Controlling an autonomous or semi-autonomous vehicle using a digital map with a behavioral data layer

Abstract

The present invention relates to a method, a computer program with instructions, and a device for controlling an autonomous or semi-autonomous vehicle. In a first step, sensor data or decisions of a motion planner are acquired (10). The current driving behavior of the vehicle is determined from the acquired data (11). The current driving behavior is compared with an expected driving behavior (12) stored in a behavior data layer of a digital map. If a deviation between the current driving behavior and the expected driving behavior is detected (13), it can first be determined (14) whether the detected deviation should be used to update the digital map, whether a teleoperator should be involved, or whether a maneuver should be initiated to bring the vehicle into a safe state.Alternatively or additionally, a notification and optionally sensor data can be sent to a backend (15). In either case, the current driving behavior is adjusted (16). The invention also relates to an autonomous or semi-autonomous vehicle, a digital map, and a method, a computer program with instructions, and a device for updating such a digital map.

Description

[0001] The present invention relates to a method, a computer program with instructions, and a device for controlling an autonomous or semi-autonomous vehicle. The invention also relates to an autonomous or semi-autonomous vehicle that uses a method or device according to the invention, as well as to a digital map for use in such a method. The invention further relates to a method, a computer program with instructions, and a device for updating such a digital map.

[0002] Automated driving, also known as autonomous driving, automated driving, or guided driving, is the movement of vehicles, mobile robots, and driverless transport systems that are largely autonomous. There are different degrees of automated driving. In Europe, various transport ministries, such as the Federal Highway Research Institute in Germany, have defined the following levels of automation: • Level 0: “Driver only”, the driver drives, steers, accelerates, brakes, etc. • Level 1: Certain assistance systems help with vehicle operation, including a cruise control system such as ACC (Automatic Cruise Control). • Level 2: Partial automation. Automatic parking, lane guidance, general longitudinal guidance, acceleration, deceleration, etc., including collision avoidance, are handled by the assistance systems. • Level 3: High automation. The driver does not need to constantly monitor the system. The vehicle performs functions such as activating the turn signal, changing lanes, and maintaining lane guidance independently. The driver can attend to other tasks but must assume control upon request within a certain warning period. • Level 4: Full automation. The system permanently assumes control of the vehicle. If the system is no longer able to handle the tasks, the driver can be asked to take over. • Level 5: No driver required. Aside from setting the destination and starting the system, no human intervention is required.

[0003] A slightly different definition of steps is provided by the Society of Automotive Engineers (SAE). In this context, reference is made to the SAE J3016 standard. Such definitions can be used as an alternative to the definitions given above.

[0004] Autonomous or semi-autonomous vehicles must exhibit driving behavior that is consistent with traffic regulations and swarm behavior. A fundamental assumption is that the driving behavior of autonomous or semi-autonomous vehicles exhibits a certain correlation with the respective environment. For example, during a tight 90° right turn, the vehicle's trajectory will most likely deviate from the ideal centerline of the roadway to the right, as is the case with most manually controlled vehicles. This implies that the behavior of autonomous or semi-autonomous vehicles at the same geographical location is similar with respect to certain environmental features, such as the road surface or the behavior of other road users.

[0005] Autonomous or semi-autonomous vehicles typically derive environmental features from high-resolution map data. High-resolution maps can, in particular, display four different layers: • Physical data layer: A set of physical real-world elements that represent the stationary environment, e.g. road markings and traffic signs. • Relational data layer: A set of semantic elements composed of physical elements, e.g., lanes formed by road markings, or construction sites formed by temporary road markings, temporary traffic signs, and temporary traffic lights. • Topological data layer: A set of links between semantic elements, e.g. predecessors, successors and neighbors of lanes. • Meta layer: Additional abstract information for physical or semantic elements, e.g. connectivity / communication, temporality, for example at construction sites.

[0006] In this context, US 2018 / 0 004 225 A1 describes a method for operating a vehicle. In the method, a number of partial maps from a collection of partial maps are retrieved from a storage medium. Each partial map represents an area of a road network for a geographical region. Each partial map is provided with an identifier, several data layers, each data layer representing a feature set of the area of the road network of this partial map, and a connection data set for connecting the partial map to another partial map representing an adjacent area to the area of the road network of this partial map. The vehicle is controlled autonomously using the retrieved partial maps.

[0007] In principle, it can be assumed that the driving behavior of autonomous or semi-autonomous vehicles is generally based on the traffic rules encoded in the high-resolution maps, e.g. with regard to lane layout, direction of travel, speed limit, right of way, etc.

[0008] Violations by autonomous or semi-autonomous vehicles of the traffic rules encoded in the high-resolution maps can have several causes: • The high-resolution maps are outdated. For example, environmental changes such as construction sites may not be encoded in the map, or errors may occur due to incomplete map creation. • Perception problems occur, e.g. due to faulty perception, occlusions or limitations of the performance of the algorithms used. • Special events, such as the formation of an emergency lane due to a rescue operation.

[0009] The causes mentioned above each lead to an anomaly in the driving behavior of the autonomous or semi-autonomous vehicle, which, in the worst case, endangers the safety of all road users. Autonomous or semi-autonomous vehicles therefore require a strategy to detect their own deviant driving behavior and thus avoid accidents.

[0010] It is an object of the invention to provide improved solutions for controlling an autonomous or semi-autonomous vehicle.

[0011] This object is achieved by a method having the features of claim 1 or 8, by a device having the features of claim 5 or 9, by an autonomous or semi-autonomous vehicle according to claim 6, by a digital map according to claim 7, and by a computer program with instructions according to claim 10. Preferred embodiments of the invention are the subject of the dependent claims.

[0012] According to a first aspect of the invention, a method for controlling an autonomous or semi-autonomous vehicle comprises the steps: - Comparing a current driving behavior of the vehicle with an expected driving behavior stored in a behavior data layer of a digital map; - Detecting a deviation between the current driving behavior and the expected driving behavior; and - Adjust the current driving behavior.

[0013] According to a further aspect of the invention, a computer program contains instructions which, when executed by a computer, cause the computer to perform the following steps for controlling an autonomous or semi-autonomous vehicle: - Comparing a current driving behavior of the vehicle with an expected driving behavior stored in a behavior data layer of a digital map; - Detecting a deviation between the current driving behavior and the expected driving behavior; and - Adjust the current driving behavior.

[0014] The term computer is to be understood broadly. In particular, it also includes control units, embedded systems and other processor-based

[0015] Data processing devices. The execution of the aforementioned steps can be performed directly by the computer or can consist of the computer controlling a component intended to execute a step.

[0016] The computer program may, for example, be made available for electronic retrieval or stored on a computer-readable storage medium.

[0017] According to a further aspect of the invention, a device for controlling an autonomous or semi-autonomous vehicle comprises: - a comparison module for comparing a current driving behavior of the vehicle with an expected driving behavior stored in a behavior data layer of a digital map and for detecting a deviation between the current driving behavior and the expected driving behavior; and - a control module to adapt the current driving behavior.

[0018] The inventive solution uses a digital map which, in addition to a relational data layer and a topological data layer, has a behavioral data layer. Expected driving behavior is stored in the behavioral data layer. This indicates the actual behavior of manually controlled vehicles on the roads with regard to specific road sections from the relational data layer and the topological data layer of the map. During the journey, the current driving behavior is compared with the expected driving behavior, i.e. with the relevant information from the behavioral data layer. If a deviation is detected, this may be due to the map deviating from the real situation, a perception problem having occurred, or a special event having occurred. In this case, the vehicle adapts its driving behavior, for example by reducing its speed.This significantly reduces the risk of an accident.

[0019] According to one aspect of the invention, the current driving behavior of the vehicle is an actual or planned driving behavior that is determined based on sensor data or decisions of a movement planner. The current driving behavior can be determined, on the one hand, by evaluating data from the vehicle's sensors, e.g., data from environmental sensors or data from sensors for recording internal vehicle variables, e.g., acceleration or steering angle sensors. In this case, the actual driving behavior is determined. Alternatively or additionally, decisions of a movement planner can also be considered, i.e., the planned driving behavior is determined. This has the advantage that a check for deviations can already be carried out before a planned driving maneuver is executed.

[0020] According to one aspect of the invention, if a deviation is detected between the current driving behavior and the expected driving behavior, a notification and optionally sensor data are sent to a backend. A mobile phone connection, for example, can be used for this purpose. The notifications can, in particular, also contain information about the current driving behavior. The backend can use the notification and the optionally sent sensor data to update the data in the behavior data layer or to influence the vehicle.

[0021] According to one aspect of the invention, if a deviation is detected between the current driving behavior and the expected driving behavior, it is determined whether the detected deviation should be used to update the digital map, whether a teleoperator should be involved, or whether a maneuver to bring the vehicle into a safe state should be initiated. These decisions can be made both in the backend and by means of a suitable device in the vehicle, e.g. using an artificial intelligence algorithm. If the decision to involve a teleoperator or to initiate a maneuver to bring the vehicle into a safe state is made in the backend, corresponding information or instructions are sent to the vehicle. In addition, an updated map can be sent to the vehicle or other vehicles if necessary.

[0022] Advantageously, an autonomous or semi-autonomous vehicle comprises a device according to the invention or is configured to carry out a method according to the invention. The vehicle can be any type of vehicle, e.g., a passenger car, a bus, a motorcycle, or a commercial vehicle, in particular a truck or an agricultural machine. Such a vehicle is characterized by the fact that the risk of accidents resulting from anomalies in driving behavior is reduced.

[0023] A digital map suitable for implementing a solution according to the invention preferably has at least one relational data layer, one topological data layer, and one behavioral data layer, with the behavioral data layer storing an expected driving behavior. The expected driving behavior can include a variety of information, e.g.: • average driving speed • average acceleration / deceleration • Steering angle / cornering behavior / deviation from the center line • Stopping, e.g. vehicles stop at a certain point due to an obstacle • Lane change despite prohibition by solid line or signage, ie taking into account rules coded in the topological data layer for the passable preceding and following lanes • Perception accuracy, e.g. to obtain information about lighting conditions • Signaling, e.g. flashing and honking • Control unit status, e.g. diagnostic data • dynamic environment, e.g. number of vehicles in the area

[0024] According to a further aspect of the invention, a method for updating a digital map comprises the steps: - Receiving notifications and optionally sensor data from a large number of autonomous or semi-autonomous vehicles regarding deviations between the current driving behavior of the respective vehicle and an expected driving behavior; - Evaluate the received notifications and sensor data; and - Updating an expected driving behavior stored in a behavioral data layer of the digital map based on a result of the evaluation.

[0025] According to a further aspect of the invention, a computer program includes instructions which, when executed by a computer, cause the computer to perform the following steps to update a digital map: - Receiving notifications and optionally sensor data from a large number of autonomous or semi-autonomous vehicles regarding deviations between the current driving behavior of the respective vehicle and an expected driving behavior; - Evaluate the received notifications and sensor data; and - Updating an expected driving behavior stored in a behavioral data layer of the digital map based on a result of the evaluation.

[0026] The term computer is to be understood broadly. In particular, it also includes workstations, distributed systems and other processor-based

[0027] Data processing devices. The execution of the aforementioned steps can be performed directly by the computer or can consist of the computer controlling a component intended to execute a step.

[0028] The computer program may, for example, be made available for electronic retrieval or stored on a computer-readable storage medium.

[0029] According to a further aspect of the invention, a device for updating a digital map comprises: - a receiving module for receiving notifications and optionally sensor data from a plurality of autonomous or semi-autonomous vehicles regarding deviations between the current driving behavior of the respective vehicle and an expected driving behavior; - an evaluation module for evaluating the received notifications and sensor data; and - an update module for updating an expected driving behavior stored in a behavior data layer of the digital map based on a result of the evaluation.

[0030] In the solution according to the invention, the vehicles of a fleet send a notification and optionally sensor data to a backend, e.g. via a mobile phone connection, if a deviation in driving behavior is detected. The notifications can in particular also contain information on the current driving behavior. The received information on the anomalous behavior is collected and processed in the backend. This makes it possible to determine whether a deviation is a problem with an individual vehicle or not. This makes it possible to distinguish map deviations from other anomalies. Since several vehicles in the same geographical area will report anomalous behavior, the expected driving behavior can also be adapted based on the reported deviations, i.e. the information in the behavioral data layer can be adapted by learning from the fleet.The correspondingly processed information is integrated into the behavioral data layer in the backend. At least the updated behavioral data layer can then be sent from the backend to the fleet's vehicles, e.g., via a mobile connection.

[0031] To generate the behavioral data layer, the vehicles in a fleet can continuously send the driving behavior information described above to a backend, e.g., via a cellular connection. This information is accumulated and processed in the backend, for example, by extracting trajectories or creating speed profiles. Both autonomous or semi-autonomous vehicles and manually controlled vehicles can be used to collect the data to generate statistics on nominal behavior.

[0032] Further features of the present invention will become apparent from the following description and the appended claims taken in conjunction with the figures. Fig. 1 schematically shows a method for controlling an autonomous or semi-autonomous vehicle; Fig. 2 shows a first embodiment of a device for controlling an autonomous or semi-autonomous vehicle; Fig. 3 shows a second embodiment of a device for controlling an autonomous or semi-autonomous vehicle; Fig. Figure 4 schematically represents a vehicle in which a solution according to the invention is implemented; Fig. 5 schematically shows a method for updating a digital map; Fig. 6 shows a first embodiment of a device for updating a digital map; Fig. 7 shows a second embodiment of a device for updating a digital map; Fig. 8 illustrates an expected driving behavior using the example of cornering; and Fig. Figure 9 schematically represents a digital map suitable for implementing a solution according to the invention.

[0033] To better understand the principles of the present invention, embodiments of the invention are explained in more detail below with reference to the figures. It is understood that the invention is not limited to these embodiments and that the described features may also be combined or modified without departing from the scope of the invention as defined in the appended claims.

[0034] Fig. Figure 1 schematically shows a method for controlling an autonomous or semi-autonomous vehicle. In a first step, sensor data or decisions from a motion planner are recorded 10. From the recorded data, the current driving behavior of the vehicle is determined 11. The current driving behavior is compared with an expected driving behavior 12, which is stored in a behavior data layer of a digital map. If a deviation between the current driving behavior and the expected driving behavior is detected 13, a notification and optionally sensor data can be sent to a backend 15. The notification can, in particular, also contain information about the current driving behavior. Alternatively or additionally, the next steps can first be determined 14, e.g.whether the detected deviation should be used to update the digital map, whether a remote operator should be involved, or whether a maneuver should be initiated to bring the vehicle into a safe state. In each case, the current driving behavior is adjusted.

[0035] Fig. Figure 2 shows a simplified schematic representation of a first embodiment of a device 20 for controlling an autonomous or semi-autonomous vehicle. The device 20 has an input 21 for receiving sensor data SD from a sensor system 41 or decisions E from a motion planner 42. A processing module 22 is configured to derive a current driving behavior F from this data. a of the vehicle. A comparison module 23 is set up to determine the current driving behavior F a with an expected driving behavior F estored in a behavioral data layer of a digital map and to determine a possible deviation A. A control module 24 is configured to determine the current driving behavior F in the event of such a deviation A. a To this end, the control module 24 can output a corresponding control signal S via an output 27 of the device 20. The comparison module 23 or the control module 24 can also be configured to send a notification B and optionally sensor data SD to a backend 80 via the output 27. The notification B can in particular also contain information on the current driving behavior. Alternatively or additionally, in addition to adapting the current driving behavior F aFirst, the next steps will be determined, e.g. whether the detected deviation A should be used to update the digital map, whether a teleoperator should be involved or whether a maneuver should be initiated to bring the vehicle into a safe state.

[0036] The processing module 22, the comparison module 23, and the control module 24 can be controlled by a control module 25. Settings of the processing module 22, the comparison module 23, the control module 24, or the control module 25 can be changed via a user interface 28. The data generated in the device 20 can be stored in a memory 26 of the device 20 if necessary, for example, for later evaluation or for use by the components of the device 20. The processing module 22, the comparison module 23, the control module 24, and the control module 25 can be implemented as dedicated hardware, for example, as integrated circuits. Of course, they can also be partially or completely combined or implemented as software running on a suitable processor, for example, a CPU or a GPU.The input 21 and the output 27 can be implemented as separate interfaces or as a combined bidirectional interface.

[0037] Fig. 3 shows a simplified schematic representation of a second embodiment of a device 30 for controlling an autonomous or semi-autonomous vehicle. The device 30 has a processor 32 and a memory 31. For example, the device 30 is a computer, a control unit, or an embedded system. Instructions are stored in the memory 31 which, when executed by the processor 32, cause the device 30 to perform the steps according to one of the described methods. The instructions stored in the memory 31 thus embody a program executable by the processor 32 which implements the method according to the invention. The device 30 has an input 33 for acquiring sensor data or decisions from a motion planner. Data generated by the processor 32 are provided via an output 34. In addition, data can be stored in the memory 31.The input 33 and the output 34 can be combined to form a bidirectional interface.

[0038] The processor 32 may include one or more processor units, such as microprocessors, digital signal processors, or combinations thereof.

[0039] The memories 26, 31 of the described embodiments can have both volatile and non-volatile memory areas and can comprise a wide variety of storage devices and storage media, for example hard disks, optical storage media or semiconductor memories.

[0040] Fig. 4 schematically illustrates an autonomous or semi-autonomous vehicle 40 in which a solution according to the invention is implemented. The vehicle 40 has a sensor system 41 for acquiring environmental information, such as cameras, radar sensors, lidar sensors, or ultrasonic sensors. A movement planner 42 is configured to plan a driving behavior of the vehicle 40. The movement planner 42 can be implemented, for example, in an assistance system or a high-performance computer of the vehicle 40. A device 20 according to the invention is configured to compare the current driving behavior of the vehicle 40, i.e., the planned or actual driving behavior, with an expected driving behavior. The expected driving behavior is stored in a behavior data layer of a digital map K, which is stored in a memory 44 of the vehicle 40. A connection to a backend 80 can be established by means of a data transmission unit 43, e.g.for transmitting sensor data SD or notifications B regarding deviations in driving behavior, for retrieving updated software for the components of the vehicle 40 or for retrieving an updated map K. The data exchange between the various components of the vehicle 40 takes place via a network 45.

[0041] Fig. Figure 5 schematically shows a method for updating a digital map. In the method, notifications and optionally sensor data are received from a plurality of autonomous or semi-autonomous vehicles regarding deviations between the current driving behavior of the respective vehicle and an expected driving behavior 50. The notifications can, in particular, also contain information about the current driving behavior. The received notifications and sensor data are evaluated 51. Based on the result of the evaluation, the expected driving behavior stored in a behavior data layer of the digital map is then updated 52. At least the updated behavior data layer, but alternatively also the entire updated digital map, can then be sent to the vehicles 53.

[0042] Fig. 6 shows a simplified schematic representation of a first embodiment of a device 60 for updating a digital map K. The device 60 has an input 61, via which a receiving module 62 receives notifications B and optionally sensor data SD from a plurality of autonomous or semi-autonomous vehicles 40 regarding deviations between a current driving behavior of the respective vehicle 40 and an expected driving behavior. The notifications B can in particular also contain information on the current driving behavior. An evaluation module 63 is configured to evaluate the received notifications B and sensor data SD. An update module 64 is configured, based on a result of the evaluation, to create a behavior data layer S V to update the expected driving behavior stored in the digital map K. The updated behavior data layer S Vor the entire updated map K can be output via an output 67 of the device 60 for further use.

[0043] The receiving module 62, the evaluation module 63, and the update module 64 can be controlled by a control module 65. Settings of the receiving module 62, the evaluation module 63, the update module 64, or the control module 65 can be changed via a user interface 68. The data generated in the device 60 can be stored in a memory 66 of the device 60 if necessary, for example, for later evaluation or for use by the components of the device 60. The receiving module 62, the evaluation module 63, the update module 64, and the control module 65 can be implemented as dedicated hardware, for example, as integrated circuits. Of course, they can also be partially or completely combined or implemented as software running on a suitable processor, for example, a CPU or GPU.The input 61 and the output 67 can be implemented as separate interfaces or as a combined bidirectional interface.

[0044] Fig. 7 shows a simplified schematic representation of a second embodiment of a device 70 for updating a digital map. The device 70 has a processor 72 and a memory 71. For example, the device 70 is a computer, a workstation, or a distributed system. Instructions are stored in the memory 71 which, when executed by the processor 72, cause the device 70 to carry out the steps according to one of the described methods. The instructions stored in the memory 71 thus embody a program executable by the processor 72 which implements the method according to the invention. The device 70 has an input 73 for receiving sensor data or notifications of deviations in driving behavior. Data generated by the processor 72 are provided via an output 74. In addition, data can be stored in the memory 71.The input 73 and the output 74 can be combined to form a bidirectional interface.

[0045] The processor 72 may include one or more processor units, such as microprocessors, digital signal processors, or combinations thereof.

[0046] The memories 66, 71 of the described embodiments can have both volatile and non-volatile memory areas and can comprise a wide variety of storage devices and storage media, for example hard disks, optical storage media or semiconductor memories.

[0047] Fig. Figure 8 illustrates, in a simplified manner, expected driving behavior using the example of cornering. It shows a road ST with an autonomous or semi-autonomous vehicle 40 on it. The road ST has a curve. The dashed line represents the center line M of the lane. Manually controlled vehicles will generally deviate from this center line M, with the trajectories driven typically lying within the area delimited by the dashed-dotted lines. This area represents the expected driving behavior. If, for example, vehicle 40 cuts the curve and follows the dotted line, there is a deviation between the current driving behavior and the expected driving behavior. In practical implementation, the expected driving behavior can include a variety of information, e.g.: • average driving speed • average acceleration / deceleration • Steering angle / cornering behavior / deviation from the center line • Stop • Changing lanes despite prohibition by solid line or signage • Perceptual accuracy • Signaling • Control unit status • dynamic environment

[0048] The information stored in the map can be specified as individual values or value ranges. When assessing a deviation from expected driving behavior, the vehicle can consider 40 tolerance ranges. These can be adjustable if necessary.

[0049] Fig. Figure 9 schematically illustrates a digital card K suitable for implementing a solution according to the invention. The card K comprises several data layers in which different information is stored. A physical data layer S Pcomprises a set of physical elements of the real world that represent the stationary environment. Examples of such physical elements are road markings and traffic signs. A relational data layer S R comprises a set of semantic elements composed of physical elements. Examples of such semantic elements are lanes formed from road markings, or construction sites formed from temporary road markings, temporary traffic signs, and temporary traffic lights. A topological data layer S T comprises a set of links between semantic elements. Examples of such links are predecessors, successors, and neighbors of lanes. A behavioral data layer S Vincludes an expected driving behavior, ie an expected target behavior. This specifies the actual behavior of vehicles on the roads with respect to specific road sections from the relational data layer S R and the topological data layer S T the card K. List of reference symbols 10 Collecting sensor data or decisions 11 Determining current driving behavior 12 Comparing current driving behavior with expected driving behavior 13 Detecting a deviation 14 Determining the next steps 15 Sending a notification, optionally sending sensor data 16 Adjusting current driving behavior 20 Device 21 Entrance 22 Processing module 23 Comparison module 24 Control module 25 Control module 26 storage 27 Exit 28 User interface 30 Device 31 storage 32 processor 33 Entrance 34 Exit 40 vehicles 41 Sensor technology 42 movement planners 43 Data transmission unit 44 storage 45 Network 50 Receive notifications and optional sensor data 51 Evaluating received notifications and sensor data 52 Updating an expected driving behavior 53 Sending the updated behavioral data layer 60 device 61 Entrance 62 Receiver module 63 Evaluation module 64 Update module 65 Control module 66 storage 67 Exit 68 User interface 70 Device 71 storage 72 processor 73 Entrance 74 Exit 80 backend A Deviation B Notification E Decision F a Current driving behavior F e Expected driving behavior K Digital Map M center line S control signal S P Physical data layer S R Relational data layer S T Topological data layer S V Behavioral data layer SD sensor data ST Street QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] US 2018 / 0 004 225 A1

[0006]

Claims

[1] Method for controlling an autonomous or semi-autonomous vehicle (40), comprising the steps: - Compare (12) a current driving behavior (F a ) of the vehicle (40) with an expected driving behavior (F e ), which is stored in a behavioral data layer (S V ) a digital card (K); - Determining (13) a deviation (A) between the current driving behavior (F a ) and the expected driving behavior (F e ); and - Adjust (16) the current driving behavior (F a ). [2] Method according to claim 1, wherein the current driving behavior (F a ) of the vehicle (40) is an actual or planned driving behavior which is determined (11) on the basis of sensor data (SD) or decisions (E) of a movement planner (42). [3] Method according to claim 1 or 2, wherein in the case of a detected deviation (A) between the current driving behavior (Fa ) and the expected driving behavior (F e ) a notification (B) and optionally sensor data (SD) are sent to a backend (80) (15). [4] Method according to one of the preceding claims, wherein in the case of a detected deviation (A) between the current driving behavior (F a ) and the expected driving behavior (F e ) is determined (14) whether the detected deviation (A) should be used to update the digital map (K), whether a teleoperator should be involved or whether a maneuver to bring the vehicle (40) into a safe state should be initiated. [5] Device (20) for controlling an autonomous or semi-autonomous vehicle (40), comprising: - a comparison module (23) for comparing (12) a current driving behavior (F a ) of the vehicle (40) with an expected driving behavior (F e ), which is stored in a behavioral data layer (S V) of a digital map (K), and for detecting (13) a deviation (A) between the current driving behavior (F a ) and the expected driving behavior (F e ); and - a control module (24) for adapting (16) the current driving behavior (F a ). [6] Autonomous or semi-autonomous vehicle (40), wherein the vehicle (40) comprises a device (20) according to claim 5 or is configured to carry out a method according to one of claims 1 to 4 for controlling the vehicle (40). [7] Digital map (K) for use in a method according to one of claims 1 to 4, wherein the digital map (K) has at least one relational data layer (S R ), a topological data layer (S T ) and a behavioral data layer (S V ), whereby in the behavioral data layer (S V ) an expected driving behavior (F e ) is stored. [8] Method for updating a digital map (K), comprising the steps: - Receiving (50) notifications (B) and optionally sensor data (SD) from a plurality of autonomous or semi-autonomous vehicles (40) regarding deviations (A) between a current driving behavior (F a ) of the respective vehicle (40) and an expected driving behavior (F e ); - evaluating (51) the received notifications (B) and sensor data (SD); and - Updating (52) a behavioral data layer (S V ) of the digital map (K) stored expected driving behavior (F e ) based on the results of the evaluation. [9] Device (60) for updating a digital map (K), comprising: - a receiving module (62) for receiving (50) notifications (B) and optionally sensor data (SD) from a plurality of autonomous or semi-autonomous vehicles (40) relating to deviations (A) between a current driving behavior (F a ) of the respective vehicle (40) and an expected driving behavior (F e ); - an evaluation module (63) for evaluating (51) the received notifications (B) and sensor data (SD); and - an update module (64) for updating (52) a behavioral data layer (S V ) of the digital map (K) stored expected driving behavior (F e ) based on the results of the evaluation. [10] A computer program comprising instructions which, when executed by a computer, cause the computer to carry out the steps of a method according to any one of claims 1 to 4 or 8.

Images