Method for creating a digital map for vehicle navigation

Abstract

A method for creating a digital map for vehicle navigation has been disclosed, comprising the following steps: K1 - Recording a variety of map information, where the map information relates to elements and / or attributes to be mapped; K2 - Recording and / or evaluating contextual information regarding individual elements and attributes recorded in step K1, whereby the contextual information assists vehicle navigation systems in performing a plausibility check of the map information based on the corresponding elements and / or attributes; K3 - Creating the digital map at least partially based on the map information, integrating the context information into the digital map.

Description

[0001] The invention relates to a method for creating a digital map for vehicle navigation. The invention further relates to a method for operating a motor vehicle, a system for operating a motor vehicle, and a navigation map for a motor vehicle. State of the art

[0002] In the context of assisted driving, map information from digital maps is increasingly used in safety-related functions. It should be noted that the digital maps themselves, as well as the vehicle sensors' ability to locate the mapped elements, are inherently susceptible to various errors that can have a safety-related impact on the vehicle navigation system based on this map information.

[0003] As a consequence of this, security concepts must be developed to minimize the risks associated with faulty maps. One such security concept involves performing plausibility checks to verify the determined vehicle position and the map data of the digital map.

[0004] The current state of the art provides no indication of the extent to which potentially prevailing local environmental conditions, which may have a negative, limiting influence on perception, may also negatively affect the plausibility of the map information, and how such negative and limiting influences can be circumvented. In the present disclosure, these influences from environmental conditions on perception are also referred to as the "context" of map information.

[0005] It is an object of the invention to provide an alternative or improved method for creating a digital map for vehicle navigation, as well as a method for operating a motor vehicle, a system for operating a motor vehicle, and a navigation map for a motor vehicle. Disclosure of the invention

[0006] The problem of the invention is solved by means of a method according to claim 1, by a method according to claim 11, by a system according to claim 18, and by a navigation map according to claim 23. Advantageous further developments, additional features and / or advantages of the invention will become apparent from the dependent claims and the following description.

[0007] It should be noted that all features mentioned in connection with the disclosed process can also be configurations of the disclosed system, and vice versa.

[0008] According to a first aspect, the present disclosure discloses a method for creating a digital map for vehicle navigation, comprising the following steps: K1 - Recording a variety of map information, where the map information relates to elements and / or attributes to be mapped; K2 - Recording and / or evaluating contextual information regarding individual elements and attributes recorded in step K1, whereby the contextual information assists vehicle navigation systems in performing a plausibility check of the map information based on the corresponding elements and / or attributes; K3 - Creating the digital map at least partially based on the map information, integrating the context information into the digital map.

[0009] The map created using the above method can advantageously support driver assistance systems in performing plausibility checks of the map information, thus increasing safety during assisted driving. Local environmental conditions that may have a negative, limiting influence on perception can be taken into account in the contextual information and circumvented.

[0010] The context information may contain instructions and / or information for performing the plausibility check of the map information based on the elements and / or attributes corresponding to the context information.

[0011] The contextual information may include physical models, in particular information on the effects of environmental influences such as weather patterns on the execution of the plausibility check.

[0012] The contextual information may include information on alternative methods for performing the plausibility check of the map information based on the relevant elements and / or attributes.

[0013] Step K2 may include evaluating the context information on a central computing unit.

[0014] The evaluation of contextual information can be based, at least in part, on data obtained through crowdsourcing.

[0015] The process may include the following step: K2b - Recording and / or evaluating metadata relating to the map information, wherein the metadata represents a reliability assessment of individual map information recorded in step K1 and / or a reliability assessment of a detection of the elements and / or attributes corresponding to the individual map information by a vehicle sensor system; wherein step K3 of creating the digital map includes integrating the metadata into the digital map.

[0016] The metadata can also be referred to as "confidence information." This indicates, more or less, how reliably the mapmaker assesses the map information itself. Limitations in reliability can arise, for example, from uncertainties in the data of the vehicle fleet or its sensors, from uncertainties regarding the satellite data used, from uncertainties in the calculation of the map data from the aforementioned information, or from uncertainties resulting from the fact that the data recording took place a long time ago and the mapped features may have changed since then, even unnoticed.

[0017] Step K2b may include evaluating the security-related meta-information on a central computing unit.

[0018] The evaluation of meta-information can be based, at least in part, on data obtained through crowdsourcing.

[0019] Step K2b may include determining an expected confidence of a detection of a mapped element and / or attribute, where the expected confidence is in particular a function of a confidence during the acquisition of the relevant map information in step K1.

[0020] According to another aspect, the present disclosure discloses a method for operating a motor vehicle, wherein the motor vehicle comprises a control unit for automated driving of the motor vehicle, the method comprising the following steps: S1 - Providing a digital map, wherein the digital map contains map information and contextual information about elements and / or attributes mapped in the digital map; S2 - Determining a current geographical position of the motor vehicle and identifying elements and / or attributes in a vehicle environment, which can be used to perform a plausibility check of at least selected map information and / or the geographical position; S3 - Check if context information is available for the elements and / or attributes identified in step S2, and make the relevant context information available; S4 - Perform the plausibility check using the elements and / or attributes identified in step S2, at least partially using the context information corresponding to the identified elements and / or attributes.

[0021] The aforementioned procedure offers the advantage, among others, that misjudgments during the plausibility check of the digital map under adverse conditions, such as lighting conditions, weather, and visibility restrictions, become less likely due to the contextual information, which includes situation-specific, geolocal plausibility information. This ultimately represents a safety improvement in assisted driving.

[0022] A further advantage of the aforementioned method, which increases safety in assisted driving, is that the additional information can also be used to predict that map information in a specific area cannot be reliably validated under current conditions, and that an alternative strategy should be chosen in this case. Examples of an alternative strategy include driving around a particular location or adjusting driving maneuvers to dynamically reduce the safety requirements for the map information.

[0023] Furthermore, taking into account contextual information that strengthens the hypothesis makes a successful, i.e., positive and correct, plausibility check more likely, which means that the availability of the plausibility check is increased.

[0024] Another advantage is that the effort required to test "corner cases" for map plausibility checks is reduced, since "edge cases" can be explicitly addressed by a model that is validatable in itself.

[0025] The context information may contain instructions and / or information for performing the plausibility check of the map information based on the elements and / or attributes corresponding to the context information.

[0026] The contextual information may include physical models, in particular information on the effects of environmental influences such as weather patterns on the execution of the plausibility check.

[0027] The contextual information may include information on alternative methods for performing the plausibility check of the map information based on the relevant elements and / or attributes.

[0028] The process may include the following step: S6 - In the event that the plausibility check in step S4 results in the map information not being plausible, the plausibility check is carried out using an alternative plausibility check procedure.

[0029] The process may include the following step: S5a - If the plausibility check in step S4 results in the map information being plausible, terminate the procedure.

[0030] The process may include the following step: S5b - In the event that the plausibility check in step S4 results in the map information being implausible, an error message is displayed.

[0031] According to another aspect, the present disclosure discloses a system for operating a motor vehicle, wherein the system comprises the following: a localization system; a memory in which a digital map is stored containing map information and contextual information about elements and / or attributes mapped in the digital map; an electronic control unit designed to receive the vehicle's current geographical position from the localization system; to identify elements and / or attributes in a vehicle's environment that can be used to perform a plausibility check of at least selected map information and / or the current geographic position; and to extract contextual information about the identified elements and / or attributes from the digital map; and to perform the plausibility check based on the identified elements and / or attributes, at least partially using the context information corresponding to the elements and / or attributes.

[0032] Contextual information is provided in a geographically restricted manner and / or only for map information for which plausibility verification is difficult and / or has proven to be difficult.

[0033] Contextual information can be provided dynamically, particularly via V2X and / or retrieval via an IP protocol-based service.

[0034] This has the advantage of reducing the amount of data in the digital map.

[0035] Contextual information can be provided in advance of entering a zone where the corresponding map information is relevant.

[0036] Contextual information can be provided depending on environmental conditions in a zone where the corresponding map information is relevant, preferably depending on weather conditions.

[0037] According to another aspect, the present disclosure reveals a navigation map for a motor vehicle, comprising: Map information for vehicle navigation regarding a variety of elements and / or attributes; Contextual information regarding the multitude of elements and / or attributes.

[0038] The context information may contain instructions and / or information for performing the plausibility check of the map information based on the elements and / or attributes corresponding to the context information.

[0039] The contextual information may include physical models, in particular information on the effects of environmental influences such as weather patterns on the execution of the plausibility check.

[0040] The contextual information may include information on alternative methods for performing the plausibility check of the map information based on the relevant elements and / or attributes. Brief description of the characters

[0041] The invention is explained in more detail below with reference to exemplary embodiments and the accompanying schematic drawing, which is not to scale. The figures (Fig.) in the drawing are merely examples and show: Fig. 1 schematically an embodiment of a method for creating a digital map for vehicle navigation; Fig. 2. Schematically, an embodiment of a method for operating a motor vehicle; and Fig. 3 schematically an embodiment of a method for plausibility check of a digital map in a motor vehicle.

[0042] Based on the Fig. 1, Fig. 2 to Fig.Section 3 below schematically describes the structure and function of a method for creating a digital map for vehicle navigation, a navigation map, as well as a method and a system for operating a motor vehicle. Corresponding reference symbols are used for corresponding features.

[0043] Fig.Figure 1 schematically illustrates the process for creating a digital map for vehicle navigation. In step K1, a large amount of map information is acquired, relating to elements and / or attributes to be mapped. Elements to be mapped can include, for example, lane markings, stop lines, traffic lights, and intersections. Corresponding attributes can include, for example, dimensions, positions, distances, and intervals. The acquisition of the map information can be carried out, for example, in a known manner, such as through crowdsourcing using suitable sensors from a vehicle fleet.

[0044] In step K2, contextual information regarding individual elements and attributes recorded in step K1 is acquired and / or evaluated. This contextual information assists vehicle navigation systems in performing a plausibility check of the map information based on the corresponding elements and / or attributes.

[0045] A plausibility check is advisable for security reasons, as maps can, in principle, contain errors, as can the recording of the mapped elements. Table 1 gives some examples of different types of errors in this regard. Table 1 Error mode Examples Mapped element is erroneously detected as positive ("false positive") Lane boundaries, relationship between lane segments, stop lines Mapped element is erroneously Lane boundaries, stop lines, 3D boundary of traffic lights, local traffic regulations not detected ("false negatives") Wrong position Lane boundaries, 3D outlines of lighting systems, masts Incorrect element type Lane type (passable / non-passable, cycle path, etc.), edge of road or carriageway, line type Incorrect value Lane number, direction of traffic flow, speed limit Incorrect assignment Adjacent roadway segments, number of the stop line assigned to a traffic light system

[0046] The contextual information can relate in particular to adverse conditions for the perception of objects and information stored in the map attributes. Examples of adverse conditions include: • Weather conditions, for example rain, snow, fog; • Weather-related phenomena, for example wet roads, snow-covered roads or objects; • Weather-related effects, for example due to wind; • Obscurations, for example by vegetation - this can also be dependent on the season; • Glare from the sun and reflections, primarily in relation to the vehicle's optical sensors; • Glare from sensors of other vehicles, primarily in relation to radar and lidar sensors; • Mechanical influences, for example from bumpy road surfaces; • Viewing angle, for example due to different sensor installations on the vehicle.

[0047] In some embodiments, the context information contains instructions and / or information for performing the plausibility check of the map information using the elements and / or attributes corresponding to the context information. The context information may also include physical models, in particular information on the effects of environmental influences such as weather patterns on the plausibility check.

[0048] The following list contains specific examples of contextual information relating to selected map information. In specific embodiments of a digital map, the information formulated as instructions or statements within the scope of this disclosure is stored in the digital map in a suitable machine-readable manner so that a driver assistance system can read and evaluate the contextual information: a. “If the mapped road section is wet, reflections must be expected”; b. “if the vehicle sensor is installed at a height of < x cm, the mapped object is difficult to detect”; c. “If the sun hits the light at a certain angle, disturbing reflections occur, so that the mapped object cannot be detected by a video sensor”; d. “If there is a strong wind, the mapped traffic light will move, so a different algorithm must be used for plausibility checks”; e. “In case of storm, the position of the mapped traffic light changes, and an alternative position and orientation of the traffic light is <... >” f. “If it is night and / or dark, then algorithm A should be used instead of B to verify the plausibility of the mapped object”; g. "If the mapped line is snow-covered, perform the plausibility check using:" i) another mapped line (with a known distance); ii) of the mapped roadside (with known distance); iii) of mapped overhead lines (with known spacing); iv) a clothoid between visible mapped landmark A and mapped landmark B"; h. “In summer the mapped object is hidden by a tree, while in winter it is clearly visible”; i. "If reflections occur on the mapped road due to wetness, then the position of the mapped stop line can be determined as follows: i) It is located 1.50m before the intersection (road edge); ii) Triangulation via two or more other (visible) landmarks (posts, etc.).

[0049] As can be seen from the examples d, e, f, g, and i above, the contextual information may include information on alternative procedures for performing the plausibility check of the map information using the relevant elements and / or attributes, or instructions on what to do if a plausibility check was unsuccessful.

[0050] Other examples of contextual information of this type include: a. "If the plausibility of the position of the mapped traffic light is only imprecisely achieved, this is not critical, since there is no other traffic light (e.g. further back) in the field of vision with which confusion would be possible"; b. "If a plausibility check of the traffic light fails, the trajectory of the ego vehicle should be based on the movement of other vehicles in the adjacent lane (optional: if the traffic density exceeds a certain threshold, establish contact with other vehicles via V2X)".

[0051] In step K3, the digital map is created at least partially based on the map information, with the context information being integrated into the digital map so that it is available for vehicle navigation.

[0052] Step K2 can include K2b, in which metadata relating to the map information is recorded and / or evaluated. This metadata represents a reliability assessment of individual map information recorded in step K1 and / or a reliability assessment of the detection of the elements and / or attributes corresponding to the individual map information by vehicle sensors. Step K3 of creating the digital map includes integrating the metadata into the digital map.

[0053] Based on the foregoing explanations, it can be understood that the method disclosed herein supplements the map information and metadata with generic or geolocal rules and models, which allow the currently prevailing conditions, such as SOTIF (Safety of Intended Functionality), weather conditions, lighting conditions, visibility, etc., to be taken into account in the plausibility check of the map information. In their combined consideration, the contextual information and the metadata thus enable a high degree of reliability for the map information, which overall increases the safety level of assisted driving.

[0054] The following will be based on Fig. 2. A method for operating a motor vehicle is described which includes a control unit for automated driving and utilizes the benefits of a digital map as explained above.

[0055] In step S1, a digital map is provided, containing map information and contextual information about the elements and / or attributes mapped in the digital map. The digital map may have been created using a method described above.

[0056] In step S2, the current geographic position of the vehicle is determined, for example, using a known method such as GPS (Global Positioning System). Furthermore, in step S2, elements and / or attributes in the vehicle's environment are identified, which can be used to perform a plausibility check of at least selected map information and / or the geographic position. Those skilled in the art will recognize that this identification depends, among other things, on the sensors available in the vehicle. The identification can be based either on a vehicle environment mapped by the sensors or on the information contained in the digital map. In the latter case, for example, a control unit of a driver assistance system uses the vehicle's own sensors to detect map attributes in the vehicle's environment.

[0057] In step S3, the control unit, for example, checks whether context information is available for the elements and / or attributes identified in step S2. If the identification of the elements and / or attributes to be used for plausibility checks in step S2 is based on the information contained in the digital map, step S3 is executed in parallel to step S2 by identifying the elements and / or attributes for the plausibility check based on the availability of the context information. After identifying these elements and / or attributes, the relevant context information is made available, for example, by loading it from the map and / or requesting it via an IP protocol-based service.

[0058] In step S4, a plausibility check is performed using the elements and / or attributes identified in step S2, at least partially using the contextual information corresponding to the identified elements and / or attributes. If the plausibility check concludes that the map information is plausible, the procedure terminates at step S5a. If the plausibility check concludes that the map information is implausible, an error message is issued in step S5b in embodiments of the procedure.

[0059] In embodiments of the method, if the plausibility check results in the map information being implausible, the plausibility check is performed in step S6 using an alternative method, the context information containing information for carrying out this alternative method. As in Fig. As indicated in 2, this can be done by using the information for carrying out the alternative procedure to execute the procedure again from step S2, i.e., if necessary, a new location determination is carried out and alternative map attributes are recognized in the vehicle environment, on the basis of which the alternative plausibility check is carried out.

[0060] Fig. Figure 3 shows an overview of the architecture of the security concept underlying the procedure.

[0061] Reference number 10 refers to the digital map or the information contained in the digital map. This includes, firstly, map information 12, i.e., the mapped elements and / or attributes. Furthermore, the map contains metadata 14, which, as described above, reflects reliability assessments of individual map information and / or reliability assessments of the detection of the elements and / or attributes corresponding to the individual map information by vehicle sensors.

[0062] Reference number 16 indicates the contextual information, which may include geolocal rules and physical models in section 17, as well as fallback positions in section 18.

[0063] As in Fig.As indicated by corresponding arrows, the map information 12 and the metadata 14 are used directly and in a manner known per se for a plausibility check 22, which is carried out within the framework of a reliability check of the digital map 10 designated by 20. For the plausibility check, a geographical position 30 of the ego-vehicle is used, as well as environmental perceptions 32 transmitted by suitable sensors, as is known per se.

[0064] The geolocal rules and physical models 17 contained in the context information 16 are presented as in the Fig.3, represented by corresponding reference numbers 16' and 17', are provided within the framework of the reliability check 20 and can be used, if necessary, in a refinement 23 of the plausibility check 22. The fallback positions 18' provided in the reliability check 20 can be used in 25 if an initial plausibility check 22 is unsuccessful or not possible due to the current geolocal environmental conditions.

[0065] At 40, the result is plausibility-checked map information with a defined security level.

[0066] The procedure described above can be used to distribute the security burden in vehicle navigation between the map provider and the user. This means that security requirements are placed on both map generation and provision, as well as on the phase in which the data is used by the respective driver assistance system. The map information is thus validated once again by the driver assistance system.

[0067] This approach avoids the need to place the full security burden on the card attributes, which would mean that the card data would have to be provided by the card provider in accordance with the required ASIL (Automotive Safety Integrity Level) and Confidence Level / Fidelity (in the sense of ISO 21448 for “SOTIF1”).

[0068] The invention is not limited to the described and illustrated embodiments. Rather, it also encompasses all further developments by skilled craftsmen within the scope of the invention defined by the claims. In addition to the described and illustrated embodiments, further embodiments are conceivable, which may include further modifications and combinations of features.

Claims

[1] Method for creating a digital map for vehicle navigation, comprising the following steps: K1 - Recording a variety of map information, where the map information relates to elements and / or attributes to be mapped; K2 - Recording and / or evaluating contextual information regarding individual elements and attributes recorded in step K1, whereby the contextual information assists vehicle navigation systems in performing a plausibility check of the map information based on the corresponding elements and / or attributes; K3 - Creating the digital map at least partially based on the map information, integrating the context information into the digital map. [2] Method according to claim 1, wherein the context information contains instructions and / or information for carrying out the plausibility check of the map information based on the elements and / or attributes corresponding to the context information. [3] Method according to claim 2, wherein the context information includes physical models, in particular information on the effects of environmental influences such as weather constellations on the execution of the plausibility check. [4] Method according to any one of claims 1 to 3, wherein the context information includes information on alternative methods for performing the plausibility check of the map information based on the relevant elements and / or attributes. [5] Method according to any one of claims 1 to 4, wherein step K2 comprises evaluating the context information on a central computing unit. [6] Method according to claim 5, wherein the evaluations of the context information are carried out at least partially on the basis of data obtained through crowdsourcing. [7] Method according to any one of claims 1 to 6, comprising the following step: K2b - Recording and / or evaluating meta-information regarding the map information, wherein the meta-information represents a reliability assessment of individual map information recorded in step K1 and / or a reliability assessment of a detection of the elements and / or attributes corresponding to the individual map information by a vehicle sensor system; Step K3 of creating the digital map includes integrating the meta-information into the digital map. [8] Method according to claim 7, wherein step K2b comprises evaluating the security-related meta-information on a central computing unit. [9] Method according to claim 8, wherein the evaluations of the meta-information are carried out at least partially on the basis of data obtained through crowdsourcing. [10] Method according to any one of claims 7 to 9, wherein step K2b comprises determining an expected confidence of a detection of a mapped element and / or attribute, and wherein the expected confidence is in particular a function of a confidence during the acquisition of the relevant map information in step K1. [11] Method for operating a motor vehicle, wherein the motor vehicle comprises a control unit for automated driving of the motor vehicle, the method comprising the following steps: S1 - Providing a digital map, wherein the digital map contains map information and contextual information about elements and / or attributes mapped in the digital map; S2 - Determining a current geographical position of the motor vehicle and identifying elements and / or attributes in a vehicle environment, which can be used to perform a plausibility check of at least selected map information and / or the geographical position; S3 - Check if context information is available for the elements and / or attributes identified in step S2, and make the relevant context information available; S4 - Perform the plausibility check using the elements and / or attributes identified in step S2, at least partially using the context information corresponding to the identified elements and / or attributes. [12] Method according to claim 11, wherein the context information contains instructions and / or information for performing the plausibility check of the map information based on the elements and / or attributes corresponding to the context information. [13] Method according to claim 12, wherein the context information includes physical models, in particular information on the effects of environmental influences such as weather constellations on the performance of the plausibility check. [14] Method according to any one of claims 11 to 13, wherein the context information includes information for alternative methods for performing the plausibility check of the map information based on the corresponding elements and / or attributes. [15] Method according to claim 14, comprising the following step: S6 - In the event that the plausibility check in step S4 results in the map information not being plausible, the plausibility check is carried out using an alternative plausibility check procedure. [16] Method according to any one of claims 11 to 15, comprising the following step: S5a - If the plausibility check in step S4 results in the map information being plausible, terminate the procedure. [17] Method according to any one of claims 11 to 16, comprising the following step: S5b - In the event that the plausibility check in step S4 results in the map information being implausible, an error message is displayed. [18] System for operating a motor vehicle, the system comprising: a localization system; a memory containing a digital map with map information and Contextual information about elements and / or attributes mapped in the digital map is stored; an electronic control unit designed to to receive the vehicle's current geographical position from the localization system; to identify elements and / or attributes in a vehicle's environment that can be used to perform a plausibility check of at least selected map information and / or the current geographic position; and to extract contextual information about the identified elements and / or attributes from the digital map; and to perform the plausibility check based on the identified elements and / or attributes, at least partially using the context information corresponding to the elements and / or attributes. [19] System according to claim 18, wherein the context information is provided in a geographically restricted manner and / or only for map information for which plausibility checks are difficult and / or have proven to be difficult. [20] System according to one of claims 18 and 19, wherein the context information is provided dynamically, in particular via V2X and / or retrieval via an IP protocol-based service. [21] System according to claim 20, wherein the context information is provided in time prior to entering a zone where the corresponding map information is relevant. [22] System according to one of claims 20 and 21, wherein the context information is provided depending on environmental conditions in a zone where the corresponding map information is relevant, preferably depending on weather conditions. [23] Navigation map for a motor vehicle, showing: Map information for vehicle navigation regarding a variety of elements and / or attributes; Contextual information regarding the multitude of elements and / or attributes. [24] Navigation map according to claim 23, wherein the context information includes instructions and / or information for performing the plausibility check of the map information based on the elements and / or attributes corresponding to the context information. [25] Navigation map according to claim 24, wherein the context information includes physical models, in particular information on the effects of environmental influences such as weather constellations on the execution of the plausibility check. [26] Navigation map according to one of claims 23 to 25, wherein the context information includes information for alternative methods for performing the plausibility check of the map information based on the corresponding elements and / or attributes.

Images