also record reference vehicle image data of the roadway and the reference map data are also generated using the reference vehicle image data.

This advantageously allows the combination of data about the height profile of the roadway with other data that are not defined as characteristics of the height profile, such as roadway markings. Furthermore, image data, in particular from a stereo camera, can also be used to record the height profile or additional data can contribute to this. For example, objects can be recognized on the roadway whose presence is not permanent, for example objects lying around, and should therefore not be taken into account as characteristic features of the recorded reference height profile. For example, permanent or temporary objects can be distinguished, for example objects on the roadway or weather-related influences, such as leaves, snow or puddles of water. In addition, image data can allow features in the vicinity of the roadway to be recognized and the determination of the actual position can be improved on the basis of these features.

In a further embodiment, vehicle image data of the roadway are also detected by the vehicle sensors and objects on the roadway are recognized using the vehicle image data. In this way, the detection of the actual height profile can advantageously be improved.

Similar to the use of reference vehicle image data described above to improve the acquisition of the reference height profile, the data recorded by the vehicle sensors can also be supplemented by vehicle image data, which, for example, enable the height profile to be cleaned of disruptive objects and weather-related influences on the road .

In a further development, when the actual height profile is compared with the reference map data, a displacement of the actual height profile in relation to the reference height profile is determined. The actual position can thus advantageously be determined in a particularly simple manner as a deviation from the reference position, in particular in the direction of the transverse extent of the roadway.

For example, it can be determined that a characteristic feature, such as a rut, is observed from a greater distance from the vehicle than from the reference vehicle. In this case, the observed distance to other features is also typically smaller. A relative displacement of the positions of the vehicles to one another can be determined by comparing the determined distances and features.

According to the invention, the reference height profile and the actual height profile are recorded along a transverse extent of the roadway and the actual position of the vehicle is determined relative to the transverse extent of the roadway. This advantageously enables lateral self-localization on the roadway.

For example, the position of the vehicle can be determined relative to the width of the roadway, or a position can be determined relative to the reference vehicle or to characteristic features of the roadway. This is particularly important for determining the lane in which the vehicle is traveling, which must be known for safe navigation of the vehicle.

In a further embodiment, the reference map data includes a large number of reference positions and reference height profiles along a longitudinal extent of the roadway, and the actual position of the vehicle is determined relative to the longitudinal extent of the roadway. This advantageously enables the actual position to be determined in the longitudinal direction of the roadway.

In particular, the method can be used in this way to improve known position determination systems that determine a position of the vehicle within a road network without necessarily taking into account the lateral position on the roadway. For example, the height profile of the roadway can be viewed as a signature and used to identify a section of road, since the irregularities typically occur in a statistically distributed manner and, with sufficiently high accuracy and data volume, a position along the longitudinal extent of the roadway can be determined with high reliability and accuracy.

In particular, the reference map data are used both for determining the lateral position and for determining the position in the longitudinal direction.

In a development of the method, the reference height profile and/or the actual height profile is recorded with a resolution of less than 20 cm, preferably less than 10 cm. In particular, this resolution is achieved in the vertical direction, ie perpendicular to the roadway, and/or laterally, ie in a lateral direction. This advantageously enables a sufficiently precise detection of the height profiles. In particular, the resolution must be sufficient to record relevant local irregularities.

In one embodiment, the reference height profile and/or the actual height profile are recorded in such a way that surface properties of the roadway are recorded, such as the roughness of the surface, which is worn down, for example, by driving on it. This can further complete the collected data.

In a further embodiment, the recorded reference height profile and/or actual height profile is also averaged. This advantageously smoothes the acquired data, for example using a moving average, to compensate for the effects of noise and small measurement errors. For example, a pitching movement of the vehicle when driving over the road can lead to deviations in the measurements. The averaging can take place, for example, in the longitudinal and/or transverse direction of the roadway, for example using a smoothing over the width of the height profile and/or over a number of height profiles that were recorded while the vehicle was driving on the roadway.

In a development, the reference map data is transmitted to an external server and stored by the external server. As a result, the reference map data can advantageously be managed and stored centrally, to make them available to the vehicle independently of the reference vehicle.

For example, the reference map data can be generated by devices of the reference vehicle and transmitted to the external server. Furthermore, the reference position and the reference height profile or related data can be transmitted to the external server and the reference map data can be generated by this. In particular, when the reference map data is generated by the external server, further data can be used, such as data recorded from other vehicles or other data that can be recorded from a database, for example.

In a further refinement, new reference map data are also generated on the basis of the actual height profile. This advantageously allows the reference map data to be corrected and/or updated using the data recorded by the vehicle.

For example, the data recorded by the vehicle, such as the actual height profile, can be transmitted to the external server and the reference map data can be updated. The updating can also be done by devices of the vehicle itself. For example, wear and tear on the roadway can be taken into account by continually adapting the reference map data to the actual state of the roadway.

Alternatively or additionally, the reference map data can be transmitted between multiple vehicles directly or via a network of vehicles. Here, too, the reference map data can be updated in a manner similar to that described above, with incremental updates being able to be transmitted between the vehicles, in particular relating to temporary changes in the height profile of the roadway.

Furthermore, a deviation between the reference height profile and the actual height profile can be determined. For example, it can be determined that a good position determination is possible in a specific area of the longitudinal extent of the roadway, while significant deviations in the recorded height profiles are detected in an adjacent area. This can occur, for example, if damage has occurred or the roadway has been renovated since the reference height profile was recorded. In this case, the vehicle can assume the role of the reference vehicle. Based on the data recorded by this vehicle, new reference map data can be generated or information can be generated that the existing reference map data in this area is not reliable and new reference map data should be generated or information as to whether a new reference elevation profile has been recorded in this area shall be.

In particular, it can be checked whether the detected deviation should be used to change the reference map data. For example, it can be determined how many vehicles have recorded deviating actual height profiles, for example in order to rule out a particularly short-term impairment of an individual vehicle. Furthermore, temporary phenomena can be taken into account, such as a mobile construction site, which leads to a change in the roadway compared to the reference map data, but is not intended to be permanent. In such a case, once the temporary change is gone, you can revert to the original reference map data.

Periodic changes in the properties of the road can also be taken into account, for example due to the change from day to night, low and high temperatures or the changing seasons. For this purpose, for example, different reference map data can be provided for changing conditions, without a completely new acquisition being necessary each time.

The system according to the invention for determining a position of a vehicle on a roadway comprises a reference vehicle that can be guided on the roadway, reference vehicle sensors being able to detect a reference position and a reference height profile of a local environment of the reference vehicle, the reference height profile includes a local unevenness of the local environment of the reference vehicle. Reference map data can be generated and transmitted to the vehicle on the basis of the reference position and the reference height profile. The vehicle can be driven on the roadway, with an actual height profile of a local area surrounding the vehicle being able to be detected by vehicle sensors. An actual position of the vehicle can be determined by comparing the actual height profile with the reference map data. The reference height profile and the actual height profile can be detected along a transverse extent of the roadway, and the actual position of the vehicle can be determined relative to the transverse extent of the roadway.

The system according to the invention is designed in particular to implement the method according to the invention described above. The system thus has the same advantages as the method according to the invention.

The reference vehicle sensors and/or the vehicle sensors can include a laser scanner, a radar sensor, an ultrasonic sensor or a camera. As a result, the height profile of the roadway can advantageously be detected in a particularly simple manner using methods known per se.

• 1 zeigt ein Ausführungsbeispiel des erfindungsgemäßen Systems,
• 2 und 3 zeigen das Ausführungsbeispiel des erfindungsgemäßen Systems auf einer Fahrbahn,
• 4 zeigt Referenz-Höhenprofile, wie sie durch ein Ausführungsbeispiel des erfindungsgemäßen Verfahrens erfasst werden, und
• 5 zeigt die Bestimmung einer Verschiebung zwischen einem Referenz-Höhenprofil und einem Ist-Höhenprofil.

The invention will now be explained using exemplary embodiments with reference to the drawings.

• 1 shows an embodiment of the system according to the invention,
• 2 and 3 show the exemplary embodiment of the system according to the invention on a roadway,
• 4 shows reference height profiles as they are detected by an embodiment of the method according to the invention, and
• 5 shows the determination of a shift between a reference height profile and an actual height profile.

In reference to 1 an embodiment of the system according to the invention is explained.

A vehicle 1 includes vehicle sensors 11 and a vehicle control unit 12 coupled to the vehicle sensors 11. The vehicle sensors 11 provide data of a vehicle height profile in front of the vehicle 1 in a vehicle detection range 10 in the direction of travel R Vehicle 1 located surface detectable.

In the exemplary embodiment, the vehicle sensors 11 comprise a LiDAR sensor in which a laser beam is swept across the vehicle detection area 10 , emitting pulses which are reflected from the surface in the vehicle detection area 10 . The transit time of the pulses from transmission to reception is measured and the distance is calculated from this. An actual height profile of the surface in the vehicle detection area 10 results from a series of such measured values. Furthermore, the characteristics of the reflected laser beam can provide further information, for example about properties of the surface.

In other exemplary embodiments, other methods and sensors known per se are used for detecting the height profile, such as RADAR or ultrasonic sensors, as an alternative or in addition.

In the exemplary embodiment, the vehicle sensors 11 also include a camera, by means of which vehicle image data is captured in the area in front of the vehicle 1 in the direction of travel R, which includes the vehicle detection area 10 in particular.

In the example shown, a reference vehicle 2 comprises reference vehicle sensors 21 of basically the same type as the vehicle sensors 11 and a reference vehicle control unit 22 coupled to them. The reference vehicle sensors 21 generate a reference height profile in a reference vehicle detection area 20 recorded in the direction of travel R in front of the reference vehicle 2. In the example, reference vehicle 2 also includes a camera that captures reference vehicle image data that includes reference vehicle capture area 20 .

The resolving power of the vehicle sensors 11 and the reference vehicle sensors 21 is selected in particular in such a way that the actual height profile 10e or the reference height profile can be detected with sufficient accuracy, as will be explained later with reference to FIG 5 is explained. In particular, the resolution in the transverse direction of the roadway 4 is less than 1 m, preferably less than 50 cm, particularly preferably less than 20 cm. This is the resolution relating to the distance between the measuring points and the vehicle sensors 11. The resolution in a direction perpendicular to the road surface, i.e. relating to the resolution of the height, is selected in such a way that typical values of the inclination of the roadway 4 as well as bumps, ruts and other characteristic features are recognizable. In particular, it is less than 20 cm, preferably less than 10 cm.

The vehicle control unit 12 and the reference vehicle control unit 22 are at least temporarily connected in terms of data to an external server 3 . This technical data connection can be established in various ways known per se, in particular by connecting to the Internet via a cellular network. Using the data connection, data can be exchanged between the external server 3 and the vehicle 1 or the reference vehicle 2 . Furthermore, the external server 3 can exchange data between the vehicle 1 and the reference vehicle 2 . Furthermore, the external server 3 can store data and data stored there can be retrieved by the vehicle 1 and the reference vehicle 2 . In a further exemplary embodiment, direct data exchange between vehicle 1 and reference vehicle 2 is provided as an alternative or in addition.

Related to the 2 , 3 and 4 an embodiment of the method according to the invention is explained. In doing so, from the above explained embodiment of the system according to the invention assumed.

The reference vehicle 2 and the vehicle 1 drive on the roadway 4 in the same direction of travel R. Reference positions x ₁ , x ₂ , x ₃ , x ₄ are indicated by dashed lines at successive positions along the roadway 4 , it being pointed out that that the spatial relationships and size relationships indicated in the exemplary representation are to be understood as purely exemplary. In the case shown, they are on roadway 4 at the same time, but they can also drive on it at different times. In any case, however, reference vehicle 2 drives on the lane in front of vehicle 1.

At the reference positions x _{1 ,} x ₂ , x ₃ , x ₄ , the reference vehicle 2 records reference height profiles 20a, 20b, 20c, 20d, 20e in the reference vehicle detection area 20. Four such reference height profiles 20a, 20b, 20c, 20d are in 4 shown. In the exemplary embodiment, the reference positions x ₁ , x ₂ , x ₃ , x ₄ are recorded in a manner known per se, for example by means of a GPS module in combination with an inertial navigation system. Based on the reference positions x ₁ , x ₂ , x ₃ , x ₄ and reference height profiles 20a, 20b, 20c, 20d, reference map data are generated which, in the example, contain the recorded data of the reference height profiles 20a, 20b, 20c, 20d, 20e with the associated reference positions x ₁ , x ₂ , x ₃ , x ₄ .

In further exemplary embodiments, the reference height profiles 20a, 20b, 20c, 20d, 20e are already pre-processed in this step, with local bumps in the roadway 4 being used to identify characteristic features and their positions relative to the transverse extent y of the roadway 4 being determined. For example, crash barriers and curbs, but also ruts, tar scars, bumps and other bumps and global properties of the surface of the roadway 4, such as an incline of the roadway 4, are determined.

In this case, the reference map data include the characteristic features and their lateral positions as an alternative or in addition to the complete reference height profiles 20a, 20b, 20c, 20d, 20e, on the basis of which these were determined.

In a similar way, the vehicle sensors 11 in the vehicle detection area 10 detect an actual height profile 10e. In this case, characteristic features of the actual height profile 10e can also be identified, analogously to the method described above for the reference vehicle 2 . In the exemplary embodiment, the vehicle sensors 11 also include a GPS module for satellite-based position determination. Using the data from the GPS module, the position is determined with an accuracy of a few meters, in particular in the direction of the longitudinal extension x of the roadway 4, and reference map data is retrieved from the external server 3, the reference height profiles 20a, 20b, 20c relevant to this area , 20d, 20e or the data on the characteristic features of the relevant reference height profiles 20a, 20b, 20c, 20d, 20e. Alternatively or additionally, an inertial navigation system is also used, in which the position of the vehicle 1 is determined based on its movement relative to a starting point.

The generated reference map data is transmitted to the vehicle 1, the transmission taking place via the external server 3, as already mentioned above with reference to FIG 1 described. The external server 3 stores the reference map data and makes them available to the vehicle 1, in particular when requested.

The exemplary embodiment also provides for the reference map data stored on the server 3 to be updated using the data recorded by the vehicle sensors 11 . For this purpose, it is determined whether and how there are deviations between the reference map data and the recorded data. The deviations can be used to determine whether the reference map data should be updated, for example to take into account the progressive wear of the roadway 4 or the changed elevation profile after a section of the roadway 4 has been renovated.

The vehicle control unit 12 compares the reference map data with the detected actual height profile 10e and an offset (offsef) Δy is determined, which indicates the difference between the lateral position of the reference vehicle 2 and the vehicle 1. The determination of the offset Δy is explained in more detail below. The offset Δy thus results in an exact localization of the vehicle in the lateral direction y of the lane 4.

As described above, image data is also captured in the illustrated embodiment. These are evaluated using image processing methods known per se and it is determined, for example, whether the height profile of roadway 4 recorded in each case is falsified by objects lying on roadway 4, particularly if these objects are only present temporarily. This can be the case with objects such as gravel, leaves or rubbish. Furthermore, weather-related influences can also be recognized and taken into account, such as snow on the road or puddles of water, which can lead to a change in the recorded height profile of the road surface. If an update of the reference map data is provided, this can also take place depending on the image data, for example if it is recognized whether a change in the actual height profile 10e compared to the reference height profile 20a, 20b, 20c, 20d, 20e recorded at an earlier point in time are permanent or temporary in nature.

In a further exemplary embodiment, the position of a lane boundary, in particular a crash barrier, of the lane 4 is detected and, using height profiles that are detected one after the other, it is determined how the lane boundary extends relative to the lane 4 . In particular, variations in the distance between a crash barrier and the roadway 4 are recorded, as they typically arise for practical reasons in road construction. The course of the distance can be detected using a plurality of detected reference height profiles 20a, 20b, 20c, 20d, 20e and used as a characteristic feature.

In further exemplary embodiments, the vehicle 1 is localized along the longitudinal extension x of the roadway 4. For this purpose, the recorded actual height profile 10a is compared with reference height profiles 20a, 20b, 20c, 20d, 20e recorded along the longitudinal extension x and a reference height profile 20a , 20b, 20c, 20d, 20e is determined. The reference position x ₁ , x ₂ , x ₃ , x ₄ assigned to this reference height profile 20a, 20b, 20c, 20d, 20e is determined as the current actual position. In particular, the potential offset Δy of the height profiles relative to one another is taken into account.

In further exemplary embodiments, to determine an actual position relative to the longitudinal extension x of the roadway 4, several actual height profiles 10e recorded at different, in particular consecutive, actual positions are combined and assigned to different reference positions x ₁ , x ₂ , x ₃ , x ₄ Reference map data compared. This can, for example, allow the actual position to be determined particularly reliably, since the data from a number of height profiles are taken into account here.

In further exemplary embodiments, the detected reference height profiles 20a, 20b, 20c, 20d, 20e and/or actual height profiles 10e are smoothed or averaged, with a moving average being formed along the direction of the width extension y and/or the longitudinal extension x of the roadway 4, for example will. In this way, noise during detection or smaller deviations, such as those caused by objects on the road, can be suppressed.

In reference to 5 the determination of a shift between a reference height profile and an actual height profile is explained. The starting point here is the system and method according to the invention explained above.

A reference height profile 20e, which was recorded by the reference vehicle 2, and an actual height profile 10e, which was recorded by the vehicle 1, are shown. The reference vehicle 2 and the vehicle 1 do not travel exactly the same trajectory on the roadway 4, but in particular their lateral positions relative to the transverse direction y of the roadway 4 differ. The height profiles, which are recorded in the direction of the transverse extent y, are therefore shifted relative to one another from the perspective of the reference vehicle 2 and the vehicle 1 .

The offset of the height profiles 10e, 20e to each other is in 5 referred to as Δy and can be determined, for example, using characteristic features 5a, 5b, 5c, which are detected both in the reference height profile 20e and in the actual height profile 10e. This is illustrated by arrows 6a, 6b, 6c, which illustrate the shift Δy of the characteristic features 5a, 5b, 5c relative to one another. The lateral position of the vehicle 1 can be determined on the basis of the displacement Δy, with a lateral position being first determined relative to the position of the reference vehicle 2 and using this relative position an absolute lateral position of the vehicle 1 being able to be determined if the absolute lateral position of the reference vehicle 2 is known.

Reference List

1

vehicle

2

reference vehicle

3

External Server

4

roadway

5a, 5b, 5c

Characteristic features

6a, 6b, 6c

arrows

10

Vehicle detection area

10e

Actual Elevation Profile

11

vehicle sensors

12

vehicle control unit

20

Reference vehicle detection area

20a, 20b, 20c, 20d, 20e

Reference elevation profile

21

reference vehicle sensors

22

Reference vehicle control unit

R

driving direction

x

longitudinal extent

y

Transverse extent, lateral direction

x1, x2, x3, x4

reference position

Δy

offset, offset

Claims (14)

Method for determining a position of a vehicle (1) on a roadway (4), a reference vehicle (2) being guided on the roadway (4), reference vehicle sensors (21) determining a reference position (x ₁ , x ₂ , x ₃ , x ₄ ) and a reference height profile (20a, 20b, 20c, 20d, 20e) of a local environment of the reference vehicle (2) are recorded, the reference height profile (20a, 20b, 20c, 20d, 20e) includes local unevenness in the local environment of the reference vehicle (2), generates reference map data based on the reference position (x ₁ , x ₂ , x ₃ , x ₄ ) and the reference height profile (20a, 20b, 20c, 20d, 20e) and are transmitted to the vehicle (1), the vehicle (1) is guided on the roadway (4), an actual height profile (10e) of a local environment of the vehicle (1) being detected by vehicle sensors (11). a comparison of the actual height profile (10e) with the reference map data, an actual position of the vehicle (1) is determined, characterized in that ss the reference height profile (20a, 20b, 20c, 20d, 20e) and the actual height profile (10e) along a transverse extent (y) of the roadway (4) are detected and the actual position of the vehicle (1) relative to the transverse extent (y) of the lane (4) is determined. procedure after claim 1 , characterized in that the reference height profile (20a, 20b, 20c, 20d, 20e) and/or the actual height profile (10e) also includes an incline of the roadway (4). procedure after claim 1 or 2 , characterized in that the actual position of the vehicle (1) is also determined using an inertial navigation system and / or a satellite-based position determination system. Method according to one of the preceding claims, characterized in that characteristic features are determined on the basis of the reference height profile (20a, 20b, 20c, 20d, 20e) and the reference map data include the determined characteristic features. Method according to one of the preceding claims, characterized in that the characteristic features include a lateral boundary and/or a groove in the roadway (4). Method according to one of the preceding claims, characterized in that reference vehicle image data of the roadway (4) are also detected by the reference vehicle sensors (21) and the reference map data are also generated using the reference vehicle image data. Method according to one of the preceding claims, characterized in that vehicle image data of the roadway (4) are also detected by the vehicle sensors (11) and objects on the roadway (4) are recognized using the vehicle image data. Method according to one of the preceding claims, characterized in that when comparing the actual height profile (10e) with the reference map data, a displacement of the actual height profile (10e) compared to the reference height profile (20a, 20b, 20c, 20d, 20e) is determined will. Method according to one of the preceding claims, characterized in that the reference map data contains a large number of reference positions (x ₁ , x ₂ , x ₃ , x ₄ ) and reference height profiles (20a, 20b, 20c, 20d, 20e) along a longitudinal extent (x ) of the roadway (4) and the actual position of the vehicle (1) relative to the longitudinal extent (x) of the roadway (4) is determined. Method according to one of the preceding claims, characterized in that the reference height profile (20a, 20b, 20c, 20d, 20e) and/or the actual height profile (10e) has a resolution of less than 20 cm, particularly preferably less than 10 cm. is detected. Method according to one of the preceding claims, characterized in that the recorded reference height profile (20a, 20b, 20c, 20d, 20e) and/or actual height profile (10e) is also averaged. Method according to one of the preceding claims, characterized in that the reference map data are transmitted to an external server (3) and stored by the external server (3). Method according to one of the preceding claims, characterized in that new reference map data are also generated on the basis of the actual height profile (10e). System for determining a position of a vehicle (1) on a roadway (4) with a reference vehicle (2) that can be guided on the roadway (4), reference vehicle sensors (21) determining a reference position (x ₁ , x ₂ , x ₃ , x ₄ ) and a reference height profile (20a, 20b, 20c, 20d, 20e) of a local environment of the reference vehicle (2) can be detected, the reference height profile (20a, 20b, 20c, 20d, 20e ) includes a local unevenness of the local environment of the reference vehicle (2), based on the reference position (x ₁ , x ₂ , x ₃ , x ₄ ) and the reference height profile (20a, 20b, 20c, 20d, 20e) reference map data can be generated and can be transmitted to the vehicle (1), the vehicle (1) can be driven on the roadway (4), an actual height profile (10e) of a local environment of the vehicle (1) being able to be detected by vehicle sensors (11), an actual position of the vehicle (1) can be determined by comparing the actual height profile (10e) with the reference map data, characterized in that d that the reference height profile (20a, 20b, 20c, 20d, 20e) and the actual height profile (10e) can be detected along a transverse extension (y) of the roadway (4) and the actual position of the vehicle (1) relative to the transverse extension (y) of the roadway (4) can be determined.

Images