Lateral guidance of a motor vehicle using a driver assistance system

Abstract

The invention relates to prioritizing swarm data for a lateral driver assistance system (2) for the control of a motor vehicle (1). For this purpose, a virtual lane center (14) for centering the motor vehicle (1) within a driving corridor (11) is determined from environmental data (U). Furthermore, a swarm trajectory (15) is determined from swarm data for the same driving corridor (11). Subsequently, it is checked whether, depending on a steering input (L) by the driver, an actual trajectory (16) results that deviates from the virtual lane center (14) according to a predefined deviation criterion (K2) and coincides with the swarm trajectory (15) according to a predefined approximation criterion (K1). Depending on the result of the check, the swarm trajectory (15) is used instead of the virtual lane center (14) for the lateral control of the motor vehicle (1).

Description

[0001] The invention relates to a method for the lateral guidance of a motor vehicle by means of a driver assistance system. The invention also relates to a driver assistance system for the lateral guidance of a motor vehicle. Furthermore, the invention relates to a motor vehicle with a corresponding driver assistance system.

[0002] Modern vehicles utilize driver assistance systems to support the driver in driving tasks and / or to fully or partially take over the steering or control of the vehicle. For example, there are lateral driver assistance systems that primarily automate the steering of the vehicle. To achieve this, the driver assistance system calculates a driving trajectory or path for a desired route, along which the vehicle's movement is to be guided. The trajectory is based, for example, on lane boundaries such as road markings, which define a driving corridor or lane for the vehicle. One option is to implement centerline guidance, in which the trajectory lies midway between the lane boundaries. This effectively steers the vehicle towards a virtual lane center.

[0003] To perceive the surroundings, the vehicle's environmental sensors are used, for example. These sensors might include a camera, whose sensor data can be evaluated by the driver assistance system to determine the trajectory. Additionally or alternatively, the trajectory can be calculated using swarm data. Swarm data contains, for example, one or more trajectories from other vehicles that have already traveled the respective section of the route and provide their collected route or driving data for use by other vehicles.

[0004] Various methods for determining trajectories from sensor data or swarm data are known from the state of the art.

[0005] For example, US 10 990 103 B2 discloses a method for operating a motor vehicle in which a trajectory to be followed is determined. This is done using a lane map showing paths traveled by other vehicles in the past in the vicinity of the motor vehicle.

[0006] From DE 10 2021 206 886 A1, a method for operating a driver assistance system of a vehicle is known. In this method, at least one trajectory created by another vehicle is used for the lateral guidance of the motor vehicle.

[0007] The object of the present invention is to improve the selection of a driving trajectory for the lateral guidance of a motor vehicle by means of a driver assistance system.

[0008] The problem is solved by the subject matter of the independent patent claims. Advantageous embodiments of the invention are disclosed by the dependent patent claims, the description, and the figures.

[0009] The invention is based on the understanding that steering a motor vehicle towards a virtual lane center can be undesirable in certain traffic situations. This includes, for example, roads without center lines, where oncoming traffic also travels between the two edge lines. The motor vehicle and oncoming traffic thus share the same lane. Such roads are found, for example, on rural roads outside of cities, at lane widenings, or at open exits.

[0010] Road damage or obstacles on the roadway can also lead to a situation where the center guidance can no longer be implemented effectively.

[0011] In such traffic situations, a driver who recognizes the situation may intervene in the steering, i.e., the lateral control of the vehicle. The driver intends to steer the vehicle onto a new trajectory that deviates from the center line. There are lateral control driver assistance systems that block, or at least temporarily interrupt or hinder, such steering interventions. The driver may lose confidence in the lateral control function, and driving comfort may be reduced.

[0012] It has proven effective to adjust the driving trajectory based on driver intervention under certain conditions. This means the driver assistance system doesn't block driver input, but rather uses the driver's activity as a starting point to, for example, find improved vehicle guidance for the respective section of the road. Swarm data is preferably used for this purpose, allowing lateral guidance, for instance, to be regulated not only by the virtual lane center but also by a swarm trajectory known from the swarm data.

[0013] According to one aspect, the invention proposes a method for the lateral guidance of a motor vehicle using a driver assistance system. First, a virtual lane center is determined for the motor vehicle from environmental data within a predefined driving corridor, particularly for a specific roadway or route segment. This virtual lane center can then be used for the lateral guidance of the motor vehicle. In particular, a target trajectory for the lateral guidance of the motor vehicle can be calculated based on the virtual lane center.

[0014] The process then uses swarm data to determine a swarm trajectory. This swarm trajectory differs from the intended trajectory, i.e., the virtual lane center. The swarm trajectory represents the trajectory of at least one other vehicle that has already traversed or passed through the specified corridor or lane section. In other words, the swarm data contains an alternative trajectory for the vehicle.

[0015] The procedure then checks whether, depending on a steering intervention or steering activity performed by the driver of the vehicle, an actual trajectory for the vehicle results that, on the one hand, deviates from the virtual lane center (i.e., the target trajectory) according to a predefined deviation criterion, and on the other hand, corresponds to the swarm trajectory according to a predefined approximation criterion. Steering intervention means that the driver intervenes in the steering, i.e., attempts to take over steering from the driver assistance system, at least temporarily.

[0016] The verification process can include a comparison of the actual trajectory with the target trajectory and a comparison of the actual trajectory with the swarm trajectory. The result of the comparison provides information on the extent to which the paths of the actual trajectory and the target trajectory, as well as the paths of the actual trajectory and the swarm trajectory, differ (deviate) or coincide (approach). In this context, coincidence means, in particular, that the deviation between the swarm trajectory and the actual trajectory is smaller than a predefined threshold. Difference means, in particular, that the deviation between the actual trajectory and the target trajectory is larger than a predefined threshold. This prevents, for example, the swarm data from being prioritized over the virtual track center when the deviations of the swarm trajectory from the target trajectory are small.

[0017] The thresholds or limits above which the deviation in trajectory is considered relevant are defined or contained in the approximation criterion and the deviation criterion. The deviation criterion, for example, includes a rule and / or an algorithm whose application or execution determines the deviation of the actual trajectory from the virtual track center. The approximation criterion, for example, includes a rule and / or an algorithm whose application or execution determines how closely the actual trajectory approaches the swarm trajectory. Examples of a concrete implementation of each rule are described in more detail later.

[0018] Depending on the result of the check, i.e., the comparison result, the swarm trajectory is then used instead of the virtual lane center for the vehicle's lateral guidance. The lateral guidance system aligns the vehicle with the swarm trajectory. This means that the swarm trajectory, and in particular the swarm data, is prioritized over the virtual lane center for the vehicle's lateral guidance. For example, driving maneuvers are determined and initiated that steer the vehicle onto the swarm trajectory and preferably keep it on that trajectory.

[0019] Thus, depending on the driver's wishes, an alternative trajectory can be used for lateral guidance. This improves ride comfort for the occupants. In particular, the selection of an alternative driving trajectory based on driver intervention is improved. Instead of blocking the driver intervention, such as steering input, the steering activity is used as a trigger to steer the vehicle onto a new trajectory.

[0020] In this context, a driver assistance system refers specifically to an electronic vehicle control system. This means that the driver assistance system is preferably designed to control the (motor) vehicle fully or semi-automatically, particularly without requiring intervention from a driver. Preferably, the present driver assistance system is configured to control the motor vehicle according to SAE Levels 1 to 4, which classify predefined levels of automation for driver assistance systems according to SAE Standard J3016.

[0021] The (semi-)automatic control system enables the vehicle to automatically perform all necessary functions, such as steering, braking, and acceleration maneuvers, as well as monitoring and perceiving the vehicle's surroundings, including road traffic, and reacting accordingly. Depending on the desired assistance function, a distinction can be made between longitudinal and lateral driver assistance systems. Alternatively, a driver assistance system with combined functionality is conceivable. Braking or acceleration maneuvers, i.e., driving maneuvers that affect the vehicle's propulsion, are implemented, for example, by a longitudinal driver assistance system. Steering maneuvers, i.e., driving maneuvers that affect the vehicle's steering, are implemented, for example, by a lateral driver assistance system.To execute the respective driving maneuver, the driver assistance system can provide a corresponding control signal to one or more actuators of the vehicle. For example, the driver assistance system can control one or more brake actuators and / or one or more steering actuators and / or one or more drive motors of the vehicle. For the control of the swarm trajectory, lateral guidance is particularly relevant in this case.

[0022] The driver assistance system uses environmental data to monitor the surroundings. To capture this data, the vehicle is equipped with environmental sensors, including one or more sensors. For example, in the context of lateral guidance, an environmental sensor could be a camera. Additionally or alternatively, a radar sensor can be used. The environmental sensor(s) provide the environmental data. In the case of a camera or radar sensor, this data consists of sensor data that creates or recreates an image or representation of the surroundings, including the road traffic around the vehicle.

[0023] The driver assistance system, in particular a control or evaluation unit of the driver assistance system, can receive and evaluate this environmental data. This evaluation allows, for example, the extraction of lane data. The lane data can describe the dimensions of the driving corridor and / or roadway. For example, the lane data can include lane markings, a road edge, or structural features at a road edge. From this lane data, the driver assistance system can calculate a lane boundary for the driving corridor, such as a lane width. The midpoint between these lane boundaries can be used as the virtual lane center to calculate the desired trajectory.

[0024] The extraction of lane data and the calculation of the lane center are performed using known methods. For example, the driver assistance system can apply an object recognition algorithm to the environmental data. The calculation of the virtual lane center based on the lane data can be performed, for example, using a predefined mathematical model, such as a third-degree clothoid polynomial. Such models for trajectory calculation are well-established.

[0025] As an alternative to sensor data, environmental data such as map or navigation data can be used, taking into account the vehicle's current geographic position. Lane data, for example, can be included in the map data. Based on the vehicle's current position, the current driving corridor or road segment can be determined from the map data, thus establishing the virtual lane center for lateral guidance. The map data can be provided to the vehicle, for example, by a location service or navigation service in the form of a road network. Another possibility is to determine or extract the lane data from swarm data.

[0026] The driver assistance system can retrieve swarm data from an external storage device or a backend server, such as a cloud server or edge cloud server. This swarm data can be contained, for example, in a so-called roadbook. The roadbook could be, for instance, a road network map, where driving data from various vehicles that have already traveled each road segment is stored for each segment. This driving data can be provided by other vehicles or by the vehicle itself.

[0027] The entirety of vehicles involved in creating the roadbook is also referred to as a vehicle swarm or vehicle fleet. Therefore, the driving data is also referred to as swarm data.

[0028] In addition to the swarm trajectory, the driving data can also include other data collected by the respective vehicle. This can include, for example, the previously described sensor data from the vehicle's environmental sensors and / or the resulting lane data. The roadbook can thus comprise a digital representation of the environment.

[0029] The swarm trajectory can be, for example, the driving trajectory of a single vehicle or a combined trajectory resulting from the driving trajectories of several vehicles. In addition to the respective road path, the swarm trajectory can also include, for example, a direction of travel and / or a speed indication for each vehicle.

[0030] The invention includes embodiments that offer additional advantages.

[0031] According to one embodiment, the method checks, particularly before using the swarm trajectory, whether center guidance of the vehicle along the virtual lane center is activated or deactivated for lateral control. Only if center guidance is activated is the virtual lane center replaced by the swarm trajectory as the control parameter for lateral control. This means that the lateral control now adjusts to the course of the swarm trajectory. Conversely, only if center guidance is deactivated is the lateral control activated directly with the swarm trajectory as the control parameter. This means that lateral control within the driving corridor is activated, and the swarm trajectory, and not the virtual lane center, is used directly as the control parameter. To check the deviation criterion, a potential virtual lane center can be determined from the lane data, for example, to which the system would adjust if lateral control were active.A distinction is therefore made between whether the center guidance is already used in the driving corridor, or whether the swarm data should be used directly for center guidance adapted to the driving situation.

[0032] According to one embodiment, it is checked whether the swarm trajectory coincides with the virtual track center according to a predefined further approximation criterion. That is, a comparison is made between the swarm trajectory and the target trajectory. The result of the comparison then provides information about the extent to which the swarm trajectory and the target trajectory deviate from each other. "Coherence" here means, in particular, that the deviation between the swarm trajectory and the target trajectory is smaller than a predefined threshold. The threshold or limit value, above which the difference in trajectory is classified as relevant, is specified or contained in the further approximation criterion.The further approximation criterion includes, for example, a rule and / or an algorithm, the application or execution of which determines the approximation of the swarm trajectory to the virtual lane center. Depending on the result of the check, i.e., depending on the comparison result, the virtual lane center is used for the lateral guidance of the vehicle. This means that the lateral guidance uses the target trajectory as a control parameter. The lateral guidance is thus adjusted to follow the target trajectory.

[0033] Preferably, the verification of the further approximation criterion takes place after the swarm trajectory has been set as the driving trajectory for lateral guidance. This means that the process can switch back from the swarm trajectory to the virtual track center for lateral guidance if the swarm trajectory and the virtual track center again approach each other according to the further approximation criterion.

[0034] According to one embodiment, the swarm trajectory is used for the lateral guidance of the vehicle only if, according to the approach criterion, the distance between the swarm trajectory and the actual trajectory along a predetermined transverse axis of the vehicle falls below a predetermined first distance threshold. Additionally, the swarm trajectory is used for the lateral guidance of the vehicle only if, according to the deviation criterion, the distance between the virtual lane center and the actual trajectory along the predetermined transverse axis of the vehicle exceeds a predetermined second distance threshold.

[0035] In other words, a distance—that is, the difference between the swarm trajectory and the actual trajectory, and between the actual trajectory and the target trajectory—can be determined and then compared with the respective distance threshold. This threshold is a limit value that indicates whether the trajectories being compared differ sufficiently from each other or coincide sufficiently to justify prioritizing the swarm trajectory. The described distance assessment can be implemented analogously for the further approximation criterion.

[0036] To prevent short-term or temporary differences in distance from directly triggering a trajectory correction, the respective distance is preferably evaluated within a predefined section of the driving corridor along one direction of travel of the vehicle or for a predefined time interval. Only if the distance within the section or for the time interval has reliably exceeded or fallen below the respective distance threshold is the comparison result classified as relevant and the swarm data prioritized. Otherwise, the original lateral guidance control is maintained. The section can, for example, be a section or segment of the roadway located behind the vehicle, i.e., opposite to the direction of travel. For example, the section can be 50 or 100 m behind the rear axle of the vehicle.

[0037] For distance determination, the respective trajectory can be considered, for example, as a chain of discrete data points, each with its own geocoordinate. These discrete data points can then be compared along the transverse axis. The transverse axis can be used, for example, as a reference axis in a coordinate system. A longitudinal axis of the vehicle can be used, for example, as a second reference axis in the coordinate system. Starting from an origin or reference point in the coordinate system where the transverse and longitudinal axes intersect, a value can be assigned to each data point for the respective trajectory. This value can, for example, represent the distance of the data point from the origin in meters.

[0038] The transverse axis can be, for example, the rear axle of the vehicle. The longitudinal axis can be, for example, the vehicle's center axis in the direction of travel. In a two-dimensional xy-coordinate system, the longitudinal axis can be represented by the x-axis, while the transverse axis is represented by the y-axis. The origin (x = 0 and y = 0) is the intersection of the longitudinal axis and the rear axle. Since only data points along the transverse axis are used for distance determination, these data points have an x-value of zero, while the y-value, measured from the origin, can be a numerical value, for example, with a positive or negative sign. The respective distance value can then be determined, for example, from the difference in the y-values ​​of the discrete data points along the rear axle.The sign indicates, in particular, whether the associated trajectory is located, for example, to the right or left of the x-axis, i.e., to the right or left of the motor vehicle.

[0039] According to one embodiment, the swarm trajectory is used for lateral guidance of the vehicle only if, according to the approximation criterion, the area between the swarm trajectory and the actual trajectory within a predetermined section of the driving corridor along one direction of travel of the vehicle falls below a predetermined first area threshold. Furthermore, the swarm trajectory is used for lateral guidance of the vehicle only if, according to the deviation criterion, the area between the virtual lane center and the actual trajectory within the predetermined section of the driving corridor along the direction of travel of the vehicle exceeds a predetermined second area threshold.

[0040] Therefore, in addition to or as an alternative to distance assessment, an area assessment can be performed to justify prioritizing swarm data over the virtual track center for lateral guidance. For this purpose, the areas between the respective trajectories are determined according to the relevant criterion and then compared with the respective area threshold. This area threshold is a limit value that indicates whether the trajectories being compared differ sufficiently from each other or coincide sufficiently to justify prioritizing the swarm trajectory. The described area assessment can be implemented analogously for the other approximation criterion.

[0041] The area assessment, as described, for example, in connection with the distance assessment, is carried out within the specified section of the driving corridor. For this purpose, the trajectories to be compared can be considered, as previously described, as a chain of data points. For the calculation of the area, for example, individual area segments A can be used. i The areas are summed up from the discrete data points along the respective trajectory. The following equations, for example, are suitable for calculating the area: A˜=∑i=1nAim <F1 A'=∑i=1nAim>F2

[0042] In the equations, Ã denotes the area between the swarm trajectory and the actual trajectory, and A' denotes the area between the swarm trajectory and the target trajectory. F1 denotes the first area threshold, and F2 denotes the second area threshold. The index n indicates the number of area segments summed. The index m represents a weighting factor for evaluating or weighting the respective area segment. For example, m could be 1, 2, 1 / 2, 1 / 3 or any other evaluation factor can be chosen.

[0043] According to one embodiment, the driver's steering input is only considered in the process if it contains a steering torque greater than a predefined threshold. This threshold is chosen to exclude steering torque caused by road conditions. In other words, any steering input made by the driver because it appears necessary due to road conditions or the state of the road surface is disregarded in the process. If the steering input is disregarded, this means that the swarm data is not prioritized for lateral guidance.

[0044] Road conditions can include, for example, road damage or defects such as potholes or obstacles on the road, such as puddles or obstacles. This prevents short-term interventions, for example due to the road surface condition, from affecting the vehicle's lateral control.

[0045] To determine the steering torque, the driver assistance system can, for example, evaluate sensor data from the steering wheel sensor, such as a capacitive sensor on the steering wheel. Such a capacitive sensor on the steering wheel can measure, for example, grip strength, points of contact, or the contact area on the steering wheel. Depending on the sensor data provided by the steering wheel sensor, the steering movement and thus the steering torque can be derived. Alternatively or additionally, it is possible to determine the steering torque using sensor data from a wheel or tire sensor on the vehicle. For example, the wheel sensor can provide the steering torque as a function of the steering angle of the respective wheel.

[0046] Preferably, the steering torque is evaluated within a predetermined period or time interval. For example, a steering intervention is only recognized as such if the steering torque, i.e., the steering angle, is maintained over the predetermined period. This allows short-term interventions, such as those required to avoid an obstacle or road damage, to be disregarded for lateral control purposes.

[0047] According to one embodiment, the actual trajectory, the swarm trajectory, and the virtual lane center are determined for each section of the driving corridor that has already been traversed. The section, and in particular its dimensions or length, is selected either based on a predetermined distance to the vehicle, especially its transverse axis, or based on a vehicle speed. This means that the prioritization of the swarm data is based on a retrospective analysis and evaluation of the respective trajectories. Thus, a section opposite the direction of travel of the vehicle, specifically behind a rear axle, is evaluated. For example, a section 5 m, 50 m, or 100 m behind the vehicle can be considered.

[0048] Alternatively, according to one embodiment, the swarm trajectory and the virtual lane center are determined in a section of the driving corridor yet to be traversed, and the actual trajectory is predicted for this section based on the steering input. The section, and in particular its dimensions or length, is selected either based on a predetermined distance to the vehicle, especially its transverse axis, or based on a vehicle speed. This allows for a predictive or anticipatory analysis and evaluation of the respective trajectories. The trajectories are thus pre-calculated or estimated for the upcoming section. Therefore, a section ahead in the direction of travel, i.e., in front of the vehicle's rear axle, can be used for the evaluation.For example, a section of 5 m, 50 m, or 100 m in front of the vehicle can be taken into account.

[0049] For use cases or application situations that may arise during the procedure and are not explicitly described here, it may be provided that, according to the procedure, an error message and / or a request for user feedback is issued and / or a default setting and / or a predetermined initial state is set.

[0050] According to one aspect, the invention also relates to a driver assistance system for the lateral guidance of a motor vehicle. This driver assistance system is configured to determine a virtual lane center for the motor vehicle within a predefined driving corridor from environmental data. This virtual lane center can then be used for the lateral guidance of the motor vehicle. The driver assistance system is also configured to determine a swarm trajectory from swarm data, which specifies the respective driving trajectory of at least one other vehicle that has already traversed the predefined driving corridor. Furthermore, the driver assistance system is configured to check whether, depending on a steering input performed by the driver of the motor vehicle, an actual trajectory results that deviates from the virtual lane center according to a predefined deviation criterion and coincides with the swarm trajectory according to a predefined approximation criterion.Depending on the result of the check, the driver assistance system is trained to use the swarm trajectory for the lateral guidance of the vehicle.

[0051] This means that the driver assistance system can be configured to perform or execute a method as described above by way of example, and / or its embodiments. For this purpose, the driver assistance system can, for example, comprise a control device or evaluation device. The control device can include a data processing device or a processor unit configured to perform an embodiment of the method according to the invention. The processor unit can, for this purpose, include at least one microprocessor and / or at least one microcontroller and / or at least one FPGA (Field Programmable Gate Array) and / or at least one DSP (Digital Signal Processor). In particular, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or an NPU (Neural Processing Unit) can be used as the microprocessor.Furthermore, the processor device can include program code configured to execute the embodiment of the method according to the invention when executed by the processor device. The program code can be stored in a data memory of the processor device. The processor device can be based, for example, on at least one circuit board and / or on at least one SoC (System on Chip).

[0052] According to one aspect, the invention also relates to a motor vehicle with a corresponding driver assistance system, as described above by way of example. As described at the outset, the driver assistance system can control the motor vehicle for lateral guidance or lateral steering. The motor vehicle can, for example, have one or more steering actuators that the driver assistance system controls to implement the lateral guidance. The motor vehicle is preferably designed as a motor vehicle, in particular as a passenger car or truck, or as a passenger bus or motorcycle.

[0053] The invention also includes further developments of the driver assistance system and the motor vehicle according to the invention, which have features already described in connection with the further developments of the method according to the invention. For this reason, the corresponding further developments of the driver assistance system and the motor vehicle according to the invention are not described again here.

[0054] The invention also includes combinations of the features of the described embodiments. The invention therefore also includes realizations that each exhibit a combination of the features of several of the described embodiments, provided that the embodiments have not been described as mutually exclusive.

[0055] The following are exemplary embodiments of the invention described. This is illustrated by: Fig. 1. A schematic representation of a motor vehicle with a driver assistance system for lateral control of the motor vehicle; and Fig. 2 a schematic process flow diagram for an exemplary procedure for operating the driver assistance system according to Fig. 1.

[0056] The exemplary embodiments described below are preferred embodiments of the invention. In these exemplary embodiments, the described components each represent individual features of the invention, which can be considered independently of one another and each further develops the invention independently. Therefore, the disclosure is intended to include combinations of features of the embodiments other than those shown. Furthermore, the described embodiments can also be supplemented by further features of the invention already described.

[0057] In the figures, identical reference symbols denote functionally equivalent elements.

[0058] Fig. Figure 1 shows a schematic representation of a motor vehicle 1 from a bird's-eye view or top view. The motor vehicle 1 is preferably designed as a passenger car. The motor vehicle 1 includes a driver assistance system 2. The driver assistance system 2 is preferably designed as a transverse driver assistance system. That is, the driver assistance system 2 is designed for the automatic or semi-automatic guidance or control of the motor vehicle 1, in particular in the transverse direction, i.e., perpendicular to a direction of travel R in which the motor vehicle 1 is moving.

[0059] To perform the lateral guidance function, the driver assistance system 2 monitors and / or observes the environment of the vehicle 1, in particular road traffic or traffic events, and automatically executes reactions to observed events in the environment, preferably without requiring intervention from the driver of the vehicle 1. In the case of lateral guidance, these reactions include, in particular, a steering maneuver. For this purpose, the driver assistance system 2, for example, controls a steering system or steering actuator 3 of the vehicle 1 with a corresponding control signal or steering signal. When the steering system 3 is supplied with the steering signal, a steering torque is generated, for example, by means of which the vehicle 1 is steered to a specific position or trajectory.

[0060] For environmental sensing, the vehicle 1 includes environmental sensors. In this document, for example, a camera 4 is shown as the environmental sensor. An alternative configuration of the environmental sensor could be, for example, a radar sensor. In this document, the camera 4 is preferably a front camera and thus captures the environment in the direction of travel R in front of the vehicle 1. The camera 4 captures an image of the vehicle 1's surroundings in a known manner. The resulting sensor data or image data, which contain the image, are provided by the camera 4 to the driver assistance system 2 for evaluation in the form of environmental data U.

[0061] The driver assistance system 2 can receive and process the environmental data U. This processing includes, for example, the detection and extraction of objects or states in the environment. For this purpose, the driver assistance system 2 can, for example, apply an object recognition algorithm to the environmental data U in a known manner. For data processing, the driver assistance system 2 preferably comprises a control device or data processing device, which, for example, includes one or more microcontrollers.

[0062] Objects in the environment can include, for example, lane data. This means that the driver assistance system 2 can extract lane data from the environmental data U. The lane data can contain information or details about a roadway 10, such as its condition or dimensions.

[0063] Fig. Figure 1 shows the motor vehicle 1 in motion, i.e., in operation, on a roadway 10. Fig. Figure 1 here represents a driving corridor 11, which can, for example, form a lane of such a roadway 10 that is suitable for use by the motor vehicle 1. The in Fig. The section of the driving corridor 11 shown in Figure 1 is designed as an example of an open exit or lane widening. This means that the driving corridor 11 widens along the direction of travel R of the vehicle 1. The driving corridor 11 thus widens in the direction of travel R. The driving corridor 11 is delimited from the rest of the roadway 10 or any other surroundings of the vehicle 1 by corridor boundaries 12. These corridor boundaries 12 can be, for example, lane markings or a road edge. The lane markings can be captured as objects from the surrounding data.

[0064] Based on the lane data, it is now possible to define a driving path for vehicle 1, along which the vehicle 1 is to move. The corridor boundaries 12 can thus be used as reference points for calculating the driving path. The driving path can, for example, be a trajectory that vehicle 1 follows when automatic lateral guidance and / or, for example, longitudinal guidance is activated. In conventional driver assistance systems 2, lateral guidance is implemented as so-called center guidance. Center guidance involves steering vehicle 1 along a virtual lane center 14, in this case, the center between the corridor boundaries 12. This virtual lane center 14 thus defines a target trajectory 17, which vehicle 1 follows in response to the center guidance.

[0065] As an alternative to center guidance based on sensor data from environmental sensors, swarm intelligence can be used for lateral guidance. In this scenario, the vehicle 1 is part of a swarm or fleet of vehicles, which may include one or more other vehicles, referred to as "external vehicles" in the following. All vehicles belonging to the swarm collect environmental data (U) and driving data while driving. Driving data can include, for example, driving settings such as speed, direction, or driving mode, which the respective vehicle uses to traverse a current lane segment or driving corridor (11). The data collected by the vehicles can be aggregated and stored as swarm data (S).The swarm data S thus includes, for every geographical position where any vehicle has been in the past, information about how the vehicle behaved at that position and what environmental conditions prevailed.

[0066] The swarm data S can be stored, for example, in the form of a so-called roadbook, i.e., a road network map, in an external storage device. This storage device could be, for example, a server such as a Cloudserver 20 or an edge cloud server. Every vehicle in the fleet can retrieve or download this swarm data S from the Cloudserver 20 and use it, for example, for lateral guidance.

[0067] The swarm data S can, for example, include a swarm trajectory 15 for the driving corridor 11. The swarm trajectory 15 is, in particular, a driving path followed by one or more other vehicles when passing through the driving corridor 11. As in Fig. As shown in Figure 1, the swarm trajectory 15 differs from the target trajectory 17. While the target trajectory 17 is oriented towards the virtual lane center 14, the swarm trajectory 15 in the present embodiment is oriented towards the right corridor boundary 12. The swarm trajectory 15 is thus closer to right-hand traffic than the target trajectory 17. Different lane widths are assumed for their calculation.

[0068] In the Fig. In the situation shown in Figure 1, with lane widening, it may happen that the underlying lane data for lane centering via the driver assistance system 2 is not suitable for situation-dependent centering. Specifically, when the lane widens, it is possible that several lanes may be created within the driving corridor 11, for example, for vehicles traveling alongside vehicle 1, where the lanes are not separated by lane markings. Centering vehicle 1 along the virtual lane center 14 would therefore result in vehicle 1 blocking other traffic.

[0069] A driver of motor vehicle 1 can usually recognize or resolve such situations more quickly than the driver assistance system 2. Therefore, it can happen that the driver steers against the center guidance. That is, the driver performs a steering intervention L and, for example, operates the steering system 3 of motor vehicle 1. As a rule, the driver will also tend to orient themselves towards right-hand traffic, so that motor vehicle 1, as shown in the present embodiment, is steered more towards the swarm trajectory 15 and thus away from the intended trajectory 17.

[0070] Fig. Figure 1 shows an actual trajectory 16 that results when the driver performs the steering input L. According to the embodiment in Fig. 1. The swarm trajectory 15 now better reflects the driver's desired driving behavior than the center guidance. If it is determined that the swarm data S reflects the driver's desired driving behavior much better, the swarm data S should be prioritized over the lane data or environmental data. The driver assistance system 2 can be operated with a suitable method for this purpose, which takes the driver's driving behavior into account for lateral guidance.

[0071] Fig. Figure 2 shows an example of a schematic process flow diagram with individual process steps for a procedure for operating the driver assistance system 2 according to Fig. 1. According to steps S1, S2, and S3 of the procedure, the steering activity (i.e., the steering intervention L), the environmental data U, and the swarm data S are used or specified as input or input data. The steering intervention L provides the current trajectory 16 (step S1). The swarm data S provide the swarm trajectory 15 (step S2), as described previously.

[0072] Based on this input data, the driver assistance system 2 can execute an algorithm, described in more detail below, to check the prioritization of the swarm data. In step S4, the algorithm specifically considers whether the center guidance, i.e., the lateral guidance function, is already active or deactivated. If the center guidance is active (Y), the process continues in step S5a. If the center guidance is deactivated (N), the process continues in step S5b.

[0073] In step S5a, the virtual lane center 14 is captured or determined from the lane data, i.e., the environmental data U. The virtual lane center 14 is then used to calculate the target trajectory 17. Thus, the actual target trajectory 17, already traveled or yet to be traveled by the vehicle 1 within the driving corridor 11, is taken into account for the further procedure.

[0074] In step S5b, a potential target trajectory 17' is determined based on the virtual lane center 14 calculated from the environmental data U. The potential target trajectory 17' is a trajectory that the vehicle 1 would theoretically follow if the center guidance system were active. Thus, instead of the actual target trajectory 17 that the vehicle 1 has traveled, the potential target trajectory 17' is assumed as a mathematical model or estimate of the target trajectory 17.

[0075] The procedure then continues in step S6. In step S6, it is checked whether the driver's steering intervention L, i.e., for example, a steering torque that the driver applies to the steering system 3 when intervening in the steering, better matches the swarm trajectory 15. That is, the aim is to determine whether the steering activity has led the driver to prefer driving based on the swarm data S and whether, through the steering intervention L, the vehicle 1 has, for example, been steered in the direction of the swarm trajectory 15. For this purpose, in step S6, the trajectories of 15, 16, and 17 are compared according to specific criteria. On the one hand, step S6 checks whether the actual trajectory 16 matches the swarm trajectory 15 according to a predefined approximation criterion K1.On the other hand, in step S6 it is checked whether the actual trajectory 16 deviates from the virtual track center 14, i.e. the target trajectory 17 or the potential target trajectory 17', according to a given deviation criterion K2.

[0076] According to the exemplary embodiments shown above Fig. 1 and Fig. 2. An area assessment is carried out for this purpose. This means that the check includes a comparison of an approximation area Ä, which results between the actual trajectory 16 and the swarm trajectory 15, and a deviation area A', which results between the actual trajectory 16 and the target trajectory 17, 17', with a predefined area threshold value F1, F2, respectively. For the calculation, the areas à and A' are divided into discrete area fractions à i and A' i subdivided and the area proportions à i and A' iThe values ​​are summed up. The index i is a number between 1 and n, where n represents the number of existing area shares used for the valuation. As in Fig. Figure 1 shows that, for this type of calculation, a section 13 of the driving corridor 11, located in the direction of travel R, is used in particular. That is, a retrospective comparison of the trajectories 15, 16, and 17 is carried out.

[0077] For the area calculation, the trajectories 15, 16, 17 are considered as point chains of discrete data points D and are in particular related along the direction of travel R to a given transverse axis, such as the rear axle 5 of the motor vehicle 1.

[0078] The respective area can be calculated, for example, using the following formula: A˜=∑i=1nA˜im <F1 A'=∑i=1nA'im>F2 where m is a rating factor for the respective area section à i and A'i F1 and F2 are the first and second area thresholds used for the assessment.

[0079] According to the embodiment in Fig. 2. The approximation criterion K1 is therefore only fulfilled if the approximation surface à i The area between the swarm trajectory 15 and the actual trajectory 16 within section 13 is smaller than the first area threshold F1. The deviation criterion K2, however, is only fulfilled if the deviation area A' iThe distance between the virtual lane center 14 and the actual trajectory 16 is greater than the second area threshold F2. The two area thresholds F1 and F2 specify respective limit values ​​that determine how far the paths of the trajectories 15, 16, and 17 must differ at least for the prioritization of the swarm data S to be taken into account when driving. This prevents, for example, even small deviations from the target trajectory 17, 17' or short-term deviations made by the driver assistance system 2 or the driver due to road conditions (e.g., potholes or ruts) from leading to a new course for the vehicle 1 or for the automatic control system of the vehicle 1.

[0080] Then, if the approach criterion K1 and the deviation criterion K2 are met, the procedure continues in step S7a or S7b. Step S7a is executed only if step S4 shows that center guidance is already active (J). In step S7a, the virtual lane center 14 for the lateral guidance of vehicle 1 is discarded and replaced by the swarm trajectory 15. That is, the swarm data S is prioritized for lateral guidance.

[0081] Step S7b is executed only if, according to step S4, it has been determined that center guidance is deactivated (N). Subsequently, in step S7b, lateral guidance, i.e., the lateral guidance function via driver assistance system 2, is activated, and the swarm trajectory 15 is used directly for the lateral guidance of the vehicle 1. This means that the swarm data S is prioritized directly over the potential target trajectory 17'.

[0082] Preferably, the swarm data S can now be prioritized over the center guidance until a further approximation criterion K3 is met. The further approximation criterion K1 indicates that the course of the swarm trajectory 15 essentially coincides with the virtual track center 14. For this purpose, for example, in step S8 of the procedure, it can be checked whether an approximation area  is smaller or larger than a third area threshold F3. The third approximation area  is the sum of individual area segments, as described previously, between the swarm trajectory 15 and the target trajectory 17, 17'. The further approximation criterion K3 is only met if the approximation area  is smaller than the area threshold. The area evaluation according to S8 is preferably again carried out within a predetermined section 13 of the driving corridor 11.

[0083] The respective area threshold value F1, F2, F3 is preferably freely applicable and can, for example, be stored in a characteristic map or in a characteristic curve depending on a length of section 13 and / or a driving speed and be retrieved or adapted accordingly for the respective driving situation.

[0084] As an alternative to area assessment, the distance between trajectories 15, 16, and 17 relative to the transverse axis 5 of vehicle 1 can be determined. Based on the respective approach criterion K1, K3, or deviation criterion K2, a corresponding distance threshold for the distance comparison can then be specified. Instead of relating the area assessment or the distance assessment to section 13, it is conceivable, for example, to consider a specific period of a few seconds for the assessment, during which vehicle 1 crossed the roadway.

[0085] If the further approximation criterion K3 is met, the procedure continues in step S9. In step S9, the virtual center 14 is again prioritized for lateral guidance, especially center guidance, since the difference between the virtual center 14 and the swarm trajectory 15 is again small. It is therefore possible to switch back from the swarm trajectory 15 to the target trajectory 17.

[0086] According to the exemplary embodiments, a retrospective analysis or evaluation of the trajectories 15, 16, 17 is provided. Alternatively, a forward-looking evaluation can be implemented. For this purpose, for example, instead of the section 13 already traversed, a section yet to be traversed, which lies ahead of the motor vehicle 1 in the direction of travel R, is used, and the trajectories 15, 16, 17 for this ahead section are estimated or determined.

[0087] Overall, the implementation examples demonstrate a possibility for prioritizing swarm data using a cross-functional assistance function. QUOTES INCLUDED IN THE DESCRIPTION

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

[0000] US 10 990 103 B2

[0005] DE 10 2021 206 886 A1

[0006]

Claims

[1] Method for the lateral guidance of a motor vehicle (1) by means of a driver assistance system (2), comprising the following steps: - determining a virtual lane center (14) in a predetermined driving corridor (11) for the motor vehicle (1) from environmental data (U), wherein the virtual lane center (14) can be used for the lateral guidance of the motor vehicle (1), - determining a swarm trajectory (15) from swarm data (S) which indicates a respective travel trajectory of at least one other vehicle which has already travelled through the specified travel corridor (11), - checking whether, depending on a steering intervention (L) carried out by a driver of the motor vehicle (1), an actual trajectory (16) results which deviates from the virtual lane center (14) according to a predetermined deviation criterion (K2) and corresponds to the swarm trajectory (15) according to a predetermined approach criterion (K2), - depending on a result of the check: using the swarm trajectory (15) for the lateral guidance of the motor vehicle (1). [2] Method according to claim 1, wherein it is checked whether a center guidance of the motor vehicle (1) along the virtual lane center (14) is activated or deactivated for the lateral guidance, wherein exactly when the center guidance is activated, the virtual lane center (14) is replaced as a control parameter for the lateral guidance by the swarm trajectory (15), and exactly when the center guidance is deactivated, the lateral guidance is activated directly with the swarm trajectory (15) as a control parameter. [3] Method according to one of the preceding claims, wherein it is checked whether the swarm trajectory (15) corresponds to the virtual lane center (14) according to a predetermined further approach criterion (K3), and depending on a result of the check, the virtual lane center (14) is used for the lateral guidance of the motor vehicle (1). [4] Method according to one of the preceding claims, wherein the swarm trajectory (15) is only used for the transverse guidance of the motor vehicle (1) if, according to the approach criterion (K1), a distance between the swarm trajectory (15) and the actual trajectory (16) along a predetermined transverse axis of the motor vehicle (1) falls below a predetermined first distance threshold value and if, according to the deviation criterion (K2), a distance between the virtual lane center (14) and the actual trajectory (16) along the predetermined transverse axis of the motor vehicle (1) exceeds a predetermined second distance threshold value. [5] Method according to one of the preceding claims, wherein the swarm trajectory (15) is only used for the lateral guidance of the motor vehicle (1) if, according to the approach criterion (K1), an area (Ä) between the swarm trajectory (15) and the actual trajectory (16) within a predetermined section (13) of the driving corridor (11) along a direction of travel (R) of the motor vehicle (1) falls below a predetermined first area threshold value (F1) and if, according to the deviation criterion (K2), an area (A') between the virtual lane center (14) and the actual trajectory (15) within the predetermined section (13) of the driving corridor (11) along the direction of travel (R) of the motor vehicle (1) exceeds a predetermined second area threshold value (F2). [6] Method according to one of the preceding claims, wherein the steering intervention (L) is only taken into account if the steering intervention (L) contains a steering torque which is greater than a predetermined steering torque threshold value, wherein the steering torque threshold value is selected such that a steering torque caused by road conditions is excluded. [7] Method according to one of the preceding claims, wherein the actual trajectory (16) and the swarm trajectory (15) and the virtual lane center (14) are determined in a section (13) of the driving corridor (11) that has already been traveled through, the section (13) being selected either as a function of a predetermined distance from the motor vehicle (1) or as a function of a vehicle speed. [8] Method according to one of the preceding claims 1 to 6, wherein the swarm trajectory (15) and the virtual lane center (14) are determined in a section of the driving corridor (11) still to be traveled through, and the actual trajectory (16) is predicted as a function of the steering intervention (L) for the section of the driving corridor (11) still to be traveled through, and the section is selected either as a function of a predetermined distance from the motor vehicle (1) or as a function of a vehicle speed. [9] Driver assistance system (2) for the lateral guidance of a motor vehicle (1), wherein the driver assistance system (2) is designed to: - to determine a virtual lane center (14) in a predetermined driving corridor (11) for the motor vehicle (1) from environmental data (U), wherein the virtual lane center (14) can be used for the lateral guidance of the motor vehicle (1), - to determine a swarm trajectory (15) from swarm data (S) which indicates a respective travel trajectory of at least one other vehicle which has already travelled through the predetermined travel corridor (11), - to check whether, depending on a steering intervention (L) carried out by a driver of the motor vehicle (1), an actual trajectory (16) results which deviates from the virtual lane center (14) according to a predetermined deviation criterion (K2) and corresponds to the swarm trajectory (15) according to a predetermined approach criterion (K1), and - depending on a result of the check: to use the swarm trajectory (15) for the lateral guidance of the motor vehicle (1). [10] Motor vehicle (1) with a driver assistance system (2) according to claim 9.

Images