Method for creating a digital street map

Abstract

The invention relates to a method for creating a digital road map, comprising the following steps: Receiving multiple data sets, where each of the multiple data sets describes a respective landmark-containing environment of multiple motor vehicles during a respective journey on the same road segment, Generating a composite dataset based on the multiple datasets, such that the generated composite dataset includes the multiple datasets; performing a SLAM optimization procedure multiple times based on the composite dataset, optimizing for different landmarks and / or landmark combinations each time to obtain respective optimization results. For each optimization result, create a digital landmark map based on the respective optimization result, whereby the landmarks included in the overall dataset are entered in the created landmark maps. Comparing the created digital landmark maps with each other, Processing the entire dataset based on the comparison to obtain a consistent overall dataset, Creating the digital street map based on the consistent overall dataset. The invention relates to a device, a computer program and a machine-readable storage medium.

Description

[0001] The invention relates to a method for creating a digital road map, a device, a computer program and a machine-readable storage medium. State of the art

[0002] The published patent application DE 10 2021 204 723 A1 discloses a method for providing a future drivable trajectory for an at least partially assisted motor vehicle by means of an assistance system.

[0003] Patent EP 3 563 265 B1 discloses an online system that creates a high-resolution map for a geographical region based on sensor data collected by a plurality of autonomous vehicles.

[0004] The published patent application US 2019 / 0390965 A1 discloses a method for detecting defects in map data. Disclosure of the invention

[0005] The object underlying the invention is to provide a concept for the efficient creation of a digital road map.

[0006] This object is achieved by means of the respective subject matter of the independent claims. Advantageous embodiments of the invention are the subject matter of the respective dependent subclaims.

[0007] According to a first aspect, a method for creating a digital road map is provided, comprising the following steps: Receiving a plurality of data sets, wherein the plurality of data sets each describe a respective landmark-containing environment of a plurality of motor vehicles during a respective journey on the same road section, Generating an overall data set based on the multiple data sets, so that the generated overall data set comprises the multiple data sets, performing a SLAM optimization process multiple times based on the overall data set, each time optimizing for different landmarks and / or landmark combinations to obtain respective optimization results, For each optimization result, create a digital landmark map based on the respective optimization result, with the landmark maps created containing the landmarks included in the overall data set, Comparing the created digital landmark maps with each other, processing the overall data set based on the comparison to obtain a consistent overall data set, Creating the digital road map based on the consistent overall data set.

[0008] According to a second aspect, a device is provided which is configured to carry out all the steps of the method according to the first aspect.

[0009] According to a third aspect, a computer program is provided, comprising instructions which, when the computer program is executed by a computer, for example by the device according to the second aspect, cause the computer to carry out a method according to the first aspect.

[0010] According to a fourth aspect, a machine-readable storage medium is provided on which the computer program according to the third aspect is stored.

[0011] The invention is based on the idea that in the real world of man-made environments, such as road traffic, changes in reality generally exhibit recurring characteristics. Changes are, for example, spatially and locally limited. Changes affect individual landmark types, for example, and not all landmarks at once. Changes occur, for example, during a construction project. Changes are therefore not gradual changes, but rather are, for example, an event at a fixed time.

[0012] Some examples: A traffic light system at an intersection is being replaced, while the road geometry remains unchanged.

[0013] A guardrail will be replaced, but the lane markings will remain.

[0014] Lane markings will be renewed, but guardrails and signs will remain.

[0015] The idea of ​​the concept described here is to systematically use these change properties to detect changes and generate a digital road map that is as error-free as possible.

[0016] The concept described here is based, in particular, on repeatedly performing SLAM optimization with different combinations of landmarks. SLAM stands for "Simultaneous Localization and Mapping," which translates into German as "simultaneous positioning and mapping."

[0017] In the previously known state-of-the-art optimization, all landmarks LM1, LM2, and LM3 are optimized. After the optimization step, a digital landmark map is generated in which all landmarks are entered.

[0018] This well-known approach fails to adequately detect changes in reality. The resulting landmark maps are therefore often incorrect.

[0019] In contrast, the concept described here proposes performing SLAM optimization multiple times for different combinations of landmarks, thus creating multiple digital landmark maps. After performing SLAM optimization multiple times, it is specifically planned to generate landmark maps based on the optimization results and compare them with each other. This comparison can detect or identify inconsistencies in the input data caused by changes in reality. This detection of inconsistencies is not possible due to the state of the art, which only performs SLAM optimization for all landmarks at once. Using this information, the comparison results, the entire data set used to create the digital road map can be made consistent. A consistent digital road map can then be generated based on the consistent data set.

[0020] This allows a digital road map to be generated or created that is as error-free as possible. This can advantageously increase the availability of driving functions of a motor vehicle that utilize such a digital road map.

[0021] In particular, the technical advantage is that a concept for the efficient creation of a digital road map is provided.

[0022] In one embodiment of the method, it is provided that the comparison of the created digital landmark maps with each other comprises a respective linking of at least two of the created digital landmark maps as operands by one or more comparison operators, wherein the overall data set is processed based on a result of the linking.

[0023] This provides the technical advantage, for example, that comparisons can be carried out efficiently.

[0024] In one embodiment of the method, it is provided that the one or more relational operators are each an element selected from the following group of relational operators: Landmark scatter comparison operator, Landmark existence comparison operator, Landmark shift comparison operator, semantic comparison operator.

[0025] This provides, for example, the technical advantage that particularly suitable comparison operators can be provided.

[0026] A landmark dispersion comparison operator compares a dispersion of the landmarks, in particular a dispersion of a position of the landmarks.

[0027] A landmark existence comparison operator compares the existence of the comparison operators, i.e. whether a landmark in one digital landmark map is found in another digital landmark map.

[0028] A landmark displacement comparison operator compares in particular a displacement of the landmarks relative to each other.

[0029] A semantic comparison operator compares in particular a semantics of the landmarks, i.e. a meaning of the landmarks.

[0030] In one embodiment of the method, it is provided that one or more changes in the road section are detected based on the comparison of the created digital landmark maps with each other, wherein the overall data set is processed based on the one or more detected changes.

[0031] This provides the technical advantage, for example, that changes in the road section can be detected efficiently.

[0032] In one embodiment of the method, it is provided that the one or more detected changes are described using one or more parameters which are selected from the following group of parameters: time of the change, spatial extent of the change, location of the change, affected landmark of the change.

[0033] This provides the technical advantage, for example, that particularly suitable changes can be detected.

[0034] In one embodiment of the method, it is provided that the processing of the overall data set comprises discarding a data set and / or a landmark of a data set that describe a reality before a detected change, so that the consistent overall data set no longer includes the discarded data set.

[0035] This provides, for example, the technical advantage that a consistent overall data set can be generated efficiently. In particular, this provides the technical advantage that the consistent overall data set contains up-to-date data.

[0036] In one embodiment of the method, it is provided that the creation of the digital road map comprises performing a SLAM optimization method based on the consistent overall data set, wherein all landmarks of the consistent overall data set are optimized to obtain an optimization result, wherein the digital road map is created based on the optimization result.

[0037] This provides the technical advantage, for example, that the digital road map can be created efficiently.

[0038] For example, the device is programmed to execute the computer program.

[0039] The method is carried out, for example, by means of the device.

[0040] Device features result analogously from process features, and vice versa. This means, in particular, that technical functionalities of the process result analogously from corresponding technical functionalities of the device, and vice versa.

[0041] For example, the method is a computer-implemented method.

[0042] The embodiments and forms of embodiment described here can be combined with each other in any way, even if this is not explicitly described.

[0043] The invention is explained in more detail below using preferred embodiments. These show: Fig. 1 a flowchart of a method according to the first aspect, Fig. 2 a device according to the second aspect and Fig. 3 a machine-readable storage medium according to the fourth aspect.

[0044] Fig. 1 shows a flowchart of a method for creating a digital road map, comprising the following steps: Receiving 101 a plurality of data sets, wherein the plurality of data sets each describe a respective landmark-containing environment of a plurality of motor vehicles during a respective journey on a same road section, Creating 103 an overall data set based on the plurality of data sets, such that the overall data set created comprises the plurality of data sets, performing 105 a SLAM optimization method based on the entire data set several times, each time optimizing for different landmarks and / or landmark combinations in order to obtain respective optimization results, for each optimization result, creating 107 a digital landmark map based on the respective optimization result, whereby the landmark maps created contain the landmarks included in the overall data set, Comparing 109 the created digital landmark maps with each other, Processing 111 the overall data set based on the comparison to obtain a consistent overall data set, Creating 113 the digital road map based on the consistent overall dataset.

[0045] Fig. 2 shows a device 201 which is configured to carry out all steps of the method according to the first aspect.

[0046] The device 201 comprises, for example, an input configured to receive the plurality of data sets.

[0047] The device 201 comprises, for example, a processor device, which may comprise one or more processors. The processor device is configured, for example, to perform the steps of generating the overall data set, repeatedly performing a SLAM optimization method, creating a digital landmark map for each optimization result, comparing the created digital landmark maps with each other, processing the overall data set, and creating the digital road map based on the consistent overall data set.

[0048] The device 201 comprises, for example, an output which is configured to output the created digital road map.

[0049] The method according to the first aspect comprises, for example, outputting the created digital road map.

[0050] Fig.3 shows a machine-readable storage medium 301 on which a computer program 303 is stored. The computer program 303 includes instructions that, when executed by a computer, cause the computer program 303 to execute a method according to the first aspect.

[0051] In a SLAM optimization process, for example, environmental sensor data from the environmental sensors (hereinafter also simply abbreviated as sensors) from a motor vehicle are simultaneously incorporated into the optimization across all data sets from multiple vehicles. Environmental sensor data, for example, describes the respective environment of the motor vehicles. Environmental sensor data, for example, describes one or more landmarks in the environment of the respective motor vehicles. Environmental sensors as defined in this description include, for example, the following environmental sensors: radar sensor, lidar sensor, image sensor of a video camera, ultrasonic sensor, infrared sensor, and magnetic field sensor.

[0052] The following explanations are intended to illustrate this by way of example: A total data set or source data set comprises or consists of D individual data sets originating from several motor vehicles: Data set 1 from vehicle 1 (time 1) Sensor1 (S1): Video camera provides the landmarks (LM) LM1: Lane markings LM2: Signs Sensor2 (S2): Radar sensor provides the landmarks LM3: Guardrails Data set 2 from vehicle 2 (time 2) Sensor1 (S1): Video camera provides the landmarks (LM) LM1: Lane markings LM2: Signs Sensor2 (S2): Radar sensor provides the landmarks LM3: Guardrails Data set DF from motor vehicle F (time Z) Sensor1 (S1): Video camera provides the landmarks (LM) LM1: Lane markings LM2: Signs Sensor2 (S2): Radar sensor provides the landmarks LM3: Guardrails

[0053] The method described here includes, for example, one or more of the following steps: Creating a single data set, including multiple data sets from multiple vehicles. Data sets from multiple vehicles are collected.

[0054] A SLAM optimization is performed multiple times (performing several SLAM optimization procedures) and in each case different combinations of landmarks are optimized, e.g.: Optimize LM1+LM2+LM3 Optimize LM1 Optimize LM2 Optimize LM3 Optimize LM1+LM2 Optimize LM1+LM3 Optimize LM2+LM3

[0055] Digital landmark maps are generated: Map 1: LM1+2+3 map based on LM1+2+3 optimization result Map 2: LM1+2+3 map based on LM1 optimization result Map K: LM1+2+3 map based on LM2+LM3 optimization result

[0056] The digital landmark maps are compared: For example, a distribution of landmarks LM1 in map 1 is compared with a distribution of landmarks LM1 in map 2.

[0057] Changes are detected, whereby the detected changes are described using parameters: Time of change, Spatial extent of the change, affected road section, Affected landmarks of the change.

[0058] The overall data set is made consistent, i.e. a consistent overall data set is determined: For example, data sets and / or landmarks that describe reality before a change are discarded.

[0059] The mapping process is repeated, creating a digital road map: based on the consistent data set, where all landmarks are used in the optimization (for example LM1+LM2+LM3).

[0060] For comparing digital landmark maps, a relational operator that compares the dispersion of landmarks is also a suitable measure. Other relational operators can be used, for example, a landmark existence comparison operator, a landmark displacement comparison operator, and a semantic comparison operator.

[0061] The concept described here advantageously creates a consistent overall data set on the basis of which the digital road map can be created so that it can be used for driving functions of a motor vehicle in a particularly reliable manner. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2021 204 723 A1

[0002] EP 3 563 265 B1

[0003] US 2019 / 0390965 A1

[0004]

Claims

[1] A method for creating a digital road map, comprising the following steps: Receiving (101) a plurality of data sets, wherein the plurality of data sets each describe a respective landmark-containing environment of a plurality of motor vehicles during a respective journey on a same road section, generating (103) an overall data set based on the plurality of data sets, such that the generated overall data set comprises the plurality of data sets, performing (105) a SLAM optimization method based on the entire data set several times, each time optimizing for different landmarks and / or landmark combinations in order to obtain respective optimization results, for each optimization result, creating (107) a digital landmark map based on the respective optimization result, wherein the landmark maps created contain the landmarks included in the overall data set, Comparing (109) the created digital landmark maps with each other, Processing (111) the overall data set based on the comparison to obtain a consistent overall data set, Creating (113) the digital road map based on the consistent overall data set. [2] The method according to claim 1, wherein comparing the created digital landmark maps with each other comprises linking at least two of the created digital landmark maps as operands by one or more comparison operators, wherein the overall data set is processed based on a result of the linking. [3] The method of claim 2, wherein the one or more comparison operators are each an element selected from the following group of comparison operators: landmark scattering comparison operator, landmark existence comparison operator, landmark displacement comparison operator, semantic comparison operator. [4] Method according to one of the preceding claims, wherein one or more changes in the road section are detected based on the comparison of the created digital landmark maps with each other, wherein the overall data set is processed based on the one or more detected changes. [5] Method according to claim 4, wherein the one or more detected changes are described using one or more parameters selected from the following group of parameters: time of the change, spatial extent of the change, location of the change, affected landmark of the change. [6] The method of claim 4 or 5, wherein processing the overall data set comprises discarding a data set and / or a landmark of a data set that describe a reality before a detected change, so that the consistent overall data set no longer includes the discarded data set. [7] Method according to one of the preceding claims, wherein the creation of the digital road map comprises performing a SLAM optimization method based on the consistent overall data set, optimizing for all landmarks of the consistent overall data set to obtain an optimization result, wherein the digital road map is created based on the optimization result. [8] Device (201) which is arranged to carry out all the steps of the method according to one of the preceding claims. [9] Computer program (303) comprising instructions which, when the computer program (303) is executed by a computer, cause the computer to carry out a method according to one of claims 1 to 7. [10] Machine-readable storage medium (301) on which the digital road map created according to one of claims 1 to 7 and / or the computer program (303) according to claim 9 are stored.

Images