METHOD FOR VERIFYING A DIGITAL MAP OF A HIGHER-LEVEL AUTOMATED VEHICLE, CORRESPONDING DEVICE AND COMPUTER PROGRAM

Abstract

A method is described for verifying a digital map of a higher-level automated vehicle (HAV), in particular a highly automated vehicle, including the steps: S1 providing a digital map, preferably a highly accurate digital map: S2 determining an instantaneous reference position and localizing the reference position in the digital map; S3 establishing at least one actual feature property of a feature in the surroundings of the reference position, the establishment being carried out with the aid of at least one information source; S4 comparing the actual feature property to a setpoint feature property of the feature and ascertaining at least one difference value as the result of the comparison. A corresponding device and a computer program are also described.

Description

FIELD

The present invention relates to a method for verifying a digital map in a higher-level automated vehicle (HAV), in particular a highly automated vehicle, and a device for this purpose.

BACKGROUND INFORMATION

With regard to an increase in the degree of automation of vehicles, more and more complex driver assistance systems are being utilized. For such driver assistance systems and functions, such as highly automated driving or fully automated driving, for example, a large number of sensors is required in the vehicle that allow for an exact detection of the vehicle surroundings.

In the following, higher-level automated is to be understood to mean all those degrees of automation that correspond to an automated longitudinal and transverse guidance at an increasing system responsibility in the sense of the Federal Highway Research Institute (Bundesanstalt für Straβenwesen, BASt), for example highly and fully automated driving.

In the related art, a plurality of options is provided to carry out a method for operating a highly automated vehicle (HAV). To enhance the localization of a highly automated vehicle (HAV) in a digital map, it is necessary to be able to guarantee the accuracy of the digital map, the problem arising in this case that road changes on short notice caused by construction sites, accidents, or other types of circumstances cannot be taken into account in the digital maps or may be taken into account only incompletely or are on short notice in such a way that a higher-level automated vehicle (HAV), in particular a highly automated vehicle, is not able to retrieve these changes on short notice sufficiently quickly and the driver must take over control of the vehicle. This may be undesirable and potentially also critical with regard to the traffic safety.

In order to control the vehicle in a higher-level automated manner in preferably all situations, it is necessary to have a largely error-free and real-time digital map.

A further relevant aspect arises from the situation in which traditional mapping methods, for example, with the aid of motor vehicles, aircraft, or satellites, are highly cost-intensive. Therefore, it is desirable to make a preferably precise assessment as to whether a re-mapping of a map section of a digital map is necessary or not. The basis for such an assessment is always a conclusion regarding the accuracy of the digital map.

With respect to the significance of assessments of how up-to-date a digital map is, based on sensor data, it is to be further noted that any type of sensor utilized for verifying a map is subject to certain specific limitations. For example, a camera is limited in that it may only capture objects, which are not concealed by other objects or are not made unrecognizable due to light effects at the point in time of the image capture. Such effects may be summarized, for example, under the concept of the robustness of a detection with the aid of a sensor.

An object of the present invention, therefore, is to provide an improved method for verifying a digital map of a higher-level automated vehicle (HAV), in particular a highly automated vehicle, and an improved device for this purpose, with the aid of which the up-to-dateness of a digital map may be reliably established and which allows for an accurate assessment as to whether a re-mapping of a map section is necessary or not, where effects resulting from sensor-specific limitations are to be largely avoided.

SUMMARY

The object may be achieved in accordance with example embodiments of the present invention. Advantageous embodiments of the present invention are described herein.

According to one aspect of the present invention, an example method is provided for verifying a digital map in a higher-level automated vehicle (HAV), in particular a highly automated vehicle, including the following steps:

• S1 providing a digital map, preferably a highly accurate digital map;
• S2 determining an instantaneous reference position and localizing the reference position in the digital map;
• S3 establishing at least one actual feature property of a feature in the surroundings of the reference position, the establishment being carried out with the aid of at least one information source; and
• S4 comparing the actual feature property to a setpoint feature property of the feature and ascertaining at least one difference value as the result of the comparison.

Preferably, the method according to the present invention includes, in a further step S5, verifying the digital map at least partially on the basis of the difference value, the digital map being classified as not up-to-date if the difference value reaches or exceeds an established threshold value of a deviation and being classified as up-to-date if the difference value remains below the established threshold value of the deviation.

In step 6, a plurality of information sources are utilized, the information sources encompassing at least one or multiple information sources from the following group of information sources:

• • car-to-infrastructure systems (C2I), which transmit data to vehicles or gather data from vehicles (for example, via WLAN, LTE);
  • car-to-car systems (C2C), which transmit data to other vehicles, preferably via a wireless communication network (WLAN) or LTE;
  • navigation systems, which have stored road courses, uphill grades, roadway lanes, and infrastructure information in the map material;
  • databases in the Internet, in which preferably road data are stored;
  • databases in the vehicles' own systems, in which data may be stored for the long term;
  • vehicle systems, preferably a head unit, which may access the Internet and may ascertain up-to-date data from databases;
  • highly accurate maps for highly automated or fully automated driving, in which data for localization tasks are stored, preferably objects including positions and dimensions; and
  • driver assistance systems, encompassing one or multiple systems from the following sub-groups:
    • lane-keeping systems (Lane-Keeping Support, LKS), which ascertain a lane preferably with the aid of camera-based systems and steer a vehicle back into the lane if the lane has been unintentionally exited;
    • road-sign systems, which ascertain a specified speed on the basis of visual systems, preferably on the basis of video cameras;
    • object-detection systems, which preferably utilize visual sensors, particularly preferably video cameras; and
    • smartphones, in particular cameras of smartphones.

The aforementioned information sources are nearly omnipresent in modern societies. Reference is made, by way of example, to the plurality of higher-level or highly automated vehicles on the roadways every day, the cameras installed at important infrastructure nodes, or the images captured worldwide by users with the aid of smartphone cameras, which are uploaded at central points of the network, for example, in a cloud. In one specific embodiment of the present invention, it is therefore provided to utilize this abundance of information, which was previously never available, for verifying a digital map and to tap several of these information sources for the verification of the digital map.

In one preferred specific embodiment, the pieces of information delivered by the plurality of information sources are filtered and combined with the aid of suitable algorithms, in order to establish the actual feature property.

Moreover, the method encompasses step S6, in which a transmission of a piece of information regarding the vehicle position and the difference value to a central server takes place.

Advantageously, in the case in which the map was classified, in step S5, as being not up-to-date, at least one of the following actions is carried out:

• • request a central server to update the digital map;
  • carry out steps S3 and S4 again; and
  • request the dispatch of a mapping vehicle, in particular a motor vehicle and/or an aircraft for mapping the surroundings of the reference position.

For the further approach, it is advantageously provided in one specific embodiment of the present invention that the at least one setpoint feature property of the at least one feature is stored in the digital map, preferably a plurality of setpoint feature properties, in turn, of a plurality of features being stored in the digital map, and, in step S3, the establishment of the at least one actual feature property takes place at least partially on the basis of the at least one setpoint feature property.

For the further approach, it is advantageously provided in one specific embodiment of the present invention that step S3 of providing the setpoint feature property of at least one feature includes selecting at least one of a plurality of possible features, the selection being carried out under consideration of the following steps.

In a step S3a, a feature model is created, the feature model describing the available information sources with the aid of which, and under which conditions, in particular at which observation angle and/or at which distance, a feature may be observed.

In a step S3b, a sensor model is created, the sensor model describing which part of the map the particular available information sources may instantaneously perceive with which specification, in particular with which resolution and/or with which noise characteristics.

Moreover, in a step S3c, a surroundings model is created, the surroundings model describing whether a feature is instantaneously detectable or concealed by static or dynamic objects with respect to the information sources; information regarding static objects being withdrawn from the digital map, while dynamic objects are ascertained from processed pieces of information of the at least one information source.

In one further specific embodiment of the present invention, the method includes the step that, from the feature model, the sensor model, and the surroundings model, an expectations hypothesis is created for a selected feature and verified in step S4.

Advantageously, the feature is a road marking, a reflector post, a guardrail, a light signal system, a road sign, a trafficable space, a traffic density, a 3D world model, and/or a speed profile.

Advantageously, the setpoint feature property as well as the actual feature property are both at least one property of the following properties of the feature: geographical position, dimensions, color, relative position with respect to an information source.

A further subject matter of the present invention is an example device for verifying a digital map of a higher-level automated vehicle (HAV), in particular a highly automated vehicle. The device encompasses at least one information source for detecting an actual feature property of a feature in the surroundings of a reference position, a memory module for storing a digital map, preferably a highly accurate digital map, the memory module being, in particular, a central server, and a control device, which is configured for exchanging data with the memory module, of the at least one information source. According to the present invention, it is provided that the control device is configured for carrying out an example method.

A computer program is also a subject matter of the present invention including program code for carrying out the method when the computer program is run on a computer.

Although the present invention is described below mainly in conjunction with passenger cars, it is not limited thereto, but rather may be utilized with any type of vehicle, truck, and/or passenger car.

Further features, possible applications, and advantages of the present invention result from the following description of exemplary embodiments of the present invention, which are represented in the figures. It should be noted that the represented features merely have a descriptive character and may also be used in combination with features of other above-described refinements and are not intended to restrict the present invention in any way.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention is explained in greater detail below with reference to a preferred exemplary embodiment, identical reference numerals being used for identical features. The figures are schematic.

FIG. 1 shows a flow chart of a first specific embodiment of the method according to the present invention; and

FIG. 2 shows an example for applying the method according to the present invention.

DETAILED DESCRIPTION OF EXAMPLE EMBODIMENTS

A flow chart of a first specific embodiment of a method according to the present invention is represented in FIG. 1. In step S1 of FIG. 1, a digital map, preferably a highly accurate digital map, is made available, which may take place, on the device side, in a memory module for storing the digital map, the memory module being, in particular, a memory module integrated into the HAV or a central server.

Step S2 includes determining an instantaneous reference position and localizing the reference position in the digital map, as is conventional in the related art. This takes place on the device side, according to the present invention, with the aid of a position module, the position module preferably being a GPS module (Global Positioning System).

The step designated as S3 in FIG. 1 encompasses establishing at least one actual feature property of a feature in the surroundings of the reference position, the establishment being carried out with the aid of at least one information source 12, 12′, 12″, 12″′.

An exemplary application of the method according to the present invention is represented in FIG. 2; for example, the reference position is the vehicle position of a mapping vehicle 10 represented in FIG. 2, the establishment of the reference position being carried out with the aid of at least one information source 12, 12′, 12″, 12″′.

Moreover, in step S3, it is provided that a plurality of information sources 12, 12′, 12″, 12″′ (FIG. 2) is utilized, the information sources 12, 12′, 12″, 12″′ encompassing at least one or multiple information source(s) from the following group of information sources:

• • car-to-infrastructure systems (C2I), which transmit data to vehicles or gather data from vehicles (for example, via WLAN, LTE);
  • car-to-car systems (C2C), which transmit data to other vehicles, preferably via a wireless communication network (WLAN) or LTE;
  • navigation systems, which have stored road courses, uphill grades, roadway lanes, and infrastructure information in the map material;
  • databases in the Internet, in which preferably road data are stored;
  • databases in the vehicles' own systems, in which data may be stored for the long term;
  • vehicle systems, preferably a head unit, which may access the Internet and may ascertain up-to-date data from databases;
  • highly accurate maps for highly automated or fully automated driving, in which data for localization tasks are stored, preferably objects including positions and dimensions; and
  • driver assistance systems, encompassing one or multiple systems from the following sub-groups:
    • lane-keeping systems (Lane-Keeping Support, LKS), which ascertain a lane preferably with the aid of camera-based systems and steer a vehicle back into the lane if the lane has been unintentionally exited;
    • road-sign systems, which ascertain a specified speed on the basis of visual systems, preferably on the basis of video cameras;
    • object-detection systems, which preferably utilize visual sensors, particularly preferably video cameras; and
    • smartphones, in particular cameras of smartphones.

Preferably, the establishment of the at least one actual feature property takes place at least partially on the basis of at least one setpoint feature property.

The feature may be, in principle, a road marking, a reflector post, a guardrail, a light signal system, a road sign, a trafficable space, a traffic density, a 3D world model, and/or a speed profile. A roadway marking 30 is represented in FIG. 2 by way of example. The setpoint feature property as well as the actual feature property may both be at least one property of the following properties of the feature: geographical position, dimensions, color, relative position with respect to an information source.

It is advantageous if the at least one setpoint feature property of the at least one feature is stored in the digital map, preferably a plurality of setpoint feature properties, in turn, of a plurality of features being stored in the digital map.

The method step of the present invention designated as step S4 encompasses comparing the actual feature property to a setpoint feature property of the feature and ascertaining at least one difference value as the result of the comparison.

In a step S5, the verification of the digital map may take place at least partially on the basis of the difference value, the digital map being classified as not up-to-date if the difference value reaches or exceeds an established threshold value of a deviation and being classified as up-to-date if the difference value remains below the established threshold value of the deviation.

On the part of the plurality of information sources, the delivered pieces of information are filtered and combined with the aid of suitable algorithms, in order to establish the actual feature property.

Furthermore, it is provided that the method also encompasses step S6, in which a piece of information regarding the vehicle position and the difference value is transmitted to a central server.

In the case in which the map was classified, in step S5, as being not up-to-date, either a central server is requested to update the digital map and/or steps S3 and S4 are carried out again and/or a request is carried out to deploy a mapping vehicle, in particular a motor vehicle and/or an aircraft for mapping the surroundings of the reference position.

Moreover, it is provided that the at least one setpoint feature property of the at least one feature is stored in the digital map, a plurality of setpoint feature properties being stored, in turn, as a plurality of features in the digital map. Moreover, the establishment of the at least one actual feature property takes place in step S3 at least partially on the basis of at least one setpoint feature property.

In an advantageous embodiment of the present invention, step S3 of providing the setpoint feature property of at least one feature encompasses selecting at least one of a plurality of possible features, the selection being carried out under consideration of the following steps:

• S3a creating a feature model, the feature model describing the available sensors with the aid of which, and under which conditions, in particular at which observation angle and/or at which distance, a feature may be observed;
• S3b creating a sensor model, the sensor model describing which part of the map the particular available sensors may instantaneously perceive with which specification, in particular with which resolution and/or with which noise characteristics;
• S3c creating a surroundings model, the surroundings model describing whether a feature is instantaneously detectable or concealed by static or dynamic objects in the surroundings of the HAV, information regarding static objects being withdrawn from the digital map, while dynamic objects are ascertained from processed sensor data of the at least one sensor.

Moreover, it is provided in a specific embodiment of the present invention that the method includes the step that, from the feature model, the sensor model, and the surroundings model, an expectations hypothesis is created for a selected feature and verified in step S4.

As represented in FIG. 2, mapping vehicle 10 is equipped with a position module, preferably a GPS module (Global Positioning System), and travels on a route section instantaneously to be traveled, which encompasses two lanes 101, 102. The establishment of the reference position takes place with the aid of at least one information source 12, 12′, 12″, 12″′. Mapping vehicle 10 may communicate, on the one hand, with the at least one information source 12, 12′, 12″, 12″′ and, on the other hand, with further motor vehicles 20, 22, 24. An advantageous effect of the present invention is, for example, that mapping services may also utilize the data of the various information sources. This allows for more efficient planning of the expensive mapping trips by the mapping service.

In this way, it may be predicted with a high likelihood, even before the beginning of the detection of the at least one actual feature property, whether a detection is to be expected and which type of detection it is.

In one specific embodiment of the present invention, it is provided that a plurality of sensors is utilized for detecting the at least one feature property. Accordingly, in this specific embodiment, step S6 encompasses verifying the digital map by carrying out a fusion of the detection results of the sensors involved in the detection.

The present invention is not limited to the described and represented exemplary embodiment. Rather, it also encompasses all expert refinements that would be understood by those skilled in the art within the scope of the present invention.

The present invention is not limited to the described and represented exemplary embodiment. Rather, it also encompasses all refinements that would be understood by those skilled in the art within the scope of the present invention.

In addition to the described and illustrated specific embodiments, further specific embodiments are possible, which may include further modifications and combinations of features.

Claims

1-13. (canceled)

14. A method for verifying a digital map of a higher-level automated vehicle, comprising the following steps: S1) providing a digital map;S2) determining an instantaneous reference position and localizing the reference position in the digital map;S3) establishing at least one actual feature property of a feature in surroundings of the reference position, the establishing being carried out using at least one information source; andS4) comparing the actual feature property to a setpoint feature property of the feature and ascertaining at least one difference value as a result of the comparison.

15. The method as recited in claim 14, wherein the higher-level automated vehicle is a highly automated vehicle

16. The method as recited in claim 14, wherein the digital map is a highly accurate digital map.

17. The method as recited in claim 14, wherein the method further includes the following step: S5) verifying the digital map at least partially on the basis of the difference value, the digital map being classified as not up-to-date when the difference value reaches or exceeds an established threshold value of a deviation and being classified as up-to-date when the difference value remains below the established threshold value of the deviation.

18. The method as recited in claim 14, wherein, in the step S3, a plurality of information sources is utilized, the information sources encompassing at least one or multiple information sources from the following group of information sources: car-to-infrastructure systems which transmit data to vehicles or gather data from vehicles; and/orcar-to-car systems which transmit data to other vehicles via a wireless communication network; and/ornavigation systems, which have stored road courses, uphill grades, roadway lanes, and infrastructure information in map material; and/ordatabases in the Internet in which road data are stored; and/ordatabases in vehicles' own systems, in which data may be stored for the long term; and/orvehicle systems which access the Internet and ascertain instantaneous data from databases; and/orhighly accurate maps for highly automated or fully automated driving, in which data for localization tasks are stored including objects including positions and dimensions; and/ordriver assistance systems, encompassing one or multiple system(s) from the following sub-groups: lane-keeping systems which ascertain a lane using camera-based systems and steer a vehicle back into the lane when the lane has been unintentionally exited, and/orroad-sign systems, which ascertain a specified speed on the basis of visual systems including video cameras; and/orobject-detection systems, which utilize visual sensors including video cameras; and/orsmartphones, including cameras of the smartphones.

19. The method as recited in claim 18, wherein the pieces of information delivered by way of the plurality of information sources are filtered and combined with the aid of algorithms, in order to establish the actual feature property.

20. The method as recited in claim 14, wherein the method further comprises the following step: S6) transmitting a piece of information regarding the vehicle position and the difference value to a central server.

21. The method as recited in claim 17, wherein, in the case in which the map was classified in step S5), as being not up-to-date, at least one of the following actions is carried out: request a central server to update the digital map; and/orcarry out the steps S3 and S4 again; and/orrequest a dispatch of a mapping vehicle for mapping the surroundings of the reference position.

22. The method as recited in claim 21, wherein the mapping vehicle is a motor vehicle and/or an aircraft

23. The method as recited in claim 14, wherein the at least one setpoint feature property of the at least one feature is stored in the digital map, and, in step S3, the establishment of the at least one actual feature property takes place at least partially based on the at least one setpoint feature property.

24. The method as recited in claim 14, further comprising the following step: providing the setpoint feature property of at least one feature, the providing including selecting at least one of a plurality of possible features, the selection being carried out under consideration of the following steps: creating a feature model, the feature model describing available information sources with the aid of which, and under which conditions, and/or which observation angle and/or at which distance, a feature may be observed;creating a sensor model, the sensor model describing which part of the map the available information sources may instantaneously discern with which specification, with which resolution and/or with which noise characteristics;creating a surroundings model, the surroundings model describing whether a feature is instantaneously detectable or concealed by static or dynamic objects with respect to the information sources, information regarding static objects being withdrawn from the digital map, while dynamic objects are ascertained from processed pieces of information of the at least one information source.

25. The method as recited in claim 24, further comprising the following step: creating and verifying, from the feature model, the sensor model, and the surroundings model, an expectations hypothesis for a selected feature.

26. The method as recited in claim 14, wherein the feature is a road marking, and/or a reflector post, and/or a guardrail, and/or a light signal system, and/or a road sign, and/or a trafficable space, and/or a traffic density, and/or a 3D world model and/or a speed profile.

27. The method as recited in claim 14, wherein the setpoint feature property and the actual feature property are both at least one property of the following properties of the feature: geographical position, and/or dimensions, and/or color, and/or relative position with respect to an information source.

28. A device for verifying a digital map of a higher-level automated vehicle (HAV), comprising: at least one information source configured to detect an actual feature property of a feature in the surroundings of a reference position;a memory module storing a digital map, the memory module being a central server; anda control device configured to exchange data with the memory module of the at least one information source, the control device being configured to: determine an instantaneous reference position and localize the reference position in the digital map;establish at least one actual feature property of a feature in surroundings of the reference position, the establishing being carried out using the at least one information source; andcompare the actual feature property to a setpoint feature property of the feature and ascertain at least one difference value as a result of the comparison.

29. A non-transitory memory medium on which is stored a computer program including program code for verifying a digital map of a higher-level automated vehicle, the computer program, when executed by a computer, causing the computer to perform the following steps: S1) providing a digital map;S2) determining an instantaneous reference position and localizing the reference position in the digital map;S3) establishing at least one actual feature property of a feature in surroundings of the reference position, the establishing being carried out using at least one information source; andS4) comparing the actual feature property to a setpoint feature property of the feature and ascertaining at least one difference value as a result of the comparison.