DE102019209707B4 - Method of operating a self-driving vehicle - Google Patents

Abstract

Method for operating a first self-driving vehicle (10), having a first communication module (20) set up for communication with second vehicles (63); a driving system (30) set up for automatic driving operation of the first vehicle (10) with a first function data set and a first control unit (40), the method having the method steps:establishing a communication connection with a second vehicle (63);receiving and determining a license level of the second vehicle (63), the range of functions of a driving system (30) of a vehicle (10, 63) increasing as the license level of the vehicle (10, 63) increases;comparison of the determined license level of the second vehicle (63) with a license level of the first vehicle ( 10);receiving a second functional data record from the second vehicle (63) if the determined license level of the second vehicle (63) exceeds the license level of the first vehicle (10); anddriving the first vehicle (10) by means of the driving system (30) using the second functional data set.

Description

The invention relates to a method for operating a self-driving vehicle, in particular for operating a driving system that controls the automatic driving of the vehicle. In the method according to the invention, data sets of the driving system are preferably retrieved from another vehicle via V2V communication.

Today's vehicles already have a large number of assistance systems that provide computer-based support to the driver in a large number of driving situations. Such assistance systems can use sensors to record a large number of measurement data, which far exceed the human sensory abilities. In addition, the speed of these assistance systems significantly exceeds human reaction times. Known driver assistance systems are, for example, lane departure warning systems, braking assistants when pedestrians are detected and adaptive cruise control, in particular for traffic jams.

By using such assistance systems, the autonomy of the driver with regard to his driving decisions is increasingly transferred to the vehicle or the control units operating in it. At the end of these developments there is a self-driving vehicle that can maneuver completely without human intervention. Fully automated passenger transport is possible using such a self-propelled vehicle.

The driving operation of a self-driving car is controlled by a driving system, which in the context of the present application is to be understood in the broadest sense as a control unit specially designed and set up for this task. This control unit is designed for automatic lateral and longitudinal guidance of the vehicle using a large number of input signals, in particular sensor data recorded by the vehicle itself and communication data received. Such control of the automatic lateral and longitudinal guidance places complex calculation requirements on the driving system, which generally has algorithms, models and control functions that are at least partially based on machine learning (artificial intelligence).

In the context of the present application, machine learning is understood to mean all methods and evaluation algorithms that are applied to ordered (clustering) or unordered amounts of input data using stochastic, probabilistic methods in order to be able to derive one- to multi-dimensional dependencies between the input data and/or intermediate data derived from them . A specific evaluation goal (training goal) is generally specified for an evaluation process. The machine learning is carried out, for example, by training a neural network, ie by determining parameters for parameterizing the neural network. If necessary, the training goal consists of relationships to be found in the form of target functions whose function parameters are to be optimized in the evaluation process. Within the scope of the present application, both the parameters of a neural network and the optimized function parameters are referred to below as a function data set.

As a rule, the driving systems used in self-driving vehicles are trained by the manufacturer using test vehicles. A vehicle equipped with the driving system is usually delivered with a first functional data record, which represents the result of this training and is intended in principle to make the functions of the driving system, such as lane departure warning, distance warning, etc. available. However, this first functional data set is not to be regarded as optimal. On the one hand, there may be differences between the vehicle and the test vehicle, for example with regard to the tires or changes made by the user (lowering, etc.). Furthermore, environmental data, such as road conditions, or target specifications, such as average vehicle distances in urban and rural areas, can vary.

In order to take these differences in the properties of the vehicles and the environment into account, an adaptation of the function data set is usually provided. In other words, the (functional) parameters should also be adjusted during operation of the vehicle. Various methods for such an adaptation are known from the prior art.

the WO 2018/139993 A1 discloses a self-driving vehicle with a machine learning (artificial intelligence)-based driving system, which is designed, for example, to recognize obstacles, to avoid collisions and to carry out autonomous driving maneuvers. The driving system controls a driving of the vehicle while the vehicle sensors collect a variety of data. After the journey, a user rating is recorded, which contains, for example, information on positions where the vehicle left the lane or a turning maneuver was carried out too quickly. The user rating and the sensor data are sent to a transmitted to the server, which adapts the driving system's Kl model, i.e. the functional data set, based on this. This solution requires the transmission of a large amount of data between each individual vehicle and the server, which can be problematic in terms of available bandwidth.

the DE 10 2017 201 226 A1 suggests using the computing resources of the vehicles themselves to adapt the AI models. The server only transmits instruction data, for example on the structure or the training goal of a neural network, to a vehicle, which adapts the AI model using this instruction data and operating data that it has recorded itself. In this case, the AI model should be adapted in particular when the computing resources of the vehicle are unused, for example while the vehicle is parked. The disadvantage is that adapting the Kl model is therefore very time-consuming and the Kl model is only available with limited functionality for a certain period of time.

WO 2019/049 141 A1 proposes the exchange of training data between vehicles, which trains an AI system h of the vehicle to map sensor data X onto a decision value Y. A test is carried out in the vehicle as to whether the received data improves the performance of an AI system, and if such an improvement is determined, the information received is integrated into the AI system.

US 2017/0 213 457 A1 describes how a vehicle V1 receives experience information relating to a route R from a vehicle V2, so that vehicle V1 can use this experience information when driving on the route R.

the DE102017212835A1 describes the segmentation of AI modules for the operation of a vehicle, for example by using separate AI models for turning right and turning left, and indicates that parameters of an AI module can be transferred from another vehicle or an external computer.

The invention is therefore based on the object of enriching the prior art and providing a method for operating a self-driving vehicle in which functional data of a driving system is updated faster than is known from the prior art and at the same time safer operation of the vehicle is possible vehicle is guaranteed.

The object of the invention is achieved by the subject matter of the main claims. Preferred developments are the subject matter of the respective dependent claims.

A first aspect of the invention relates to a method for operating a self-driving vehicle, in particular a method carried out by this vehicle. A vehicle carrying out the method has in particular a first communication module set up for communication with other vehicles. The communication module is set up in particular for vehicle-to-vehicle communication (vehicle-to-vehicle—V2V) via an air interface, for example WLAN, Bluetooth, radio, etc. In addition, the first communication module is preferably designed for communication with a network server, with a GPS satellite and access control devices.

A vehicle carrying out the method according to the invention also has a driving system set up for automatic driving operation of the vehicle. The driving system is set up in particular for fully automatic longitudinal and/or lateral guidance of the vehicle. For this purpose, the driving system preferably accesses a large number of sensor data recorded by the vehicle and communication data received by the vehicle. The driving system is designed in particular to carry out driving assistance functions up to fully automatic driving. For this purpose, the driving system has control algorithms that are at least partially based on machine learning and have target functions and/or neural networks, for example. To operate the driving system, functional data are therefore necessary which, for example, have parameters of a neural network or optimized functional parameters. This functional data is stored in the vehicle's driving system as a functional data set. At the beginning of the method according to the invention, a (relatively) first functional data set is always stored in the driving system of the vehicle.

A (first) vehicle carrying out the method according to the invention also has a control unit. This control unit is preferably designed to communicate with the other components of the vehicle, in particular with the first communication module and with the driving system and possibly with sensors, and to coordinate their functions. The control unit also preferably allocates to or controls a memory.

At least the following method steps take place in the method according to the invention. The vehicle first establishes communication, preferably under the control of the first control unit and by means of the first communication module cation connection with a second vehicle. This communication connection is preferably via V2V communication, but can also be switched via smart infrastructure or a mobile network.

The first vehicle receives and determines a license level of the second vehicle via the communication connection established with the second vehicle. The license level is received or called up by the second vehicle before the second functional data set. The license level is particularly preferably received in the form of a digital signature, for example a private PGP key or other cryptographic means. In the method according to the invention, the vehicle also compares the determined or received license level of the second vehicle with a current license level of its own. The own license level is preferably stored in a memory of the vehicle, for example also in the form of a digital signature or the like.

According to the invention, the vehicle receives a second functional data set from the second vehicle when the license level determined for the second vehicle exceeds the current license level of the vehicle itself. In other words, the vehicle receives functional data suitable for the operation of the driving system from the second vehicle. The second functional data set is preferably functional data used by the second vehicle for the operation of its driving system. This second function data is preferably formatted identically to the function data of the first function data set. Alternatively, the functional data of the second functional data set are transmitted in a format that is independent of the vehicle or type. The functional data of the second functional data set are preferably parameters of a neural network or optimized functional parameters of a specific target function. This ensures that functional data is only imported from vehicles with higher certification.

In the method according to the invention, driving operation of the vehicle is mediated by the driving system, that is to say at least partially automatic, using the second function data set. In a preferred embodiment, the first function data set is replaced by the second function data set. The first function data record is also preferably adapted or adjusted by the second function data record. For example, a neural network of the vehicle is configured with the parameters (coefficients) of the second functional data set, or a target function of the vehicle is configured with the functional parameters of the second functional data set.

The method according to the invention is based on the knowledge that the updating or adaptation of the functional data used by the driving system is necessary for the optimal functioning of driving systems based on machine learning. Only in this way can the individual properties of the respective vehicle and/or local peculiarities of the driving environment be mapped in the functional data. However, updating the functional data in each vehicle itself is time-consuming and resource-consuming. The vehicles, in particular their driving systems, therefore learn from one another in the method according to the invention by exchanging the functional data records. Since the function data sets are transmitted directly between the vehicles, data routing via a backend server is unnecessary and updated (adapted) function data are received with a high probability in the local environment. A vehicle can advantageously update its functional data efficiently in the method according to the invention.

The method according to the invention advantageously makes it possible that a disadvantageous transfer of information between vehicles involved in the method does not take place. In other words, a deterioration in the performance of the vehicle's driving system as a result of the second functional data can thus advantageously be avoided. This form of implementation thus makes it possible for a vehicle to only learn from vehicles which themselves have already reached a higher update stage of their driving system, or of the data based on machine learning on which this is based.

In a preferred embodiment, the method according to the invention is carried out by a vehicle that has at least one first sensor, such as a distance sensor (LIDAR, ultrasound) or a camera, set up to record environmental data. This vehicle also preferably has at least one second sensor configured to acquire vehicle data, such as a wheel speed sensor or an immersion depth sensor of a shock absorber. The operation of the driving system using the second function data set then also preferably takes place on the basis of sensor data recorded by means of the first and/or second sensors.

In a preferred embodiment, identification information of the second vehicle is received in addition to the license level of the second vehicle. These characterize preference know the type of the second vehicle. As already explained in the introduction, the driving systems can be trained, at least in part, for specific vehicle types. As a rule, a typification for a parking assistant is more decisive than for a system for automatic distance control when driving. According to this embodiment, the second function data set is received and/or used depending on the license level of the second vehicle and on the basis of the identification information of the second vehicle. For example, functional data can only be received from vehicles of the same type. Alternatively, certain function data are preferably also received and/or used by vehicles that are not of the same type, in particular for functions that are independent of type.

In a further preferred embodiment of the method according to the invention, the vehicle is driven by means of the driving system depending on the license level of the vehicle. In other words, the license level affects not only the transfer of knowledge between the vehicles but also the function of the vehicle itself. The license level therefore plays the role of a degree of experience or adaptation of a vehicle in the method according to the invention and in this respect influences the provision of automated driving functions of the Vehicle, especially the driving system. The degree of external control, be it by a user or by a backend server, particularly preferably decreases as the license level increases. Parameters of the driving system are also preferably adjusted as a function of the license level, for example by allowing vehicles with higher license levels to drive automatically at higher speeds.

In a particularly preferred embodiment, the range of functions of the driving system increases as the license level of the vehicle increases. In other words, the degree of automation of driving operations increases as the license level of the vehicle increases. The license levels of the vehicle are particularly preferably correlated with different levels of autonomous driving. A vehicle particularly preferably has a license level of 0 when it is delivered. Furthermore, the license level 0 preferably corresponds to a degree of automation level 0 (“driver only”) according to which the driver permanently performs the longitudinal and lateral guidance and the driving system does not actively intervene in the driving operation.

In a preferred embodiment of the method according to the invention, license level 1 corresponds to a degree of automation level 1 (“assisted”), according to which a driver permanently assumes longitudinal or lateral guidance and the driving system assumes the other function. License level 2 also preferably corresponds to a level 2 automation level (“partially automated”), according to which a driver must constantly monitor the vehicle and the driving system takes over the longitudinal and lateral guidance in the specific application.

Furthermore, license level 3 preferably corresponds to a level 3 automation level (“highly automated”), according to which a driver does not have to monitor the vehicle permanently, but must be potentially able to take control of the vehicle. According to level 3 of automation, the driving system also takes over the longitudinal and lateral guidance in a specific application, also recognizes the limits of the driving system independently and prompts the driver with sufficient time reserve to take control. License level 4 preferably corresponds to a degree of automation level 4 (“fully automated”) according to which no driver is required in the specific application and the driving system handles the situation completely autonomously in the specific application.

License level 5 also preferably corresponds to a level 5 automation level (“driverless”) according to which no driver is required from the beginning to the end of a journey and the driving system takes over the driving task in full in all use cases. The above definitions correspond to the "Five Levels of Autonomous Driving" as defined by the Association of the German Automotive Industry, VDA. Use case includes road types, speed range and environmental or environmental conditions.

In a further preferred embodiment of the method according to the invention, information on the license level and on the function data set of the vehicle is also transmitted to a network server, for example to a backend server of the vehicle manufacturer or service partner. Furthermore, additional data on the driving behavior of the vehicle are preferably transmitted to the network server. According to this embodiment, the network server is in particular a licensing server or a server set up and authorized to issue license levels. According to this implementation, the network server also checks the data received from the vehicle to determine whether it qualifies the vehicle for a higher license level. Exemplary implementation forms of these tests are explained in detail below. As a result of these checks, the vehicle also receives certificate information for a higher license level from the network server in the course of the vehicle's method according to the invention. This In the method according to the invention, a higher license level means that the vehicle can pass on its function data set to more vehicles and that the vehicle has additional automatic driving functions available, as described above.

In a preferred embodiment of the method according to the invention, the information about the function data record transmitted from the vehicle to the network server includes details about an update of the function data record carried out by the vehicle. This information can contain, for example, based on sensor data of how many kilometers driven and/or driving time the vehicle has updated the functional data set. This information can likewise preferably have a computing time for the update carried out or the number of update runs carried out. This information about updates made by the vehicle can be compared in the network server, for example, with empirical values to determine whether the vehicle is likely to qualify for a higher license level.

According to this embodiment, the information about the function data record transmitted from the vehicle to the network server also preferably includes details about the license level of the second vehicle. As already explained above, in the method according to the invention the vehicle only learns from vehicles with higher qualifications. Thus, for example, a connection with a minimum number of higher-qualified second vehicles with a higher license level can be made a prerequisite for the granting of a higher license level to the vehicle. It is then assumed that updating the function data using the second function data sets is advantageous.

In an embodiment that is also preferred, the additional information includes information about a driving time, a driving route and/or a damage-free status with the current license level of the vehicle. In other words, the information for updating the function data is not transmitted directly to the network server, but rather indirectly. This makes sense, for example, when the functional data in the vehicle is updated continuously or with a predetermined periodicity. Therefore, for example, kilometers driven and/or driving time since the last increase in the license level are meaningful for the further development of the vehicle's artificial intelligence. As described above, a vehicle is preferably delivered with a license level of 0. An increase in the license level to license level 1 is then preferably only possible after 500 km, also preferably only after 1000 km or alternatively preferably only after 5000 km. Particularly preferably, a license increase is only possible if all or a predetermined portion of this route has been traveled through without an accident. Likewise, an upgrade to the next higher license level 2 can be made dependent on a route that has been driven at least with license level 1.

In a likewise preferred embodiment, some of the license levels or at least one highest license level can also be made dependent on a real driving test of the self-driving vehicle. Such a test can be, for example, a course provided by the manufacturer, which the self-propelled vehicle must pass through without errors. The manufacturer or course operator can precisely define the driving tasks assigned to the vehicle and thus estimate whether the automatic driving system is already sufficiently adapted for automatic driving operation. Such a driving test is particularly preferably a mandatory requirement in order to become a second vehicle of the method according to the invention, ie to be able to pass on its functional data record to other vehicles. In a preferred embodiment, such a driving test is only necessary to enable additional functions, such as automatic driving at higher speeds.

A further aspect of the invention relates to a vehicle, in particular a passenger car with an internal combustion engine, an electric motor or a hybrid motor which is set up for carrying out the steps of a vehicle in the method according to the invention. For this purpose, the vehicle has at least one first sensor set up for acquiring data on the surroundings and at least one second sensor set up for acquiring vehicle data.

The at least one first sensor is designed to detect sensor signals relating to the surroundings of the vehicle. The at least one second sensor is designed to detect sensor signals relating to the vehicle itself. In this case, an environment signal received by means of the at least one first sensor preferably enables the vehicle to obtain information about its environment and preferably depicts a large amount of environmental information. A status signal received by means of the at least one second sensor preferably puts the vehicle in a position to obtain information about its own status and, for this purpose, preferably depicts a large number of items of status information. The first sensor is an imaging sensor, for example Sensors, such as cameras, or distance sensors, such as LIDAR. The second sensors are, for example, wheel speed sensors and the like.

The vehicle according to the invention also has a first communication module set up for communication with a network server and other vehicles. In addition, the communication module can be set up for communication with other devices, such as GPS satellites or a smart infrastructure. The communication module preferably has a radio, mobile radio, WLAN, and/or Bluetooth transceiver or alternative wireless communication devices.

The vehicle also has a driving system set up for automatic driving operation of the vehicle. The driving system is designed in particular for partially or fully automatic transverse and/or longitudinal guidance of the vehicle. For this purpose, the vehicle is preferably connected to the first sensors, to the second sensors and to the communication module and receives their sensor data as input data. The driving system also has models, neural networks or target functions that are at least partially based on machine learning and on adaptable functional data sets. These function data sets are the data sets already described with reference to the method according to the invention.

The vehicle also has a first control unit that is set up to communicate with the at least one first sensor, with the at least one second sensor, with the first communication module and with the driving system. The control unit is also set up to carry out the steps of the vehicle in the method according to the invention or to instruct the other components of the vehicle to carry out the steps of the method according to the invention. The first control unit is designed in particular to control the first communication module for establishing a communication connection with a second vehicle and for receiving a second functional data set from the second vehicle, and also designed to enable the driving system of the vehicle to operate the vehicle using the second functional data set check.

The control unit is also designed to determine a license level of the second vehicle, in particular based on information on the license level of the second vehicle received by means of the communication module. The control unit is also designed to compare the determined license level with a license level of the vehicle and to receive, preferably also use, the second function data set only if the determined license level of the second vehicle exceeds the license level of the vehicle. In a likewise preferred embodiment, the control unit is designed to control the driving system for carrying out the driving operation of the vehicle depending on the license level of the vehicle. The control unit is designed to control the driving system in such a way that the range of functions of the driving system increases as the license level of the vehicle increases.

A further aspect of the invention relates to a method of a control unit of a vehicle, which has at least one first sensor set up to record data on the surroundings, at least one second sensor set up to record status data of the vehicle and a communication module, a driving system set up to drive the vehicle automatically, and the Has a control unit, the method having at least the following steps: Establishing a communication link with a second vehicle, receiving and determining a license level of the second vehicle, the range of functions of a driving system of a vehicle increasing as the license level of the vehicle increases, comparison of the determined license level of the second vehicle with a license level of the first vehicle, receiving a second functional data record from the second vehicle if the determined license level of the second vehicle exceeds the license level of the first vehicle, and driving drives the vehicle by means of the driving system using the second functional data set.

In a preferred embodiment, the method also has the following steps: using the second function data record if the license level determined exceeds the license level of the vehicle. The method also preferably has the method step of driving the vehicle using the driving system depending on the license level of the vehicle. The range of functions of the driving system increases as the license level of the vehicle increases.

A further aspect of the present invention relates to a network server, for example a network server from a vehicle manufacturer or from its service partner. The network server according to the invention has a second communication module set up for data communication with a plurality of vehicles and a second control unit. In this case, the second control unit of the network server is set up to receive information from the vehicle about a license level, about a functional data record and about a driving behavior of the vehicle. Further, the second Control unit designed to check the authorization of the vehicle for a higher license level based on the information received and a defined catalog of requirements. The catalog of requirements is preferably defined by the manufacturer and has, for example, the criteria described above, such as distance traveled, travel time and number of contacts with vehicles with a higher license. Finally, the network server is designed to transmit certificate information for a higher license level to a vehicle authorized for the higher license level, ie depending on the result of the test carried out. According to the invention, the range of functions of a vehicle's driving system set up for automatic driving operation of the vehicle increases as the license level of the vehicle increases.

A further aspect of the present invention relates to a system comprising a vehicle according to the invention, as described above, and a network server according to the invention, as described above. All method steps and aspects of the method according to the invention can advantageously be implemented in the system according to the invention. Preferred developments of the method according to the invention correspond to the preferred developments described with reference to the vehicle according to the invention or with reference to the server according to the invention.

A further aspect of the invention relates to a computer program, comprising instructions which, when the program is executed by a computer, such as a control unit of a vehicle, cause the latter to carry out the steps of the vehicle in the method according to the invention, in particular the steps: establishment of a communication connection with a second vehicle, receiving a second functional data set from the second vehicle, driving the vehicle by means of the driving system using the second functional data set.

A further aspect of the invention relates to a computer program, comprising instructions which, when the program is executed by a computer, such as a control unit of a vehicle, cause it to carry out the steps of the vehicle in the method according to the invention, in particular the steps: receiving information about a license level, to a functional data set and to a driving behavior of the vehicle from the vehicle; Checking an authorization of the vehicle for a higher license level based on the information received and a defined catalog of requirements; and communicating certificate information for a higher license level to a vehicle eligible for the higher license level.

The method steps of the method according to the invention can be implemented by electrical or electronic parts or components (hardware), by firmware (ASIC) or by executing a suitable program (software). The method according to the invention is also preferably realized or implemented by a combination of hardware, firmware and/or software. For example, individual components for carrying out individual method steps are designed as a separate integrated circuit or are arranged on a common integrated circuit. Individual components set up to carry out individual method steps are also preferably arranged on a (flexible) printed circuit board (FPCB/PCB), a tape carrier package (TCP) or another substrate.

The individual method steps of the method according to the invention are also preferably designed as one or more processes that run on one or more processors in one or more electronic computing devices and are generated when one or more computer programs are executed. The computing devices are preferably designed to work together with other components, for example a communication module, and one or more sensors or cameras in order to implement the functionalities described herein. The instructions of the computer programs are preferably stored in a memory, such as a RAM element. However, the computer programs can also be stored in a non-volatile storage medium such as a CD-ROM, a flash memory or the like.

It is also apparent to those skilled in the art that the functionalities of multiple computers (data processing devices) can be combined or combined in a single device or that the functionality of a specific data processing device can be distributed across a large number of devices in order to carry out the steps of the method according to the invention without deviate from the method according to the invention.

Unless stated otherwise in the individual case, the various embodiments of the invention mentioned in this application can advantageously be combined with one another.

• 1 eine schematische Darstellung eines erfindungsgemäßen Systems aus einem erfindungsgemäßen Fahrzeug und aus einem erfindungsgemäßen mobilen Netzwerkserver gemäß einer Ausführungsform; und
• 2 ein Ablaufdiagramm eines im erfindungsgemäßen System durchgeführten erfindungsgemäßen Verfahrens gemäß einer Durchführungsform.

The invention is explained below in exemplary embodiments with reference to the associated drawings. Show it:

• 1 a schematic representation of a system according to the invention from a vehicle according to the invention and from an invention mobile network server of the invention according to one embodiment; and
• 2 a flow chart of a method according to the invention carried out in the system according to an embodiment.

1 FIG. 12 shows a schematic illustration, in particular a block diagram, of an exemplary vehicle 10, in particular a two-track vehicle with an internal combustion, electric or hybrid engine. The vehicle 10 includes a plurality of first sensors, in particular a first sensor 11, a second sensor 12, and a third sensor 13. The first sensors 11, 12, 13 are set up to capture environmental data of the vehicle 10 and include, for example, a camera for capturing an image of an environment immediately surrounding vehicle 10, distance sensors, such as ultrasonic sensors or LIDAR, for detecting distances to objects surrounding vehicle 10. The first sensors 11, 12, 13 transmit the environmental signals they detect to a first control unit 40 and a driving system 30 of the vehicle 10.

The vehicle 10 also has a plurality of second sensors, in particular a fourth sensor 51 , a fifth sensor 52 and a sixth sensor 53 . Second sensors 51, 52, 53 are sensors for determining status data relating to vehicle 10 itself, such as current position and movement information of the vehicle. The second sensors are consequently, for example, speed sensors, acceleration sensors, inclination sensors, sensors for measuring an immersion depth of a shock absorber, wheel speed sensors or the like. The second sensors 51, 52, 53 transmit the status signals they detect to the first control unit 40 of the vehicle 10. In addition, the second sensors 51, 52, 53 transmit their measurement results directly to a driving system 30 of the vehicle 10.

The vehicle 10 also has a first communication module 20 with a memory 21 and one or more transponders or transceivers 22 . The transponders 22 are radio, WLAN, GPS or Bluetooth transceivers or the like. The transponder 22 communicates with the internal memory 21 of the first communication module 20, for example via a suitable data bus. The first communication module 20 also communicates with the first control unit 40. In addition, the first communication module 20 is set up to communicate with a mobile network server 70, in particular a backend server of a vehicle manufacturer, contractual partner or fleet operator. The communication takes place in particular with a second communication module 90 of the network server 70. The first communication module 20 is preferably set up to communicate via an LTE mobile network. Furthermore, the first communication module 20 is set up to communicate with other vehicles 63, in particular the communication modules of other, similar vehicles 63, in particular via vehicle-to-vehicle communication.

The vehicle 10 also has the driving system 30 which is set up for fully automatic driving operation, in particular for autonomous longitudinal and lateral guidance, of the vehicle 10 . The driving system 30 has a navigation module 32 which is set up to calculate routes between a starting point and a destination and to determine the maneuvers to be carried out by the vehicle 10 along this route. The navigation module 32 is also preferably configured to perform specific maneuvers of the vehicle 10, such as parking and exiting maneuvers. In addition, the driving system 30 includes an internal memory 31 which communicates with the navigation module 32, for example via a suitable data bus. In particular, a model, a neural network and/or a target function, which are at least partially based on machine learning, are stored on the internal memory 31 . Furthermore, functional data for the model, the neural network and/or the target function are stored in the internal memory 31 as a functional data set. The functionality of the driving system 30 is preferably controlled by the control unit 40 and enabled in particular as a function of a license level of the vehicle 10 .

The vehicle 10 also has a first control unit 40 which is set up to carry out the steps of the vehicle 10 in the method according to the invention. In this case, the control unit 40 carries out the steps itself or controls the other components of the vehicle 10 accordingly. For this purpose, the first control unit 40 has an internal memory 41 and a CPU 42, which communicate with one another, for example via a suitable data bus. In addition, first control unit 40 is in communication with at least first sensors 11, 12, 13, second sensors 51, 52, 53, first communication module 20 and driving system 30, for example via one or more respective CAN connections, one or multiple respective SPI connections, or other suitable data connections.

The network server 70 has a second control unit 80 which is used to carry out the Steps of the network server 70 is set up in the inventive method. For this purpose, the second control unit 80 has an internal memory 81 and a CPU 82, which communicate with one another, for example via a suitable data bus. The network server 70 also has a second communication module 90 . The second communication module 90 has a memory 92 and one or more transponders or transceivers 91 . The transponders 91 are radio, WLAN, GPS or Bluetooth transceivers or the like. The transponder 91 communicates with the internal memory 92 of the second communication module 90, for example via a suitable data bus. The second communication module 90 is preferably set up to communicate via an LTE mobile radio network.

2 shows a schematic flow chart of a method according to the invention carried out in the system 100 according to the invention, according to one embodiment. The system 100 according to the invention consists of a (first) vehicle 10, a second vehicle 63 configured analogously to vehicle 10, and a network server 70.

In a first step S100 of the method according to the invention, the network server 70 transmits a basic function data record for the driving system 30 of the vehicle 10 together with certificate information for a license level 0 to the vehicle 10. The vehicle 10 receives this data and loads it into a model, into a neural network or in a target function of the driving system 30. The license level of the vehicle 10 is set to zero by the certificate information. Thus, the driving system 30 is enabled only for a degree of automation level 0 (“driver only”) and the driver continuously carries out the longitudinal and lateral guidance without the driving system actively intervening in the driving operation.

In step S200, vehicle 10 drives with license level 0 and automation level 0, adapting the basic data set received from network server 70 based on sensor data recorded by first and second sensors 11, 12, 13, 51, 52, 53. After a predetermined driving distance, for example 1000 km, the vehicle 10 upgrades to license level 1 in a step S201. According to the embodiment described, no contact with a network server 70 or another vehicle 63 is necessary for this. Driving the 1000 km thus only serves to adapt the basic data record to significant deviations of the vehicle 10 from the test vehicle. After the upgrade, the vehicle now has license level 1 and is therefore in automation level 1 ("assisted"), according to which the driver permanently takes over the longitudinal or lateral guidance and the driving system takes over the other function.

Meanwhile, in step S300, a second vehicle 63 drives with license level 4 and thus already with a degree of automation level 4 (“fully automated”), according to which no driver is required in the specific application and the driving system in the specific application handles the situation completely autonomously. Use cases designate road conditions, speed ranges and environmental conditions. Before step S401, the first vehicle 10 and the second vehicle 63 approach each other.

In step S401, the vehicle 10 uses its first communication module 20 to transmit a request for identification information to the second vehicle 63. The second vehicle 63 receives this request and then transmits its identification information, in particular information about its license level 4, to the vehicle 10. In step S403 compares The vehicle then changes its license level, i.e. license level 1, to the license level of the second vehicle 63, i.e. license level 4. The result is that the second vehicle has already activated a higher degree of automation and a functional data record from the second vehicle 63 is therefore available is advantageous.

In step S404, the vehicle 10 therefore requests a second functional data set from the second vehicle 63, whereupon the second vehicle 63 transmits its second functional data set to the first vehicle 10 in step S405. In step S406, the first vehicle 10 then installs the second functional data set in its driving system 30 or adapts its driving system 30 with the second functional data set.

In step S500, first vehicle 10, whose first functional data set has now been adapted using the second functional data set of second vehicle 63, drives further kilometers while driving system 30 and/or control unit 40 use the model on which driving system 30 is based, neural network, or the target function, i.e. the underlying function data set, is adapted or updated.

After the vehicle has covered a predetermined distance or a predetermined driving time, it transmits a request for an upgrade of its license level to the network server 70 in step S501. With the request, the vehicle transmits information about its license level 1, its functional data record and its driving behavior ten to the network server 70. The data relating to its functional data record include, in particular, information about the contact with the second vehicle 63 with license level 4 and the information about driving behavior include, in particular, the driving time covered without an accident and the computing time used to adapt the functional data.

In step S502, the network server 70 uses the information received from the vehicle 10 to check whether this information authorizes the vehicle 10 to upgrade its license level. If the result of this check is positive, the network server 70 transmits certificate information to the vehicle 10 in step S503, which entitles it to an upgrade to license level 2. In step S600, the vehicle 10 upgrades to license level 2 based on the received certificate information. Thus, in the driving system 30 of the vehicle 10, the degree of automation level 2 (“partially automated”) is enabled. In the subsequent step S601, the vehicle drives with automation level 2, with the driver having to monitor the vehicle continuously, but the driving system 30 already takes over the longitudinal and lateral guidance in a specific application, with a further adaptation of the functional data set of the driving system 30 taking place.

Reference List

10

vehicle

11

first sensor

12

second sensor

13

third sensor

20

communication module

21

Storage

22

transponder

30

driving system

31

Storage

32

navigation module

40

control unit

41

Storage

42

CPU

51

fourth sensor

52

fifth sensor

53

sixth sensor

61

satellite

62

base station

63

other vehicle

70

network server

80

control unit

81

Storage

82

CPU

90

communication module

91

transponder

92

Storage

100

system

Claims (10)

Method for operating a first self-driving vehicle (10), having a first communication module (20) set up for communication with second vehicles (63); a driving system (30) set up for automatic driving operation of the first vehicle (10) with a first functional data set and a first control unit (40), the method having the method steps: establishing a communication link with a second vehicle (63); Receiving and determining a license level of the second vehicle (63), the range of functions of a driving system (30) of a vehicle (10, 63) increasing as the license level of the vehicle (10, 63) increases; Comparison of the determined license level of the second vehicle (63) with a license level of the first vehicle (10); receiving a second feature record from the second vehicle (63) if the determined license level of the second vehicle (63) exceeds the license level of the first vehicle (10); and Driving operation of the first vehicle (10) by means of the driving system (30) using the second functional data set. procedure after claim 1 , further comprising the steps of: receiving identification information of the second vehicle (63); Receiving and/or using the second functional data record based on the determined license level of the second vehicle (63) and based on the identification information of the second vehicle (63). procedure after claim 2 , further comprising the steps of: determining a vehicle type of the second vehicle (63) based on the received identification information; Comparison of the determined vehicle type of the second vehicle (63) with a vehicle type of the first vehicle (10); and receiving the second capability record from the second vehicle (63) when the first drive vehicle (63) and the second vehicle (63) are of the same type; or using the second function data record for type-independent functions of the driving system (30). Procedure according to one of Claims 1 until 3 ,Further comprising the method step: driving operation of the first vehicle (10) by means of the driving system (30) depending on the license level of the first vehicle (10). Procedure according to one of claims 2 until 4 , further comprising the method steps: transmission of information on the license level, on the function data set and on the driving behavior of the first vehicle (10) to a network server (70); receiving certificate information for a higher license level from the network server (70). procedure after claim 5 , wherein the information on the function data record has information on an update carried out by the first vehicle (10) and/or on the license level of the second vehicle (63). procedure after claim 5 or 6 , wherein the additional information has information about a driving time, a driving distance and/or a damage-free status with the current license level of the first vehicle (10). Vehicle (10), having a first communication module (20) set up for communication with other vehicles (63); a driving system (30) set up for automatic driving operation of the vehicle (10) and a driving system for carrying out the method according to one of Claims 1 until 7 furnished first control unit (40). Network server (70), having a second communication module (90) set up for data communication with a plurality of vehicles (10); and a second control unit (80), the second control unit (80) being arranged to: receive information from the vehicle (10) about a license level, about a functional data record and about a driving behavior of the vehicle (10); on the basis of the information received and a defined catalog of requirements to check whether the vehicle (10) is authorized for a higher license level; and transmit certificate information for a higher license level to a vehicle (10) authorized for the higher license level, the range of functions of a driving system (30) of the vehicle (10, 63) set up for automatic driving operation of the vehicle increasing as the license level of the vehicle (10, 63) increases. Computer program, comprising instructions which, when the program is executed by a computer, cause the latter to carry out the steps of the method according to one of Claims 1 until 7 to execute.

Images