DE102023205600A1 - Driving autonomous or semi-autonomous vehicles - Google Patents

Abstract

The present invention relates to a device for driving an autonomous or partially autonomous vehicle (12), comprising: an input interface (24) for receiving an action signal from an external monitoring device (16) in an environment of the vehicle, wherein the action signal comprises information about a driving maneuver to be carried out by the vehicle; an evaluation unit (26) for determining a control signal for an actuator (20) of the vehicle based on the action signal; a checking unit (28) for checking whether the driving maneuver to be carried out impairs vehicle safety of the vehicle based on the action signal; and an output interface (30) for controlling the actuator to carry out the driving maneuver based on the control signal if the driving maneuver to be carried out does not impair vehicle safety. The present invention further relates to a system (10) for driving an autonomous or semi-autonomous vehicle (12) and to a method for driving an autonomous or semi-autonomous vehicle (12) and a method for monitoring an autonomous or semi-autonomous vehicle (12).

Description

The present invention relates to a device for driving an autonomous or partially autonomous vehicle. The present invention further relates to an external monitoring device for monitoring an autonomous or partially autonomous vehicle. In addition, the present invention relates to a system for driving an autonomous or partially autonomous vehicle as well as a vehicle and two methods.

Modern vehicles (cars, trucks, motorcycles, etc.) have a large number of sensors (radar, lidar, camera, ultrasound, etc.) that provide information to a driver or a vehicle control system. Such environmental sensors record the vehicle's surroundings, objects in this environment (infrastructure objects, moving objects, etc.) and other road users (other vehicles, pedestrians, cyclists, etc.). Based on the data recorded, a model of the vehicle's environment can be created and changes in this vehicle environment can be responded to. In particular, a driver can be warned of dangers or a driving function of the vehicle can be carried out partially or fully autonomously. A driver assistance system (Advanced Driver Assistance System, ADAS) is a device in a motor vehicle that supports the driver in driving situations and/or carries out individual driving functions fully automatically.

One trend in this context is so-called sensorless vehicles. Both vehicle manufacturers and researchers are working on doing away with sensors arranged on the vehicle and instead relying on sensors in the environment and corresponding communication approaches (Vehicle2X, Vehicle2Vehicle, etc.). By doing away with sensors and evaluation electronics in the vehicle, costs can be saved and complexity reduced. On the other hand, new challenges arise in connection with the necessary communication and the reliability of the sensors in the environment.

In this area of conflict between the costs of integrating more and more sensors and corresponding evaluation hardware on the one hand and the most extensive possible autonomous operation of a vehicle, the present invention is aimed at providing an approach for driving an autonomous or partially autonomous vehicle. On the one hand, the most extensive possible autonomy when driving the vehicle should be enabled, i.e. the most comprehensive possible support should be offered to passengers of the vehicle or to a driver of the vehicle. On the other hand, the sensors used for this should be minimized as much as possible in order to reduce costs and also avoid possible errors due to sensor errors. In particular, the autonomous or partially autonomous operation of a vehicle should be enabled without this vehicle having its own medium and long-range environmental sensors.

• einer Eingangsschnittstelle zum Empfangen eines Aktionssignals von einer externen Überwachungseinrichtung in einer Umgebung des Fahrzeugs, wobei das Aktionssignal Informationen zu einem von dem Fahrzeug auszuführenden Fahrmanöver umfasst;
• einer Auswerteeinheit zum Ermitteln eines Steuersignals für einen Aktor des Fahrzeugs basierend auf dem Aktionssignal;
• einer Überprüfungseinheit zum Überprüfen, ob das auszuführende Fahrmanöver eine Fahrzeugsicherheit des Fahrzeugs beeinträchtigt basierend auf dem Aktionssignal; und
• einer Ausgangsschnittstelle zum Ansteuern des Aktors zum Ausführen des Fahrmanövers basierend auf dem Steuersignal, wenn das auszuführende Fahrmanöver die Fahrzeugsicherheit nicht beeinträchtigt.

To achieve this object, the present invention relates in one aspect to a device for driving an autonomous or semi-autonomous vehicle, comprising:

• an input interface for receiving an action signal from an external monitoring device in an environment of the vehicle, the action signal comprising information about a driving maneuver to be performed by the vehicle;
• an evaluation unit for determining a control signal for an actuator of the vehicle based on the action signal;
• a checking unit for checking whether the driving maneuver to be performed impairs vehicle safety based on the action signal; and
• an output interface for controlling the actuator to execute the driving maneuver based on the control signal if the driving maneuver to be executed does not compromise vehicle safety.

• einer Sensorschnittstelle zum Empfangen eines Sensorsignals mit Informationen zu einem Fahrzeug von einem Umgebungssensor;
• einer Bewertungseinheit zum Ermitteln eines Aktionssignals mit Informationen zu einem durch das Fahrzeug auszuführenden Fahrmanöver basierend auf dem Sensorsignal; und
• einer Sendeeinheit zum Übermitteln des Aktionssignals an das Fahrzeug.

Another aspect of the present invention relates to an external monitoring device for monitoring an autonomous or semi-autonomous vehicle, comprising:

• a sensor interface for receiving a sensor signal with information about a vehicle from an environmental sensor;
• an evaluation unit for determining an action signal with information about a driving maneuver to be performed by the vehicle based on the sensor signal; and
• a transmitting unit for transmitting the action signal to the vehicle.

A further aspect of the present invention relates to a system for controlling an autonomous or semi-autonomous vehicle with a device as described above and an external monitoring device as described above.

A further aspect of the present invention relates to a vehicle with a device as described above, wherein the vehicle has no medium or long-range environmental sensors arranged in the vehicle.

Further aspects of the invention relate to methods designed in accordance with the device and the external monitoring device for driving or monitoring an autonomous or partially autonomous vehicle, as well as computer program products with program code for carrying out the steps of the methods when the program code is executed on a computer. In addition, aspects of the invention relate to storage media on which computer programs are stored which, when executed on a computer, cause the methods described herein to be carried out.

Preferred embodiments of the invention are described in the dependent claims. It is understood that the features mentioned above and those to be explained below can be used not only in the combination specified in each case, but also in other combinations or on their own, without departing from the scope of the present invention. In particular, the device, the external monitoring device, the system, the vehicle, the methods and the computer program product can be designed in accordance with the embodiments described for the device and the external monitoring device in the dependent claims.

According to the invention, an action signal is received from the external monitoring device in the vehicle or in the device. The external monitoring device is arranged in an environment of the vehicle and has corresponding sensors to detect the vehicle. The environmental sensors, for example radar, lidar or camera sensors, are therefore not arranged on the vehicle itself, but in the environment. The received sensor signal is also evaluated in the external monitoring device in order to determine a driving maneuver to be carried out by the vehicle. This driving maneuver is then communicated to the device and used there to control a corresponding actuator of the vehicle. In the device itself, it is only checked whether the driving maneuver to be carried out impairs the safety of the vehicle. For example, it can be checked whether the driving maneuver will lead to an accident, provided this can be determined on the basis of the available information. In particular, a check can be carried out in the form of a so-called sanity check. In other words, an action that the vehicle should carry out is determined in a unit arranged outside the vehicle. The vehicle then carries out the determined action, whereby an action corresponds in particular to a driving maneuver (in particular steering, braking, accelerating or combinations thereof). A driving maneuver is understood here in particular to mean an adjustment/change of a movement of the vehicle.

In comparison to approaches in which environmental sensors are provided on the vehicle itself, the invention makes it possible to largely dispense with environmental sensors on the vehicle. In this respect, cost savings can be achieved. In contrast, the sensors are provided on a monitoring device arranged externally to the vehicle, which is located in the environment. The evaluation is also carried out externally to the vehicle. In comparison to recording sensor data inside the vehicle, computing power in the vehicle can be dispensed with, which can lead to further cost savings. It is possible for a single monitoring device to take over the data acquisition and evaluation for a large number of vehicles. In this respect, added value can be realized, since redundant sensors on different vehicles can be dispensed with. Finally, it is possible to provide higher processor and evaluation capacities on externally arranged monitoring devices, which can also improve traffic safety if necessary. In addition, the centralized determination of action signals for several road users makes it possible to coordinate and coordinate driving maneuvers of several road users. By dispensing with sensors and evaluation in the vehicle, cost savings can be achieved on the one hand and improved evaluation and increased safety can be made possible on the other. In addition, the approach according to the invention enables greater accuracy, since a large number of different monitoring devices can be accessed at critical points, for example. For example, different cameras that record a traffic situation at an intersection from different angles or sensor signals from different cameras can be used.

In a preferred embodiment of the device, the checking unit for checking whether the driving maneuver to be carried out impairs the safety of the vehicle is designed based on an artificial intelligence approach. For example, an artificial neural network can be used. A learning algorithm can thus achieve improved safety for the vehicle and other road users involved.

In a preferred embodiment, the evaluation unit is designed to determine the control signal based on an artificial intelligence approach. Additionally or alternatively, the evaluation unit is designed to determine the control signal for a controller and/or a brake of the vehicle. For example, an artificial neural network or some type of learning algorithm can be used in the evaluation unit. By using an artificial intelligence approach, efficient data processing can be achieved. In addition, a high level of precision can be achieved when determining the control signal, since a learning effect is made possible on the basis of previous traffic situations. By directly controlling a controller, i.e. a unit by which the vehicle is controlled, and/or a brake, a direct driving intervention is carried out. The movement of the vehicle is directly influenced.

In a preferred embodiment, the input interface is designed to receive the action signal via direct short-range communication. Additionally or alternatively, the input interface is designed to receive the action signal from an intelligent infrastructure unit. For example, a low-energy short-range communication standard can be used as direct short-range communication. For example, data can be transmitted via the 2.4 gigahertz frequency band. The action signal can be received by a correspondingly designed communication chip. It is also possible for transmission to take place via a short-range communication unit that is already provided on the vehicle and for the input interface to be connected to a corresponding interface of this short-range communication unit. Transmission errors can be avoided by using direct short-range communication. Secure data transmission can be provided, particularly in the case of time-critical reactions to changes in the environment. An intelligent infrastructure unit is understood to mean in particular an infrastructure object, such as a traffic light, a lane boundary, a traffic sign, etc. Sensors and corresponding data evaluation devices can be embedded in such intelligent infrastructure objects or infrastructure units. By communicating with such intelligent infrastructure objects, access to their sensors is possible.

In a preferred embodiment, the input interface is designed to receive a current vehicle position of a vehicle from the external monitoring device. The evaluation unit is designed to determine the control signal based on the current vehicle position. In particular, it is possible that in addition to the action signal, a current vehicle position is also received, which can possibly be determined by the external monitoring device with comparatively high precision. By using a current vehicle position determined by the external monitoring device, highly precise localization can be made possible. In particular, precise localization in relation to other road users and also in relation to infrastructure objects can be made possible. This results in efficient localization and efficient determination of the control signal.

In a preferred embodiment of the external monitoring device, the sensor interface is designed to receive the sensor signal from a sensor arranged on the monitoring device, in particular a radar, lidar or camera sensor. The sensor is advantageously arranged directly on the monitoring device. A corresponding signal is therefore recorded directly by a sensor in the environment and the action signal is determined accordingly. Both the environmental sensors and the evaluation of the sensor signal from the environmental sensors take place in the immediate vicinity of the vehicle. A high level of safety is achieved by avoiding the need to connect additional components.

In a preferred embodiment of the external monitoring device, the evaluation unit is designed to determine the action signal based on an artificial intelligence approach. In particular, a pre-trained artificial neural network can be used. By using an artificial intelligence approach, an action signal can be determined efficiently that is appropriate to the situation. In addition, a precise prediction is made regarding the driving maneuver to be carried out by the vehicle.

In a preferred embodiment, the transmitting unit is designed to transmit the action signal to the vehicle via direct short-distance communication. In particular, the transmission can be based on a low-energy short-distance communication standard, for example. Transmission via several intermediate stations is avoided. This results in a high level of security in the data transmission and thus a high level of certainty that the action signal is available promptly and that the vehicle can initiate the driving maneuver to be carried out.

The term “environment” of a vehicle is understood here to mean, in particular, the immediate surroundings of the vehicle. The environment of the vehicle includes, in particular, the area visible through sensors arranged on the vehicle. The environment also includes, in particular, the area within which a sensor on an infrastructure unit or on another monitoring device can detect the vehicle. The environment of the vehicle can be defined, for example, by specifying a radius. However, it is also possible for the environment of the vehicle to be an irregular area. An autonomous or partially autonomous vehicle is, in particular, a vehicle that can at least partially carry out driving maneuvers autonomously without the need for a driver to intervene. Vehicle safety is particularly impaired if the driving maneuver can pose a risk to the vehicle and/or the vehicle’s surroundings and other road users. To estimate whether a driving maneuver impairs vehicle safety, a comparison with a predefined threshold value can be made, for example. Road users can be, in particular, pedestrians and cyclists as well as other vehicles.

• 1 eine schematische Darstellung eines erfindungsgemäßen Systems zum Führen eines autonomen oder teilautonomen Fahrzeugs;
• 2 eine schematische Darstellung einer erfindungsgemäßen Vorrichtung;
• 3 eine schematische Darstellung einer erfindungsgemäßen externen Überwachungseinrichtung;
• 4 eine schematische Darstellung einer beispielhaften Verkehrssituation unter Verwendung des erfindungsgemäßen Ansatzes zum Führen eines autonomen oder teilautonomen Fahrzeugs;
• 5 eine schematische Darstellung einer E/E-Architektur, die den erfindungsgemäßen Ansatz umsetzt;
• 6 eine schematische Darstellung eines erfindungsgemäßen Verfahrens zum Führen eines autonomen oder teilautonomen Fahrzeugs; und
• 7 eine schematische Darstellung eines erfindungsgemäßen Verfahrens zum Überwachen eines autonomen oder teilautonomen Fahrzeugs.

The invention is described and explained in more detail below using some selected embodiments in conjunction with the accompanying drawings. They show:

• 1 a schematic representation of a system according to the invention for driving an autonomous or semi-autonomous vehicle;
• 2 a schematic representation of a device according to the invention;
• 3 a schematic representation of an external monitoring device according to the invention;
• 4 a schematic representation of an exemplary traffic situation using the inventive approach for driving an autonomous or semi-autonomous vehicle;
• 5 a schematic representation of an E/E architecture implementing the inventive approach;
• 6 a schematic representation of a method according to the invention for driving an autonomous or semi-autonomous vehicle; and
• 7 a schematic representation of a method according to the invention for monitoring an autonomous or semi-autonomous vehicle.

In the 1 a system 10 according to the invention for driving an autonomous or partially autonomous vehicle 12 is shown schematically. The system 10 comprises a device 14 and an external monitoring device 16. The illustration is to be understood as a schematic side view of a traffic situation. In the example shown, the device 14 is integrated into the vehicle 12. The monitoring device 16 is integrated into an intelligent infrastructure unit 18, a traffic light in the exemplary embodiment shown, or is attached to the traffic light. An actuator 20 of the vehicle 12 is controlled by means of the device 14 according to the invention in order to carry out a driving maneuver. For example, a control or a brake of the vehicle can be controlled in order to initiate or carry out a driving maneuver in this way. The vehicle according to the invention can in particular be a vehicle that has no (or only selected) medium or long-distance environmental sensors and in this respect does not have its own sensor system for detecting other road users.

The system 10 according to the invention provides that the vehicle 12 and its surroundings are detected externally, for example within a field of view 21 of a sensor of the external monitoring device 16, and a situation-appropriate reaction of the vehicle 12 is determined. The driving maneuver to be carried out is not determined in the vehicle 12 itself, but in the external monitoring device 16 and transmitted to the vehicle 12 in the form of an action signal. The vehicle 12 carries out the driving maneuver.

In the illustrated embodiment in 1 For example, an object 22 lying on the road can be detected by the external monitoring device 16 and an evasive maneuver for the vehicle 12 can be determined. The action signal with information about this evasive maneuver is then transmitted from the external monitoring device 16 to the vehicle 12 or the device 14. By means of the device 14, a corresponding actuator 20 of the vehicle 12 is controlled in such a way that the driving maneuver is implemented. For example, a trajectory for the evasive maneuver of the vehicle can be specified in a vehicle coordinate system and the evasive maneuver can then be carried out based on a corresponding control of actuators of the vehicle.

Compared to previous approaches, the approach according to the invention offers the advantage that sensors and evaluation electronics can be dispensed with within the vehicle 12. For example, a reduced sensor set can be used in the vehicle 12. In particular, it is advantageous if approaches of artificial intelligence implemented, as these often place comparatively high demands on available sensors and evaluation electronics. By arranging the corresponding sensors and evaluation electronics in the external monitoring device 16, costs for the vehicle can be saved.

In the 2 In this context, a device 14 according to the invention for driving an autonomous or partially autonomous vehicle is shown schematically. The device comprises an input interface 24, an evaluation unit 26, a checking unit 28 and an output interface 30. The units and interfaces can be partially or completely implemented in software and/or hardware. In particular, the units can be designed as a processor, processor modules or also as software for a processor. The device 14 can in particular be designed in the form of a control unit or a central computer of an autonomous or partially autonomous vehicle or as software for such a control unit or such a central computer. For example, the device 14 can be implemented in software that is executed on a central computer of the vehicle and can be connected to communication devices and actuators of the vehicle via corresponding interfaces.

An action signal is received via the input interface 24. For this purpose, the input interface 24 is connected, for example, to a corresponding communication module of the vehicle. The connection can be made, for example, via a bus system of the vehicle. It is particularly advantageous if short-distance communication is used and direct communication is established with the monitoring device 16. For example, a 2.4 gigahertz communication connection can be established. However, it goes without saying that a connection via a mobile data connection can also be used in principle. The action signal can be received in particular in the form of a digital signal with a predefined format. The driving maneuver to be carried out is preferably stored in the action signal according to a predefined syntax. This predefined syntax can then be evaluated in the device.

In the evaluation unit 26, a control signal for an actuator of the vehicle is determined based on the action signal. The action signal is evaluated or processed. The control signal corresponds to a control of an actuator of the vehicle. In the evaluation unit 26, for example, vehicle-specific adjustments or vehicle actuator-dependent settings are made. In particular, an artificial intelligence approach can be implemented in the evaluation unit 26 to determine the control signal. For example, the use of an artificial neural network that has already been pre-trained can be provided. In addition, it is also possible for a type of experience-based learning to be provided in order to achieve increasing accuracy in determining the control signal over the duration of use of the device according to the invention. The evaluation unit 26 can be designed to determine the control signal, for example for a control system, i.e. a steering system of the vehicle and/or a brake of the vehicle.

The checking unit 28 checks whether the driving maneuver to be carried out compromises the safety of the vehicle. The action signal is used as a basis for this. The checking unit 28 can also use other data that may be available for this check, for example data regarding the current movement of the vehicle. The check can be carried out in particular in the form of an estimate of a probability of compromise of vehicle safety with predefined criticality. For example, it can be determined that the driving maneuver will lead to a collision with a probability of 80%. By applying a further criterion, such a high probability of a collision can then lead to the determination that the driving maneuver compromises vehicle safety. If necessary, an alternative or modified driving maneuver can then be initiated. For example, emergency braking can be triggered. The checking unit 28 can use other data for the check. Before the actuator is activated, it is therefore checked whether the driving maneuver does not compromise vehicle safety. The verification unit 28 also preferably implements an artificial intelligence approach, for example an artificial neural network that is pre-trained.

An actuator of the vehicle is controlled by means of the output interface 30. For this purpose, the output interface 30 is connected to the various actuators of the vehicle, for example to brakes, control devices, etc., via a bus system of the vehicle. The driving maneuver is implemented by controlling the actuator.

In the 3 An embodiment of a monitoring device 16 according to the invention is shown schematically. The monitoring device 16 comprises a sensor interface 32, an evaluation unit 34 and a transmission unit 36. The units and Interfaces of the external monitoring device 16 can be partially or completely implemented in software and/or hardware.

The sensor interface 32 is used to receive the sensor signal. For example, the sensor interface can be connected to an environmental sensor for this purpose. For example, data can be received from a camera, lidar, radar or ultrasound sensor. In particular, pre-processed data can be used for which parts of the evaluation have already been carried out. The sensor interface 32 can in particular be in communication with a sensor arranged on an infrastructure unit.

An action signal is determined in the evaluation unit 34. For this purpose, the sensor signal is evaluated. It goes without saying that several sensor signals can be evaluated in order to determine an action signal with greater significance. The action signal can be determined in particular in the form of a signal with predefined syntax. In other words, the driving maneuver to be carried out can be described abstractly. The evaluation unit 34 can also implement an artificial intelligence approach, for example a pre-trained artificial neural network.

The transmitting unit 36 serves to transmit the action signal to the vehicle. For example, the transmitting unit 36 can be connected to a corresponding short-range communication unit. The transmission can be designed according to the communication standards previously described with regard to the device.

In 4 the approach according to the invention is shown schematically. In particular, according to the approach of the present invention, an E/E architecture for use in vehicles without medium or long-range environmental sensors is described. The communication between vehicles according to the invention can, for example, be as in 4 shown, whereby 4 is to be understood as a schematic plan view of a traffic situation. In particular, 4 two vehicles 12 according to the invention are shown, which are in communication with an external monitoring device 16. In the embodiment shown, the external monitoring device 16 receives its information directly from various sensors or other infrastructure objects. On the right-hand side, the various data sources or intelligent infrastructure objects are shown, which can be used to record the environment using environmental sensors (for example satellites, intelligent traffic lights, intelligent gas stations, cloud systems, ...).

In the exemplary embodiment shown, the additional vehicles 37 shown in dashed lines represent conventional vehicles that can optionally have vehicle-to-vehicle or vehicle-to-infrastructure communication devices. The external monitoring device 16 executes artificial intelligence algorithms based on the measurements or information received from the various data sources on the right-hand side. The data is processed in the external monitoring device 16 and the result is then transmitted to the vehicle 12 according to the invention in the form of an action signal. According to the invention, the data acquisition and data processing are therefore arranged outside the vehicle. In the vehicle 12 itself, only post-processing is used to ensure that the vehicle safety of the vehicle is not compromised when the proposed driving maneuver is carried out.

In 5 a possible E/E architecture of a vehicle 12 according to the invention is shown schematically. It is indicated that the vehicle 12 in the exemplary embodiment shown is in wireless communication with the external monitoring device 16. In particular, the action signal and possibly further information are transmitted wirelessly. The action signal can, for example, include a trajectory for the vehicle that the vehicle should follow. The action signal can also include information about a desired vehicle behavior of the vehicle 12.

The communication between the components of the vehicle 12 and the externally arranged units (sensors, intelligent infrastructure devices, etc., cf. 4 ) is limited and takes place exclusively via the external monitoring device 16. This requires a reduction in the number of necessary communication channels. In addition, internal vehicle access to raw data is reduced. Within the vehicle 12, the number of processor units can consequently be reduced. This results in reduced complexity with regard to the number of connected processor units. In the exemplary embodiment shown, a control ECU 38 (Electronic Control Unit), a suspension ECU 40, a communication ECU 42 and a brake ECU 44 are arranged within the vehicle. It is understood that further ECUs can be provided. The various ECUs are connected. In addition, different elements 46, such as actuators, valves, lights, switches, etc., are connected to the various ECUs.

In the external monitoring device 16, all non-time-critical tasks are carried out and the digital environment is also created. In addition, some cybersecurity functions are implemented. Information that is not relevant to the movement of the vehicle (for example infotainment) can also be processed outside the vehicle. In addition, time-critical tasks and, if necessary, kinematics-based localization can also be carried out in the external monitoring device 16. It is also possible for information relating to the movement of the vehicle to be processed. Safety-critical or time-critical decisions can be made within the vehicle 12. In addition, cybersecurity requirements can also be partially met internally.

In 6 A method according to the invention for driving an autonomous or partially autonomous vehicle is shown schematically. The method comprises steps of receiving S10 an action signal, determining S12 a control signal, checking S14 whether a driving maneuver impairs vehicle safety, and controlling S16 an actuator of the vehicle.

In 7 a method according to the invention for monitoring an autonomous or partially autonomous vehicle is shown. The method comprises steps of receiving S20 a sensor signal, determining S22 an action signal and transmitting S24 the action signal.

Both methods can in particular be implemented in software, whereby the software can be executed on a processor of a vehicle or a processor of an intelligent infrastructure unit.

The invention has been fully described and explained with reference to the drawings and the description. The description and explanation are to be understood as exemplary and not restrictive. The invention is not limited to the disclosed embodiments. Other embodiments or variations will become apparent to those skilled in the art upon use of the present invention and upon careful analysis of the drawings, the disclosure and the following claims.

In the claims, the words "comprising" and "having" do not exclude the presence of further elements or steps. The undefined article "a" or "an" does not exclude the presence of a plurality. A single element or unit can perform the functions of several of the units mentioned in the claims. An element, unit, interface, device and system can be partially or completely implemented in hardware and/or software. The mere mention of some measures in several different dependent claims is not to be understood as meaning that a combination of these measures cannot also be used advantageously. A computer program can be stored/distributed on a non-volatile data carrier, for example on an optical memory or on a semiconductor drive (SSD). A computer program can be distributed together with hardware and/or as part of hardware, for example via the Internet or via wired or wireless communication systems. Reference signs in the claims are not to be understood as limiting.

Reference symbols

10

system

12

vehicle

14

contraption

16

external monitoring device

18

Infrastructure unit

20

Actuator

21

Field of view

22

object

24

Input interface

26

Evaluation unit

28

Verification unit

30

Output interface

32

Sensor interface

34

Valuation unit

36

Transmitter unit

37

additional vehicle

38

Control ECU

40

Suspension ECU

42

Communication ECU

44

Brake ECU

46

elements

Claims (14)

Device for driving an autonomous or semi-autonomous vehicle (12), comprising: an input interface (24) for receiving an action signal from an external monitoring device (16) in an environment of the Vehicle, wherein the action signal comprises information about a driving maneuver to be carried out by the vehicle; an evaluation unit (26) for determining a control signal for an actuator (20) of the vehicle based on the action signal; a checking unit (28) for checking whether the driving maneuver to be carried out impairs vehicle safety of the vehicle based on the action signal; and an output interface (30) for controlling the actuator to carry out the driving maneuver based on the control signal if the driving maneuver to be carried out does not impair vehicle safety. Device according to Claim 1 , wherein the checking unit (28) is designed to check whether the driving maneuver to be carried out impairs vehicle safety of the vehicle (12) based on an artificial intelligence approach. Device according to one of the preceding claims, wherein the evaluation unit (26) is designed to determine the control signal based on an artificial intelligence approach; and/or is designed for a control and/or a brake of the vehicle (12). Device according to one of the preceding claims, wherein the input interface (24) is designed to receive the action signal via a direct short-range communication; and/or is designed by an intelligent infrastructure unit. Device according to one of the preceding claims, wherein the input interface (24) is designed to receive a current vehicle position of the vehicle (12) from the external monitoring device (16); and the evaluation unit (26) is designed to determine the control signal based on the current vehicle position. External monitoring device (16) for monitoring an autonomous or partially autonomous vehicle (12), with: a sensor interface (32) for receiving a sensor signal with information about a vehicle from an environmental sensor; an evaluation unit (34) for determining an action signal with information about a driving maneuver to be carried out by the vehicle based on the sensor signal; and a transmission unit (36) for transmitting the action signal to the vehicle. External monitoring device (16) according to Claim 6 , wherein the sensor interface (32) is designed to receive the sensor signal from a sensor arranged on the monitoring device, in particular a radar, lidar or camera sensor. External monitoring device (16) according to one of the Claims 6 until 7 , wherein the evaluation unit (34) is designed to determine the action signal based on an artificial intelligence approach, in particular based on a pre-trained artificial neural network. External monitoring device (16) according to one of the Claims 6 until 8th , wherein the transmitting unit (36) is designed to transmit the action signal to the vehicle (12) via direct short-range communication. System (10) for driving an autonomous or semi-autonomous vehicle (12), comprising a device according to one of the Claims 1 until 5 and an external monitoring device (16) according to one of the Claims 6 until 9 . Vehicle (12) with a device (14) according to one of the Claims 1 until 5 , whereby the vehicle does not have any medium or long-range environmental sensors arranged on the vehicle. Method for driving an autonomous or partially autonomous vehicle (12), comprising the steps of: Receiving (S10) an action signal from an external monitoring device (16) in an environment of the vehicle, wherein the action signal comprises information about a driving maneuver to be carried out by the vehicle; Determining (S12) a control signal for an actuator (20) of the vehicle based on the action signal; Checking (S14) whether the driving maneuver to be carried out impairs vehicle safety of the vehicle based on the action signal; and Controlling (S16) the actuator to carry out the driving maneuver based on the control signal if the driving maneuver to be carried out does not impair vehicle safety. Method for monitoring an autonomous or partially autonomous vehicle (12), comprising the steps of: receiving (S20) a sensor signal with information about a vehicle from an environmental sensor; determining (S22) an action signal with information about a driving maneuver to be carried out by the vehicle based on the sensor signal; and Transmitting (S24) the action signal to the vehicle. Computer program product with program code for carrying out the steps of the method according to one of the Claims 12 or 13 when the program code is executed on a computer.

Images