Method and device for creating a positioning map

Abstract

A method (300) and a device for creating a positioning map (340), the method comprising the steps of: receiving (310) environmental data values, partitioning (320) the environmental data values into at least two independent data groups, creating (330) a plurality of partial maps, creating (340) a positioning map, and providing (350) the positioning map.

Description

Technical Field

[0001] The present invention particularly relates to a method for creating a positioning map, the method having the following steps: receiving environmental data values; dividing the environmental data values into at least two independent data groups; creating a plurality of partial maps; creating a positioning map based on the plurality of partial maps; and providing the positioning map. Summary of the Invention

[0002] The method for creating a positioning map according to the present invention includes the following steps: receiving environmental data values representing the surrounding environment of at least one vehicle; dividing the environmental data values into at least two independent data groups; and creating a plurality of partial maps based on the at least two independent data groups. The method further includes the following steps: creating a positioning map based on the plurality of partial maps according to a consistency check of the plurality of partial maps; and providing the positioning map.

[0003] For example, the environmental data values are detected by means of an environmental sensing device of the vehicle and then transmitted for reception (by means of a server, etc.; here by means of a device). The environmental sensing device is understood to be at least one video sensor and / or at least one radar sensor and / or at least one lidar sensor and / or at least another sensor, which is configured to detect the surrounding environment of the vehicle in the form of environmental data values. In a possible embodiment, the environmental sensing device includes, for example, a computing unit (processor, working memory, hard disk) with suitable software and / or is connected to such a computing unit.

[0004] The environmental data values are understood to be, for example, data values detected by more than one environmental sensing device (of different vehicles) and / or detected at different times. In one embodiment, the environmental data values include, for example, video data and radar data of the surrounding environment (which are comparable), where the video data and radar data are detected by the environmental sensing devices of the vehicle or different vehicles.

[0005] Here, the surrounding environment is understood to be, for example, an area that can be detected by means of an environmental sensing device. In one embodiment, the surrounding environment of at least one vehicle is understood to be an area where at least partial regions overlap, where each partial region corresponds to the surrounding environment of the vehicle.

[0006] Creating a map (here: a plurality of partial maps and / or a positioning map) is understood to be, for example, adding environmental data values to a base map. In one embodiment, creating includes, for example, adding environmental features included in the surrounding environment to the base map. The base map is understood to be, for example, map raw data and / or an existing (digital) map.

[0007] The creation of multiple partial maps is particularly understood as: creating at least two partial maps based on at least two independent data sets, where the partial maps represent different aspects of the surrounding environment (different sensors [such as video and radar, etc.], different detection times [day and night, etc.], different weather conditions, different traffic conditions, and / or traffic density, etc.).

[0008] The consistency check of at least two partial maps is understood as: checking whether at least two partial maps can be fused into a common map (here: a localization map). Here, for example (based on the respective environmental features), it is checked whether at least two partial maps are consistent (within a pre-given tolerance range) and / or can be made consistent. The result of the consistency check is analyzed and processed to identify errors in map creation. If the consistency check fails, it can be inferred that there is an error in cartography. Therefore, this step is part of a safety concept in the sense of ISO 26262 / ISO / PAS 21448 and also helps to increase the error detection rate and thus the quality of the created localization map even within the scope of non-safety-critical applications.

[0009] The creation of a localization map based on the consistency check of multiple partial maps is understood as: fusing all partial maps (that can be fused based on the performed consistency check) into a localization map. In one implementation, for example, this can be understood as integrating all environmental features included in exactly these partial maps into a common map, which then corresponds to the localization map.

[0010] Environmental features are understood as: objects (traffic signs, infrastructure features [guardrails, bend directions, tunnels, bridges, etc.], buildings, etc.) that can be detected and / or classified or assigned, for example, by means of the vehicle's environmental sensing devices. In one implementation, environmental features are, for example, additionally or alternatively understood as the road alignment (number of lanes, bend radius, etc.) and / or the pattern of multiple (e.g., repeating) objects (such as the characteristic sequence of traffic signs, etc.).

[0011] A map is, for example, understood as a digital map that exists in the form of (map) data values on a storage medium. The map is, for example, constructed such that it includes one or more map layers, where the map layer, for example, shows the map (the alignment and position of roads, buildings, landscape features, etc.) from a bird's-eye view. This corresponds, for example, to the map of a navigation system. Another map layer, for example, includes a radar map, where the localization features included in the radar map store radar signatures. Another map layer, for example, includes a lidar map, where the localization features included in the lidar map store lidar signatures.

[0012] The positioning map is structured in such a way that it is suitable for the navigation of a vehicle, in particular an automated vehicle. This is understood, for example, to mean that the positioning map is structured to determine the highly precise position of the (automated) vehicle by comparing the positioning map with sensor data values of the (automated) vehicle. To this end, the positioning map includes, for example, environmental features with highly precise position specifications (coordinates).

[0013] An automated vehicle is understood to be a vehicle that is structured according to SAE levels 1 to 5 (cf. the standard SAE J3016).

[0014] A highly precise position is understood to be a position that is so accurate within a predefined coordinate system (for example, WGS84 coordinates) that the position does not exceed a maximum permitted ambiguity. Here, the maximum ambiguity can depend, for example, on the surrounding environment. In addition, the maximum ambiguity can depend, for example, on whether the vehicle is being operated manually, semi-automatically, highly automatically or fully automatically (corresponding to one of SAE levels 1 to 5). In principle, the maximum ambiguity is so low that the reliable operation of the (automated) vehicle is particularly ensured. For the fully automated operation of an automated vehicle, the maximum ambiguity is, for example, in the order of approximately 10 centimeters.

[0015] Providing the positioning map is understood to mean that the positioning map can, for example, be called up or stored and transmitted in a transmissible manner. In one embodiment, providing is understood, for example, to mean transmitting the positioning map to at least one (automated) vehicle and / or another server and / or the cloud.

[0016] The method according to the invention advantageously solves the following task: providing a method for creating a positioning map. This task is solved by means of the method according to the invention in such a way that the received environmental data values are divided into at least two independent data groups, and then a plurality of partial maps are created based on the at least two independent data groups, and a positioning map is created based on the plurality of partial maps. The following advantage emerges here: the creation of the positioning map has a favorable impact on the reliability of map generation and vehicle positioning. This is an important safety measure for feature-based vehicle positioning within the scope of a safety concept.

[0017] Preferably, the division is carried out according to at least one division criterion. At least one division criterion is selected, in particular, according to the environmental characteristics of the surrounding environment.

[0018] Partitioning according to at least one partitioning criterion, where in particular the at least one partitioning criterion is selected according to the environmental characteristics of the surrounding environment, is understood as follows: For example, filtering or searching for environmental data values according to a pre-given criterion (environmental characteristics), and then storing the environmental data values independently according to this criterion. Possible partitioning criteria are, for example, the different aspects described above. The partitioning criterion is particularly understood as the detection time (day or night; pre-given detection time period [morning; 6 - 10 o'clock, etc.]) and / or weather conditions (sunny, clear, foggy, etc.) and / or sensor type (one data group includes all video data, another data group includes all radar data, etc.) and / or traffic conditions (one data group includes (according to a pre-given criterion) environmental data values detected at high traffic density, another data group includes (according to a pre-given criterion) environmental data values detected at low traffic density, etc.) and / or other criteria.

[0019] Preferably, at least one partitioning criterion is selected according to the weighting of environmental characteristics, where the weighting is based on a pre-given criterion and / or based on a machine learning scheme.

[0020] Weighting is understood, for example, as follows: Preferably, a certain partitioning criterion is used, and other partitioning criteria are only used when, for example, certain requirements for environmental data values are met. For example, if there is no data detected at night, then partitioning according to day and night is meaningless.

[0021] Weighting according to a pre-given criterion is understood, for example, as that the weighting has been derived from empirical data (inspection, analysis, testing, etc.) in advance.

[0022] Weighting based on a machine learning scheme is understood, for example, as follows: (especially relative to a pre-given result) according to the actual configuration of environmental data values (such as the number of vehicles, the frequency of each sensor type, the frequency of each detection time or time period, etc.) by means of a corresponding "machine learning algorithm" for weighting.

[0023] Preferably, a consistency check is performed by means of a similarity measure (especially by means of a similarity measure and by means of the identification and / or minimization of the influence of mapping errors based on the similarity measure).

[0024] The similarity measure is understood, for example, as the Hausdorff and / or OSPA measure and / or other measures.

[0025] The device (especially the computing unit) according to the present invention is arranged to perform all steps of the method according to the present invention.

[0026] In a possible implementation, the device includes a computing unit (processor, working memory, hard disk) and suitable software for implementing the method according to the present invention. For this purpose, the device includes, for example, a sending and / or receiving unit, which is configured to transmit and / or receive environmental data values and / or positioning maps (especially by means of a vehicle and / or an external server or cloud), or the sending and / or receiving unit is configured to connect to a sending and / or receiving device by means of a suitable interface. In another implementation, the device is configured as a server or cloud (i.e., a complex of servers or computing units).

[0027] Furthermore, a computer program is claimed, which includes instructions that, when the computer program is run by a computer, cause the computer to implement the method according to the present invention. In one implementation, the computer program corresponds to the software included by the device.

[0028] Furthermore, a machine-readable storage medium is claimed, on which the computer program is stored.

[0029] Advantageous extensions of the present invention are listed in the following description. Description of the Drawings

[0030] Embodiments of the present invention are shown in the drawings and further elaborated in the following description. The drawings show:

[0031] Figure 1 An embodiment of the method according to the present invention is shown in the form of a flowchart. Detailed Description of the Invention

[0032] Figure 1 An embodiment of a method 300 for creating a 340 positioning map is shown.

[0033] Method 300 starts in step 301.

[0034] In step 310, environmental data values representing the surroundings of at least one vehicle are received.

[0035] In step 320, the environmental data values are divided into at least two independent data groups.

[0036] In step 330, a plurality of partial maps are created based on at least two independent data groups.

[0037] In step 340, a positioning map is created based on the plurality of partial maps according to a consistency check of the plurality of partial maps.

[0038] In step 350, the positioning map is provided.

[0039] Method 300 ends in step 360.

Claims

1. A method (300) for creating a positioning map (340), the method comprising: receiving (310) environmental data values representing the surroundings of at least one vehicle, wherein the surroundings of the at least one vehicle include regions of at least partial overlap of partial regions, and wherein each partial region corresponds to the surroundings of a vehicle; dividing (320) the environmental data values into at least two independent data groups; creating (330) a plurality of partial maps based on the at least two independent data groups; creating (340) the positioning map based on the plurality of partial maps according to a consistency check of the plurality of partial maps; providing (350) the positioning map, wherein the creation of the plurality of partial maps and / or the positioning map includes adding environmental features included in the surroundings to a base map, wherein the consistency check of the plurality of partial maps includes checking whether the plurality of partial maps can be fused into a common positioning map, wherein it is checked whether the plurality of partial maps are consistent and / or can be made consistent within a pre-given tolerance range based on respective environmental features, and wherein the result of the consistency check is analyzed to identify errors in the creation of the plurality of partial maps, and if the consistency check fails, it is inferred that there are errors in the creation of the plurality of partial maps.

2. The method (300) according to claim 1, characterized in that the dividing (320) is performed according to at least one dividing criterion, and wherein the at least one dividing criterion is selected according to the environmental characteristics of the surroundings.

3. The method (300) according to claim 2, characterized in that the at least one dividing criterion is selected according to a weighting of the environmental characteristics, wherein the weighting is based on a pre-given criterion and / or based on a machine learning scheme.

4. The method (300) according to claim 1, characterized in that the consistency check is performed by means of a similarity measure and by means of the identification and minimization of the influence of cartographic errors based on the similarity measure.

5. A device configured to perform all steps of the method (300) according to any one of claims 1 to 4.

6. The device according to claim 5, wherein the device includes a computing unit.

7. A computer program product comprising instructions which, when the computer program product is run by a computer, cause the computer to perform the method (300) according to any one of claims 1 to 4.

8. A machine-readable storage medium having stored thereon a computer program which includes instructions that, when the computer program is run by a computer, cause the computer to perform the method (300) according to any one of claims 1 to 4.

Images