Method and system for generating a positioning map comprising at least one single feature

By collecting data and detecting the combination of single-sense features by fleet vehicles driving on designated traffic road sections, the ambiguity problem in the positioning of automated and auxiliary vehicle systems is solved, and the reliability and accuracy of positioning are improved.

CN120014865APending Publication Date: 2025-05-16ROBERT BOSCH GMBH
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202411633085.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2023-11-16
Filing Date
2024-11-15
Publication Date
2025-05-16

AI Technical Summary

Technical Problem

Existing automated and auxiliary vehicle systems face ambiguity problems when positioning, resulting in inaccurate positioning and low reliability.

Method used

Data is collected by traveling on designated traffic road sections by fleet vehicles, and a combination of single-sense features is detected using environmental sensing devices and stored in a positioning map to improve the reliability and accuracy of vehicle positioning.

Benefits of technology

The reliability and accuracy of the vehicle's own positioning is significantly improved, ensuring that only the combination of reliable detection of uni-sense features is used for positioning, and improving the overall performance of the system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120014865A_ABST
    Figure CN120014865A_ABST
Patent Text Reader

Abstract

The invention relates to a method for generating a positioning map for an automated and / or auxiliary vehicle system (2) for operating vehicles (3a, 3b), comprising at least one combination of undefined features, in which fleet data is collected by fleet vehicles (3b) of a fleet traveling on a traffic road section associated with the positioning map, a positioning map is generated from the collected fleet data and at least one individual feature combination is ascertained from the collected fleet data, based on the individual feature combination and the fleet data during each single travel on the traffic road section. It is determined whether the individual feature combination can be detected by the vehicle (3b) of the fleet or by the environment sensor (4) of the vehicle (3a) if the individual feature combination can be detected by the environment sensor (4), or by comparing the individual feature combination with a model of the sensor detection range of the vehicle (3a) which does not belong to the fleet along the traffic road section, or by comparing the individual feature combination with a model of the sensor detection range of the vehicle (3a) which does not belong to the fleet. And if so, storing the single-meaning feature combination in a temporary positioning map. The invention also relates to a corresponding system (1).
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to a method for generating a positioning map containing at least one unambiguous feature combination for operating an automated and / or auxiliary vehicle system of a vehicle, wherein fleet data are collected by driving fleet vehicles of the fleet on a traffic road section assigned to the positioning map, a positioning map is generated from the collected fleet data, and at least one unambiguous feature combination is determined from the collected fleet data. In addition, the invention also relates to a system for generating a positioning map containing at least one unambiguous feature combination for operating an automated and / or auxiliary vehicle system of a vehicle, the system comprising a plurality of fleet vehicles, each of which has at least one environmental sensor device, and at least one central data processing unit, which can be connected to the fleet vehicles for collecting fleet data by driving the fleet vehicles on a traffic road section assigned to the positioning map, and the central data processing unit is configured to generate a positioning map from the collected fleet data and to determine at least one unambiguous feature combination from the collected fleet data. Background Art

[0002] The detection of the vehicle's surroundings using the vehicle's own perception system based on environmental sensor devices, which can have, for example, radar sensors and / or video cameras, is an essential component of modern automated vehicle systems (AD systems) and auxiliary vehicle systems (DA systems). The surrounding environment model used here is often supplemented by additional map information, such as lane directions. The map data used for this purpose (for example, in the form of map products such as planning maps, behavior maps and / or positioning maps) can be provided to the corresponding vehicle system via a map service.

[0003] The determination of the vehicle position relative to the map used is the basis for the use of map data. Feature-based localization of the vehicle, in particular the ego-localization, enables an estimation of the vehicle position and orientation relative to the map, wherein the estimation is based on the localization map and vehicle-side radar feature measurements. In particular, ambiguities in the localization map are a challenge for real-time feature-based localization solutions.

[0004] Crowdsourcing of fleet data for the purpose of map creation is already known. For this purpose, fleet data are requested and transmitted for traffic road areas to be mapped and / or monitored without targeted selection. In addition, a large number of algorithms for creating, extending and updating positioning maps are known. In addition, solutions are known that identify local univocal feature combinations in positioning maps.

[0005] DE 10 2018 008 988 A1 relates to a method for updating map data, which are used for navigation of an autonomously operated motor vehicle. According to the method, a route segment in the environment of the motor vehicle is detected by an environmental sensor device of the motor vehicle in that detection data characterizing the route segment is generated by the environmental sensor device. Furthermore, if the detection data differ from the map data, a deviation data set is determined that characterizes the difference between the detection data and the map data with respect to the route segment. Furthermore, the deviation data set is transmitted to an external computing facility. Furthermore, the map data is updated based on the deviation data set with the aid of the external computing facility. Summary of the invention

[0006] The subject of the present invention is a method for generating a positioning map containing at least one unambiguous feature combination for operating an automated and / or auxiliary vehicle system of a vehicle. The reliability of a feature-based vehicle positioning system can be improved by this method, in that the unambiguous feature combination is processed in a new way and stored in a positioning map. In particular, by comparing the corresponding unambiguous feature combination with either "the fleet data at each single time of driving on the traffic road section" or "the modeling of the sensor detection range of vehicles not belonging to the fleet along the traffic road section" according to the present invention, only unambiguous feature combinations that can also be detected by the environmental sensor devices of the fleet vehicles or the environmental sensor devices of the vehicles not belonging to the fleet are selected and stored in the positioning map. Only such detectable unambiguous feature combinations can be used for positioning. Therefore, the vehicle's own positioning can be performed by the unambiguous feature combination that can be reliably detected, which significantly improves the reliability and accuracy of the own positioning. In contrast, in traditional self-positioning, the detectability of the corresponding unambiguous feature combination by the vehicle's environmental sensor device is not taken into account.

[0007] According to the method of the present invention, the corresponding univocal feature combination can be compared with the "convoy data at each single driving on the traffic road section", so that for each driving, it can be determined whether the univocal feature combination can be detected by means of the environmental sensor device of the convoy vehicle. In this optional scheme, it is assumed that the vehicle providing the convoy data (convoy vehicle) is also a vehicle using a positioning map with at least one univocal feature combination. In an alternative to this, the vehicle using the positioning map with at least one univocal feature combination can be different from the vehicle providing the convoy data, especially different in terms of the sensor detection range of their respective environmental sensor devices. In this case, at least one univocal feature combination is compared with the "modeling of the sensor detection range of the vehicle not belonging to the convoy along the traffic road section (for example, along the lane direction of the traffic road section)". Here, for example, a simple model of a vehicle not belonging to the convoy can be used. Here, the vehicle can be a vehicle of the convoy that is additionally constructed in terms of the corresponding sensor detection range, that is, a vehicle of such a convoy: the vehicle of the convoy is different from the vehicle of the convoy providing the convoy data in terms of the corresponding sensor detection range. The result of comparing the at least one univocal feature combination with the modeling of the sensor detection range of vehicles not belonging to the convoy along the traffic route section can then be used to generate a localization map that is valid for all vehicles of the additionally constructed convoy.

[0008] According to the method according to the invention, fleet data is first collected by driving the fleet vehicles of the fleet on a traffic road section assigned to a positioning map. In this case, the corresponding fleet vehicle is driven on the traffic road section so that data about the environment of the fleet vehicle can be detected with the help of an environmental sensor device. The environmental sensor device can have, for example, at least one camera and / or at least one radar sensor and / or at least one lidar sensor. The fleet vehicles can be designed to be identical, in particular with regard to their respective environmental sensor devices. In addition, the fleet vehicles can also be designed to be essentially identical in construction in other respects. The collected fleet data can, for example, be transmitted to a cloud system, that is, to a fleet cloud. The map data can, for example, be further transmitted from the fleet cloud to a fleet data map creation cloud and processed there, in particular in a conventional manner, to form a positioning map.

[0009] At least one unambiguous feature combination can then be determined from the fleet data, in particular from the fleet data associated with the positioning map or the traffic road section. This is described, for example, in the article "Comparison of 3D interest point detectors and descriptors for point cloud fusion", R. et al.,ISPRSAnnals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2014, Volume II-3, Pages 57-64.

[0010] A localization map which can be generated by the method according to the invention and which contains at least one unambiguous combination of features can be used in particular for operating an automated vehicle system (AD system) and / or an assistive vehicle system (DA system) of the vehicle.

[0011] According to an advantageous embodiment, based on the result of the comparison, it is determined from which lane or lanes of the traffic road section the environment sensor device can detect the univocal feature combination. For this purpose, information from the lane data layer of the generated positioning map can be used. If the individual trips on the traffic road section for generating the positioning map have been assigned to the individual lanes, this information is already available and can be used to determine from which lane or lanes of the traffic road section the environment sensor device can detect the univocal feature combination.

[0012] According to another advantageous configuration, using the detection rate determined as a result of the comparison and specific to the unambiguous feature combination and / or the visibility of the unambiguous feature combination from at least two different areas of the traffic road section determined as a result of the comparison and / or the real-time nature of the fleet data, at least one quality value is determined for the unambiguous feature combination and stored in the positioning map. The detection rate can be specified, for example, as a percentage or as a ratio: how many times the unambiguous feature combination was actually detected by the fleet vehicles during driving on the traffic road section. The visibility of the unambiguous feature combination can be specified, for example, by the number of lanes from which the unambiguous feature combination can be detected. The quality value is preferably continuous.

[0013] Another subject of the invention is a system for generating a localization map containing at least one univocal feature combination for operating an automated and / or auxiliary vehicle system of a vehicle, the system comprising a plurality of convoy vehicles, each of which has at least one environmental sensor device, and at least one central data processing unit, the central data processing unit being connectable to the convoy vehicles for collecting convoy data by driving the convoy vehicles on a road section assigned to the localization map, and the central data processing unit being configured to generate a localization map from the collected convoy data and to determine at least one univocal feature combination from the collected convoy data. The central data processing unit is configured to determine, based on the result of a comparison of the univocal feature combination with the convoy data at each single driving on the road section or with a modeling of the sensor detection range of a vehicle not belonging to the convoy along the road section, whether the convoy vehicle or the environmental sensor device of the vehicle can detect the univocal feature combination, and, if the environmental sensor device can detect the univocal feature combination, to store the univocal feature combination in a temporary localization map.

[0014] The advantages mentioned above with respect to the method are correspondingly associated with the system. Advantageous configurations of the method can correspond to advantageous configurations of the system. The central data processing unit can be cloud-based, for example.

[0015] According to an advantageous embodiment, the central data processing unit is configured to determine, based on the result of the comparison, from which lane or lanes of the traffic road section the environment sensor system can detect the univocal feature combination. This embodiment is correspondingly associated with the advantages mentioned above with respect to the corresponding embodiment of the method.

[0016] According to another advantageous embodiment, the central data processing unit is configured to determine at least one quality value for the univocal feature combination using the detection rate determined from the result of the comparison and specific to the univocal feature combination and / or the visibility of the univocal feature combination from at least two different areas of the traffic road section determined from the result of the comparison and / or the real-time nature of the fleet data, and to store the quality value in the positioning map. This embodiment is correspondingly associated with the advantages mentioned above with respect to the corresponding embodiment of the method.

[0017] According to another advantageous embodiment, the central data processing unit is configured to generate a separate detection request for each unambiguous feature combination and to send this detection request to the convoy vehicle in order to preferably or exclusively detect the corresponding unambiguous feature combination by means of the environmental sensor device of the corresponding convoy vehicle during driving on the traffic road section. By means of this detection request, also referred to as a work request, the detection of an unambiguous feature combination can be requested in a targeted manner. In this way, a high real-time performance and accuracy of the unambiguous feature combination can be ensured, in particular by means of a large number of feature statistics, while at the same time maintaining a low data rate. Thus, efficient crowdsourcing of the unambiguous feature combinations is possible.

[0018] According to another advantageous embodiment, the central data processing unit is configured to determine a route and trajectory plan for operating an automated and / or auxiliary vehicle system, with which a unique feature combination can be detected. This ensures that the surrounding sensor system can reliably detect the corresponding unique feature combination, in particular enabling precise localization of the vehicle.

[0019] According to another advantageous configuration, an automated and / or auxiliary vehicle system of at least one convoy vehicle of the system or at least one vehicle not belonging to the convoy is configured to perform self-localization using a localization map containing at least one unambiguous feature combination, and only such unambiguous feature combinations are considered here: their quality values ​​stored in the localization map are above a predetermined lower limit value. Therefore, the vehicle system can use this quality value in order to ensure a minimum quality of the unambiguous feature combination for a given application case, such as the initialization of self-localization. Therefore, this quality value can be taken into account when activating or initializing self-localization, so that, for example, only high-quality unambiguous feature combinations can be used in applications with high error rate requirements. This makes it possible to further improve the efficiency and reliability of self-localization. Therefore, the system can be used as part of a safety concept for feature-based vehicle self-localization. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The present invention is exemplarily described below based on preferred embodiments with reference to the drawings, wherein the features explained below may be considered individually or in different combinations of at least two of these features to demonstrate advantages and / or developments of the present invention.

[0021] The accompanying drawings show:

[0022] Figure 1 is a schematic representation of an embodiment of a system according to the present invention; and

[0023] Figure 2 is a flow chart of an embodiment of a method according to the present invention. DETAILED DESCRIPTION

[0024] In the accompanying drawings, the same reference numerals are provided for components that are identical or have the same functions. In detail, repeated descriptions of such components may be omitted to avoid unnecessary repetition.

[0025] Figure 1 A schematic representation of an exemplary embodiment of a system 1 according to the invention for generating a localization map containing at least one unambiguous feature combination for operating an automated and / or auxiliary vehicle system 2 of a vehicle 3 a or 3 b is shown.

[0026] The system 1 has a plurality of convoy vehicles 3b, each of which has at least one environmental sensor device 4. In addition, the system 1 has a central data processing unit 5, which can be connected to the convoy vehicles 3b in order to collect convoy data by the convoy vehicles 3b driving on traffic road sections assigned to the positioning map. The central data processing unit 5 is configured to generate a positioning map from the collected convoy data and to determine at least one unambiguous feature combination from the collected convoy data.

[0027] The central data processing unit 5 is also configured to determine whether the univocal feature combination can be detected by the environmental sensor device 4 of the convoy vehicle 3b or the vehicle 3a based on the result of comparing the univocal feature combination with "the convoy data at each single time of driving on the traffic road section" or with "the modeling of the sensor detection range along the traffic road section of the vehicle 3a that does not belong to the convoy", and if the environmental sensor device 4 can detect the univocal feature combination, the univocal feature combination is stored in a temporary positioning map.

[0028] Furthermore, central data processing unit 5 is configured to determine, based on the result of the comparison, from which lane or lanes of the traffic road section surroundings sensor system 4 was able to detect the univocal feature combination.

[0029] Furthermore, the central data processing unit 5 is further configured to determine at least one quality value for the univocal feature combination using the detection rate determined as a result of the comparison and specific to the univocal feature combination and / or the visibility of the univocal feature combination from at least two different areas of the traffic road section and / or the real-time nature of the fleet data, and store the quality value in a positioning map. Here, the automated and / or auxiliary vehicle system 2 of at least one fleet vehicle 3b of the system 1 or at least one vehicle 3a not belonging to the fleet is configured to perform its own positioning using a positioning map containing the at least one univocal feature combination, and only considers the univocal feature combinations whose quality values ​​stored in the positioning map are above a predefined lower limit value.

[0030] The central data processing unit 5 is also configured to generate a separate detection request for each univocal feature combination and send the detection request to the convoy vehicle 3b so that the corresponding univocal feature combination can be preferably or only detected by the environmental sensor device 4 of the corresponding convoy vehicle 3b during driving on the traffic road section.

[0031] Furthermore, the central data processing unit 5 can also be configured to determine a route and trajectory plan for operating the automated and / or auxiliary vehicle system 2 , using which a unique feature combination can be detected.

[0032] Figure 2 A flow chart of an exemplary embodiment of the method according to the invention is shown for generating a localization map containing at least one unambiguous feature combination for operating an automated and / or auxiliary vehicle system 2 of a vehicle (not shown).

[0033] In method step 100, a convoy vehicle of the convoy drives on a traffic road section assigned to a positioning map, during which the traffic road section is detected by means of environmental sensor devices of the corresponding convoy vehicle. Thus, convoy data assigned to the traffic road section are generated in method step 100. In method step 101, the convoy data generated in method step 100 are collected centrally.

[0034] In method step 102, a position map is generated from the collected fleet data and stored in method step 103. In method step 104, at least one unambiguous feature combination is ascertained from the collected fleet data.

[0035] In step 105, based on the result of comparing the univocal feature combination with "the fleet data at each single driving on the traffic road section" or with "the modeling of the sensor detection range of vehicles not belonging to the fleet along the traffic road section", it is determined whether the fleet vehicle or the vehicle's environmental sensor device can detect the univocal feature combination, wherein if the environmental sensor device can detect the univocal feature combination, the univocal feature combination is stored in a temporary positioning map.

[0036] Furthermore, in method step 105 , it can also be determined based on the result of the comparison from which lane or lanes of the traffic road section the surroundings sensor system was able to detect the univocal feature combination.

[0037] Furthermore, in method step 105, at least one quality value is determined for the univocal feature combination using the detection rate determined from the result of the comparison and specific to the univocal feature combination and / or the visibility of the univocal feature combination from at least two different areas of the traffic road section and / or the real-time nature of the fleet data. In method step 103, the quality value is stored in the positioning map.

Claims

1. A method for generating a positioning map containing at least one unambiguous feature combination for operating an automated and / or auxiliary vehicle system (2) of a vehicle (3a, 3b), wherein: Fleet data is collected by having a fleet vehicle (3b) of the fleet travel on a traffic road section assigned to the positioning map, a positioning map is generated from the collected fleet data, and at least one univocal feature combination is obtained from the collected fleet data, characterized in that based on the result of comparing the univocal feature combination with "the fleet data at each single time of driving on the traffic road section" or with "the modeling of the sensor detection range of the vehicle (3a) not belonging to the fleet along the traffic road section", it is determined whether the environmental sensor device (4) of the fleet vehicle (3b) or the vehicle (3a) can detect the univocal feature combination, and if the environmental sensor device (4) can detect the univocal feature combination, the univocal feature combination is stored in a temporary positioning map.

2. The method according to claim 1, characterized in that Based on the result of the comparison, it is determined from which lane or lanes of the traffic road section the surroundings sensor system (4) was able to detect the univocal feature combination.

3. The method according to claim 1 or 2, characterized in that: Using a detection rate determined from a result of the comparison that is specific to the unambiguous feature combination and / or a visibility of the unambiguous feature combination from at least two different areas of the traffic road section determined from a result of the comparison and / or the real-time nature of the fleet data, at least one quality value is determined for the unambiguous feature combination and stored in the positioning map.

4. A system (1) for generating a positioning map containing at least one unambiguous feature combination for operating an automated and / or auxiliary vehicle system (2) of a vehicle (3a, 3b), the system comprising a plurality of convoy vehicles (3b) and at least one central data processing unit (5), the convoy vehicles each having at least one environmental sensor device (4), the central data processing unit being connectable to the convoy vehicles (3b) for collecting convoy data by the convoy vehicles (3b) driving on a traffic road section assigned to the positioning map, and the central data processing unit being configured to generate a positioning map from the collected convoy data and to determine at least one unambiguous feature combination from the collected convoy data, characterized in that The central data processing unit (5) is configured to determine whether the convoy vehicle (3b) or the environmental sensor device (4) of the vehicle (3a) can detect the univocal feature combination based on the result of comparing the univocal feature combination with "convoy data at each single time of driving on the traffic road section" or with "modeling of the sensor detection range of the vehicle (3a) not belonging to the convoy along the traffic road section", and if the environmental sensor device (4) can detect the univocal feature combination, the univocal feature combination is stored in a temporary positioning map.

5. The system (1) according to claim 4, characterized in that The central data processing unit (5) is configured to determine, based on a result of the comparison, from which lane or lanes of the traffic road section the surroundings sensor system (4) was able to detect the univocal feature combination.

6. System (1) according to claim 4 or 5, characterized in that The central data processing unit (5) is configured to determine at least one quality value for the univocal feature combination using a detection rate determined as a result of the comparison and specific to the univocal feature combination and / or a visibility of the univocal feature combination in at least two different areas of the traffic road section determined as a result of the comparison and / or the real-time nature of the fleet data, and to store the quality value in the positioning map.

7. System (1) according to any one of claims 4 to 6, characterized in that The central data processing unit (5) is configured to generate a separate detection request for each univocal feature combination and to send the detection request to the convoy vehicle (3b) so that the corresponding univocal feature combination is preferably or only detected by means of the environmental sensor device (4) of the corresponding convoy vehicle (3b) during driving on the traffic road section.

8. System (1) according to any one of claims 4 to 7, characterized in that The central data processing unit (5) is configured to determine a route and trajectory plan for operating an automated and / or auxiliary vehicle system (2), using which route and trajectory plan univocal feature combinations can be detected.

9. System (1) according to any one of claims 6 to 8, characterized in that An automated and / or auxiliary vehicle system (2) of at least one convoy vehicle (3b) or at least one vehicle (3a) not belonging to the convoy of the system (1) is configured to perform its own localization using a localization map containing the at least one unambiguous feature combination, and only takes into account such unambiguous feature combinations whose quality values ​​stored in the localization map are above a predefined lower limit value.

Citation Information

Patent Citations

  • Procedures for updating map data

    DE102018008988A1