Method and device for operating an environmental sensor system

Abstract

Method (300) and apparatus for providing (340) an optimization function for operating an environmental sensor system, comprising a step of receiving (310) a plurality of environmental data sets, wherein each environmental data set was acquired by means of sensors of a vehicle, wherein each environmental data set represents an environment of the respective vehicle, wherein each environmental data set includes at least a time and / or position information when acquiring the corresponding environment, a step of creating (320) a digital map, depending on the environmental data sets, wherein the creation (320) of the digital map comprises several sub-steps, wherein at least some sub-steps comprise monitoring algorithms, each of which generates a quality indicator of the correspondingly executed sub-step, a step of aggregating (330) the quality indicators and determining the optimization function,depending on the aggregated quality indicators, and a step of providing (340) the optimization function to operate an environmental sensor system.

Description

[0001] The present invention relates, inter alia, to a method for providing an optimization function for operating an environmental sensor system. In this method, a plurality of environmental data sets are received, and a digital map is created based on these data sets. The creation of the digital map comprises several sub-steps, at least some of which include monitoring algorithms, each of which generates a quality indicator for the respective sub-step. Subsequently, the quality indicators are aggregated, and the optimization function is determined based on the aggregated quality indicators and made available for operating the environmental sensor system. Disclosure of the invention

[0002] The method for providing an optimization function for operating an environmental sensor system comprises a step of receiving a multitude of environmental data sets, wherein each environmental data set was acquired by means of sensors of a vehicle, wherein each environmental data set represents an environment of the respective vehicle, and wherein each environmental data set includes at least a time and / or position information when acquiring the corresponding environment.The procedure further comprises a step of creating a digital map, depending on the environmental data value sets, wherein the creation of the digital map comprises several sub-steps, wherein at least some sub-steps comprise monitoring algorithms, each of which generates a quality indicator of the correspondingly executed sub-step, a step of aggregating the quality indicators and determining the optimization function, depending on the aggregated quality indicators, and a step of providing the optimization function to operate an environmental sensor system.

[0003] Environmental sensors are defined as at least one video sensor, at least one radar sensor, at least one lidar sensor, at least one ultrasound sensor, and / or at least one other sensor, designed to detect the environment of a vehicle—in particular, the environmental characteristics of that environment—in the form of sensor data. In one possible embodiment, the environmental sensors may, for example, include a processing unit (processor, RAM, hard drive) with suitable software and / or be connected to such a processing unit.

[0004] Vehicle sensors include video, radar, lidar, ultrasound, and / or other sensors designed to capture a vehicle's environment in the form of sensor data.

[0005] An environmental feature here can be understood to mean, for example, an infrastructure feature (road markings, guardrails, etc.) and / or a traffic sign (road signs, traffic lights, etc.) and / or a structural feature (buildings, bridges, tunnels, etc.) and / or another feature that can be detected by means of environmental sensors.

[0006] Each environmental data set includes at least one time and / or position value recorded when the corresponding environment was captured. Environmental data sets can also contain additional information, such as traffic information, weather information, and / or contextual information (e.g., detailed information about the driving scenario, possibly using detailed map information), etc.

[0007] A digital map is a map that exists as (map) data values ​​on a storage medium. The map is structured, for example, by comprising one or more map layers. One map layer might show a bird's-eye view (the course and position of roads, buildings, landscape features, etc.). This corresponds, for example, to a map used in a navigation system. Another map layer might include a radar map, where environmental features covered by the radar map are overlaid with a radar signature. Yet another map layer might include a lidar map, where environmental features covered by the lidar map are overlaid with a lidar signature.

[0008] In one embodiment, the map is designed as a high-precision map. This high-precision map is specifically designed to be suitable for navigating an automated vehicle. This means, for example, that the high-precision map is designed to determine a highly accurate position of the automated vehicle by comparing stored environmental features with sensor data from the vehicle. For this purpose, the high-precision map includes, for example, these environmental features with highly accurate positional data (coordinates). Here, "map" is understood to mean, in particular, a globally accurate map.

[0009] A high-precision position is one that, within a given coordinate system, such as WGS84 coordinates, is so accurate that it does not exceed a maximum permissible level of uncertainty. This maximum uncertainty can depend on factors such as the surrounding environment. Furthermore, it can depend on whether a vehicle is operated manually or semi-, highly, or fully automatically (corresponding to one of SAE levels 1 to 5). Generally, the maximum uncertainty is low enough to ensure the safe operation of automated vehicles. For fully automated operation, the maximum uncertainty is typically around 10 centimeters.

[0010] An automated vehicle is defined as a partially, highly or fully automated vehicle according to one of the SAE levels 1 to 5 (see standard SAE J3016).

[0011] Creating a digital map means, for example, integrating environmental features encompassed by the environmental data values ​​into a base map according to their position and / or removing or adjusting environmental features already included in the base map. A base map in this context refers specifically to a digital map (as described above).

[0012] Providing the optimization function for operating environmental sensors means, for example, that this optimization function is provided in the form of digital data in such a way that it can be requested and received by a vehicle with corresponding environmental sensors - for example, via a radio connection.

[0013] An optimization function, for example, refers to signals that lead to an adjustment of the settings of one or more sensors in the environmental sensor system. The optimization function can include, for instance, adjusting the transmission and / or evaluation frequency of a lidar / radar sensor and / or the light sensitivity of a video sensor. Furthermore, the optimization function can also include commands for the visual and / or acoustic output of signals representing the state of the environmental sensor system within the corresponding vehicle.

[0014] The method according to the invention advantageously solves the problem of providing a method for supplying an optimization function for operating environmental sensors. To ensure high quality of the (digital) map, various monitoring algorithms are used along the processing chain for map creation (here, the respective sub-steps), each of which generates a quality indicator for the corresponding sub-step. The aforementioned problem is now solved by the method according to the invention, in particular, by providing these quality indicators or information from the executed monitoring algorithms as a service, for example, to optimize the corresponding environmental sensors. This advantageously allows, for example, an evaluation of the environmental sensor performance of individual vehicles to be carried out without further measures on the vehicle side.

[0015] Preferably, the quality indicators include information about filtered-out and / or inconsistent data.

[0016] These monitoring algorithms include, for example, consistency checks (filtering (sensor) data that is inconsistent compared to the majority of the available data) and quality assessments (evaluating the quality of each output step in the processing chain or mapping pipeline for creating the (digital) map). The quality indicators thus represent the results of the respective monitoring algorithms as digital information.

[0017] These results from the monitoring algorithms can thus provide, for example, indications of (fundamental) problems with environmental sensors or individual sensors, such as inconsistencies between individual environmental data sets and other environmental data sets, false positives (a large proportion of other environmental data sets do not contain a particular object) and inconsistent semantics of (individual) objects.

[0018] Preferably, in one of the sub-steps, environmental characteristics are extracted from the environmental data values. The corresponding quality indicator represents a statistical frequency of the environmental characteristics.

[0019] In this process, for example, the available environmental data sets are associated and transformed into a common coordinate system. The transformed environmental data sets are then fused (e.g., 100 measurements of a traffic sign are fused into a map attribute). Based on this, statistical frequencies or statistics for the sensors used can be generated in the Fleet Data Statistics calculation. The following useful statistics are proposed: information about false negatives (i.e., the environmental sensors failed to detect an object), information about false positives (i.e., the environmental sensors detected a non-existent object), and other statistics such as a misclassification rate.

[0020] The device, in particular a computing unit, is configured to perform all steps of the method for providing an optimization function for operating environmental sensors. A computing unit can be, for example, a server, a server network, or a cloud.

[0021] The device or computing unit comprises a processor, main memory, a storage medium, and suitable software for carrying out the method according to one of the method claims. Furthermore, the device comprises an interface for transmitting and receiving data values, via a wired and / or wireless connection, for example with corresponding vehicle components (control units, communication devices, environmental sensors, navigation systems, etc.) and / or other vehicle-external devices (servers, cloud, etc.).

[0022] Furthermore, a computer program is claimed, comprising instructions which, when executed by a computer, cause the computer to execute a method according to one of the method claims for providing an optimization function for operating an environmental sensor system. In one embodiment, the computer program corresponds to the software comprised of the second device.

[0023] Furthermore, a machine-readable storage medium on which the computer program is stored is required.

[0024] Advantageous embodiments of the invention are specified in the dependent claims and listed in the description. Drawings

[0025] Exemplary embodiments of the invention are shown in the drawing and are explained in more detail in the following description. It shows: Fig. 1 An embodiment of the inventive method for providing an optimization function for operating an environmental sensor in the form of a flowchart. Embodiments of the invention

[0026] Fig. Figure 1 shows an embodiment of a method 300 for operating 340 an environmental sensor system.

[0027] In step 301, procedure 300 starts.

[0028] In step 310, a multitude of environmental data sets are received, wherein each environmental data set was acquired by means of sensors of a vehicle, wherein each environmental data set represents an environment of the respective vehicle, and wherein each environmental data set includes at least a time and / or position information when acquiring the corresponding environment.

[0029] In step 320, a digital map is created based on the environmental data sets. This step comprises several sub-steps, at least some of which include monitoring algorithms, each of which generates a quality indicator for the respective sub-step.

[0030] In step 330, the quality indicators are aggregated and the optimization function is determined based on the aggregated quality indicators.

[0031] In step 340, the optimization function for operating an environmental sensor system is provided.

[0032] In step 350, procedure 300 ends.

Claims

[1] Method (300) for providing (340) an optimization function for operating an environmental sensor system, comprising: - Receiving (310) a plurality of environmental data sets, wherein each environmental data set was acquired by means of sensors of a vehicle, wherein each environmental data set represents an environment of the respective vehicle, wherein each environmental data set includes at least a time and / or position information when acquiring the corresponding environment; - Creating (320) a digital map, depending on the environment data value sets, wherein the creation (320) of the digital map comprises several sub-steps, wherein at least some sub-steps comprise monitoring algorithms, each of which generates a quality indicator of the correspondingly executed sub-step; - Aggregating (330) the quality indicators and determining the optimization function, depending on the aggregated quality indicators; and - Providing (340) the optimization function for operating an environmental sensor system. [2] Method (300) according to claim 1, characterized by that the quality indicators include information about filtered-out and / or inconsistent data. [3] Method (300) according to claim 1, characterized by , that in one of the sub-steps environmental characteristics are extracted from the environmental data values ​​and the corresponding quality indicator represents a statistical frequency of the environmental characteristics. [4] Device, in particular a computing unit configured to perform all steps of the method (300) according to any one of claims 1 to 3. [5] Computer program comprising instructions which, when the computer program is executed by a computer, cause it to execute a method (300) according to any one of claims 1 to 3. [6] Machine-readable storage medium on which the computer program according to claim 5 is stored.

Images