METHOD AND PROCESSOR CIRCUIT FOR UPDATING A DIGITAL ROAD MAP

Abstract

A method of updating a digital road map which contains spatially resolved description values of at least one mapped environmental characteristic for at least one mapped road. A virtual journey along the at least one mapped road may be simulated, on basis of the road map to be updated, by a processor circuit before the updating process. The road map may be used to determine, for different virtual geopositions, an associated virtual measurement value of the at least one mapped environmental characteristic The virtual measurement values may be used to generate, for the virtual geopositions, a concatenated data record which describes the virtual journey. Newly received concatenated data records of real journeys of measuring motor vehicles and the concatenated data records of the virtual journey may be combined. The updated road map may be calculated from the combined received concatenated data records of real journeys and the virtual concatenated record.--

Claims

1-6. (canceled)

7. A method of updating a digital road map which respectively contains spatially resolved description values of at least one mapped environmental property for at least one mapped road, comprising: by a processor circuit, respectively receiving, for at least one measuring motor vehicle, at least one graph of first concatenated data sets which describe a respective real journey of the motor vehicle, a graph, from among the at least one graph, of the first concatenated data sets, a first data set, from among the first concatenated data sets, indicates, for a received geo-position which was passed during the real journey, at least one measured value captured at the received geo-position for a new environmental property or the at least one mapped environmental property in the digital road map; before the updating of the digital road map, simulating a virtual journey along the at least one mapped road by, determining a respective virtual measured value of the new environmental property or the at least one mapped environmental property in the digital road map, from the digital road map for different virtual geo-positions, the virtual measured value corresponding to a description value from the digital road map or to an interpolation value from a plurality of the description values, generating a second concatenated data set describing the virtual journey from the virtual measured value for a virtual geo-position, from among the different virtual geo-positions, combining the first concatenated data sets for the respective real journey of the at least one motor vehicle and the second concatenated data set for the virtual journey, calculating the updated road map from the combined first and second concatenated data sets by, calculating updated, spatially resolved description values of the new environmental property or the at least one mapped environmental property in the digital road map, respectively described by the at least one captured measured value and the virtual measured value, from the at least one captured measured value and the virtual measure value and the received geo-position and the different virtual geo-positions of the first and second concatenated data sets, in which at a respective virtual geo-position, from among the different virtual geo-positions, a mean value of the at least one captured measured value and the virtual measured value is calculated to be assigned to the received geo-position as measured at the received geo-position or in a region around the real geo-position, or a statistical distribution of the at least one captured measured value or the virtual measured value is determined, and the respective description value is determined from the calculated mean value or the statistical distribution; and iteratively updating the digital road map on basis of the first concatenated data sets received in temporal succession.

8. The method as claimed in claim 7, wherein the processor circuit transmits the updated road map to at least one motor vehicle using the digital road map.

9. The method as claimed in claim 7, wherein a respective weighting factor is applied to the first and second concatenated data sets when combining the first and second concatenated data sets.

10. The method as claimed in claim 7, wherein the first concatenated data set for the real journey already taken into account is deleted by the processor circuit.

11. A processor circuit having at least one microprocessor (P) and having a data memory (MEM) which stores program instructions which, during execution by the at least one microprocessor (P), cause the at least one microprocessor to carry out a process to update a digital road map which respectively contains spatially resolved description values of at least one mapped environmental property for at least one mapped road, the process comprising: respectively receiving, for at least one measuring motor vehicle, at least one graph of first concatenated data sets which describe a respective real journey of the motor vehicle, a graph, from among the at least one graph, of the first concatenated data sets, a first data set, from among the first concatenated data sets, indicates, for a received geo-position which was passed during the real journey, at least one measured value captured at the received geo-position for a new environmental property or the at least one mapped environmental property in the digital road map; before the updating of the digital road map, simulating a virtual journey along the at least one mapped road by, determining a respective virtual measured value of the new environmental property or the at least one mapped environmental property in the digital road map, from the digital road map for different virtual geo-positions, the virtual measured value corresponding to a description value from the digital road map or to an interpolation value from a plurality of the description values, generating a second concatenated data set describing the virtual journey from the virtual measured value for a virtual geo-position, from among the different virtual geo-positions, combining the first concatenated data sets for the respective real journey of the at least one motor vehicle and the second concatenated data set for the virtual journey, calculating the updated road map from the combined first and second concatenated data sets by, calculating updated, spatially resolved description values of the new environmental property or the at least one mapped environmental property in the digital road map, respectively described by the at least one captured measured value and the virtual measured value, from the at least one captured measured value and the virtual measure value and the received geo-position and the different virtual geo-positions of the first and second concatenated data sets, in which at a respective virtual geo-position, from among the different virtual geo-positions, a mean value of the at least one captured measured value and the virtual measured value is calculated to be assigned to the received geo-position as measured at the received geo-position or in a region around the real geo-position, or a statistical distribution of the at least one captured measured value or the virtual measured value is determined, and the respective description value is determined from the calculated mean value or the statistical distribution; and iteratively updating the digital road map on basis of the first concatenated data sets received in temporal succession.

12. The processor circuit as claimed in claim 11, wherein the processor circuit transmits the updated road map to at least one motor vehicle using the digital road map.

13. The processor circuit as claimed in claim 11, wherein a respective weighting factor is applied to the first and second concatenated data sets when combining the first and second concatenated data sets.

14. The processor circuit as claimed in claim 11, wherein the first concatenated data set for the real journey already taken into account is deleted by the processor circuit.

15. A storage medium (MEM) having a program code which is configured, during execution by a processor circuit, to cause the processor circuit to carry out a process to update a digital road map which respectively contains spatially resolved description values of at least one mapped environmental property for at least one mapped road, the process comprising: respectively receiving, for at least one measuring motor vehicle, at least one graph of first concatenated data sets which describe a respective real journey of the motor vehicle, a graph, from among the at least one graph, of the first concatenated data sets, a first data set, from among the first concatenated data sets, indicates, for a received geo-position which was passed during the real journey, at least one measured value captured at the received geo-position for a new environmental property or the at least one mapped environmental property in the digital road map; before the updating of the digital road map, simulating a virtual journey along the at least one mapped road by, determining a respective virtual measured value of the new environmental property or the at least one mapped environmental property in the digital road map, from the digital road map for different virtual geo-positions, the virtual measured value corresponding to a description value from the digital road map or to an interpolation value from a plurality of the description values, generating a second concatenated data set describing the virtual journey from the virtual measured value for a virtual geo-position, from among the different virtual geo-positions, combining the first concatenated data sets for the respective real journey of the at least one motor vehicle and the second concatenated data set for the virtual journey, calculating the updated road map from the combined first and second concatenated data sets by, calculating updated, spatially resolved description values of the new environmental property or the at least one mapped environmental property in the digital road map environmental property, respectively described by the at least one captured measured value and the virtual measured value, from the at least one captured measured value and the virtual measure value and the received geo-position and the different virtual geo-positions of the first and second concatenated data sets, in which at a respective virtual geo-position, from among the different virtual geo-positions, a mean value of the at least one captured measured value and the virtual measured value is calculated to be assigned to the received geo-position as measured at the received geo-position or in a region around the real geo-position, or a statistical distribution of the at least one captured measured value or the virtual measured value is determined, and the respective description value is determined from the calculated mean value or the statistical distribution; and iteratively updating the digital road map on basis of the first concatenated data sets received in temporal succession.

16. The storage medium as claimed in claim 15, wherein the processor circuit transmits the updated road map to at least one motor vehicle using the digital road map.

17. The storage medium as claimed in claim 15, wherein a respective weighting factor is applied to the first and second concatenated data sets when combining the first and second concatenated data sets.

18. The storage medium as claimed in claim 15, wherein the first concatenated data set for the real journey already taken into account is deleted by the processor circuit.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0026] Examples of the invention are described. In this respect, these and other aspects and advantages will become more apparent and more readily appreciated from the following description of the example embodiments, taken in conjunction with:

[0027] FIG. 1 shows a drawing for illustrating a known algorithm for generating model data relating to a road map from concatenated data sets, as can be used according to an example;

[0028] FIG. 2 shows a drawing for illustrating a graph of concatenated data sets; and

[0029] FIG. 3 shows a schematic illustration of one embodiment of the processor circuit according to an example.

DESCRIPTION

[0030] In the described examples, the described components of the examples are each individual features which should be considered independently of one another and which also each may be developed independently of one another. Therefore, the disclosure is also intended to include combinations other than the illustrated combinations of the features of the examples. Furthermore, the described examples can also be supplemented by further features of the examples which have already been described.

[0031] In the figures, identical reference signs each denote functionally identical elements.

[0032] FIG. 1 shows a processor circuit 10 which may be provided, for example, in an Internet server and may be implemented by a computing center, for example. The processor circuit 10 may be based on one or more microprocessors (so-called CPUs - Central Processing Units). A digital road map 12 can be calculated from sensor data 11 by the processor circuit 10. The processor circuit 10 may receive the sensor data 11 from motor vehicles 13 which may be situated in a road network of an environment 14 and may describe the motor vehicles’ 13 measured values 15 for at least one environmental property of the environment 14, for example, the current temperature and/or a road condition and/or a position of lanes of the respective road. The measured values 15 may be represented by the sensor data 11. The measured values 15 of a journey of each of the motor vehicles 13 may be respectively combined overall to form a graph 16 of concatenated data sets, which will also be explained in more detail below in connection with FIG. 2.

[0033] The processor circuit 10 may be based on at least one microprocessor (P) and a data memory (MEM) which is coupled to the latter and may contain program instructions for the method described here.

[0034] The road map 12 can be calculated in a manner known per se from such graphs 16 of concatenated data sets by an algorithm 17 known per se from the related art, that is to say the road map 12 can be calculated overall from the measured values 15 in a data format which can be stipulated by a predefinable data format. For example, it may be possible to choose a road format from a map provider wishing to use the road map 12. Such a road map 12 may describe, for example, lane courses and/or center line courses and/or lane boundaries and/or landmarks. A local distribution of at least one environmental property may be described by description values in the road map.

[0035] FIG. 2 again illustrates the structure of the measured values 15 in the graphs 16. FIG. 2 illustrates how a motor vehicle 13 can carry out a journey 20 along a road 19 in the environment 14. The road 19 may be intended to be mapped in the road map 12. Along the route or road 19, the motor vehicle 13 can respectively measure, at different geo-positions 21, the geo-position 21 itself and optionally at least one further measured value 15 (also see FIG. 1) during the journey 20 in a manner known per se using an on-board sensor circuit. The measurement of the geo-position 21 itself may also constitute a measured value 15 in the sense of the examples.

[0036] A graph 16 of concatenated data sets 22 may now be formed from the measured values 15. In this case, each data set 22 represents the measurement or the measurement result at one of the geo-positions 21. The data set 22 may here describe, for example, measurement data relating to a pose 23 of the motor vehicle 13, as held by the motor vehicle at the geo-position 21, a measurement of the geo-position 21 itself, for example, expressed as coordinates of a GNSS (Global Navigation Satellite System), for example, of a GPS (Global Positioning System). In addition, a measured value 15 may optionally also be obtained, for example, from another sensor of the motor vehicle 13, for example a temperature value and/or an air quality value and/or a brightness value, to mention as examples. For the sake of clarity, only one data set 22 is completely provided with reference signs in FIG. 2. The temporal sequence in which the motor vehicle 13 has passed the geo-positions 21 in succession may result in the order or sequence of the data sets 22 in the graph 16. Such a graph 16 may be represented by sensor data 11 and may be received by the processor circuit 10 from the motor vehicle 13 (for example via an Internet connection) and, together with further graphs 16 for further journeys and/or from further motor vehicles, may be converted into the model data 18 for the road map 12 by the processor circuit 10 in the described manner by the algorithm 17. In this case, the individual measured values may then be combined in order to obtain description values of the road 19 and/or of a further environmental property therefrom, which description values represent, for example, a parameterized line or another format known per se. The individual measured values 15 may then actually no longer be necessary for the road map 12.

[0037] The algorithm 17 described in connection with FIG. 1 and as may be known per se from the related art may only be able to calculate a road map 12 on the basis of raw data, that is to say the graphs 16 containing the concatenated data sets or precisely on the basis of the measured values 15. If there is already a finished road map 12, no further additional new information from measured values 15 received at a later time can consequently be integrated in these map data or these model data 18 relating to the road map 12 by the algorithm 17 alone.

[0038] FIG. 3 illustrates how an updated road map 12′ containing updated model data 18′ can now be calculated on the basis of the model data 18 relating to the road map 12 (already existing description values) if new additional sensor data 11′ containing additional graphs 16′ of concatenated data sets, that is to say additional measured values, arrive or are received.

[0039] The algorithm 17 itself may need not be adapted or changed for this purpose. Rather, the processor circuit 10 can use the graphs 16′ and can combine the graph 16′ with artificially generated virtual measured values 15′ which are combined to form a graph 24 which can be structured in the same manner as the graph 16 described in FIG. 2, that is to say the graph 24 need not differ from the newly received graphs 16′ in terms of data structure. For this purpose, the processor circuit 10, in a first operation S10, can operate a simulation module 25 which can simulate a virtual journey 27 on the roads 26 which have already been mapped on the basis of the already existing road map 12, that is to say the road map’s 12 model data 18 with the description values contained therein, by respectively generating or carrying out at least one artificial measured value 15′, that is to say a virtual measurement, for different virtual geo-positions 28. The simulation module 25 may be a computer program. In this case, the respective artificial measured value 15′ describes the respective description value of the respective environmental property which has already been mapped in the road map 12, which description value is stored or described in the model data 18 relating to the road map 12. The described interpolation may also be provided.

[0040] The result is therefore artificial measured values 15′ which can be combined to form a graph 16 of concatenated data sets in the manner described in FIG. 2. Therefore, these artificially generated graphs 24 also have the structure which can also be generated by a motor vehicle 13 on a real journey. For the algorithm 17, there is therefore no difference between new information and the information which has already been mapped, and, in operation S11, the graphs 24 and the newly received graphs 16′ can be processed together in the same manner by the algorithm 17, thus generating the model data 18′ relating to the updated road map 12′. For this purpose, it may not be necessary to hold the graphs 16 originally taken as a basis (see FIG. 1). These may be deleted from the processor circuit 10 after the first version of the road map 12 has been generated.

[0041] The road map 12 and the updated road map 12′ may each be made available to at least one motor vehicle using a road map in operation S12 after the road maps 12, 12′ have been finished. Measured values 15 can therefore be iteratively or repeatedly received from motor vehicles 13, may be processed iteratively or in succession by the processor circuit 10, and respective updated road maps 12′ can be output or transmitted to motor vehicles using the updated road maps 12′. A continuous or stepwise update of the road map 12 in the motor vehicles using the updated road maps 12′ on the basis of measuring motor vehicles 13 can therefore be continuously operated or enabled. The motor vehicles 29 using the updated road maps 12′ may therefore be coupled to the measuring motor vehicles 13 via the processor circuit 10 and may be supplied with updated model data 18′ relating to an updated road map 12′ in stages or steps.

[0042] This makes it possible for many vehicle manufacturers to create map material with their own observations of the vehicle sensor set (sensor circuits of the measuring motor vehicles), to keep the map material up-to-date or to update changes in order to create a new version of a base map, for example. The observations from the sensor environment are converted into a graph representation by way of a processing. However, for navigation assistance in using motor vehicles, the information to be updated must be in a format which can be interpreted by the calculating entity of the base map (road map to be updated). The tried and tested mapping process can be gathered from FIG. 1. This FIG. 1 process can be assumed to be the starting point for the examples described below.

[0043] In order to generate the model of this base map, it may be necessary to process the vehicle sensor information of different origins (for example different vehicle manufacturers, derivatives, to mention examples) in the map section to be calculated. This situation is due to the correspondence search which constitutes the basis for determining identity factors between the items of information which appear. In this case, it may now no longer be necessary to recurrently resort to established data sets or subsets of the latter. This saves an enormous volume of resources (memory, RAM and CPU load). In addition, a massive amount of computing effort and time, which needs to be expended to resolve the correspondence search and the optimization problem (SLAM method), may be prevented. In addition, from the point of view of data protection law, the purpose limitation of personal information is taken into account since the data set concerned can be deleted after initial processing.

[0044] In order to minimize the amount of computing effort, time, resources and complexity, a conversion of the calculated initial map which already exists (own generated maps or from third-party providers) into a graph representation is provided. The examples describe how virtual observations and/or virtual geo-positions, which constitute the basis for virtual journeys, can be created as an input from a base map. These virtual variables may then be converted into a graph structure, which may includes virtual journeys which have already been processed and/or optimized. The corresponding processes can be gathered from FIG. 3.

[0045] On the basis of the described examples, new sessions (data relating to an added recording journey) can be individually and directly processed, which may significantly accelerate the correspondence search and the subsequent graph optimization and therefore enormously saves resources. In addition, the raw data could be deleted after processing with the overall graph since the raw data may no longer be required to calculate a map model again.

[0046] As already described, the virtual graph variables are generated from an imported map model. The generated variables may be fixed during graph optimization. The generated model can therefore be considered to be ground truth. It is particularly advantageous to ensure that a sufficient information content of a generated graph is available for the respective correspondence search. The following elements or operations can be used for the generation: [0047] import of reference map model according to the data structure of the calculating entity/library; [0048] interpolation of polyline geometries in the case of continuous objects such as polylines (linear observations exist only from very few supporting points, with the result that description values are now generated here for additional virtual geo-positions); [0049] calculation of virtual geo-positions by sampling a reference path; [0050] calculation of virtual observed objects; [0051] linking of virtual geo-positions to associated virtual objects.

[0052] Overall, the examples show how a map model can be converted into a graph representation.

[0053] A description has been provided with particular reference to examples, but it will be understood that variations and modifications can be effected within the spirit and scope of the claims, which may include the phrase “at least one of A, B and C” as an alternative expression that refers to one or more of A, B or C, contrary to the holding in Superguide v. DIRECTV, 358 F3d 870, 69 USPQ2d 1865 (Fed. Cir. 2004).