WO2017167801A1

WO2017167801A1 - Driver assistance system for supporting a driver when driving a vehicle

Info

Publication number: WO2017167801A1
Application number: PCT/EP2017/057400
Authority: WO
Inventors: Peter SCHÖGGL; Jürgen Holzinger; Uwe Dieter GREBE; Mario OSWALD
Original assignee: Avl List Gmbh
Priority date: 2016-03-29
Filing date: 2017-03-29
Publication date: 2017-10-05
Also published as: AT518489B1; AT518489A2; AT518489A3; DE102016205152A1

Abstract

The invention relates to a driver assistance system for supporting a driver when driving a vehicle. The driver assistance system preferably has at least one sensor which is designed to at least partly detect a driving situation of a vehicle and at least one first data interface for reading traffic data of a route lying ahead. Furthermore, the driver assistance system preferably comprises at least one second data interface for reading lane data, in particular the topography and/or lane course of a lane section lying ahead and/or a preceding lane section. A prediction module of the driver assistance system is preferably designed to dynamically simulate at least one future driving scenario on the basis of the current driving situation, the traffic data, and the lane data and to dynamically simulate and output possible trajectories of the vehicle on the basis of the at least one future driving scenario. An optimization module of the driver assistance system is preferably designed to select and output one of the possible trajectories using at least one specified boundary condition which characterizes a driving style attribute of the driver assistance system, and a control module is preferably connected to the longitudinal system, the brake system, and/or the drive system of the vehicle so as to guide the vehicle on the basis of the selected trajectory.

Description

Driver assistance system for assisting a driver in driving a vehicle

The invention relates to a driver assistance system for assisting a driver in driving a vehicle, which has at least one sensor that is configured to at least partially detect the vehicle. Furthermore, the driver assistance system has at least one first data interface for reading in traffic data of a preceding route and a second data interface for reading in roadway data of a preceding and / or preceding roadway section. A control module, which is connected to the steering system, the brake system and / or the drive system of the vehicle, serves to guide the vehicle.

The invention also relates to a corresponding method for assisting a driver.

From the prior art, a variety of different sensor systems for detecting the environment of a vehicle is known. By means of these sensor systems it is possible, for example, to detect lane boundaries in traffic or to determine the distance to a vehicle in front and thus a relative position of the vehicle in front. Also, by means of the known sensor systems, a relative speed to the vehicle in front can be determined.

On the basis of this information collected by the sensor systems, vehicles can be operated in automated operating modes. In these operating modes, in particular In particular, in a highly automated or fully automated operating mode, the driver is partially or even completely removed from driving the vehicle.

A highly automated vehicle here is a vehicle in which a driver assistance system takes over the lateral and longitudinal guidance for a certain period of time and / or in specific situations, wherein the driver does not monitor the system, but if necessary, to run the vehicle with sufficient time reserve is asked. Limits of the driver assistance system are preferably detected by the driver himself. In particular, however, the driver assistance system is not able to bring about the risk-minimum state from every starting situation.

A fully automated vehicle here is a vehicle in which a driver assistance system takes over the lateral and longitudinal guidance completely in a defined application, the driver does not have to monitor the driver assistance system. Before leaving the application, the driver assistance system requests the driver, preferably with sufficient time reserve, to take over the guidance of the vehicle. If this is not the case, the driver assistance system is preferably returned to the minimum risk state. Limits of the driver assistance system are preferably recognized by the driver assistance system itself. The driver assistance system is preferably able, even in all situations, to bring about a risk minimum state.

Driver assistance systems for highly automated or fully automated vehicles combine a variety of functions, such as lane departure warning, lane change assistant, lane change assistance, traffic sign recognition, emergency braking system, emergency stop system, adaptive cruise control, etc., to perform a transverse and longitudinal guidance reliably.

EP 2 942 765 A1 relates to a system for assisting a driver of a vehicle in potential lane change operations. The method performed by the system includes the steps of generating sensor data by at least one sensor that physically senses the environment of the vehicle, predicting the future motion behavior of at least one sensed vehicle and determining whether a gap exists on an adjacent lane of the vehicle. If a neighboring lane of the vehicle is more suitable for the predicted motion behavior, a recommendation is made regarding the feasibility of lane change of the vehicle to this more appropriate lane, the result of determining the presence of a gap and the future motion behavior of the at least one detected vehicle be combined. Based on this recommendation information, a notification is issued to the driver of the vehicle when lane change is feasible. Document WO 2013/138000 A1 relates to a method for operating a vehicle in an autonomous operating mode. The method includes the steps of: determining, using a computer system, a current state of a vehicle, the vehicle configured to operate in an autonomous mode; Determining, using the computer system, an instantaneous state of the environment of the vehicle, the environment of the vehicle having at least one other vehicle; Determining, using the computer system, a predicted behavior of the at least one other vehicle based on the at least one current state of the vehicle and the current state of the environment of the vehicle; Determining, using the computer system, a confidence level, wherein the confidence level comprises a probability that the at least one other vehicle is performing the predicted behavior, and wherein the confidence level is at least the predicted behavior, the current state of the vehicle, and the current state of the environment of the vehicle dependent; and controlling, using the computer system, the vehicle in the autonomous mode based on the predicted behavior, the confidence level, the current state of the vehicle, and the current state of the environment of the vehicle.

The document US 9,248,843 B1 relates to a computer-implemented method to detect objects in the vicinity of a vehicle and to respond to it. An object can be identified in the surroundings of the vehicle, the object having a direction of movement and a location. A set of possible actions may be generated for the object using map information representing the Describe the environment of the vehicle and the direction of movement and the location of the object. A set of possible future trajectories of the object is generated based on the possible actions, and a probability value of each trajectory of the set of possible future trajectories is determined based on context information including a status of the detected object. A final future trajectory is determined based on the determined probability values for each trajectory of the set of possible future trajectories. The vehicle is then manipulated in such a way as to avoid the final future trajectory and the object.

The document WO 2015/032508 A1 relates to a method for optimizing a driver assistance system, which has the working steps: determining at least one driver assistance system A to be optimized; Determining at least one vehicle parameter function that characterizes an operating condition of a vehicle and at least one environmental parameter function that characterizes the environment of the vehicle; Calculating at least one driving situation score function that characterizes a driving situation of the vehicle, based at least on the at least one vehicle parameter function and / or at least one environment parameter function, calculating at least one control engagement score function characterizing the activity of the driver assistance system A; Calculating a correction function that depends on the at least one driving situation characteristic value function and characterizes a subjective perception of the driving situation by at least one vehicle occupant, based at least on the at least one control intervention characteristic function and based on the at least one vehicle parameter function and / or the at least one environmental parameter function ,

Document DE 10 2014 208 311 A1 relates to a driver assistance system with an operating mode for fully automated vehicle guidance of a motor vehicle, wherein the fully automated vehicle guidance is individualized in that it is adapted to the individual needs of a vehicle driver. The publication DE 10 2006 039 583 A1 relates to a driver assistance system with assistance functions that can be determined by parameters, wherein the driver assistance system is adaptively designed by means of variable parameters. Document DE 10 2013 225 057 A1 relates to a method for controlling a vehicle from a current position to a target position, comprising determining the risk of driving a plurality of positions in an environment of the vehicle at a current time and at several times following the current time Timing points and determining a trajectory for the vehicle that connects or approximately connects the current position and the target position, taking into account calculated risks of driving (based for example on collision probabilities, traffic rules) as well as driving dynamics and comfort parameters. The vehicle is then controlled along the modified trajectory. It is an object of the invention to provide an improved driver assistance system and method for assisting a driver in driving a vehicle. In particular, it is an object of the invention to improve a subjective perception of the driver with respect to the guidance of the vehicle by the driver assistance system.

This object is achieved by a driver assistance system and a method for assisting a driver in driving a vehicle according to the independent claims. Advantageous embodiments are the subject of the dependent claims. The teaching of the claims is expressly made a part of the description.

A first aspect of the invention relates to a driver assistance system for assisting a driver in driving a vehicle, wherein the driver assistance system preferably has at least one sensor which is set up to at least partially detect a driving situation of a vehicle and at least one first data interface for reading in traffic data of a preceding route , Furthermore, the driver assistance system preferably comprises at least one second data interface for reading in roadway data, in particular topography and / or roadway course of a preceding roadway section and / or one past road section. A predictive module of the driver assistance system is preferably configured to dynamically simulate at least one future driving scenario based on the current driving situation, the traffic data, and the roadway data, and dynamically simulate and output possible trajectories of the vehicle based on the at least one future driving scenario. An optimization module of the driver assistance system is preferably set up to select and output one of the possible trajectories based on at least one predetermined boundary condition characterizing a driving style attribute of the driver assistance system and a control module is preferably connected to the longitudinal system, the braking system and / or the drive system of the vehicle Way connected to guide the vehicle based on the selected trajectory.

A second aspect of the invention relates to a driver assistance system for assisting a driver in driving a vehicle, which preferably has at least one sensor which is set up to at least partially detect a driving situation of the vehicle and at least one first data interface for reading in traffic data of a preceding route , Furthermore, the driver assistance system preferably has a second data interface for reading in road data, in particular topography and / or road course, of a preceding roadway section and / or of a preceding roadway section. A prediction module of the driver assistance system is preferably set up to dynamically simulate and output at least one future driving scenario from the current driving situation, the traffic data and the roadway data. An optimization module of the driver system is preferably configured to calculate and output a trajectory of the vehicle based on the at least one future driving scenario and at least one predetermined constraint characterizing a driving style attribute of the driver assistance system, and a control module is connected to the steering system, the braking system, and / or the drive system of the vehicle connected in such a way to guide the vehicle based on the calculated trajectory.

A third aspect of the invention relates to a method for assisting a driver when driving a vehicle, which preferably has the following working steps: - detecting at least one input parameter by which a current driving scenario in which the vehicle is located is characterized;

Guiding the vehicle by a driver assistance system, in particular based on the current driving scenario;

dynamically simulating, in parallel with guiding the vehicle, at least one future driving scenario based on the current driving scenario and simulating possible trajectories of the vehicle based on the at least one future driving scenario; and

Selecting one of the possible trajectories based on at least one predetermined boundary condition characterizing a driving style attribute of the driver assistance system, wherein the selected trajectory forms the basis for further guiding the vehicle.

A fourth aspect of the invention relates to a method for assisting a driver in driving a vehicle, which preferably has the following working steps:

- detecting at least one input parameter by which a current driving scenario in which the vehicle is located is characterized;

- dynamically simulating, in parallel with guiding the vehicle, at least one future driving scenario based on the current driving scenario; and

Calculating a trajectory of the vehicle based on the at least one future driving scenario and at least one predetermined constraint characterizing a driving style attribute of the driver assistance system, wherein the calculated trajectory forms the basis for further guiding the vehicle.

Output within the meaning of the invention means providing data to a further work step or a further module. Output is in particular via an interface. Traffic data in the sense of the invention relate to the absolute and / or relative position of other road users to the vehicle as well as context information relating to these positions. For example, it can be concluded from the positions on a traffic density and thus, for example, on an expected traffic incident. Furthermore, traffic data may also include the speed and / or acceleration of the other road users as well as environmental data, for example on the weather, in a relevant area for the vehicle area.

A driver assistance system according to the invention enables highly automated or fully automated driving of the vehicle.

A driving situation in the sense of the invention includes information about the state of the vehicle, in particular longitudinal speed, lateral speed, longitudinal acceleration, lateral acceleration, steering angle, throttle position, lane, and the condition of the road users in the immediate vicinity of the vehicle, i. those road users, which are theoretically perceptible, in particular visible, from the perspective of the driver's tool. A driving situation here is in particular a state consideration at a time. Therefore, at different times, different driving scenarios are preferably present according to the invention, even if the constellation of other road users around the vehicle does not change at the different times.

A driving scenario in the sense of the invention describes an interaction of a vehicle with its surroundings. In particular, the driving scenario includes information about the driving situation. Preferably, the driving scenario further comprises information about traffic, weather and / or road data on the road section relating to the vehicle. A driving scenario is preferably a holistic view of a plurality, in particular of all parameters relevant to the movement of the vehicle. A driving scenario here is in particular a state consideration at a time. Therefore, at different times, different driving scenarios are preferably present according to the invention, even if the constellation of other road users around the vehicle does not change at the different times. A control parameter in the context of the invention is an adjustment of a manipulated variable on the vehicle, which serves to control the driving operation. Adjustment parameters are, in particular, the throttle or accelerator pedal position, a brake pressure or brake signal, a gear selection, etc.

Lane data within the meaning of the invention have at least information about the topography of a section of road ahead. Roadway data preferably also contains information about the roadway course of the respective road section.

A module in the sense of the invention is a component of a computer system. In this case, a module can be embodied in particular as hardware and / or software. A trajectory in the context of the invention is the time course of the movement of a physical body, in particular of a vehicle or other road user. Driving style in the sense of the invention is the way in which a vehicle is guided. A driving style is characterized by the behavior of the vehicle driver or a vehicle-guiding driver assistance system in different driving scenarios.

Preferably, these are behaviors that relate to driving scenarios in which the driver must make changes to the vehicle state, for example, the initiation of an overtaking process, a lane change, etc.

Control data in the sense of the invention are data which can be used to control a vehicle. The control data comprise at least one assignment rule, in particular a function or table which specifies the at least one boundary condition.

A boundary condition preferably refers to at least one constellation of at least input parameters to which values of setting parameters and / or body parameters parameters are assigned. The values of the adjustment parameters and / or body parameters here contain information about the subjective perception of the occupant and / or their likely behavior in relation to a driving scenario. A driving style attribute in the sense of the invention is suitable for characterizing the subjective perception of a driver or a driver group in relation to the driving style of a driver assistance system. Driving style attributes are, in particular, the driving time, ie the speed with which the driver assistance system attempts to cover a distance, perceived, ie perceived, safety, perceived efficiency, driving dynamics and drivability, ie a subjective perception of the driving behavior of the vehicle as a reaction on actions by the driver assistance system. Further driving style attributes are preferably those characteristics which do not produce a subjective perception or impression in the driver, but at whose objective values the driver can be interested. These include, for example, the emission as well as the actual energy consumption.

The invention is based in particular on the recognition that the evaluation of a driver assistance system in automated or fully automated driving of a vehicle in the future will depend to a very significant degree on how the subjective perception of the driver fails in the driving actions of the driver assistance system. Therefore, it is advantageous to take into account not only legal requirements such as speed limitation, overtaking ban, etc., and safety aspects for the prevention of accidents when defining a destination by the driver assistance system, but also to take into account one or more constraints that are relevant to the driver's experience a vehicle occupant and / or a final assessment of a covered by the driver assistance system route are crucial. This is ensured according to the invention by the inclusion of boundary conditions which relate to at least one driving style attribute. A further finding which has flowed into the invention is that driver assistance systems are able, by means of the possibility of digital information processing, to include in the planning of the trajectory of the vehicle, in contrast to the driver, information which far exceeds the current one driving situation go out. By exchanging data with other vehicles (car-to-car) or with the infrastructure (car-to-infrastructure) as well as using topographical data and road-course data, predictions can be made about future driving scenarios. According to the invention, these future driving scenes are simulated dynamically, preferably in real time. On the basis of these future driving scenarios, all possible trajectories which the driver assistance system could calculate are calculated, preferably taking account of physical, legal and safety specifications. From these possible trajectories, that trajectory is then selected on the basis of boundary conditions which best suit the driving style of the driver or a group of drivers, so that a positive evaluation of the driver is to be expected. These boundary conditions are preferably stored in the form of driving style attributes in the driver assistance system. Alternatively, an optimal trajectory can be calculated directly taking into account the boundary conditions.

It should be noted here that the driver assistance system generally has more information than the occupants of the vehicle, especially as the driver. Knowing the topography or even the traffic density or braking a non-visible vehicle, the vehicle can select a trajectory which has been optimized in terms of safety and energy efficiency. Since the driver does not have this information and is normally not able to process an amount of parallel information corresponding to the driver assistance system, such a driving style of the driver assistance system optimized according to purely objective criteria can lead to uncertainty or dissatisfaction on the part of the driver. because he can not understand these decisions of the driver assistance system.

Against this background, the invention proposes to consider driving style attributes in the selection or calculation of a planned trajectory in order to adapt an ideal trajectory according to objective criteria to the human expectation.

The combination of the dynamic simulation of the traffic situation for a predictive driving style of the driver assistance system in conjunction with the consideration of boundary conditions, which characterize driving style attributes, can highly automated or fully automated driving of the vehicle can be simultaneously optimized in terms of both objective and subjective criteria.

In an advantageous embodiment of the driver assistance system according to the invention, the second data interface is connected to a first data memory on which the roadway data are stored.

In a further advantageous embodiment of the driver assistance system according to the invention, the second data interface is set up in order to obtain the traffic data of a route ahead via a data connection from a central server and / or from preceding road users. In this embodiment, the prediction module can simulate a particularly precise simulation of future driving scenarios, since data is available in the vehicle about road sections which the vehicle will reach in the immediate future.

In a further advantageous embodiment, the driver assistance system has a driving style selector module which is set up to detect a selection relating to the driving style attribute, in particular an indication of a value range for the driving style attribute. If a vehicle is operated in a highly automated or fully automated manner, it is not possible for a driver assistance system to learn the driving style of a vehicle occupant or driver. Therefore, according to the invention, a driving style selector module is provided, in which the driver of the vehicle can enter via a user interface what expectations he has with respect to various criteria on the vehicle. From this, the boundary conditions are derived.

In a further advantageous, alternative embodiment of the driver assistance system according to the invention, this has a driving style selector module, wherein the driving style selector module has a second data memory and is adapted to drive attitude data, traffic data and lane data, which at least partially characterize a driving scenario, and data from at least one control parameter to access the vehicle, wherein the driving style selector module is arranged to store values of the at least one setting parameter and the sensor data in the second data memory and at least one boundary condition with respect to the at least one Setting parameters depending on the stored data. Alternatively or in addition to an input via a user interface, the driving style selector module can therefore provide that the driving style or the driving habits of the occupant or driver are learned by the driver assistance system, in particular via a correlation analysis.

In a further advantageous embodiment of the driver assistance system according to the invention, the at least one boundary condition characterizes a, in particular a single, driving style attribute from the following group: driving time, emission, energy consumption, perceived efficiency, perceived safety, driving dynamics, drivability.

The features and advantages described above in relation to the first and second aspects of the invention and its advantageous embodiments also apply to the third and fourth aspects of the invention and its advantageous embodiments and vice versa.

In an advantageous embodiment of the method according to the invention, the latter has the operating step of evaluating a traveled trajectory of the vehicle and / or of at least one past driving scenario, the result of the evaluating entering into a definition of the at least one boundary condition, in particular a deviation from the driving style attribute to correct. By this function, a self-learning mechanism is implemented according to the invention, which allows a continuous improvement of the simulation for the calculation of an optimal trajectory. The resulting control loop corresponds to a type of model-based optimization by which the driver assistance system automatically adapts its driving style to a new vehicle or changes to the vehicle, for example different payloads.

In a further advantageous embodiment of the method according to the invention, the method comprises the steps of calculating at least one characteristic value, which characterizes travel time, emission, energy consumption, safety, performance and / or driveability of the selected or calculated trajectory, on the basis of the evaluation. As a result, this characteristic value is output. Based on the characteristic value For example, a performance of the driver assistance system can be evaluated, for example by comparison with reference values.

In a further advantageous embodiment of the method according to the invention, a drive control and / or at least one vehicle control, in particular steering control, brake control and / or suspension are adapted on the basis of the at least one future driving scenario and / or the selected calculated trajectory. As a result, the vehicle can be optimally adapted to the track conditions. For example, bumps or potholes can be anticipated and compensated in this way.

In a further advantageous embodiment of the method according to the invention, the at least one future driving scenario and / or at least one past driving scenario has information that is captured by infrastructure and / or other road users, in particular preceding or following road users.

In a further advantageous embodiment of the method according to the invention, the at least one future driving scenario and / or at least one past driving scenario has information about a preceding driving section and the preceding roadway section and / or traffic information over a route lying ahead.

In a further advantageous embodiment of the method according to the invention, the at least one future driving scenario and / or at least one past driving scenario has information about absolute positions, relative positions, speed and / or acceleration of the vehicle and other, in particular preceding or following road users and / or weather in the area of the vehicle and in the area of other, in particular preceding or following, road users.

In a further advantageous embodiment of the method according to the invention, the at least one boundary condition characterizes a, in particular a single, driving style attribute from the following group: vehicle, emission, sensed energy consumption, objective energy consumption, perceived safety, driving dynamics, drivability.

In a further advantageous embodiment of the method according to the invention, the at least one boundary condition imitates a driving style of the driver or of a driver group.

In a further advantageous embodiment of the method according to the invention, the at least one boundary condition recreates an adapted driving style of the driver or of a driver group which reflects the different sensation of the driver or a group of drivers between automatic guidance and manual guidance of the vehicle. Depending on risk affinity and technical confidence, different occupants or different drivers of a driver assistance system expect a driving style that differs from the individual driving style of the driver or the occupant. Thus, drivers who are sensitive to the risk or inmates can expect that a driver assistance system will spread out physical limits more than the driver would expect for himself, for example because of a lack of driving experience. This function of the invention is particularly important against the background that the driver's own driving experience by manual driving is expected to decrease sharply in the future.

In a further advantageous embodiment of the method according to the invention, several boundary conditions are considered whose driving style attributes are weighted differently. As a result, some driving style attributes can be disproportionately considered in the optimization and other driving style attributes disproportionately considered.

In a further advantageous embodiment of the method according to the invention, the selection or calculation takes place on the basis of a cost function (in which the at least one boundary condition is received). In a further advantageous embodiment of the method according to the invention, for determining the at least one boundary condition during a training phase, the latter also has the following working steps: recording values of at least one adjustment parameter for guiding the vehicle and values of at least an input parameter which at least partially characterizes a driving scenario; and establishing at least one constraint for adjusting the at least one adjustment parameter in response to the at least one input parameter based on the sensed values.

In a further advantageous embodiment of the method according to the invention, the simulation takes place in real time, in particular on the basis of real-time data of the present driving scenario. In a further advantageous embodiment of the method according to the invention, the simulation is carried out periodically, preferably with a periodicity of about one second to about 10 minutes, preferably from about 10 seconds to about one minute, and most preferably about one second, about 10 seconds or about one Minute. In a further advantageous embodiment of the method according to the invention, the simulating covers a future period of about one second to about 10 minutes, preferably even about 10 seconds to about one minute, and most preferably about one second, about 10 seconds or about one minute , Other features, advantages and applications of the invention will become apparent from the following description of exemplary embodiments in conjunction with the figures. It shows at least partially schematically:

1 shows a vehicle with a driver assistance system according to the invention;

FIG. 2 shows a flowchart which represents a possible sequence of a method according to the invention;

FIG. 3 shows a representation of a first example of a driving scenario from a bird's point of view;

Figure 4 is an illustration of a second example of a driving scenario from a bird's eye view; FIG. 5 shows a further illustration of the first example of a driving scenario according to FIG. 3 from a lateral plan view; Figure 6 is a representation of a third driving scenario from a bird's eye view; and

Figure 7 is an illustration of a fourth example of a driving scenario from a bird's eye view. 1 shows a vehicle 2 with an exemplary embodiment of the driver assistance system 1 according to the invention. The driver assistance system 1 has a plurality of sensors 3a-3d, and in the example shown in FIG. 1, these are a rear-facing camera 3a, a front-facing camera 3b. a forward-looking radar system 3d and a rear-facing radar system 3c. The data acquired by the individual sensors are preferably transmitted to a prediction module 6 of the driver assistance system 1 wirelessly or by cable. Further elements of the driver assistance system 1 are a first data interface 4, which, for example via a mobile radio transmitter, can receive a data connection to an infrastructure, in particular a central traffic server 12, and a second data interface 5, which preferably has a second data memory 11 via a data connection is connected, in which further preferably lane data are stored. Via the first data interface 4, which is preferably designed as a radio interface, traffic data can thus be received and / or retrieved directly from other road users 13a-13g via the infrastructure, for example the data server 12 or even traffic data. Lane data, in particular topography, road course, information on the infrastructure etc. of a relevant roadway section in the driver assistance system 1 can be read in via the second data interface 5 and processed there. The second data interface can in principle also be embodied as a radio interface and can receive the lane data from the data server 12 or from another source.

In addition or as an alternative to the sensors 3a-3d, a large number of other sensors are possible, for example ultrasound and / or lidar sensors. By means of the sensors, the driving situation of the vehicle can be monitored in traffic. Thus, for example, the proximity of the vehicle, for example when parking, can be monitored with an ultrasonic sensor and the distance and the relative speed, in particular acceleration, to other vehicles in the field of vision of the vehicle 2 can be determined with a radar system. With a lidar sensor objects in the vicinity of the vehicle 2 and distances to other vehicles can be detected and recognized with the camera, the lane and traffic signs or objects around the vehicle 2 and possibly even identified.

The driver assistance system 1 is preferably designed in this way in order to guide the vehicle 2 in a highly automated or even fully automated manner. For this purpose, those information are primarily relevant, which can detect the driver assistance system 1 with its sensors installed in the vehicle 3a - 3d (on-board). This is particularly important if the driver assistance system 1 has no data connection to the central data server 12, other infrastructure or other road users 13a-13g, since the driver assistance system 1 the vehicle 2 autonomously in this case, ensuring the greatest possible safety for the vehicle 2 and its occupants must lead.

If, in addition to the information of the sensors 3a-3d associated with the driver assistance system 1, data is transmitted by the infrastructure or other road users 13a-13g, in particular preceding road users, the driver assistance system 1 can realize an even more predictable driving style than is already possible according to the invention is possible only with the data of the driver assistent sensors 3a-3d.

The driver assistance system 1 according to FIG. 1 furthermore has a prediction module 6, which is set up to simulate future driving scenarios on the basis of a current driving scenario and / or past driving scenarios. Furthermore, traffic data and lane data go into the simulation of the prediction module 6 as well as data on the weather at the location of the vehicle 2 or another location on a relevant stretch of road.

On the basis of the future driving scenario, the prediction module 6 subsequently simulates a multiplicity of possible trajectories of the vehicle. These are output to an optimization module 7, which in turn can select one of the trajectories 10a, 10b as ideal trajectories. This type of simulation can be visualized as a view into the crystal ball, ie a foreseeing of a highly probable driving scenario or such a driving situation. For example, a sequence of starting and braking in a traffic jam more generally known as the concertina effect, as well as the propagation of this concertina effect through the traffic jam, can be predicted and the trajectory of the own vehicle 1 intelligently adapted to this development. A control module of the driver assistance system is in particular connected via a data link to the controls of the steering system, the braking system and / or the drive system of the vehicle 2 in order to execute the corresponding trajectories 10a, 10b, 10c. The driver assistance system 1 shown in FIG. 1 further has a driving style selector module 14. This driving style selector module 14 is used, in particular, to detect an indication of the driver or of the occupants in relation to the driving style of the driver assistance system they desire. The driving style selector module 14 therefore has in particular a user interface, for example a touch-sensitive display. Alternatively or additionally, it can also be provided that a mobile telephone or other electronic device which can be connected to the driver assistance system 1 via a data connection is used as the user interface. Preferably, the driving style selector module 14 is configured to autonomously generate driving style attributes relating to its driving style during manual driving by a driver or during automated driving. For this, the driving style selector module 14 can access driving situation data, traffic data and roadway data which at least partially characterize a driving scenario. Furthermore, the driving style selector module 14 may be responsive to data relating to at least one actuating parameter for guiding. Ren the vehicle 2 access. Value constellations of the driving scenarios to the control parameters are preferably stored in a second data memory 15, which is associated with the driving style selector module 14: In this way, the driving style selector module 14 can create boundary conditions which characterize in the sequence driving style attributes of the driver assistance system 1. In the second data memory 15 thus correlations between different driving scenarios and corresponding constellations of Stellparameterwerten are stored, which can be applied in the corresponding driving scenarios. The individual modules of the driver assistance system 1 are preferably part of a computing device in the vehicle 1, in particular of one or more on-board computers. The individual modules are designed as hardware or software components. As sensors 3a-3d of the driver assistance system 1, in particular also sensors can be used, which are additionally assigned to other systems of the vehicle 2 or are part of these other systems.

The method according to the invention, whose exemplary embodiments are shown schematically in the flowchart of FIG. 2, will be explained below with reference to FIGS. 2 to 6.

FIG. 3 shows an example of a driving scenario in which a vehicle 2 is located on the left side of the picture on a three-lane road. In front of the vehicle, another vehicle 13f travels in the middle lane and in turn three vehicles 13c, 13d, 13e next to each other on all three lanes. Further ahead drive two vehicles 13a, 13b, also next to each other on the left and the middle lane. This current driving scenario is detected 102, in particular by means of the sensors 3a-3d arranged in the vehicle 2. Since the vehicle 2 with the driver assistance system 1 has a higher speed than the preceding vehicle 13f in the middle lane, the driver assistance system 1 becomes typical in this driving scenario initiate an overtaking maneuver with respect to the vehicle 13f and to change the lane, as indicated by the arrow 10. The vehicle 2 is guided by the driver assistance system 1 105. According to the invention, further future driving scenarios are now simulated 106a, 106b by the driver assistance system 1 parallel to the guidance of the vehicle 2, in particular in real time. In particular, the simulation is carried out dynamically, ie previously simulated driving scenarios are replaced by the current one.

On the basis of the velocities of the vehicles in the immediate vicinity of the vehicle 2 determined by means of the radar sensor 13d, the driver assistance system 1 can determine by simulation 106a, 106b that the vehicle in the left lane, i. the fast lane, traveling vehicle 13c has a much lower speed than the adjacent vehicles 13d, 13e. If the vehicle 2 with the driver assistance system 1, as shown in FIG. 3, initiate an overtaking maneuver at high speed and thus follow the trajectory 10, the driver assistance system 1 determines by the simulation 106a, 106b that the vehicle 2 is in the driver assistance system 1 in a future driving scenario would be in a wedged position between the preceding vehicle 13c and the vehicle 13f next to the vehicle 2.

This is shown in Fig. 4. Thus, in this driving scenario, the vehicle 2 would have to make a strong deceleration to adapt to the speed of the preceding vehicle 13c in the passing lane, and could not complete the passing operation with respect to the vehicle 13f. The driving maneuver corresponding to the trajectory 10 will therefore cause a subjective perception in an occupant of the vehicle 2 that the driver assistance system 1 drives with little foresight, and therefore a poor judgment with respect to the drivability achieved by the driver assistance system 1, i. produce the driving behavior, which is subjectively perceived by an occupant, generated by the driver assistance system 1. Also, the occupant will be aware that the maneuver initiated by any acceleration or at least the high speed and abrupt deceleration behind the vehicle 13c will result in low energy efficiency or high energy consumption, which also gives a negative impression to the occupant will be left.

The driver assistance system 1 therefore preferably further simulates a plurality of different possible trajectories 106a on the basis of the current driving Scenario of Fig. 3 and taking into account the future driving scenario of Fig. 4 and selects that trajectory of 107 a, which causes the occupants a positive overall impression of the driving style of the driver assistance system 1. For example, in the current driving scenario of FIG. 3, the driver assistance system 1 could follow the preceding vehicle 13f, not initiate an overtaking maneuver or merely initiate a lane change at a reduced speed in the passing lane to finally overtake the vehicle 13c after a possible lane change thereof to the middle lane to be able to. Alternatively, an ideal trajectory can be calculated directly taking into account a future driving scenario 107b.

Preferably, the driver assistance system 1 of the vehicle 2 can also take into account how a preceding roadway section 9a runs or, as shown in FIG. 5, which topography is to be expected on the preceding roadway section 9a and if further factors exist which may be due to the infrastructure should be considered, such as a speed limit, as also shown in Fig. 5. Taking into account the topography shown in FIG. 5, the driver assistance system 1 of the vehicle 2 selects a lower acceleration or speed since the simulation 106a, 106b, 106c receives the information that the preceding roadway section 9a has a sloping gradient and that In addition, a speed limit is prescribed for this gradient.

Furthermore, the driver assistance system 1 includes real-time data in the simulation of driving scenarios 106a, 106b, which is detected by infrastructure and / or other, in particular preceding and / or following road users 13a, 13b, 13c, 13ds, 13e, 13f, 13g It can simulate future driving scenarios even more accurately and thereby incorporate information that can not be derived from the detection and evaluation of the current driving situation 102.

For example, a modified driving scenario of FIG. 3 is shown in FIG. 6, in which the vehicles 13a and 13b, which drive ahead of the vehicle 2 with the driver assistance system 1, cause a rear-end collision. If this information is provided by the vehicles gen 13a, 13b or the vehicles 13c, 13d, 13e, which have a direct view of the driving situation of the two vehicles 13a, 13b, forwarded directly or via an infrastructure to the vehicle 2, which is guided by the driver assistance system 1, so may the vehicle 2 incorporate this event into its simulation. As illustrated in FIG. 6, the driver assistance system 1 can choose between different trajectories 10 a, 10 b, 10 c in order to master a future driving scenario resulting from the driving scenario of FIG. 6.

Such a future driving scenario is shown in FIG. 7. The vehicle 2 has been guided for the trajectory 10b and the speed is reduced, since the driver assistance system 1 has predicted during simulating 106a, 106b that all vehicles in the right lane must drive past the vehicles 13a, 13b blocked by the accident. In the then current driving scenario of FIG. 7, the driver assistance system 1 of the vehicle 2 can again choose whether it remains in the right lane and follows the preceding vehicle 13f or again makes a lane change in order to overtake the preceding vehicle 13f during the zippering process ,

When selecting 107a of the ideal trajectory or calculating 107b of the ideal trajectory, according to the invention, it is preferable to consider, in addition to the result of the simulation 106a, 106b, boundary conditions which characterize driving style attributes. The driving style attributes preferably indicate objective criteria, such as an occupant or a group of occupants who are identified, for example, by age or gender, will perceive a driving style of the driver assistance system 1 that manifests itself in the respective trajectory. Such driving style attributes can be, for example, the driving time, the perceived energy consumption, the objective energy consumption, the perceived safety, the driving dynamics and / or the drivability. Ideally, the boundary conditions simulate a driving style of the occupant or occupant group. In this case, however, it may also be preferably taken into account that an occupant as a passenger of a driver assistance system 1 will make other demands on the driving style than if the same person were driving himself. at In the calculation, different driving style attributes are preferably weighted differently and, in particular, the optimization of a cost function can be used to achieve an overall optimum. In order to adapt to changing environmental conditions or even properties of the vehicle 2, it may further be provided that the driver assistance system 1 evaluates trajectories and / or driving scenarios covered. The boundary conditions can be modified accordingly to detect deviations from predefined target corridors with respect to the driving style attributes changed conditions.

In addition, a characteristic value can be calculated 109, which evaluates the performance of the driver assistance system 1. The finding of an ideal trajectory for coping with a future driving scenario by the driver assistance system 1 is preferably supplemented by communication of the driver assistance system 1 with selected vehicle controls in order to prepare the vehicle 2 for conditions on a preceding route section 9a. For example, the steering control can be informed that a violent steering movement is imminent, or the brake control can be prepared to impose a severe braking. The brake control can then provide, for example, a hydraulic pressure at the right time. The suspension can be prepared for example on bumps, so that they can be compensated in the ideal case. The simulation is preferably done in approximately one second increments, and more preferably, a period of the next 10 seconds to about one minute is covered.

As already described in relation to the driver assistance system 1 according to FIG. 1, the invention as a further aspect has a learning of the boundary conditions with regard to the driving style attributes. This training is preferably carried out during a training phase, during which the driver controls at least the longitudinal and lateral control of the vehicle 2 manually. More preferably, it is in particular a continuous training: Whenever the driver leads the vehicle 2 himself, the driver assistance system 1 changes into the learning mode.

In the training phase of the driver assistance system 1, values of at least one adjustment parameter for guiding the vehicle, in particular parallel to values of the at least one input parameter, are recorded 101 and stored in the second data memory 15. From the respective values of the parameters at the same time or in a same time period, correlations result, which reflect driver reactions in different driving scenarios and therefore contain information about the driving style of the driver. Boundary conditions are set up on the basis of this information. The driver assistance system 1 accesses these predefined boundary conditions in highly automated or fully automated driving of the vehicle 2, in order to achieve as pleasant a perception of the vehicle as possible for the driver. Additionally or alternatively, these values or correlations are output 104 as boundary conditions for guiding the vehicle.

In addition or as an alternative to the values of the setting parameter, values of at least one body parameter which reflects the body function of a vehicle occupant, in particular of the driver, can also be recorded 101. This can be achieved, for example, with an intelligent device, in particular an intelligent watch (smart device, smart watch). The objective values of the body parameter are chosen so that they can characterize a subjective perception of the driving style of a driver assistance system 1. In particular, the heart rate, the blood pressure, the adrenalin level and / or the respiratory activity of an occupant come into question. Also the values of the body parameters are compared with values of the input parameters, i. correlated with the various driving scenarios and derived from this boundary conditions that serve as a rule when driving the vehicle. The picking up 101 of body parameters preferably continues during highly automated or fully automated driving, so that further data for evaluation, past driving scenarios and / or trajectories traveled are obtained.

The correlations or updated correlations enter into the selection 107a or the evaluation 107b of the ideal trajectory. The invention makes possible a comprehensive optimization of a driving operation performed by a driver assistance system 1. Here, not only legal requirements and safety-relevant specifications are included in the management of the vehicle, but also by a special driver or a special driver group preferred driving styles, which are defined by the driving style attributes. The driver assistance system 1 is in this case able to independently adapt to the respective driver and the vehicle or modifications to the vehicle. This ensures that the driver assistance system 1 has an optimum driving style for the respective vehicle 2 or its vehicle configuration and / or for the respective driver. Depending on the requirements of the current or future driving scenarios and the occupants of the vehicle 2, in particular an optimization of the energy consumption of the vehicle 2 can be carried out. For example, all specifications based on the driving scenarios and the boundary conditions can be included in an energy cost function. On the basis of energy requirements of different aggregates of the vehicle 2 for momentary and future driving scenarios, an energy price can be determined. By allocating value quotas to the various aggregates of the vehicle 2, each aggregate may decide whether it is consuming or feeding energy available in the vehicle 2.

LIST OF REFERENCE NUMBERS

Driver assistance system

vehicle

a, 3b, 3d sensor

first data interface second data interface

Forecast Module

optimization module

Control module, 9b lane section 0, 10a, 10a, 10c trajectory

1 first data memory 2 central server 3a, 13b, 113c, 13d, 13e, 13f, 13g road user 4 driving style selector module 5 second data memory

Claims

claims

A driver assistance system (1) for assisting a driver in driving a vehicle (2), comprising

at least one sensor (3a, 3b, 3c, 3d), which is set up to at least partially detect a driving situation of the vehicle (2),

at least one first data interface (4) for reading in traffic data of a preceding route,

at least one second data interface (5) for reading in roadway data, which in particular characterize the topography and / or roadway course of a preceding roadway section (9a) and / or past roadway section (9b),

a prediction module (6) arranged to simulate at least one future driving scenario based on the current driving situation, the traffic data and the roadway data and possible trajectories (10a, 10b, 10c) of the vehicle (2) based on the at least one future driving scenario and spend

an optimization module (7) configured to select and output one of the possible trajectories (7a, 7b) on the basis of at least one predetermined boundary condition which characterizes a driving style attribute of the driver assistance system (1), and

a control module (8) connected to the steering system, braking system and / or drive system of the vehicle (2) for guiding the vehicle (2) based on the selected trajectory (10a, 10b, 10c) ,

characterize at least a first data interface (4) for reading in traffic data of a preceding route, at least one second data interface (5) for reading in road data, which in particular characterize the topography and / or road course of a preceding roadway section (9a) and / or a preceding roadway section (9b),

a prediction module (6), designed to simulate and output at least one future driving scenario from the current driving situation, the traffic data and the roadway data,

an optimization module (7) arranged to calculate and output a trajectory (10) of the vehicle (2) based on the at least one future driving scenario and at least one predetermined constraint characterizing a driving style attribute of the driver assistance system (1) Control module (8), which is connected to the steering system, the braking system and / or the drive system of the vehicle (2) in such a way to guide the vehicle (2) on the basis of the calculated trajectory (10).

Driver assistance system (1) according to one of the preceding claims, wherein the second data interface (5) is adapted to the traffic data of a preceding route via a data link from a central server (12) and / or preceding road users (13a, 13b, 13c, 13d , 13e, 13f).

Driver assistance system (1) according to one of the preceding claims, which has a driving style selector module (14) which is adapted to detect a selection with respect to the driving style attribute, in particular an indication of a value range for the driving style attribute, via a user interface.

Driver assistance system (1) according to one of the preceding claims, which has a driving style selector module (14), wherein the driving style selector module (14) has a second data memory (15) and is adapted to Fahrsituations- onsdaten, traffic data and lane data, which at least partially a driving scenario and access data from at least one control parameter for guiding the vehicle (2), wherein the driving style selector module (14) is set up, values of the at least one control parameter and the sensor stored in the second data memory (15) and set up at least one boundary condition with respect to the at least one adjusting parameter as a function of the stored data.

A method (100) for assisting a driver in driving a vehicle (2), comprising the following steps:

Detecting (102) at least one input parameter by which a current driving scenario in which the vehicle (2) is located is characterized;

Guiding (105) the vehicle (2) by a driver assistance system (1);

Simulating (106a) parallel to driving the vehicle (2), at least one future driving scenario based on the current driving scenario, and simulating possible trajectories (10a, 10b) of the vehicle (2) based on the at least one future driving scenario; and

Selecting (107a) from one of the possible trajectories (10a, 10b, 10c) based on at least one predetermined boundary condition characterizing a driving style attribute of the driver assistance system (1), the selected trajectory (10a, 10b) providing the basis for further guidance of the vehicle (2) forms.

A method (100) for assisting a driver in driving a vehicle, comprising the steps of:

Guiding (105) the vehicle by a driver assistance system (1);

Simulating (106b), in parallel with guiding the vehicle (2), at least one future driving scenario based on the current driving scenario; and

Calculating (107b) a trajectory (10) of the vehicle (2) based on the at least one future driving scenario and at least one predetermined constraint characterizing a driving style attribute of the driver assistance system (1), the calculated trajectory (10) providing the basis for further guiding the vehicle (2) forms. The method (100) of claim 6 or 7, further comprising the following operation:

Evaluating (108) a traveled trajectory of the vehicle (2) and / or at least one past driving scenario, the result of the evaluating entering into a definition of the at least one boundary condition, in particular in order to correct a deviation from the driving style attribute.

Method (100) according to one of claims 6 to 8, wherein a drive train control and / or at least one vehicle control, in particular steering control, brake control and / or suspension, on the basis of at least one future driving scenario and / or the selected or calculated trajectory (10a , 10b) are adapted.

Method (100) according to any one of claims 6 to 9, wherein the current driving scenario and / or at least one past driving scenario includes / contains information in addition to the driving situation, which is provided by infrastructure and / or other, in particular preceding or following, road users (13a, 13b, 13c, 13d, 13e, 13f, 13g) is detected.

Method (100) according to one of Claims 6 to 10, wherein the at least one future driving scenario and / or at least one past driving scenario has information about a preceding roadway section (9a) and / or preceding roadway section (9b) and / or traffic information over a preceding route ,

Method (100) according to one of claims 6 to 1 1, wherein the at least one future driving scenario and / or at least one past driving scenario includes information about absolute position, relative position, speed and / or acceleration of the vehicle (2) and other, in particular preceding o - the following, road users (13a, 13b, 13c, 13d, 13e, 13f, 13g) and / or has a weather in the area of the vehicle and other, in particular preceding or following, road users.

13. Method (100) according to claim 6, wherein the at least one boundary condition characterizes a, in particular a single, driving style attribute from the following group: driving time, emission, perceived efficiency, energy consumption, perceived safety, driving dynamics, drivability.

14. The method according to claim 6, wherein the at least one boundary condition simulates a driving style of the driver or of a driver group.

A method (100) according to any one of claims 6 to 14, wherein the at least one constraint replicates an adapted driving style of the driver or a group of drivers reflecting the different sensation of the driver or group of drivers between automatic guidance and manual guidance of the vehicle (2).

Method (100) according to one of claims 6 to 15, wherein a plurality of boundary conditions are considered and their driving style attributes are weighted differently.

Method (100) according to one of claims 6 to 16, which for setting the at least one boundary condition during a training phase further comprises the following steps:

Detecting (101) values of at least one setting parameter for guiding the vehicle (2); and

Establishing (103) at least one boundary condition for setting the at least one adjustment parameter as a function of the at least one input parameter on the basis of the recorded values.

The method (100) of any one of claims 6 to 17, wherein simulating comprises a future time period of from about 1 second to about 10 minutes, more preferably from about 10 seconds to about 1 minute, and most preferably about 1 second, 10 seconds or 1 second min covers.

A computer program comprising instructions which, when executed by one or more computers, cause it to perform the steps of a method according to any one of claims 6 to 18.

20. Computer-readable medium on which a computer program according to claim 19 is stored.