DE102019215490A1 - Method for providing a digital localization map for a vehicle - Google Patents

Abstract

Method for providing a digital localization map (400) for a vehicle (100), comprising the steps of: - receiving a request for creating the digital localization map (400) by the vehicle (100); - creating the digital localization map (400) from cyclically provided Environment information in the environment of the vehicle (100); and - transmitting localization features of the digital localization map (400) to the vehicle (100) as a function of the time and location of the requesting vehicle (100).

Description

The present invention relates to a method for providing a digital localization map for a vehicle. The invention also relates to a cloud-based device for providing a digital localization map for a vehicle. The invention also relates to a computer program. The invention also relates to a machine-readable storage medium.

State of the art

Localization systems with feature-based, digital localization maps for determining vehicle position and orientation are known, central system components of automated driving functions.

A large number of approaches for creating digital localization maps from feature measurements are known. Features, for example in the simplest case clusters (i.e. combinations of several features) or edges of objects, are extracted here as measurements of an environment sensor system. The step of the feature extraction required for this is known per se.

Furthermore, model-based approaches are known which allow a minimum required feature density for digital localization maps to be derived, which leads to the required localization accuracy. Also known are approaches for reducing the feature density in order to obtain a representation of the environment in the form of a digital localization map that is as memory-efficient as possible. These approaches lead to a homogeneous reduction in the feature density over the entire digital localization map.

Furthermore, the environment is subject to constant changes, be it e.g. through weathering, accidents, but also through conscious human influences, such as structural measures. In these cases, a change in the environment leads to obsolescence of the digital map modeling the environment. Since it is not always possible to determine such a change directly and / or to update the digital maps of all vehicles immediately, detection of such changes in the environment is very important for future assistance and driving functions.

DE 10 2016 224 329 A1 discloses a method and associated system for locating a vehicle.

Disclosure of the invention

It is an object of the invention to provide an improved method for providing a digital localization map for a vehicle.

• - Empfangen einer Anforderung zum Erstellen der digitalen Lokalisierungskarte durch das Fahrzeug;
• - Erstellen der digitalen Lokalisierungskarte aus zyklisch bereitgestellter Umfeldinformation im Umfeld des Fahrzeugs; und
• - Übermitteln von Lokalisierungsmerkmalen der digitalen Lokalisierungskarte an das Fahrzeug in Abhängigkeit von Zeit und Ort des anfordernden Fahrzeugs.

According to a first aspect, the object is achieved with a method for providing a digital localization map for a vehicle, with the steps:

• - Receipt of a request for the creation of the digital localization map by the vehicle;
• - Creation of the digital localization map from cyclically provided environmental information in the surroundings of the vehicle; and
• - Transmission of localization features of the digital localization map to the vehicle as a function of the time and location of the requesting vehicle.

This enables improved localization with reduced transmission effort, computing load and electrical energy for a cloud-based device. The accuracy of the localization is advantageously improved because the digital localization map is specifically adapted to local phenomena in the environment of the user vehicle. This advantageously creates, provides and uses a digital collision map with a feature density that is adapted as required. When creating the localization map, additional information is used for this purpose that allows conclusions to be drawn about the required feature density. This information includes, for example, an expected or actual traffic density (high traffic density potentially results in higher occlusion rates) or a frequency of changes in the surroundings (results in outdated localization maps).

According to a second aspect, the object is achieved with a method for requesting a digital localization map by a vehicle, the digital localization map being requested as a function of defined local and temporal circumstances of the vehicle, the digital localization map being received by a cloud-based device.

• - eine Kartierungseinrichtung zum Erstellen der digitalen Lokalisierungskarte;
• - eine Anforderungseinrichtung zum Verarbeiten einer Anforderung der digitalen Lokalisierungskarte durch das Fahrzeug; und
• - eine Anpassungseinrichtung zum Anpassen von Lokalisierungsmerkmalen der digitalen Lokalisierungskarte in Abhängigkeit von der Anforderung des Fahrzeugs.

According to a third aspect, the object is achieved with a cloud-based device for providing a digital localization map for a vehicle, comprising:

• - a mapping device for creating the digital localization map;
• a request device for processing a request for the digital localization map by the vehicle; and
• an adaptation device for adapting localization features of the digital localization map as a function of the requirements of the vehicle.

According to a fourth aspect, the object is achieved with a computer program comprising instructions which, when the computer program is executed by a computer, cause the computer to carry out a proposed method.

According to a fifth aspect, the object is achieved with a machine-readable storage medium on which the computer program is stored.

Advantageous further developments of the method are the subject of the dependent claims.

An advantageous further development of the method provides that environmental features detected by sensors are used as environmental information. As a result, the digital localization map can be optimally adapted to the localization requirements of user vehicles.

A further advantageous development of the method provides that a traffic density is computationally determined from the information on the surroundings. In this way, mapping vehicles are advantageously used to create additional information in the form of a traffic density.

Another advantageous development of the method provides that information provided relating to traffic density in the vicinity of the vehicle is used as information about the surroundings. This means that external service providers who offer traffic density information (e.g. Google Maps® and the like) can also be used.

Another advantageous development of the method provides that at least one of the following information is used as information about the surroundings: time of day, time of year, light situation.

Another advantageous development of the method provides that meta-information of at least one localization feature of the digital localization map is used as information about the surroundings. In this way, an additional criterion for the defined selection of the localization feature can advantageously be provided.

A further advantageous development of the method provides that localization features to be transmitted to the vehicle are selected from a digital location map completely filled with location features and transmitted to the vehicle. Advantageously, based on a completely created digital location map, localization features can be adapted and transmitted to the vehicle.

A further advantageous development of the method provides that the environment information is taken into account by appropriately adapted provision of localization data for the cloud-based device. This advantageously supports the fact that less mapping data from the fleet of mapping vehicles has to be loaded to the cloud-based device. Due to the lower volume of data, the mapping and adaptation of the digital localization map can be made simpler, e.g. with regard to storage space, computing power, electrical energy consumption, etc.

• 1 ein Übersichtsdiagramm mit einer Darstellung eines konventionellen Verfahrens einer Verwendung einer digitalen, merkmalbasierten Lokalisierungskarte für ein Fahrzeug;
• 2 ein Blockschaltbild einer Ausführungsform der erfindungsgemäßen cloudbasierten Vorrichtung;
• 3 eine Darstellung zur Erläuterung einer prinzipiellen Wirkungsweise des vorgeschlagenen Verfahrens; und
• 4 einen prinzipiellen Ablauf eines vorgeschlagenen Verfahrens zum Bereitstellen einer digitalen Lokalisierungskarte für ein Fahrzeug.

Further measures that explain and improve the invention are explained in more detail below together with the description of preferred exemplary embodiments of the invention with reference to several figures. Brief description of the figures
In the accompanying figures shows:

• 1 an overview diagram with an illustration of a conventional method of using a digital, feature-based localization map for a vehicle;
• 2 a block diagram of an embodiment of the cloud-based device according to the invention;
• 3 a representation to explain a basic mode of operation of the proposed method; and
• 4th a basic sequence of a proposed method for providing a digital localization map for a vehicle.

Embodiments of the invention

1 shows a basic system diagram of a conventional method of providing a digital, feature-based localization map for a vehicle, in particular for an automated vehicle. For the operation of automated driving functions, there are generally high or strict requirements for functional safety. These high requirements are carried through to the vehicle's localization system. The main input signals from feature-based In addition to current sensor measurements, localization systems are digital, feature-based localization maps. In modern sensor systems there are many examples of sensors that have to meet functional safety requirements (e.g. ASIL according to ISO 26262 regarding clearly defined safety goals).

The dashed division of 1 shows the creation and transmission of the digital localization map in the left area. In the right area is the use of the digital localization map within a user vehicle 100 shown.

Is shown in 1 a basic sequence of a method for providing a digital localization map, wherein in one step 1 recognition of static environmental features (“static perception”) and vehicle movement estimation using one or more mapping vehicles 110 is carried out. In one step 2 the mapping vehicles are uploaded 110 data recorded by sensors in the cloud. In one step 3 For example, a high-performance cloud-based electronic computer device (not shown) is used to create and / or update and / or expand the digital localization map.

In one step 4th a radio-based transmission of the digital localization map to a, preferably automated, user vehicle took place 100 . For example, this can be done for each section of the route, so that highly up-to-date map data from the digital localization map is sent to the user vehicle for subsections of a route 100 be transmitted.

The one in the crotch 4th wirelessly transmitted digital location map 200 is therefore in the user vehicle 100 available in electronic form and can be accessed by the user vehicle 100 be used in a manner known per se for self-localization.

On the user vehicle 100 For this purpose, a per se known localization device (not shown) is implemented, by means of which in one step 5 a sensory detection of static environmental features (not shown) is carried out. The user vehicle records 100 static environmental features (e.g. buildings, traffic signs, infrastructure objects, etc.) by means of at least one active or passive sensor (e.g. radar sensor, lidar sensor, ultrasonic sensor, camera, etc.) and estimates vehicle movement. In one step 6th a highly accurate position of the user vehicle is achieved through cooperation with map data from the feature-based localization map 100 determined.

In one step 7th For this purpose, the map data is merged with the environment data previously recorded by sensors, which in one step 8th a position and an orientation (“vehicle pose”) of the user vehicle 100 be determined. In a subsequent step 9 the determined localization data can be used by a higher-level function of the vehicle (e.g. automated driving function).

It is therefore not provided in the explained conventional method that an adaptation of a feature density of the digital localization map of additional information relating to the local surroundings of the vehicle is carried out.

The present invention proposes an adaptation of the feature density of the digital localization map as a function of specific additional information, such as traffic density, accident frequency, weather situation, time, season, etc.

The invention proposes an improved provision or use of a feature-based localization map for a vehicle.

2 shows a block diagram of an embodiment of a proposed cloud-based (ie arranged in the cloud) device for providing a digital localization map for a vehicle. You can see a fleet of mapping vehicles 110 , the sensor-recorded environment information to a mapping device 310 a cloud-based device 300 that will be used to generate a digital location map. It should be mentioned that the mapping vehicle 110 also as a user vehicle 100 can act and vice versa. Furthermore, the mapping facility 310 from an external database 200 Additional information relating to an environment of a user vehicle 100 provided. The ones from the mapping vehicles 110 and from the database 200 The additional information provided is transmitted cyclically, for example according to a predetermined time grid, in order to thereby ensure a high level of detail of a local image of the surroundings of the user vehicle 100 to enable. The additional information can, for example, provide information on lighting conditions, position of the sun, traffic density, traffic jam situations, etc. in the vicinity of the user vehicle 100 include.

To a request facility 320 can be a requirement of a user vehicle 100 are fed. The request is made to an adaptation facility 330 transmitted, which then depends on the time and location of the requesting user vehicle 100 adapted features of the Location map 200 to the user vehicle 100 transmitted.

In this way, the mapping changes (creation of the digital localization map) and adapts to changing spatial phenomena (e.g. traffic density). The digital localization map and the mapping are accordingly functions of time and place.

For example, the mapping can be dependent on the current traffic density: Specifically, this can mean that more localization features are entered in the digital location map when the traffic density is high, since significantly more features are often hidden by other road users in traffic jams than in normal traffic density. The quality of the localization depends, among other things, on the number and constellation of the visible features, whereby the required accuracy is supported by an adapted mapping.

Mapping can also be dependent on the respective time of day and / or season: At night, a passive sensor (e.g. camera) can only use features in an actively illuminated area. Features in the distance or to the side are not visible and therefore do not need to be mapped. During the day, however, you can make an important contribution and should be available in the digital localization map.

Mapping can also be dependent on the position of the sun (light situation): The quality of the detection of localization features from sensor data can depend heavily on the camera being dazzled by the sun. The position of the sun is known and can therefore be used to map those localization features that are not in the glare area of the sun.

Required information about the traffic density can be obtained from external service providers, for example weather services, traffic services and the like, externally or from mapping vehicles 110 are recorded. For example, the number of other vehicles in the direct vicinity of the mapping vehicle is conceivable 110 on the basis of environment sensors and the mapping function of the mapping device 310 additionally to be made available in order to determine the traffic density in this way.

The invention advantageously enables an optimal selection of the localization features (e.g. in the form of maximum compression of the required information), which leads to an optimized use of the radio-based communication channel and thus to minimal costs for data transmission given the requirements for the availability and accuracy of the localization .

Requirements for the accuracy of the localization can be met in a targeted manner by adapting the mapping to local phenomena. Accuracy requirements depend on the target function and environment. For example, the requirements for the longitudinal accuracy of the localization on motorways are lower than in urban environments, so that map contents of the digital localization map can be adapted accordingly.

Parameters that can be changed during the creation and provision of the digital localization map can be, for example: number of localization features, constellation or spatial distribution of localization features, meta-information or additional attributes of localization features (e.g. degree of coverage of objects in winter by snow , Glare effects of objects due to a specific position of the sun, reflection effects of objects due to moisture, etc.).

As a result, an amount of data to be transmitted from the digital localization map can advantageously be used 200 be reduced and the user vehicle 100 can change despite the reduced scope of localization features M ₁ ... _{localize M n} with high precision depending on the specific situation regarding place and time.

In this way, the proposed method supports reliable vehicle localization while at the same time reducing the amount of data transferred.

This makes it possible, for example, to send a high-resolution digital localization map to the user vehicle in a traffic jam situation in which there are usually a lot of occlusions 100 is transmitted. In contrast, if there is no traffic jam situation, a low-resolution digital localization map (with few localization features M ₁ ... M _n ) to the requesting vehicle 100 transmitted and thus resources are saved.

It is advantageous with the cloud-based device according to the invention 300 possible to create, provide and use a digital localization map with a need-based feature density. When creating the digital localization map, additional information is used for this purpose, which allows conclusions to be drawn about the required feature density. This information includes, for example, the expected traffic density (high Traffic density potentially leads to higher obscuration rates) or frequency of changes in the surroundings (lead to an outdated localization map).

1. (i) Die digitale Lokalisierungskarte wird auf cloudbasierten Vorrichtung 300 mit einer maximalen Merkmalsdichte (vollständige digitale Lokalisierungskarte) erzeugt. Beim Abruf der digitalen Lokalisierungskarte durch ein Anwenderfahrzeug 100 wird die Merkmalsdichte derart angepasst, dass lediglich eine benötigte Datenmenge übertragen und vom Anwenderfahrzeug 100 verarbeitet werden muss.

On the cloud-based device 300 the fleet data are processed into a digital localization map, whereby the local adjustment of the feature density can be carried out in two advantageous forms:

1. (i) The digital location map is on a cloud-based device 300 generated with a maximum feature density (complete digital localization map). When a user vehicle calls up the digital localization map 100 the feature density is adapted in such a way that only a required amount of data is transmitted and from the user vehicle 100 must be processed.

• (ii) Bei einer zweiten Variante wird eine Merkmalsdichte bereits bei der eigentlichen Kartenerstellung bedarfsgerecht angepasst. Dadurch brauchen von den Kartierungsfahrzeugen 110 vorteilhaft weniger Daten betreffend sensorisch erfasste Umfeldmerkmale an die cloudbasierte Vorrichtung 300 hochgeladen werden.

This approach has the advantage that the feature density can be adapted on the basis of current and possibly dynamic information (e.g. weather, traffic data, time information, etc.). A possible disadvantage of reduced storage efficiency on the cloud-based device 300 Due to the large amount of data, this can be neglected in comparison.

• (ii) In the case of a second variant, a feature density is adapted as required during the actual map creation. This will need from the mapping vehicles 110 advantageously, less data relating to environmental features detected by sensors to the cloud-based device 300 uploaded.

This approach has the advantage that the storage efficiency is on the cloud-based device 300 may be increased due to reduced data volume. In addition, must from user vehicles 100 less data to the cloud-based device in local regions with lower feature density requirements 300 are transmitted and processed there, which is advantageous, for example, on a data rate and / or a required computing power and / or an electrical energy consumption of the cloud-based device 300 can affect. On the other hand, there is a possible disadvantage that the feature density is determined on the basis of “static”, ie rarely / never changing additional information.

The advantages and disadvantages mentioned are then weighed against each other when selecting the appropriate variant in each case.

3 shows schematically a greatly simplified principle of operation of the proposed method.

A total of four quadrants can be seen A. , B. , C. and D. with localization features M ₁ ... M _{n of} a digital localization map.

In one quadrant A. one recognizes localization features M ₁ ... M ₅ the digital localization map, with an environment with a high rate of change or obscuration, for example due to traffic jams, construction sites, etc. quadrant B. shows the localization features of the digital localization map in a situation with a low rate of change or concealment.

The quadrants C. , D. show, in principle, localization features that are actually in the digital localization map for the user vehicle 100 Are available.

In quadrant C. it can be seen that the user vehicle 100 all localization features M1 ... M5 of the localization card are available.

In contrast, it is in the quadrant D. Due to a low rate of change or obscuration, all localization features are not required M ₁ ... M ₅ to the user vehicle 100 to transfer, which is why the digital localization map in the user vehicle 100 only the localization features M1 , M4 , M5 be available.

As a result, only those localization features are required M ₁ ... M _{n of} the digital localization map to be transmitted to the user vehicle, which leads to a highly precise localization of the user vehicle 100 are required, whereby the transmitted features are adapted to local conditions in the environment of the user vehicle.

4th shows a basic sequence of a method for providing a digital localization map for a vehicle.

In one step 500 a request to create the digital localization map is received 400 through the vehicle 100 .

In one step 510 a creation of the digital localization map 400 from cyclically provided information about the surroundings of the vehicle 100 carried out.

In one step 520 a transmission of localization features of the digital localization map 400 to the vehicle 100 depending on the time and place of the requesting vehicle 100 carried out.

It is advantageously possible to implement the method in software, whereby a efficient and easy adaptability of the method is supported.

In summary, the present invention provides a method and a device with which a targeted adaptation of feature densities in a digital localization map is possible. A trade-off between high feature density for high localization accuracy and robustness as well as storage and data rate efficiency is focused.

A local increase in the feature density in digital localization maps can lead to an advantageous influence on the robustness of the localization system with regard to masking effects and obsolete maps. This point is of significant importance for the entire automated driving function, as high system availability and security can only be guaranteed with a robust and safe localization solution.

When implementing the invention, the person skilled in the art will also implement embodiments not explained above.

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 102016224329 A1 [0006]

Claims (12)

Method for providing a digital localization map (400) for a vehicle (100), comprising the steps: - Receiving a request to create the digital localization map (400) by the vehicle (100); - Creating the digital localization map (400) from cyclically provided information about the surroundings in the surroundings of the vehicle (100); and - Transmission of localization features of the digital localization map (400) to the vehicle (100) as a function of the time and location of the requesting vehicle (100). Procedure according to Claim 1 , environmental features detected by sensors being used as environmental information. Procedure according to Claim 2 , whereby a traffic density is computationally determined from the information about the surroundings. Procedure according to Claim 2 or 3 wherein information provided relating to traffic density in the vicinity of the vehicle (100) is used as information about the surroundings. Method according to one of the preceding claims, at least one of the following information is used: time of day, time of year, light situation. Method according to one of the preceding claims, meta-information of at least one localization feature of the digital localization map (400) being used as environment information. Method according to one of the preceding claims, wherein localization features to be transmitted to the vehicle (100) are selected from a digital location map (400) completely filled with location features and transmitted to the vehicle (100). Method according to one of the preceding claims, wherein the information about the surroundings is taken into account by appropriately adapted provision of localization data for the cloud-based device (300). A method for requesting a digital localization map (200) by a vehicle (100), wherein the request for the digital localization map (200) is carried out as a function of defined local and temporal circumstances of the vehicle (100), the digital localization map (200) from a cloud-based device (300) is received. Cloud-based device (300) for creating a digital localization map (200) for a vehicle (100, 110), comprising. - A mapping device (310) for creating the digital localization map (400); - a request device (320) for processing a request for the digital localization map (400) by the vehicle (100, 110); and - an adaptation device (330) for adapting localization features (M ₁ ... M _n ) of the digital localization map (400) as a function of the requirements of the vehicle (100). Computer program comprising instructions which cause the computer program to be executed by a computer, a method according to one of the Claims 1 to 8th to execute. Machine-readable storage medium on which the computer program is based Claim 11 is stored.

Images