Deployment system for the supportive operation of an at least partially autonomously operable ego-vehicle in a predefined road section

Abstract

Provisioning system (1,1a) for supporting the operation of an at least partially autonomously operable ego vehicle (10) in a given road section, characterized in that at least one first module (2) is provided which has at least one first interface (3), wherein the first interface (3) is configured to receive current signal data relating to the road section, wherein the signal data is provided by a road infrastructure (8), wherein the signal data includes at least current signal data for controlling road traffic in accordance with the road traffic regulations, wherein the first interface (3) is configured for receiving raw environmental data relating to the road segment, wherein the raw environmental data at least partially encompass the road segment, wherein the raw environmental data are provided by an infrastructure (5), wherein the raw environmental data include at least camera raw data and / or radar raw data, and wherein the first interface (3) is further configured to receive raw environment server data relating to the road segment, which are provided by an external server (6), wherein the raw environment server data include at least current traffic condition information, and wherein the first module (2) comprises a processor (13) configured to create a fully up-to-date digital environment of the relevant road segment based on at least the relevant environment server raw data, the relevant environment raw data and the current signal data and wherein the provisioning system (1,1a) further comprises a second module (14) for generating various target raw trajectories for the relevant road segment based on the digital environment, wherein the second module (14) comprises a second interface (25) for transmitting such a target raw trajectory to a corresponding ego vehicle (10) in the road segment, wherein the target raw trajectory is designed as a target trajectory to be adapted and / or validated by the corresponding ego vehicle (10), and wherein the second module (14) is designed to transmit the target raw trajectory adapted to the respective autonomy level to the respective ego vehicle (10) based on the autonomy level transmitted by different ego vehicles (10).

Description

[0001] The invention relates to a provisioning system for supporting the operation of an at least partially autonomously operable ego vehicle in a predetermined road section and to an ego vehicle.

[0002] An autonomous / semi-autonomous vehicle is a vehicle that operates without a driver, or where the driver only intervenes in emergencies. The vehicle drives autonomously by, for example, independently recognizing the road layout, other road users, or obstacles, calculating the corresponding control commands within the vehicle, and transmitting these to the vehicle's actuators, thereby correctly influencing the vehicle's course.

[0003] Advanced driver assistance systems (ADAS) and autonomous driving (AD) are well-known, market-leading technologies in current automotive research and on the market. Both ADAS and AD rely heavily on information about the vehicle's surroundings, which in the current state of technology is referred to as "perception," "digital environment," "vehicle context," or simply "environment."

[0004] This information is collected in two ways: either exclusively with the onboard sensors, as in the case of adjusting the vehicle speed based on the distance to the vehicle in front, or with the onboard sensors, especially lidar / radar sensors, together with information from external systems such as the cloud, GPS, map information, etc.

[0005] However, such sensors, for example radar sensors, ultrasonic sensors, lidar sensors, and cameras, are very expensive. Retrofitting them is often impossible or uneconomical, especially in older vehicles.

[0006] DE 10 2020 211970 A1 discloses a method for controlling an autonomously driving vehicle, comprising: recording environmental sensor data of a vehicle's environment by means of an environmental sensor of the vehicle, wherein the environmental sensor data are summarized as data points in a first three-dimensional point cloud with a first resolution of environmental sensor data; generating a second three-dimensional point cloud of environmental sensor data based on the first three-dimensional point cloud of environmental sensor data by means of a first trained neural network, wherein the second three-dimensional point cloud comprises data points of the first three-dimensional point cloud, and wherein the second three-dimensional point cloud has a point area with a second resolution that is higher than the first resolution;and the recognition of an object in the vicinity of the vehicle based on the data points of the point area of ​​the second three-dimensional point cloud with the second resolution by a second trained neural network. DE 10 2020 200 433 A1 discloses a method for calculating optimized trajectories by at least one control unit, wherein measurement data from at least one infrastructure unit and / or at least one vehicle are received, an intention, for example in the form of reference trajectories, demand trajectories and alternative trajectories, is received from at least one vehicle or determined by evaluating measurement data, and the received measurement data and the intention received or determined from at least one vehicle are used to calculate at least one regionally optimized trajectory for the at least one vehicle.

[0007] It is therefore an object of the present invention to provide a provisioning system for supporting the operation of an ego-vehicle. Furthermore, it is an object to provide such a supported ego-vehicle.

[0008] This problem is solved by a provisioning system with the features of claim 1 and an ego vehicle with the features of claim 12. Further advantageous measures are listed in the dependent claims, which can be suitably combined to achieve further advantages.

[0009] The task is solved by a provisioning system for the supportive operation of an at least partially autonomously operable ego-vehicle in a given road section, wherein at least a first module is provided which has at least a first interface, wherein the first interface is configured to receive current signal data relating to the road section, wherein the signal data is provided by a road infrastructure, and wherein the signal data includes at least current signal data for controlling road traffic in accordance with the road traffic regulations. and wherein the first interface for receiving raw environmental data relating to the road segment is designed, wherein the raw environmental data at least partially covers the road segment, wherein the raw environmental data is provided by an infrastructure, wherein the raw environmental data includes at least camera raw data and / or radar raw data, and wherein the first interface is further configured to receive raw environmental server data relating to the road segment, which is provided by an external server, wherein the raw environmental server data includes at least current traffic condition information, and wherein the first module comprises a processor which is configured to create a fully up-to-date digital environment of the relevant road segment based on at least the relevant environment server raw data, the relevant environment raw data and the current signal data, and and wherein the provisioning system further comprises a second module for generating various target raw trajectories for the relevant road segment based on the digital environment, wherein the second module comprises a second interface for transmitting such a target raw trajectory to a corresponding ego vehicle in the road segment, wherein the target raw trajectory is designed as a target trajectory to be adapted and / or validated by the corresponding ego vehicle and wherein The second module is trained to transmit the target raw trajectory, adapted to the autonomy level, to the respective ego vehicle based on the autonomy level transmitted by various ego vehicles.

[0010] Raw data refers to unprocessed or only partially pre-processed data. Signal data can include, for example, traffic light red / green, construction site warnings, etc.

[0011] Environmental raw data can, for example, show the road section as a camera image / radar recordings or include lidar sensor data of the road section.

[0012] Such cameras can be installed, for example, on the sides of roads.

[0013] Raw environmental server data includes information such as road closures, upcoming construction sites, and the weather along the route. Furthermore, other data relating to the road segment can also be transmitted.

[0014] A section of road can be, for example, an intersection, or a longer stretch of road, or a city, or several cities, etc.

[0015] By outsourcing the generation of the digital environment to, for example, an external computer system as the first module, rapid generation using a large amount of data can be achieved, particularly through increased computing capacity, more powerful processors, etc. Computationally intensive AI (Artificial Intelligence) algorithms can be used for this purpose, enabling the rapid and reliable creation of a digital environment.

[0016] Furthermore, the provisioning system according to the invention can generate various trajectories as target-raw trajectories. The target-raw trajectories can be generated by coupling them to externally autonomously driving ego vehicles that have previously transmitted a destination, or all target-raw trajectories for the road segment can be generated in general.

[0017] A target-raw trajectory is essentially a target trajectory that requires only minor adaptation and / or validation in the ego-vehicle, for example, by precisely adjusting the target-raw trajectory to the exact position of the ego-vehicle using it. Such an exact position can be determined by kinematics-based localization by the ego-vehicle or by transmitting a global position to the ego-vehicle via a global positioning system.

[0018] Furthermore, this allows the target-raw trajectory to be adapted / validated. Based on such an adaptation and / or validation of the target-raw trajectory, the ego vehicle can operate autonomously even without the internal creation of a digital environment; that is, the control units can generate the control variables / parameters for the actuators based on the target trajectory.

[0019] Such a provisioning system eliminates the need for expensive, environment-generating sensors, computationally intensive operations, and corresponding hardware modules in autonomous vehicles.

[0020] Furthermore, according to the invention, the second module is configured to transmit the target raw trajectory, adapted to the respective autonomy level, to the respective ego vehicle based on the autonomy level transmitted by different ego vehicles. Thus, for example, faster transmission is possible using a reduced target raw trajectory, which is sufficient, for instance, for an ego vehicle with autonomy level two or three.

[0021] In further training, the second module is designed to generate the various target raw trajectories for autonomously operating ego vehicles with autonomy level four or higher. The ego vehicle can then request the target raw trajectories or, if connected to the provisioning system, receive them automatically. This allows computationally intensive target raw trajectories for a highly automated ego vehicle to be generated externally. The ego vehicle can thus avoid computationally intensive and therefore hardware-intensive calculations, which require not only a sufficiently complete digital environment model but also the computationally intensive calculation of the target trajectory.

[0022] In further training, the second module features a receiving interface for receiving target information from a specific ego-vehicle. This second module is configured to use the second interface to transmit the corresponding target raw trajectory, based on the target information, to the specific ego-vehicle. The second interface can be used for transmitting the target information, or a separate receiving interface can be provided. This allows for the transmission of a target raw trajectory tailored to the ego-vehicle, eliminating the need for the complete data set of all target raw trajectories and thus enabling faster data transmission.

[0023] In further training, the second module is trained to automatically transmit the target / raw trajectory to the specific ego-vehicle at continuous time intervals. This ensures a smooth journey for the ego-vehicle.

[0024] In a further configuration, the deployment system is designed as multiple edge clouds for a given road segment, with each edge cloud containing the first and second modules. This allows virtually each edge cloud to generate a digital environment and create target / raw trajectories for the respective road segment. These edge clouds, which can be configured as external computers, can be installed in or along the corresponding road segment, enabling transmission via radio / NFC (Near Field Communication). This ensures secure transmission, independent of an existing internet / network connection.

[0025] In further training, each of the edge clouds can be configured to receive V2X data generated by other road users concerning the environment of the specified road segment via at least one first interface of at least one first module. The processor of at least one first module can be configured to create the fully up-to-date digital environment of the relevant road segment, based at least on the relevant environment server raw data, the relevant environment raw data, the current signal data, and the received V2X data. A more precise digital environment can be created using the additional V2X data.

[0026] In further training, each edge cloud is positioned within the area of ​​the road segment responsible for that particular edge cloud. Furthermore, the edge clouds can be configured to transmit the planned raw trajectory to the corresponding ego vehicles via V2X communication. This ensures reliable transmission even in areas with radio interference.

[0027] In a further embodiment, the first interface of the first module is configured to receive SmartHome data from a Smart Home of a user of an ego vehicle, wherein the processor is further configured to generate a fully up-to-date digital environment at least on the basis of the relevant environment server raw data, the relevant environment raw data, the current signal data, and the SmartHome data, and wherein the second module is configured to generate a personalized target raw trajectory and transmit it to the user's ego vehicle.

[0028] This allows, for example, the control of the ego vehicle on private property, such as parking in a garage / carport or parking the ego vehicle in a designated space on the property.

[0029] With further training, the deployment system can also be configured to connect to one or more ego vehicles and to at least partially ensure data security for the connected ego vehicles. This allows for the partial external implementation of cybersecurity firewalls. Additionally, all non-time-critical tasks and information irrelevant to the ego vehicles' movement domain, such as infotainment and body data, can be outsourced to the deployment system and processed externally. This saves computing capacity and thus hardware in the ego vehicle.

[0030] In further training, the first interface of the first module is designed to receive off-road data based on off-road elements in the specified road section, and wherein the processor of the first module is designed to create a fully up-to-date digital environment of the relevant road section, at least based on the relevant environment server raw data, the relevant environment raw data, the current signal data, and the off-road data, and wherein the second module is designed to accomplish the generation of target raw trajectories for the relevant road section based on the digital environment, including the off-road data.

[0031] Off-road elements can include, for example, petrol stations or car washes / workshops.

[0032] The off-road data includes the services offered, such as the provision of gasoline, car washes, repairs, etc.

[0033] Furthermore, the task is solved by an ego-vehicle having a bidirectional communication module for communication with a provisioning system as described above, wherein the communication module is configured to receive a target raw trajectory at least on the basis of an input target position, and wherein the communication module is configured to receive position data for determining the position of the ego-vehicle by a global satellite navigation system, and / or the ego-vehicle is configured to calculate position data for determining the position of the ego-vehicle using kinematics-based localization, and wherein the ego-vehicle is further configured to adapt and / or validate the target raw trajectory as a target trajectory on the basis of the position data in order to achieve at least partial autonomous operation in a given road segment, and wherein the ego-vehicle is configured toto transmit the autonomy level to the provisioning system using the bidirectional communication module, in order to receive a target raw trajectory aligned with the autonomy level.

[0034] Furthermore, according to the invention, the ego vehicle is configured to transmit the autonomy level to the deployment system via the bidirectional communication module, in order to receive a target raw trajectory adapted to the autonomy level. This allows, for example, an ego vehicle with autonomy level two to receive a data-reduced target raw trajectory, enabling faster data transmission and simpler data processing.

[0035] In further training, the ego vehicle is trained to receive, via the communication module, the current signal data relating to the road segment for the entered route, wherein the signal data is provided by a road infrastructure, wherein the signal data includes at least current signal data for controlling road traffic in accordance with the road traffic regulations; and wherein the ego vehicle is trained, via the communication module, to receive the raw environmental data relating to the road segment for the entered route, wherein the raw environmental data includes at least part of the road segment, wherein the raw environmental data is provided by an infrastructure, wherein the raw environmental data includes at least camera raw data and / or radar raw data; and wherein the ego vehicle is trained, via the communication module, to receive the raw environmental server data relating to the road segment, which is provided by an external server.where the environment server raw data includes at least current traffic condition information to be received, and wherein the ego vehicle is trained to represent the target raw trajectory as the target trajectory to adapt and / or validate at least on the basis of the received signal data, the received environment raw data, the received environment server raw data, and the exact position.

[0036] This enables rapid adaptation and / or validation of the target raw trajectory for the safe autonomous operation of an ego vehicle that does not have an expensive sensor system to record the environment.

[0037] Furthermore, V2V data transmitted by other vehicles can also be used for the adaptation and / or validation of the target-raw trajectory.

[0038] Such a transmission allows the autonomous vehicle to operate autonomously within the specified road segment, without generating a digital environment. This can be an autonomous vehicle equipped with driver assistance systems (i.e., Level 2 autonomy) or a more highly automated vehicle. Using its precise position, the autonomous vehicle can adapt and validate the transmitted target trajectory and then use this target trajectory to generate the necessary control parameters / vehicle parameters via the various control units. With such an autonomous vehicle, featuring external processing of sensor data and internal processing of the received target trajectory, safe and rapid autonomous operation can be achieved without expensive sensors or complex sensor processing with costly hardware components.

[0039] The communication module can include a gate for distributing specific data from the received target trajectory (raw) and / or the adapted and / or validated target trajectory to the relevant control units. Such a gate allows the transmitted data to be quickly distributed internally within the vehicle.

[0040] In further development, the ego vehicle incorporates a sensor system for acquiring ego vehicle data relating to its surroundings. The ego vehicle is trained to adapt or validate the target trajectory based on this data. This can involve only a few sensors, serving solely for comparison or adaptation of the target trajectory. This enables simple and cost-effective adaptation and validation of the target trajectory, ensuring, for example, that an autonomously operated ego vehicle can safely stop even in the event of a communication failure.

[0041] Further features and advantages of the present invention will become apparent from the following description with reference to the accompanying figures. These show: Fig. 1: a provisioning system according to an initial design, Fig. 2: further training of such a provisioning system, Fig. 3: an ego vehicle according to the invention, Fig. 4: an ego-vehicle architecture,

[0042] Fig. Figure 1 shows a provisioning system 1 according to a first design.

[0043] This module features a server / computer module as its first module 2. This first module 2 includes a first interface 3. This first interface 3 can encompass several different interfaces 3, which, for example, are based on NFC technology, radio frequency, or 5G technology and can receive data via this technology. The first interface 3 can be configured for both data reception and data transmission.

[0044] The first module 2 receives, via the first interface 3, the current signal data relating to a specified road segment, which is provided by road infrastructure 8, such as a traffic light. This signal data includes at least current signal data for controlling road traffic in accordance with the road traffic regulations. This could, for example, be the current traffic light status (green / red).

[0045] Such a road section can be a predefined area, for example a city or a section thereof, or one or more streets.

[0046] Furthermore, the raw environmental data pertaining to the road segment is received by the first module 2. This data encompasses the surroundings. The raw environmental data can be provided by an infrastructure 5, which includes at least camera raw data and / or radar raw data. The infrastructure 5 could, for example, consist of cameras / LiDAR sensors / radar sensors located at the roadside / near buildings, providing an environmental image as raw environmental data. This raw environmental data can be partially pre-processed / compressed to facilitate transmission. Thus, other road users, objects, pedestrians, obstacles, and cyclists in the vicinity of the road segment can be detected within the raw environmental data.

[0047] Furthermore, raw data from the environment server, provided by an external server 6, can be received via the first interface 3.

[0048] The external server 6 can, for example, be configured as a cloud. The raw environmental server data includes current traffic condition information. Furthermore, this data can also include weather / temperature data or other environmental data.

[0049] Furthermore, the provisioning system 1 can communicate with one or more location systems, for example a global satellite navigation system 7, to locate, for example, the individual ego vehicles 10 in the relevant road section.

[0050] Furthermore, the first interface 3 can receive the current off-road data relevant to the road segment from a service provider 9 as an off-road element. The off-road data from a service provider 9 includes the respective service offered. Service providers are primarily gas stations. Other service providers can include, for example, car washes or repair shops.

[0051] Thus, by connecting to service providers in conjunction with, for example, a low fuel level indicator, autonomous driving to the nearest gas station can be achieved.

[0052] Additional up-to-date information data for the road section, provided as off-road data by Internet of Things devices, can also be received.

[0053] This could include, for example, information about parking at a major event or parking in a parking garage.

[0054] Furthermore, the deployment system 1 includes a processor 13, which is designed to create a fully up-to-date digital environment of the relevant road segment, at least based on the relevant environment server raw data, the relevant environment raw data, the current signal data, the current information data, the off-road data, and the V2X data. For this purpose, the processor 13 can employ machine learning methods such as AI (Artificial Intelligence) algorithms, for example, a neural network.

[0055] The data can be received wirelessly. Such a digital environment digitally maps the surroundings, i.e., the road section including its objects, etc.

[0056] Furthermore, deployment system 1 has a second module 14. This module is designed to generate a target raw trajectory based on the digital environment. This is a preliminary trajectory that must later be adapted and / or validated in the ego vehicle 10.

[0057] This generates the target raw trajectory for an Ego Vehicle 10 with autonomy level four or higher.

[0058] The provisioning system 1 can be configured to generate all target raw trajectories for the digital road segment or, when coupled to ego vehicles 10, only those target raw trajectories that are required by the corresponding ego vehicles 10.

[0059] For this purpose, the target information of a specific ego-vehicle 10 must be transmitted to the provisioning system 1. After generating the target-raw trajectory, which is aligned with the target, the corresponding target-raw trajectory is transmitted to the specific ego-vehicle 10 via the second interface 25.

[0060] Furthermore, a reduced target trajectory tailored to the Egovehicle 10 can also be transmitted. For example, only a reduced target trajectory can be transmitted for an Egovehicle 10 with autonomy level 3, as fully autonomous operation at this level is not possible for an Egovehicle 10. This data reduction enables faster data transmission.

[0061] Furthermore, when SmartHome data generated by a user's Smart Home 4 is received, a personalized target raw trajectory can be generated for the user's Egofahrzeug 10 and transmitted to the user's Egofahrzeug 10.

[0062] This makes it possible, for example, to control the Egofahrzeug 10 on private property, such as parking it in a garage / carport or parking it in a designated spot on the property.

[0063] Furthermore, when connecting the Ego Vehicles 10 to the Deployment System 1, additional external cybersecurity firewalls for the Ego Vehicles 10 can be implemented in the Deployment System 1. Information that is irrelevant to the movement domain of the Ego Vehicles 10, such as infotainment and bodywork, or other non-time-critical tasks, can also be processed by the second module 14, so that the computing capacity does not have to be provided by the Ego Vehicle 10.

[0064] Fig. Figure 2 shows a further configuration of such a deployment system 1a. This comprises several edge clouds 15a, 15b. These can be configured for a specific road segment, with each edge cloud 15a, 15b having a first module 2 and a second module 14. The first module 2 can create a digital environment for the road segment responsible for the respective edge cloud 15a, 15b, and the second module 14 can create a target raw trajectory for the responsible road segment and transmit it to the respective ego vehicle 10.

[0065] The edge clouds 15a, 15b can be positioned in the relevant road section in such a way that the target raw trajectory can be transmitted to the corresponding ego vehicles 10 via V2X communication, for example by radio. This enables reliable transmission even in areas with radio interference, a so-called "dead zone".

[0066] Furthermore, a large traffic area can thus be covered, in which adjacent road segments are covered by neighboring edge clouds 15a, 15b. For example, a partial overlap 16 between neighboring edge clouds 15a, 15b and adjacent road segments can be provided. This means that when an ego vehicle 10 changes from one road segment with an edge cloud 15a to another road segment, an automated login, i.e., a takeover of the neighboring edge cloud 15b, occurs, so that a seamless generation and transmission of the intended raw trajectory from the start to the desired destination can be achieved by the edge clouds 15a, 15b.

[0067] This enables the reliable transmission of the target raw trajectory to the Ego vehicles 10 from start to finish over a long distance.

[0068] The edge clouds 15a and 15b can each be configured as external computing modules positioned within the road segment. These are coupled or coordinated with each other to create a target raw trajectory across a wide traffic area.

[0069] Fig. Figure 3 shows an ego vehicle 10 according to the invention. This vehicle has a gate 17 as a bidirectional communication module for communication with a provisioning system 1.

[0070] The target raw trajectory can be received from the Ego Vehicle 10 via Gate 17.

[0071] In this process, the destination, which for example has been entered by the user into a navigation device 23, may have been transmitted in advance and the target raw trajectory may be aligned with the destination.

[0072] Furthermore, the position data for determining the position of the ego vehicle 10 can be received via the gate 17 by a global satellite navigation system / location system 7, or the ego vehicle 10 can determine the position data using a kinematics-based localization.

[0073] Current signal data from a road infrastructure 8 for the road segment can also be received, as well as the raw environmental data relating to the ego-vehicle 10 from an infrastructure 5 for the road segment, and the raw environmental server data relating to the ego-vehicle 10 from a server 6. Furthermore, additional data such as current information data for the road segment, generated by Internet of Things devices 12, off-road data generated by a service provider 9, and smart home data generated by a smart home 4, as well as V2V data from other road users, can be received.

[0074] Based on at least the received signal data, the received raw environmental data, the received raw environmental server data, and the exact position, as well as, if applicable, off-road data, smart home data, information data, and V2V data, the target raw trajectory can now be adapted and validated as the target trajectory. A computing module 18 may be available for this purpose, which, for example, is also based on a machine learning method.

[0075] The computing module 18 can also handle all time-critical tasks as well as kinematics-based localization and vehicle safety, i.e., partially take over the cybersecurity firewalls, and process other information relevant to the movement domain of the ego vehicle 10.

[0076] Based on the target trajectory and the vehicle's own onboard sensors, such as temperature sensors, the actuators for operating the actuators and other vehicle parameters for autonomous operation can be generated.

[0077] All non-time-critical tasks, such as vehicle parameters for the infotainment system and bodywork, as well as the generation of the target raw trajectory, can be handled by provisioning system 1. Additionally, cybersecurity firewalls, meaning partial vehicle security, can be handled by provisioning system 1 for the ego vehicle 10. The additional vehicle parameters generated in this way, as well as the target raw trajectory, are transmitted to the ego vehicle 10 and further processed internally by an ego vehicle architecture.

[0078] Fig. Figure 4 schematically shows such an ego-vehicle architecture.

[0079] This includes Gate 17, which receives the target raw trajectory from provisioning system 1 as well as other time-non-critical data from provisioning system 1. Furthermore, Gate 17 is designed to receive current signal data, raw environmental data, raw environmental server data from which the positions of surrounding objects, traffic information, etc., can be determined, as well as precise position data and, if applicable, off-road data, information data, and smart home data.

[0080] Based on the received information, the computing module 18 can now validate or adapt the target / raw trajectory. Computing module 18 can be implemented in gate 17.

[0081] Based on the validated target trajectory and the data received by the provisioning system 1 and the received current signal data, the environment raw data, the received environment server raw data, the position data, the off-road data, the gate 17 can now forward corresponding information to the corresponding control units for further processing.

[0082] The control units can be a brake control unit 19 and a steering control unit 20 or a higher-level domain control unit 21, or other component control units 22, which generate the control commands and vehicle parameters for setting the actuators.

[0083] This specifies a decentralized E / E architecture, which includes domain control units 21 and component control units 22 as well as the computing module 18, which can also be decentralized, for validating and adapting the target raw trajectory into a target trajectory and for generating all control commands and setting parameters for actuators etc.

[0084] Such an ego-vehicle 10 allows for the development of a sensorless ego-vehicle 10 using machine learning methods, which can quickly generate and validate a target trajectory. External processing of the sensor data enables rapid processing of all sensor data for generating a digital environment, as well as a raw target trajectory which then only needs to be adapted and validated by machine learning methods within the ego-vehicle 10. Reference symbol list 1 Deployment system 2 first module 3 first interface 4 Smart Home 5 Infrastructure 6 external servers 7 Satellite navigation system 8 Road infrastructure 9 service providers 10 Ego vehicles 11 road users 12 Internet of Things devices 13 processor 14 second module 15a,15b Edge Clouds 16 Overlap 17 Gate 18 Calculation module 19 Brake control unit 20 Steering control unit 21 Domain control unit 22 component control units 23 Navigation device 25 second interface

Claims

Provisioning system (1, 1a) for supporting the operation of an at least partially autonomously operable ego vehicle (10) in a specified road section, characterized in that at least a first module (2) is provided which has at least a first interface (3), wherein the first interface (3) is configured to receive current signal data relating to the road section, wherein the signal data is provided by a road infrastructure (8), wherein the signal data includes at least current signal data for controlling road traffic in accordance with the road traffic regulations, wherein the first interface (3) is configured to receive raw environmental data relating to the road section, wherein the raw environmental data includes at least part of the road section, wherein the raw environmental data is provided by an infrastructure (5), and wherein the raw environmental data includes at least camera raw data and / or radar raw data.and wherein the first interface (3) is further configured to receive raw environment server data relating to the road segment, which is provided by an external server (6), wherein the raw environment server data includes at least current traffic condition information, and wherein the first module (2) has a processor (13) configured to create a fully up-to-date digital environment of the road segment in question based on at least the raw environment server data relating to the road segment, the raw environment data relating to the road segment, and the current signal data, and wherein the provisioning system (1, 1a) further has a second module (14) for generating various target raw trajectories for the road segment in question based on the digital environment, wherein the second module (14) has a second interface (25) for transmitting such a target raw trajectory to a corresponding ego vehicle (10) in the road segment,wherein the target raw trajectory is configured as a target trajectory to be adapted and / or validated by the corresponding ego vehicle (10), and wherein the second module (14) is configured to transmit the target raw trajectory adapted to the respective autonomy level to the respective ego vehicle (10) based on the autonomy level transmitted by different ego vehicles (10). Provisioning system (1,1a) according to claim 1, characterized in that the second module (14) is configured to generate the various target raw trajectories for autonomously operable ego vehicles (10) with autonomy level four or higher. Provisioning system (1, 1a) according to claim 1 or 2, characterized in that the second module (14) has a receiving interface for receiving target information of a specific ego vehicle (10) and wherein the second module (14) is configured to transmit the corresponding target raw trajectory with respect to the target information to the specific ego vehicle (10) via the second interface (25). Provisioning system (1,1a) according to claim 3, characterized in that the second module (14) is configured to automatically transmit the target raw trajectory to the specific ego vehicle (10) at continuous time intervals. Provisioning system (1,1a) according to one of the preceding claims, characterized in that the provisioning system (1,1a) is configured as at least several edge clouds (15a,15b) for a given road section, wherein each of the edge clouds (15a,15b) has a first module (2) and a second module (14). Provisioning system (1, 1a) according to claim 5, characterized in that each of the edge clouds (15a, 15b) is configured to receive V2X data relating to the environment of the specified road segment, generated by other road users (11), via the at least one first interface (3) of the at least one first module (2), and the processor (13) of the at least one first module (2) is configured to create the fully current digital environment of the relevant road segment, at least on the basis of the relevant environment server raw data, the relevant environment raw data, the current signal data, and the received V2V data. Provisioning system (1,1a) according to claim 5 or 6, characterized in that each of the edge clouds (15a,15b) is arranged in the area of ​​the road section responsible for the edge cloud (15a,15b). Provisioning system (1,1a) according to claim 7, characterized in that the edge clouds (15a,15b) are configured to transmit the target raw trajectory to the corresponding ego vehicles (10) via V2X communication. Provisioning system (1,1a) according to one of the preceding claims, characterized in that the first interface (3) of the first module (2) is configured to receive SmartHome data from a Smart Home (4) of a user of an ego vehicle (10) and the processor (13) is configured to generate a fully up-to-date digital environment at least on the basis of the relevant environment server raw data, the relevant environment raw data and the current signal data and the SmartHome data, and wherein the second module (14) is configured to generate a personalized target raw trajectory and transmit it to the ego vehicle (10) of the user. Provisioning system (1,1a) according to one of the preceding claims, characterized in that the provisioning system (1,1a) is further configured to provide coupling to one or more ego vehicles (10) and to provide at least partial data security for the coupled ego vehicles (10). Provisioning system (1, 1a) according to one of the preceding claims, characterized in that the first interface (3) of the first module (2) is configured to receive off-road data based on off-road elements in the specified road section, and the processor (13) of the first module (2) is configured to create a fully up-to-date digital environment of the relevant road section, at least based on the relevant environment server raw data, the relevant environment raw data, the current signal data, and the off-road data, and wherein the second module (14) is configured to accomplish the generation of target raw trajectories for the relevant road section based on the digital environment, including the off-road data. Ego vehicle (10) comprising a bidirectional communication module for communication with a provisioning system (1, 1a) according to one of the preceding claims, characterized in that the communication module is configured to receive a target raw trajectory at least on the basis of an input target position, and wherein the communication module is configured to receive position data for determining the position of the ego vehicle (10) by a global satellite navigation system (7) and / or the ego vehicle (10) is configured to calculate position data for determining the position of the ego vehicle (10) on the basis of kinematics-based localization, and wherein the ego vehicle (10) is further configured to adapt and / or validate the target raw trajectory on the basis of the position data as a target trajectory for the purpose of achieving at least partial autonomous operation in a given road segment.and wherein the ego vehicle (10) is trained to transmit the autonomy level to the provisioning system (1,1a) via the bidirectional communication module, in order to receive a target raw trajectory adapted to the autonomy level. Ego vehicle (10) according to claim 12, characterized in that the ego vehicle (10) is configured to receive, by means of the communication module, the current signal data relating to the road segment for the entered route, wherein the signal data are provided by a road infrastructure (8), wherein the signal data include at least current signal data for controlling road traffic in accordance with the road traffic regulations, and wherein the ego vehicle (10) is configured to receive, by means of the communication module, the raw environment data relating to the road segment for the entered route, wherein the raw environment data at least partially encompass the road segment, wherein the raw environment data are provided by an infrastructure (5), wherein the raw environment data include at least camera raw data and / or radar raw data, and wherein the ego vehicle (10) is configured toby means of the communication module to receive the raw environmental server data relating to the road segment, which are provided by an external server (6), wherein the raw environmental server data include at least current traffic condition information, and wherein the ego vehicle (10) is designed to adapt and / or validate the target raw trajectory as a target trajectory at least on the basis of the received signal data, the received raw environmental data, the received raw environmental server data, and the exact position. Ego vehicle (10) according to claim 12 or 13, characterized in that the communication module comprises a gate (17) for distributing certain data of the received target raw trajectory and / or the adapted and / or validated target trajectory to the relevant control devices.

Images