DE102020206239B4 - Method and system for determining a motor vehicle's own position and a motor vehicle equipped with such a system - Google Patents

Abstract

Method for determining an inherent position (2) of a motor vehicle (3), wherein a data processing unit (4) is provided with a map data set (10) of an area (11) of the motor vehicle (3) and by means of a sensor system (5) of the motor vehicle (3) at least an environmental element (12) in the environment (11) is at least partially detected and a sensor data set (13) at least partially characterizing the environment (11) is provided to the data processing unit (4), so that the data processing unit (4) from the sensor data set (13) and the Map data set (10) generates and provides a self-position data set (14),
characterized in that
the data processing unit (4) is provided with a context data record (15), on the basis of which the own position data record (14) is subjected to a plausibility check by means of the data processing unit (4),
at least one item of movement information (16) of the environmental element (12) being added to the context data record (15) for the plausibility check.

Description

According to patent claim 1, the invention relates to a method for determining an inherent position of a motor vehicle. Furthermore, the invention according to patent claim 8 relates to a system designed to carry out this method for determining an inherent position of a motor vehicle. In addition, the invention according to patent claim 9 relates to a motor vehicle that is equipped with such a system.

Today, there are far-reaching efforts in automotive engineering to continue to automate motor vehicles, in particular motor vehicles, ie passenger cars and/or trucks, in order to continue to relieve the driver of the motor vehicle or passenger car or truck of driving tasks. According to SAE J3016, the respective degrees of automation or degrees of autonomy are divided into five levels. Especially in connection with levels 3, 4 and 5, where the driver of the motor vehicle may at least temporarily turn away from a current driving task and/or from a current traffic situation, there is of course a need for levels 3, 4 and/or 5 to train automated motor vehicles in a particularly safe and reliable manner.

In the case of highly automated motor vehicles (Level 3), the driver is allowed, for example, to turn away from the driving tasks at least temporarily and/or in certain predetermined or specific applications of the motor vehicle. This means that the motor vehicle then drives independently or performs the driving task itself. For example, the motor vehicle that is highly automated according to level 3 is designed to control a steering and drive unit of the motor vehicle independently or automatically, that is to say without any action on the part of the driver, in order to complete a driving task.

In a fully automated motor vehicle (Level 4), the driver is allowed to leave the completion of the driving task entirely to the motor vehicle at least temporarily, for example over the entire duration of a specific application - the driver is then merely a passenger in the motor vehicle. In the case of autonomous vehicles (level 5), no human driver is provided at all. This means that only passengers are transported by means of the autonomous motor vehicle according to level 5, with these passengers at no time controlling or supervising/monitoring a driving task of the motor vehicle.

For a particularly safe and reliable functioning of the motor vehicle that can be operated or moved automatically from level 3, there is a need for a particularly safe or autonomous navigation of the motor vehicle. This requires a particularly reliable determination or determination of a current position of the corresponding motor vehicle. The prior art offers systems for this purpose, for example based on a comparison of image data of landmarks (such as ground markings, signs, spatial structures, etc.) captured by a vehicle camera with a high-precision digital map. There are also approaches based on lidar point clouds and radar features.

For example, the DE 10 2016 203 723 A1 a method in which a vehicle determines its pose—that is, its position and/or spatial orientation—using information about the environment and uses environmental sensors to determine additional information about dynamic objects in the environment and uses it to determine the pose.

Furthermore, the DE 10 2013 003 117 A1 a method for self-localization of a vehicle, images of the environment being captured and analyzed by means of an image capture unit in order to compare image features with comparison features stored in a database in order to determine a position of the vehicle therefrom.

the DE 10 2017 211 632 A1 shows a method for operating a highly automated vehicle, in the context of which a vehicle position of the vehicle is checked for plausibility by comparing a current actual traffic density with an expected target traffic density.

the DE 10 2015 206 342 A1 shows a method for correcting a position of a vehicle, in the context of which at least one object in the area surrounding the vehicle is identified, the position of which can be referenced on a digital map, and a real distance from the object is determined.

The object of the present invention is to further improve the determination of a motor vehicle's own position.

This object is achieved by a method having the features specified in patent claim 1, by a system having the features specified in patent claim 8 and by a motor vehicle having the features specified in patent claim 9. Features, advantages and advantageous configurations of the method according to the invention are to be regarded as features, advantages and advantageous configurations of the system according to the invention or of the motor vehicle according to the invention and vice versa. Features, Benefits and Beneficial Configurations of the system according to the invention are to be regarded as features, advantages and advantageous configurations of the motor vehicle according to the invention and vice versa. The invention also includes the combinations of features of the described embodiments.

The invention thus relates (in a first aspect) to a method for determining an inherent position of a motor vehicle. The motor vehicle can be designed as a motor vehicle, in particular a passenger car and/or truck. Furthermore, the motor vehicle can be a motorcycle or a self-propelled work machine. In particular, it is provided that the motor vehicle is designed to be highly automatic, fully automatic and/or autonomously operable or movable according to one of the initially mentioned levels according to SAE J3016, in particular from level 3.

In the method, a data processing unit is provided with a map data record of an area surrounding the motor vehicle. The data processing unit is, in particular, a computer device that can be embodied in the vehicle and/or external to the vehicle. The data processing unit or the computer device is designed to process and/or further process data provided by the data processing unit.

In the method, at least one environmental element in the environment is at least partially detected by means of a sensor system of the motor vehicle and a sensor data set at least partially characterizing the environment is provided to the data processing unit. This means that the motor vehicle has the sensor system, which includes at least one sensor unit. The respective sensor unit in turn has at least one sensor, for example an ultrasonic sensor, a laser sensor or laser scanner, a lidar sensor, a camera sensor, a radar sensor, etc. It can also be provided that the sensor system includes the data processing unit, at least a vehicle-internal part of the data processing unit. In any case, the respective sensor unit and the data processing unit are coupled to one another in terms of data technology, so that a respective sensor value of the sensor unit can be made available to the data processing unit in data form.

The surroundings of the motor vehicle are to be understood as meaning the surroundings of the motor vehicle, in particular a close-up area, with the portion of the surroundings or surroundings relevant to the method being limited or defined, for example, by a sensor range of the sensor system. In other words, the close range is the portion of the environment or surroundings that is relevant for the method.

In the method, the sensor system therefore includes or detects or senses at least one feature or element that is arranged in the environment or in the near field. This feature or element is called the "environmental element". The environmental element, which can also be referred to as an environmental feature, is, for example, part of an infrastructure or an infrastructure element, in particular a building element (wall, ceiling, pillar, floor, floor marking, sign, door/gate frame, door/gate leaf, window, etc.) and/or a traffic infrastructure element (road, lane, road marking, junction, crossing, cycle path, pedestrian path, traffic-calmed zone, curb, green verge, plant, roundabout, signage, traffic light, railroad track, turnpike, Etc.). These lists are only to be understood as examples and as not exhaustive.

The sensor data record has at least one piece of information about the detected environmental element, for example a relative position of the environmental element to the motor vehicle. The map data set is a map in digital form or in data form. In particular, the map data set is a digital topographical map that is suitable for enabling navigation, for example by means of a navigation system. This is because the map data record has data or information about a traffic network and other natural and man-made objects that are firmly connected to the earth's surface. This means that information about the environmental elements can be stored in the map data set.

The sensor data set and the map data set are both made available to the data processing unit, so that it generates and makes available a personal position data set from the sensor data set and the map data set. This means that the data processing unit is designed to process or further process data from the sensor dataset and data from the map dataset in order to generate the own position dataset or ego position dataset and make them available to the navigation system, for example.

The own position data record or ego position data record comprises at least one piece of information about a geographic longitude and a geographic latitude of the position at which the motor vehicle, which can also be referred to as the ego vehicle, is currently located on the globe. Furthermore, the self-position data set can contain information about a height in relation to normal height zero point. In particular, the own position data record also has information about an alignment of the host vehicle, i.e. in which compass direction (for example specified in degrees) a forward direction of travel parallel to a longitudinal axis of the host vehicle is currently pointing or at the current position. Information about an inclination of the longitudinal axis, a transverse axis and/or a vertical axis of the host vehicle in relation to a horizon is also conceivable for the own position data set, for example to determine whether the host vehicle or the motor vehicle is on/on a slope.

In order to further improve the determination of the ego position of the ego vehicle, the invention provides that the data processing unit is provided with a context data set, which is used by the data processing unit to subject the own position data set to a plausibility check. In this case, it is provided in particular that the data of the map data set and the data of the sensor data set are digitally linked to one another. This is preferably done on the basis of heuristics, which are used to link the information or information about the at least one environmental element with the data of the map dataset.

Put simply, the ego vehicle estimates its ego location or ego position and generates, for example by means of the data processing unit, the own position data set, which is made available to the navigation system, for example, for further navigation or for further navigation of the ego vehicle. The contextual record represents contextual knowledge used to reduce estimation uncertainty. For example, a mathematical algorithm is used for estimating the ego position, on the result side of which there is a large number of possible solutions or a solution space having this large number. These solutions are, for example, a large number of location hypotheses or possible own position data sets. The context knowledge or the context data record is then used for improved delimitation of the possible solutions or the possible own positions of the host vehicle in which the solution space is further restricted than is known from the prior art. This results in a more effective or more precise and more efficient or faster solution of the mathematical algorithm and consequently a more accurate and faster determination of the motor vehicle's own position.

In the plausibility check of the own position data set, it is thus determined whether the ascertained or determined ego position of the motor vehicle, ie the own position data set, is plausible. If, during the plausibility check, the data processing unit comes to the conclusion, for example, that a surrounding element detected by the sensor system cannot usually be present at the current position of the motor vehicle, it can be provided, for example, that the method - at least partially - is run through again or is carried out, for example, until the plausibility check returns a positive result. This is particularly advantageous for highly automatic, fully automatic and/or autonomous driving tasks, for example for a fully automatic and/or autonomous parking process for the motor vehicle. In particular, if the surroundings of the host vehicle or its immediate area are not well mapped in the digital map dataset and/or only a few environmental elements are stored in the map dataset with regard to the immediate area of the motor vehicle, determining the own vehicle's position can be restricted. It would then only be possible to a limited extent, for example, to efficiently link the sensor data set and the map data set to one another using the heuristics in order to generate the own position data set. The context data record or the context knowledge, which is used in the data processing unit to check the own position data record for plausibility, intervenes here.

According to the invention, at least one movement information item of the environmental element is added to the context data record for the plausibility check of the own position data record. In other words, it is determined or ascertained which state of motion the environmental element to be detected or currently detected has. It can thus be determined in the method or for the method whether the environmental element is a dynamic, ie currently moving environmental element, or a static, ie stationary or stationary environmental element. Accordingly, the context data record then includes context knowledge that characterizes, for example, whether the environmental element currently detected or recorded by the sensor system of the motor vehicle in the vicinity of the host vehicle is a stationary infrastructure element, for example a house wall, etc., or a currently moving environmental element, for example another Road users, for example a pedestrian, another motor vehicle, etc.

For the plausibility check of the own position data record, information can then be used, for example, that another motor vehicle detected by the sensor system of the host vehicle, which is arranged next to the host vehicle in relation to the longitudinal axis of the latter, is stationary (i.e. parked, for example) or is moved (e.g. drives). It can thus be distinguished whether the host vehicle is passing a parking bay in its current position or own position, in which the other motor vehicle is parked, or whether the host vehicle is passing an adjacent lane in which the other motor vehicle is driving. As a result, the plausibility check is particularly reliable, so that a particularly accurate own position data record results, or so that the position hypotheses can be selected particularly efficiently.

Alternatively or additionally, at least one item of distance information relating to a distance relating to the environmental element can be added to the context data record for a plausibility check. This distance relating to the surrounding element can, for example, include an overall height of the surrounding element, for example a height between a fastening surface to which the surrounding element is attached or fastened, and a prominent element of the surrounding element, for example an upper edge, in which case this distance relating to the surrounding element, for example the overall height is compared with a comparative value during the plausibility check. If, for example, it is determined that the distance relating to the surrounding element and the comparison value deviate excessively from one another, the plausibility check supplies a negative result. Instead, the plausibility check delivers a positive result if the distance relating to the surrounding element and the comparison value at least essentially correspond to one another. In this case, the own position of the host vehicle is then plausible.

This distance information enables the motor vehicle's own position to be determined even more efficiently and/or more precisely.

A distance measure between the environmental element and the motor vehicle or host vehicle can be added to the distance information. This again increases a particularly efficient and precise determination of the motor vehicle's own position. If, for example, a rough determination of the ego vehicle's own position (where, for example, there is more than one location hypothesis), in particular based on the map data set, it is determined that it is so close to an infrastructure element that another environmental element cannot be arranged between the motor vehicle and the infrastructure element , but another infrastructure element between the motor vehicle and the infrastructure element, for example by means of sensors, is detected, the plausibility check returns a negative result. In other words, the plausibility check provides that either the own position of the motor vehicle or the environmental element and/or the additional environmental element are/is incorrectly recorded. As already mentioned above, if the result of the plausibility check is negative, the vehicle's own position is determined again, for example, or another of the possible own position data sets is selected, thus preventing an erroneous or imprecise determination of the own vehicle's own position.

Alternatively or additionally, a distance measure between the surrounding element and at least one further surrounding element can be added to the distance information. In this way, the determination of the motor vehicle's own position is even more reliable and safer. For example, contextual knowledge is added to the context data record as to whether there can be another surrounding element between the surrounding element and the further surrounding element, for example, if the three surrounding elements named as examples are designed and/or arranged according to standard standards. For example, based on the plausibility check, it can be ruled out that, for example, a pedestrian is staying or moving between two surrounding elements that are arranged too close together for a pedestrian to fit between them.

It has also proven to be advantageous if arrangement information for the environmental element is added to the context data record for the plausibility check, using which a probability is assessed as to whether the environmental element is arranged as intended in the current environment or in the vicinity of the motor vehicle. In other words, the plausibility check then evaluates or checks whether it is at all probable that the environmental element, which was detected by the sensor system of the motor vehicle, for example, is actually located in the vicinity of the host vehicle. This is particularly advantageous in parking facilities, especially multi-storey car parks, especially when an entrance and an exit of the multi-storey car park are close together, often only a few meters apart. This is because it is then particularly difficult or time-consuming, when determining the own vehicle's own position, to ascertain and/or to determine whether the host vehicle is located at the entrance or at the exit of the multi-storey car park. If the sensor system detects a surrounding element, for example an element of the multi-storey car park, which can only be found at the entrance to the multi-storey car park, a different location hypothesis, which characterizes that the motor vehicle is located at the exit of the multi-storey car park, can be rejected.

In a further advantageous embodiment of the method, it is provided that the context data set is at least partially created or expanded by the sensor system of the motor vehicle. This means that the sensor system of the motor vehicle is designed to at least partially detect the surroundings of the motor vehicle in such a way that the surrounding elements can be detected or detected or sensed by means of the sensor system. In particular, the sensor system of the motor vehicle is designed to record the movement information of the environmental element and, for example, to make it available to the data processing unit. Furthermore, the sensor system of the motor vehicle can be designed to record the at least one piece of distance information, in particular the distance between the surrounding element and the motor vehicle and/or the distance between the surrounding element and the at least one other surrounding element, and to make it available to the data processing unit, for example. Furthermore, the sensor system of the motor vehicle can be designed to detect the corresponding environmental element in such a way that the arrangement information of the environmental element is made available to the data processing unit, for example.

If the context data record or the context knowledge is created and/or expanded using the sensor system of the motor vehicle, a separate sensor system for carrying out the method can be dispensed with. In this case, the sensor system of the motor vehicle then fulfills at least one double functionality, namely - firstly - the sensor system then provides the functionality for carrying out the method for determining the vehicle's own position and - secondly - the sensor system provides a functionality for driving the motor vehicle, for example for at least a driver assistance system of the motor vehicle ready. A correspondingly designed motor vehicle is then designed to be particularly mass-efficient or lightweight, as a result of which the motor vehicle can be operated in a particularly fuel- or energy-efficient manner and/or with low emissions.

Finally, according to a further embodiment of the method, a text of the environmental element is recorded by means of a semantic processor unit of the sensor system and its semantic content is added to the context data record. For this purpose, the sensor system, in particular the semantic processor unit, has a text recognition unit, which can be formed, for example, by a traffic sign recognition system of the motor vehicle. This means that in the method, a text of the environmental element, for example a sign, is read by machine and evaluated, with the corresponding evaluation result representing further context knowledge that is added to the context data record. To use the example of the motor vehicle again in connection with the multi-storey car park, it can be provided that the motor vehicle, in particular its sensor system, records and evaluates a text “Entrance” so that this is used to check the own position data set for plausibility. Consequently, the host vehicle's own position data set is assessed as plausible if the rough determination of its own position, for example based on the map data set, shows that the host vehicle is located at the entrance to the multi-storey car park. In this way, accuracy when determining the motor vehicle's own position is increased even further.

The invention is based on the idea that purely assigning observation features to map features is not always unambiguous and robust. For example, repetitive structures in parking environments, such as in the parking garage, can result in an environmental element being misassigned. This can be the case, for example, when the motor vehicle detects a large number of identical or at least similar columns, wall elements, signs, etc. and possibly incorrectly assigns them to one of the possible location hypotheses or one of the possible own position data sets. The motor vehicle's own position determination would then be incorrect or at least inaccurate. If, however, in the method--as proposed according to the invention--in addition to detecting the environmental elements, the context knowledge, for example about dynamic objects and/or distances, is used, certain ambiguities can be resolved. For example, a floor line observed or detected by the sensor system cannot be located directly on a wall if a dynamic object is observed behind this floor line during the method or, for example, a larger free space is reported by means of an ultrasonic sensor than would be possible from the physical conditions. In other words, the method effectively prevents the floor line detected by the sensor system of the motor vehicle from being (wrongly) interpreted as a floor line arranged directly on the wall.

The present method also relates to a particularly accurate method for self-localization or self-position determination. This means that the own position data record determines the own position of the motor vehicle particularly precisely, in particular in the centimeter range. In this context, one also speaks of so-called HD localization (“high definition”). It is to be understood in this context that a detection of dynamic objects does not serve, for example, the existence of a Estimate the cycle path and use this information when comparing it with the digital map or with the map data set. Instead, information about the existence of objects at specific locations in the ego vehicle environment is used without an object classification and/or without assuming conditions such as lanes or cycle paths. This is used in order to limit the solution space of the optimization problem (estimation of the motor vehicle's own position and/or selection from more than one possible location hypothesis) more efficiently. A further example of this would be that a pedestrian or passer-by is detected laterally of the host vehicle, for example by means of the sensor system of the motor vehicle, whereby a (mathematical) secondary condition for the estimation of the own position or for the optimization problem is generated. This constraint states that the self-position cannot be directly on a wall, since the pedestrian would not fit between the wall and the motor vehicle.

The invention also relates (in a second aspect) to a system for determining a motor vehicle's own position, the system being designed to carry out the method. The second aspect of the invention also includes developments of the system according to the invention which have features as have already been described in connection with the features of the method according to the invention. For this reason, the corresponding developments of the system according to the invention are not described again here.

Finally, the invention relates (in a third aspect) to a motor vehicle that is equipped with a system for determining the motor vehicle's own position. The third aspect of the invention also includes developments of the motor vehicle according to the invention, which have features as have already been described in connection with the features of the method according to the invention and/or the system according to the invention. For this reason, the corresponding developments of the motor vehicle according to the invention are not presented again here.

• 1 eine schematische Ansicht eines Kraftfahrzeugs, mit einem System zum Bestimmen einer Eigenposition des Kraftfahrzeugs;
• 2 eine schematische Draufsicht einer Verkehrssituation des Kraftfahrzeugs;
• 3 eine schematische Draufsicht einer anderen Verkehrssituation des Kraftfahrzeugs und
• 4 eine schematische Draufsicht noch einer weiteren Verkehrssituation des Kraftfahrzeugs.

An exemplary embodiment of the invention is described below. For this shows:

• 1 a schematic view of a motor vehicle, with a system for determining an inherent position of the motor vehicle;
• 2 a schematic plan view of a traffic situation of the motor vehicle;
• 3 a schematic top view of another traffic situation of the motor vehicle and
• 4 a schematic plan view of yet another traffic situation of the motor vehicle.

The exemplary embodiment explained below is a preferred embodiment of the invention. In the exemplary embodiment, the described components of the embodiment each represent individual features of the invention to be considered independently of one another, which also develop the invention independently of one another and are therefore also to be regarded as part of the invention individually or in a combination other than the one shown. Furthermore, the embodiment described can also be supplemented by further features of the invention already described.

Elements with the same function are each provided with the same reference symbols in the figures.

A method and a system 1 for determining an inherent position 2 of a motor vehicle 3 and the motor vehicle 3 equipped with the system 1 are described together below.

In 1 shows the motor vehicle 3 in a schematic view, which is equipped with the system 1 for determining the own position 2 . The system 1 is designed to carry out the method for determining the own position 2 of the motor vehicle 3 . Accordingly, the system 1 has means that can be used or are used to carry out the method.

The system 1 or the motor vehicle 3 equipped with the system 1 has a data processing unit 4 and a sensor system 5 . The data processing unit 4 is in particular a computer device, for example a (vehicle-internal) computing unit 6 of a vehicle navigation system (not shown in detail), in particular a satellite navigation system. Alternatively or additionally, the computer device or the data processing unit can be part of a server device 7, which is in particular designed as a (vehicle-external) server device 7, for example a cloud server device. In this case, the motor vehicle 3 and the server device 7 each have a corresponding data transceiver 8 for wireless data communication.

The motor vehicle 3 is designed as a passenger car in the present example. However, it is just as conceivable for the motor vehicle 3 to be in the form of a truck, a motorcycle or a self-propelled work machine. The motor vehicle 3 or the passenger car Gen has a steering and drive unit 9, which provides at least one highly automated, fully automated and/or autonomous operating mode of the motor vehicle 3. This means that the motor vehicle 3 or the passenger car is designed to be highly automatic, fully automatic and/or autonomous. For example, a driver of the motor vehicle 3 is allowed to turn away from corresponding driving tasks at least temporarily and/or in certain predefined or specific applications of the motor vehicle 3 if the motor vehicle 3, in particular the steering and drive unit 9, according to level 3 of the SAE J3016 is formed. If the steering and drive unit 9 is designed according to level 4 of SAE J3016, at least one fully automated operating mode is provided for the motor vehicle 3 . The driver of the motor vehicle 3 is then allowed to leave the completion of the driving task entirely to the motor vehicle 3 at least temporarily, for example over the entire duration of a specific application - the driver is then merely a passenger of the motor vehicle 3. If the steering and drive unit 9 or If the motor vehicle 3 is designed according to level 5 of SAE J3016, at least one autonomous operating mode of the motor vehicle 3 is enabled. In this case, no human driver is provided at all to intervene in a driving operation of the motor vehicle 3 . This means that only passengers (and possibly freight) are transported by means of the autonomously drivable motor vehicle 3 according to Level 5, with the passengers at no time controlling or supervising/monitoring a driving task of the motor vehicle 3 .

In modern motor vehicles, such as motor vehicle 3, it is already provided from level 3 that at least one driving task can be carried out by motor vehicle 3, in particular by its steering and Drive unit 9 is executed or done. A prominent example of this is highly automated, fully automated and/or autonomous parking and/or parking out of the motor vehicle 3, for example by means of a parking assistant of the motor vehicle 3.

For a particularly safe execution of the highly automated, fully automated and/or autonomous driving task, it is necessary for at least the steering and drive unit 9 of the motor vehicle 3 to be provided with the particularly precisely determined own position 2 of the motor vehicle 3 . Put simply, the vehicle must “know” where it is at the time the highly automated, fully automated and/or autonomous driving task is started and during this time.

For this purpose, the data processing unit 4 is provided with a map data set 10 in which an area 11 of the motor vehicle 3 is stored in digital form, ie in data form, for example stored. It can then be provided, for example, that in the method for determining the own position 2 of the motor vehicle 3 using the map data set 10, which is a topographic map, for example, a rough determination of the own position first takes place.

Furthermore, in the method, at least one environmental element 12 is at least partially detected by means of the sensor system 5 of the motor vehicle 3 . The sensor system 5 is designed to generate a sensor data set 13 or a data set originating from the sensor system 5 based on a corresponding detection result, which at least partially characterizes the environment 11, in particular the at least one environmental element 12. Furthermore, the sensor system 5 is designed to provide the sensor system data set 13 to the data processing unit 4 , for example to send it or to have it available for download by the data processing unit 4 .

In the further course of the method, the data processing unit 4 combines the map data set 10 and the sensor data set 13 in such a way that the data processing unit 4 generates a personal position data set 14 from the sensor data set 13 and the map data set 10 and makes it available to the vehicle navigation system and/or the steering and drive unit 9, for example delivers. On the basis of or after the rough determination of one's own position, there is, for example, more than one location hypothesis or more than one own position data record, each of which represents a possible own position 2 of motor vehicle 3 .

Especially in driving tasks, for example driving maneuvers that require a particularly high trajectory fidelity of the motor vehicle 3, in particular when parking the motor vehicle 3 in and/or out of a parking space, there is a need to determine the own position 2 particularly precisely and particularly quickly. For this purpose, the data processing unit 4 is provided with a context data record 15, which is used by the data processing unit 4 to subject the own position data record 14 to a plausibility check. For example, if the data processing unit 4 has determined or estimated two or more location hypotheses that are equivalent to one another and that are based on a heuristic evaluation or combination of the sensor data set 13 and the map data set 10, the context data set 15 is used to check the respective location hypothesis for plausibility check fen. On the basis of the plausibility check, at least one of the location hypotheses then falls away as a possible own position 2 if it fails the plausibility check. A number of possible solutions when estimating the own position of the motor vehicle 3 is particularly small, which is why the estimation or determination of the own position of the motor vehicle 3 proceeds more efficiently. Because ultimately it has to be selected from fewer local hypotheses.

2 shows a traffic situation of the motor vehicle 3 in a schematic top view, with the method for determining or determining the own position 2 (see 1 ) the context record 15 (see 1 ) At least one item of movement information 16 of the corresponding environmental element 12 is added for the plausibility check. A final list of what is considered or understood as a surrounding element 12 in the method would go beyond the scope of this description. Therefore, only a few types of surrounding elements 12 are discussed, which are described as representative of all conceivable surrounding elements 12 . In particular, each object in the surroundings 11 of the motor vehicle 3 can be a surrounding element 12 if the corresponding object can be sensed or detected by means of the sensor system 5 . In the figures, too, not all conceivable surrounding elements 12 are provided with their own respective reference symbols, in order to ensure a particularly good overview.

According to 2 the corresponding environmental element 12 can be another motor vehicle 17, for example. Another example of the at least one environmental element 12 are road markings 18, plants - in particular trees 19, traffic signs 20, other road users, such as pedestrians 21, etc.

The movement information 16 is used in the method for determining the own position 2 of the motor vehicle 3 to determine whether the corresponding environmental element 12 is currently moving or whether the corresponding environmental element 12 is currently stationary.

It's further in 2 shown that the context data record 15 (see 1 ) At least one item of distance information relating to a distance 22 relating to the corresponding environmental element 12 is added for the plausibility check. Referring again to 1 , where the corresponding environmental element 12 is shown as a traffic sign 20, the distance 22 can be a height 23 of the environmental element 12 or of the traffic sign 20, for example. In addition, the distance information can have a distance measure 24 that characterizes a spatial distance between the corresponding environmental element 12 and the motor vehicle 3 . It is conceivable that the corresponding distance 24 is or will be measured parallel to a vehicle longitudinal axis of the motor vehicle 3 or parallel to a vehicle transverse axis of the motor vehicle 3 . Alternatively, it is conceivable to measure the distance directly between the motor vehicle 3 and the corresponding surrounding element 12 . In 2 Only a few of the possible distances 24 or distances 22 are shown.

3 shows a different traffic situation of the motor vehicle 3 in a schematic plan view, wherein the motor vehicle 3 is arranged in one of several lanes 25, in particular driving on it. In addition, in 3 a safety strip 25a drawn, which is commonly referred to as "hard shoulder". The safety strip 25a or the hard shoulder can also represent a surrounding element 12 . Equally, the respective lane 25 or the respective lane can form a respective surrounding element 12 . they are in 3 a few of the distance dimensions 24 drawn in, which characterize the respective distance between the motor vehicle 3 and the corresponding surrounding element 12, in particular the respective further motor vehicle 17. In addition, in 3 some of further distance dimensions 26 drawn in, which characterize a spatial distance between the respective surrounding element 12 and at least one further of the surrounding elements 12 .

The distance information, i.e. the distance measure 24 and/or the distance measure 26, is assigned to the context data record 15 (see 1 ) provided for the plausibility check of the own position data set 14, for example by means of the data processing unit 4 (see 1 ) added. A corresponding measurement of the respective distance 24, 26 and/or a corresponding measurement of the distance 22, in particular the height 23, is preferably carried out by means of corresponding sensors of the sensor system 5.

Furthermore, in the traffic situation as described in 3 is shown, for the plausibility check of the own position data record 14 the respective movement information 16 of the other motor vehicles 17 or at least one of the motor vehicles 17 are used. In this case, the sensor system 5 is designed to record the movement information 16 of the corresponding motor vehicle 17 or the corresponding environmental element 12, for example by means of a radar sensor and/or laser scanner etc.

4 shows a further traffic situation of the motor vehicle 3 in a schematic top view 4 the motor vehicle 3 is shown driving on a parking deck 27, in particular a multi-storey car park 28, in particular in a fully automated, highly automated and/or autonomous manner, for example due to the operation of the parking steering assistance system of the motor vehicle 3. Wall elements 29 of the multi-storey car park 28 are shown in section for this purpose. This results in a plan view of the floor of the parking deck 27, with the motor vehicle 3 being set up on the floor as intended and in particular rolling or driving over it. they are in 4 further examples of a respective surrounding element 12 are drawn in, for example a respective one of the wall elements 29 can represent a surrounding element 12 . Furthermore, an opening 30 in one of the wall elements 29 or between two wall elements 29 can form one of the surrounding elements 12 . The opening 30 can be a window opening, a door opening, a gate opening, etc., for example. In multi-storey car parks in particular, it is customary for safety reasons to arrange a blocked area 31 on the floor of the parking deck 27 in front of openings 30 through which a passer-by or pedestrian can step onto the parking deck 27 . Such a blocking surface 31 can also represent a surrounding element 12 . Furthermore, such a blocked area 31 can be arranged in many other places, in particular in the public traffic area, for example before and after public transport stops at/in crossing areas, for lane closures, etc.

Using the traffic situation as an example 4 it is described that the context data record 15 for plausibility check an arrangement information of at least one of the environmental elements 12, which are within range of the sensors 5, is added. On the basis of the arrangement information—in particular by means of the data processing unit 4—a probability is evaluated as to whether the corresponding environmental element 12 is arranged as intended in the current environment 11 of the motor vehicle 3 . In a synopsis of the 4 with the 2 it becomes clear, for example, that a plausibility check for the hypothesis that the motor vehicle 3 is located in the multi-storey car park 28 or on the parking deck 27 delivers a negative plausibility check result if the sensor system 5 detects or captures a surrounding element 12 that is not in the context of the parking garage 28 or the parking deck 27 fits. Very striking: captures that in 4 shown motor vehicle 3 on the parking deck 27 or in the parking garage 28 a tree 19, the plausibility check would deliver a negative plausibility check result.

The plausibility check is even more efficient if the data processing unit is provided with a semantic content of the corresponding environment element 12 or environment elements 12 . For this purpose it can be provided that the motor vehicle 3, in particular its sensor system 5, has a semantic processor unit which is designed to detect a text of the environmental element 12 and to provide the semantic content of the text. If, for example, the text "Exit" is attached to an orientation sign 32, a correspondingly different location hypothesis, namely that the motor vehicle 3 would be in the area of an entrance to the multi-storey car park 28, can be rejected, since this (wrong) location hypothesis fails at the latest in the plausibility check.

Reference List

1

system

2

own position

3

motor vehicle

4

data processing unit

5

sensors

6

unit of account

7

server setup

8th

data transceiver

9

Steering and drive unit

10

card record

11

vicinity

12

environment element

13

sensory record

14

Own Position Record

15

context record

16

movement information

17

motor vehicle

18

lane marking

19

tree

20

traffic sign

21

pedestrian

22

Distance

23

Height

24

distance measure

25

lane

25a

security stripes

26

distance measure

27

parking deck

28

parking garage

29

wall element

30

opening

31

restricted area

32

orientation sign

Claims (9)

Method for determining an inherent position (2) of a motor vehicle (3), wherein a data processing unit (4) is provided with a map data set (10) of an area (11) of the motor vehicle (3) and by means of a sensor system (5) of the motor vehicle (3) at least an environmental element (12) in the environment (11) is at least partially detected and a sensor data set (13) at least partially characterizing the environment (11) is provided to the data processing unit (4), so that the data processing unit (4) from the sensor data set (13) and the card data record (10) generates and provides a personal position data record (14), characterized in that the data processing unit (4) is provided with a context data record (15), on the basis of which the personal position data record (14) is subjected to a plausibility check by means of the data processing unit (4), the Context data record (15) for plausibility check at least one item of movement information (16) of the U environment element (12) is added. procedure after claim 1 , characterized in that the context data record (15) for plausibility check at least one piece of distance information relating to the environmental element (12) distance (22) is added. procedure after claim 2 , characterized in that a distance measure (24) between the surrounding element (12) and the motor vehicle (3) is added to the distance information. procedure after claim 2 or 3 , characterized in that a distance measure (26) between the surrounding element (12) and at least one further surrounding element (12) is added to the distance information. Method according to one of the preceding claims, characterized in that, for the purpose of a plausibility check, information on the arrangement of the environmental element (12) is added to the context data record (15), on the basis of which a probability is assessed as to whether the environmental element (12) is in the current environment (11) as intended of the motor vehicle (3) is arranged. Method according to one of the preceding claims, characterized in that the context data record (15) is at least partially created or expanded by the sensor system (5) of the motor vehicle (3). procedure after claim 6 , characterized in that a text of the environmental element (12) is detected by means of a semantic processor unit of the sensor system (5) and its semantic content is added to the context data set (15). System (1) for determining an inherent position (2) of a motor vehicle (3) by means of a method designed according to one of the preceding claims. Motor vehicle (3) with a after claim 8 trained system (1) for determining an inherent position (2) of the motor vehicle (3).

Images