Method for guiding a motor vehicle across a road with at least two lanes, and motor vehicle

Abstract

Method for guiding a motor vehicle (14) across a road (10), in which - a lane marking (28) assigned to a driver's side of the motor vehicle (14) is assigned to a lane type (V1b) based on sensor data (32) representing an environment of the motor vehicle (14), wherein the lane type characterizes whether the lane marking (28) is assigned to an ego lane on which the motor vehicle (14) is to be driven, or to an adjacent lane near the ego lane, - the lane marking (28) assigned to the driver's side of the motor vehicle (14) is assigned to the lane type (V1a) based on swarm data (18) received from other vehicles, - the lane type assigned to the lane marking (28) is determined as the lane type determined based on the swarm data (18) if the lane type determined on the basis of the sensor data (32) and the lane type determined on the basis of the swarm data (18) differ (V4), and - the motor vehicle (14) is assisted by means of a control device depending on the specified lane type of the lane marking (28) assigned to the driver's side of the motor vehicle (14) by means of a control device, is steered laterally (V4).

Description

[0001] The present invention relates to a method for guiding a motor vehicle laterally on a road with at least two lanes and to a motor vehicle.

[0002] WO 2021 / 043507 A1 discloses a method for the lateral guidance of a vehicle, in which environmental data of the vehicle is recorded while traveling on the road. Furthermore, stored environmental data is retrieved that was recorded by a plurality of other vehicles, not currently traveling on the road, while traveling on the road. This stored environmental data is validated against the recorded environmental data. The lateral guidance of the method is carried out based on the validated environmental data. The environmental data captures at least one lane marking or boundary on the left (from the vehicle's perspective), one lane marking or boundary on the right (from the vehicle's perspective), or at least one outer edge of the roadway as an environmental feature.

[0003] Furthermore, US patent 10,955,855 B1 discloses a method for autonomous navigation in which journey segments are generated from a starting point to a destination. This involves detecting a highway entrance, exit lane, or median structure based on road markings using a camera and a sensor. A 3D model is generated in real time based on the data output from the camera and sensor. This real-time 3D model is validated at the current location using a high-resolution map database. If the journey segment passes the highway entrance or exit based on the validated real-time 3D model, it follows the existing lane without exiting. Otherwise, it follows the highway entrance or exit.

[0004] German patent DE 10 2019 213 185 A1 discloses a method for the lateral guidance of a vehicle in which environmental data of the vehicle are recorded while traveling along a route. Furthermore, stored environmental data collected by multiple other vehicles traveling along the route are retrieved. The stored environmental data is validated against the recorded environmental data, and the lateral guidance of the vehicle is carried out based on this validated environmental data.

[0005] Furthermore, DE 10 2020 202 163 A1 discloses a method for detecting objects and / or structures in the vicinity of a vehicle. Moreover, DE 10 2016 220 647 A1 discloses a method for virtually representing at least one supplementary differential roadway equipment for a roadway to be traveled on by a motor vehicle.

[0006] The object of the present invention is to provide a solution that enables safe lateral guidance of a motor vehicle when swarm data and sensor data of the motor vehicle diverge.

[0007] This problem is solved by the subject matter of the independent claims. Further possible embodiments of the invention are disclosed in the dependent claims, the description, and the figures.

[0008] The invention relates to a method for guiding a motor vehicle laterally on a road. This method is particularly advantageous on a two-lane road when the center line is missing or obscured, for example by dirt or snow, and therefore cannot be detected by the vehicle's sensors. Especially when there is oncoming traffic on this road, which has at least two lanes, it is important that the motor vehicle is guided safely in its lane to avoid a collision with oncoming traffic.

[0009] The method involves assigning a lane marking associated with the driver's side of the vehicle to a lane type based on sensor data representing the vehicle's surroundings. The lane type characterizes whether the lane marking is assigned to a dedicated lane in which the vehicle must be driven or to an adjacent lane. This means that the lane marking associated with the driver's side of the vehicle—located to the left of the vehicle in right-hand traffic and to the right in left-hand traffic—is detected by a sensor device in the form of sensor data representing the lane marking.Based on this sensor data, an electronic processing unit determines whether the detected lane marking defines the driver's lane or the adjacent lane and is therefore assigned to the driver's lane or the adjacent lane. The process further involves assigning the lane marking associated with the driver's side of the vehicle to the lane type based on swarm data received from other vehicles. Thus, the swarm data determines whether the lane marking associated with the driver's side belongs to the driver's lane or the adjacent lane. The swarm data describes environmental data that has been recorded by other vehicles in the past and made available to the vehicle. This swarm data can be provided to the vehicle, for example, via a server.

[0010] The procedure further stipulates that the lane type assigned to the lane marking is determined based on swarm data if the lane type determined based on sensor data and the lane type determined based on swarm data differ for the lane marking assigned to the driver's side. In this case, the electronic computing unit can perform a comparison between the lane type determined via swarm data for this lane marking and the lane type determined via sensor data. If the comparison reveals that the lane type determined based on sensor data and the lane type determined based on swarm data for this lane marking are not identical, then the lane type of the lane marking is specified as the lane type determined based on swarm data.This results in an override of the lane marking type determined based on sensor data. In this case, it is assumed that there is an error in the sensor data, which causes the lane marking type determined based on the sensor data to differ from the lane marking type determined based on swarm data. This ensures that the vehicle can be safely laterally steered even in the event of a sensor data error.

[0011] The procedure further stipulates that the vehicle is laterally steered by a control unit, depending on the defined lane type of the lane marking assigned to the driver's side of the vehicle. Depending on whether the lane marking, determined based on swarm data, defines the vehicle's own lane or the adjacent lane, the vehicle is steered accordingly to minimize the risk of a collision with oncoming traffic in the adjacent lane. This allows for particularly safe operation of the vehicle. The vehicle can be laterally steered based on at least one predefined condition, using the defined lane type of the lane marking assigned to the driver's side of the vehicle.For example, the distance of a vehicle to a lane marking assigned to the passenger side of the vehicle can only be adjusted according to the defined lane type if, as a predefined condition, the lane type determined based on sensor data and the lane type determined based on swarm data for the lane marking assigned to the driver's side differ for at least a predefined distance traveled and / or a predefined elapsed time. This prevents frequent, unwanted changes in the vehicle's distance to the lane marking assigned to the passenger side. To ensure the vehicle remains centered on the road, the lane marking assigned to the driver's side and the lane marking assigned to the passenger side must each be defined as ego markings.In this process, the vehicle is guided from one side of the road to the center when the predetermined distance has been covered and / or the predetermined time has elapsed. A primary source for this lateral guidance is a live-detected shape of the lane marking corresponding to the passenger side of the vehicle.

[0012] In a possible further development of the invention, the vehicle is steered laterally based on sensor data. This sensor data can be provided by a camera, radar, lidar, or ultrasonic sensor. Lateral steering of the vehicle based on sensor data enables particularly safe control of the vehicle using real-time data representing its surroundings. This allows the vehicle to be guided along the road with exceptional safety, especially in rapidly changing conditions.

[0013] In another possible alternative embodiment of the invention, the vehicle is laterally steered using swarm data. This means that both the lane type is determined using swarm data, and the lateral steering of the vehicle is controlled by swarm data, by using the swarm data instead of the sensor data for lateral steering. Particularly in the event of an error in the sensor data, it is especially safe to laterally steer the vehicle using swarm data, especially since the swarm data is verified across multiple vehicles.

[0014] In a further possible embodiment of the invention, it is provided that the lane type assigned to the lane marking is defined as the lane type determined based on the sensor data if, for the lane marking assigned to the passenger side of the vehicle, a lane type determined based on the sensor data and a lane type determined based on the swarm data differ. In other words, the lane type assigned to the lane marking is defined as the lane type determined based on the sensor data if, for both the lane marking assigned to the driver's side and the lane marking assigned to the passenger side, the respective lane types determined based on the sensor data and the lane types determined based on the swarm data for the respective lane marking differ and are therefore not identical.If it is determined that both the lane markings assigned to the driver's side and those assigned to the passenger's side are assigned to different lane types via the swarm data and the sensor data, then it is determined that there is most likely an error in the swarm data, since the sensor data is generally considered more reliable than the swarm data. For example, a past change to the road's lane configuration may be present, which is already reflected in the sensor data but not yet in the swarm data.If the lane type determined for the lane marking assigned to the driver's side differs between the sensor data and the swarm data, but the lane type determined for the lane marking assigned to the passenger side is identical for both sensor data and the swarm data, then the lane type assigned to the lane marking is defined as the lane type determined based on the swarm data. This allows the lane type of the lane marking assigned to the driver's side to be determined with a particularly high probability of accuracy, thereby minimizing the risk of collision between the vehicle and oncoming traffic or another vehicle.

[0015] In a further possible embodiment of the invention, the sensor data includes camera data, which is used to analyze the lane markings associated with the driver's side of the vehicle via an image recognition method. Furthermore, the lane markings associated with the passenger side of the vehicle can be analyzed using the same image recognition method. The camera data can, in particular, include images depicting the vehicle's surroundings, especially the lane markings, which can be examined within the framework of the image recognition method to determine the presence of lane markings or to identify the type and thus the lane type of lane markings.The camera data enables a particularly comprehensive recording of the vehicle's surroundings, which, through the image recognition process, especially by means of the electronic computing device, allows for a particularly informed decision to be made about the lane type of the respective road marking, in particular the road marking assigned to the driver's side of the vehicle.

[0016] In a further possible embodiment of the invention, the lane type assigned to the road marking is determined based on the swarm data if the swarm data meets at least one predefined quality criterion. If the swarm data does not meet any predefined quality criterion, the lane type assigned to the road marking is determined based on the sensor data. The at least one quality criterion describes how reliable the swarm data is with regard to determining the lane type of the road marking. Thus, the swarm data is only used to determine the lane type of the road marking assigned to the driver's side if this swarm data has proven to be sufficiently reliable, accurate, and accurate based on the at least one predefined quality criterion.

[0017] In this context, a further possible embodiment of the invention may provide that the quality criterion is a correlation rate between characteristic values ​​determined from the swarm data and characteristic values ​​determined from the sensor data for a defined section of the road, and / or that a localization accuracy of the vehicle's location on the road based on the swarm data is specified as a quality criterion. This means that for a defined, predetermined section of the road, the sensor data is compared with the swarm data, and the quality criterion is considered fulfilled if the swarm data yields the same result as the sensor data for at least one predetermined characteristic value within a predetermined tolerance range. This characteristic value could, for example, be the distance of the vehicle to a predetermined lane marking and / or an angular error of the vehicle.The defined section of the route could, for example, be a section of road recently traveled by the vehicle, approximately 100 meters in length. This involves a historical analysis of the swarm data in comparison with the sensor data. If the quality criterion is the localization accuracy of the vehicle's position on the road based on the swarm data, then it is checked whether the localization accuracy for the vehicle is sufficiently high. Localization accuracy describes the degree of precision with which the vehicle's position is located using the swarm data, particularly on a digital map. This can involve examining the vehicle's position along a specific direction of travel. For example, it can be analyzed whether the vehicle's position can be located with meter, centimeter, or millimeter accuracy using the swarm data.The localization accuracy of the swarm data can be verified, for example, using the respective curves of the road. In particular, the sensor data can be used to determine the relative position of the vehicle on the road relative to a curve, and it can be checked whether this relative position, determined using the swarm data, corresponds to the vehicle's position on the road relative to the curve, especially within a predefined tolerance range. If the localization accuracy of the vehicle determined using the swarm data is greater than a predefined minimum localization accuracy, then this quality criterion is considered met. The respective predefined quality criteria ensure that the swarm data is only used to determine the lane type of the respective road markings if it exhibits sufficient accuracy.

[0018] In a further possible embodiment of the invention, it is provided that the lane type assigned to the road marking is determined as the lane type based on the swarm data, while a predetermined time interval is running and / or while the motor vehicle travels a predetermined distance and / or while the motor vehicle is moving in the direction of travel at a speed within a predetermined speed range and / or when the road has a curvature above a predetermined curvature limit and / or when a road having the road marking has a width within a predetermined width range and / or when the motor vehicle has a distance to a further road marking assigned to a passenger side of the motor vehicle within a predetermined distance range.This means that the lane type assigned to the lane marking is determined based on sensor data when the time interval has elapsed and / or the vehicle is outside the specified lane and / or the vehicle is traveling at a speed outside the specified speed range in the direction of travel or against the direction of travel and / or the curvature of the road is below or equal to the specified curvature limit and / or the road surface bearing the lane marking has a width outside the specified width range and / or the vehicle is at a distance from another lane marking assigned to the passenger side of the vehicle that is outside the specified distance range. Thus, specific boundary conditions can be defined for the overriding of the sensor data by the swarm data when determining the lane type.If these boundary conditions are not met, then the lane type determined based on sensor data for the lane marking assigned to the driver's side of the vehicle can be prevented from overriding the lane type determined based on swarm data. Clear limits can thus be defined as to when the lane type determined based on swarm data may override the lane type determined based on sensor data. Outside these limits, the lane type of the lane marking assigned to the driver's side is always determined by the sensor data, as it is assumed that the sensor data is particularly accurate due to its up-to-dateness. This minimizes the risk of an erroneous override of the lane type determined based on sensor data for the lane marking assigned to the driver's side by the lane type determined based on swarm data.

[0019] In a further embodiment of the invention, the lane type assigned to the driver's side of the vehicle is defined as the lane type determined based on the sensor data if the lane type determined using the sensor data and the lane type determined using the swarm data are identical. If the same lane type is determined for the driver's side of the vehicle using both the swarm data and the sensor data, then it is assumed that there is no error in the sensor data and therefore the sensor data can be used to determine the lane type.Regardless of whether the same lane type is determined for a lane marking assigned to the passenger side of the vehicle via the sensor data as well as via the swarm data, if the identical lane type is determined for the lane marking assigned to the driver's side, it is thus stipulated that the sensor data should not override each other for determining the lane type.In particular, since it is assumed that the sensor data are especially reliable for determining the lane type of the lane marking assigned to the passenger side, this method is designed so that the sensor data are only overridden by the swarm data regarding the assessment of the lane type of the lane marking assigned to the driver's side if the assessment of the lane type for the lane marking assigned to the driver's side of the vehicle differs between the swarm data and the sensor data, but not in the assessment of the lane type for the lane marking assigned to the passenger side. The risk of an incorrect assessment of the lane type of the lane marking assigned to the driver's side of the vehicle can thus be kept particularly low. Consequently, the risk of the vehicle colliding with other road users can be kept particularly low.

[0020] The invention further relates to a motor vehicle which is equipped to be operated within the framework of the previously described method according to the invention or a further development thereof. Advantages and advantageous further developments of the method according to the invention are to be regarded as advantages and advantageous further developments of the motor vehicle according to the invention and vice versa.

[0021] Further features of the invention may become apparent from the following description of the figures and from the drawings. The features and combinations of features mentioned above in the description, as well as the features and combinations of features shown below in the description of the figures and / or in the figures themselves, can be used not only in the combinations specified, but also in other combinations or individually, without departing from the scope of the invention.

[0022] The drawing shows in: Fig. 1 a schematic top view of a road with two lanes, on which a motor vehicle is driving in an ego lane and another road user is driving in an adjacent lane, wherein the motor vehicle has a sensor device comprising a camera device by means of which sensor data representing an environment of the motor vehicle can be recorded, and the motor vehicle receives swarm data from a higher-level server device, whereby the motor vehicle can be assisted to steer laterally on the road; Fig. 2. A procedure scheme for a method for lateral steering of the motor vehicle, wherein a lane type of a road marking assigned to a driver's side of the motor vehicle is determined on the basis of both sensor data and swarm data, and the motor vehicle is laterally steered depending on the respective lane types determined; and Fig. 3 a procedure scheme for guiding the motor vehicle laterally on the road.

[0023] Identical or functionally equivalent elements are marked with the same reference symbols in the figures.

[0024] In Fig. Figure 1 shows a road 10 in a top view, where the road 10 has two lanes 12. A center line to separate the lanes 12 is missing. Therefore, it can be difficult for road users traveling on the road 10 to determine how many lanes 12 the road 10 has. In this case, a motor vehicle 14 and another road user 16 are traveling on the road 10. The motor vehicle 14 is traveling in its own lane, while the other road user 16 is traveling in the opposite direction in an adjacent lane.

[0025] The vehicle 14 is configured to receive swarm data 18 from a higher-level server 20. This swarm data 18 describes driving data of other vehicles that were previously traveling along road 10, particularly in the ego lane. Furthermore, the vehicle 14 includes a camera system 22 as part of a sensor system, which is configured to record camera data 32 representing the vehicle 14's surroundings. For this purpose, the camera system 22 has a field of view 24 directed towards the vehicle 14's surroundings, specifically towards the area in front of the vehicle 14 in the direction of travel. This allows the camera system 22 to detect the lane markings 26 that define the outer boundaries of road 10.The first of the lane markings 26, which is subsequently identified by reference sign 28, delineates the road 10 on the side corresponding to the driver's side of the motor vehicle 14. The second lane marking 30 delineates the road 10 on the side corresponding to the passenger side of the motor vehicle 14.

[0026] To prevent a collision between the vehicle 14 and another road user 16 during at least assisted lateral steering, it is essential that the vehicle 14 can distinguish whether the road 10 has only a single lane 12, allowing the vehicle 14 to be driven in the center of the road 10, or whether the road 10 has multiple lanes 12, requiring the vehicle 14 to be kept safely in its designated lane to avoid a collision with another road user 16 traveling in the adjacent lane. For this purpose, the vehicle 14 includes an electronic computing unit that uses sensor data 32 and / or swarm data 18 to determine whether the first lane marking 28 corresponds to the adjacent lane or the designated lane.If it is determined that the first lane marking 28 is assigned to the ego lane, then the vehicle 14 can be safely guided in the center of road 10, since road 10 only has one lane 12. If the electronic computing device determines that the first lane marking 28 is assigned to the adjacent lane, then the vehicle 14 must be kept safely in the ego lane during assisted lateral steering to avoid a collision between the vehicle 14 and the other road user 16, who might be in the adjacent lane, since road 10 has at least the ego lane and the adjacent lane.

[0027] Future driver assistance systems can use swarm data 18 in addition to live lane detection. This swarm data 18 can encompass a wealth of highly current information and be stored in a cloud. Furthermore, this swarm data 18 can be based on data detected by current production vehicles, particularly via a front camera, and is stored cumulatively in the backend. In this way, numerous vehicles detect specific lane boundaries and upload these detected lanes to the cloud. There, the data is analyzed and made available to users, in this case, the vehicle 14. In addition to this swarm data 18, the vehicle 14 can primarily utilize live lanes detected by the camera system 22, which are represented by sensor data 32. Based on this, lateral guidance for the vehicle 14 can be provided.With live detection by the camera system 22, it can be determined, particularly using neural networks, whether detected lane markings are assigned to the ego lane or the adjacent lane. Specifically, a driver assistance system without swarm data 18 may require two lane markings 26 assigned to the ego lane, based on which the vehicle 14 can be kept in the center of this lane 12. If the first lane marking 28 assigned to the driver's side of the vehicle 14 is recognized as belonging to the adjacent lane, then lateral guidance of the vehicle 14 can be initiated, taking hysteresis into account. This means that an incorrectly detected or incorrectly assigned lane marking 26 can directly affect the driving behavior of the vehicle 14. This behavior can be significantly improved with the help of swarm data 18.

[0028] The motor vehicle 14 is equipped with a sensor device which may include a camera device 22 and / or radar and / or ultrasonic sensors, and which can additionally receive swarm data 18. Furthermore, the motor vehicle 14 is configured to be laterally steered with assistance from a control unit of the motor vehicle 14.

[0029] A problematic issue is the incorrect assignment of a neighboring lane marking. Furthermore, the system's driving behavior on narrow country roads could be incorrect or inconsistent, and therefore incomprehensible, depending on the accuracy of live lane detection at any given moment. In the current state of the art, live-detected lanes 12 are not overridden by swarm data 18. This means that live lane detection via sensor data 32, which controls the vehicle 14, is given higher priority in the algorithm than swarm data 18. To resolve these problems, in certain situations, information from the live detection by the camera system 22 should be overridden by swarm data 18. This allows the swarm data 18 to stabilize an erroneously detected lane detection by the camera system 22.

[0030] If the adjacent lane is recognized as such, the vehicle 14 is controlled correctly, whereby the vehicle 14 can be kept in its designated lane based on sensor data 32 and / or swarm data 18, and can be guided, for example, approximately 15 centimeters away from the second lane marking 30 assigned to the passenger side of the vehicle 14. If the adjacent lane is incorrectly recognized as its designated lane, the vehicle 14 may behave incorrectly, driving in the center of the lane 10.

[0031] In Fig. Figure 2 shows a process diagram for a method for guiding the motor vehicle 14 laterally. In a first process step V1a, the first lane marking 28 assigned to the driver's side of the motor vehicle 14 is assigned to a lane type based on the received swarm data 18. Furthermore, in the first process step V1a, the second lane marking 30 assigned to the passenger side of the motor vehicle 14 is assigned to a lane type based on the swarm data 18. Here, the lane type characterizes whether the respective lane marking 26 is assigned to the ego lane in which the motor vehicle 14 is to be guided, or to the adjacent lane.

[0032] In a further first procedural step V1b of the procedure, it is provided that, based on the sensor data 32 representing the environment of the motor vehicle 14, both the first lane marking 28 assigned to the driver's side of the motor vehicle 14 and the second lane marking 30 assigned to the passenger side of the motor vehicle 14 are each assigned to a lane type. In a second procedural step V2, a comparison of the respective determined lane types for the respective lane markings 26 is carried out. In a first step, it is checked whether the determined lane type assigned to the first lane marking 28, which was determined depending on the swarm data 18, corresponds to the lane type determined for the first lane marking 28, which was determined depending on the sensor data 32.In right-hand traffic, the first step checks whether the lane marking 28 on the left (first) lane, determined based on swarm data 18 and sensor data 32, including lane type and lane assignment, is identical. In a second step of the comparison, the lane type determined based on sensor data 32 and the lane type determined based on swarm data 18 are compared for the second lane marking 30 assigned to the passenger side of the vehicle 14. In right-hand traffic, the second step of the procedure checks whether the lane markings 30 on the right (second) lane, determined based on swarm data 18 and sensor data 32, including lane type and lane assignment, are identical.

[0033] If, during the comparison in the second process step V2, it is determined that identity exists in both the first and second steps of the process, then, in a third process step V3, the lane type assigned to the first lane marking 28 for the vehicle 14 is determined as the lane type calculated based on the sensor data 32, and the vehicle 14 is steered laterally based on the sensor data 32. If, during the comparison, it is determined that no identity exists in either the first or second step, then the vehicle 14 is also steered laterally in the third process step V3 based on the sensor data 32, whereby the lane type for the first lane marking 28 is determined based on the sensor data 32.If, during the comparison, it is determined that identity exists in the first step and no identity exists in the second step, then in the third procedure step V3 the motor vehicle 14 is steered laterally based on the sensor data 32, whereby the lane type assigned to the first lane marking 28 is determined as the lane type determined depending on the sensor data 32.

[0034] If the comparison reveals no identity in the first step, but identity exists in the second step, a fourth process step V4 follows, in which the lane type assigned to the first lane marking 28 is determined as the lane type based on the swarm data 18. Here, the vehicle 14 is laterally steered by the control unit, depending on the determined lane type of the first lane marking 28. Specifically, the vehicle 14 is laterally steered based on the sensor data 32. Alternatively, the vehicle 14 can be laterally steered based on the swarm data 18.

[0035] After the specified distance-time limits have elapsed, the motor vehicle 14 can be further controlled in the third process step V3, following the fourth process step V4. In the fourth process step V4, the lane type assigned to the first lane marking 28 can only be determined as the lane type based on the swarm data 18 if the swarm data 18 fulfills at least one specified quality criterion. Two quality criteria are specified here. One of the quality criteria is a required agreement rate between the characteristic values ​​determined from the swarm data 18 and the characteristic values ​​determined from the sensor data 32 for a defined section of the road. The second quality criterion is a required localization accuracy for the localization of the motor vehicle 14 on the road 10 based on the swarm data 18.

[0036] Distance-time limits can, in particular, specify a time interval and / or a distance to be covered and / or a speed range for the motor vehicle 14 and / or a curvature limit and / or a width range for the road 10 and / or a distance range for the distance of the motor vehicle 14 to the second lane marking 30.This means that, within the fourth process step V4, the lane type assigned to the first lane marking 28 is determined as the lane type calculated based on the swarm data 18, while a specified time interval is running and / or while the motor vehicle 14 travels the specified distance and / or while the motor vehicle 14 is moving at a speed within a specified speed range in the direction of travel and / or if the road 10 has a curvature above a specified curvature limit and / or if the road 10, which has the lane markings 26, has a width within a specified width range and / or if the motor vehicle 14 is within the specified distance range of the second lane marking 30.

[0037] In principle, good live data for lane detection, in this case the sensor data 32 provided by the camera system 22, are generally preferable to potentially outdated map data and thus to swarm data 18. In one method, the use of swarm data 18 for reconciling the live data is therefore restricted, while still solving the problem described.

[0038] In Fig. 3 is an excerpt of the procedure from Fig.Figure 2 is shown in more detail. This figure illustrates the case in which, based on swarm data 18, it was determined that the first lane marking 28 belongs to the adjacent lane, while based on sensor data 32, it was determined that the first lane marking 28 belongs to the ego lane, as determined during the comparison in the second process step V2. Subsequently, in the fourth process step V4, the lane type determined based on sensor data 32 is overridden by the swarm data 18, provided that the confidence level of the swarm data 18 is sufficiently high and thus the swarm data 18 meets at least one predefined quality criterion. A multidimensional characteristic map 34 allows for the definition of specific application parameters for the override of sensor data 32 by swarm data 18 to determine the lane type.Application parameters can thus include a maximum distance and / or a maximum time and / or a speed range and / or a minimum confidence level and / or a curvature range and / or a distance range. These application parameters determine whether the lane type determined from sensor data 32 is overridden by the lane type determined from swarm data 18. These application parameters can define the distance-time limits. Up to a situation 36 in which the distance-time limits are reached, or in particular exceeded, the vehicle 14 is laterally steered by the control unit, depending on the lane type of the first lane marking 28 determined from swarm data 18. The vehicle 14 can also be oriented using the second lane marking 30 during lateral steering.If situation 36 occurs, in which the distance-time limits are reached, then the motor vehicle 14 is subsequently laterally controlled in the third process step V3, in which the motor vehicle 14 is controlled on the basis of the sensor data 32, in particular depending on the lane type of the first lane marking 28 determined on the basis of the sensor data 32.

[0039] Overall, the invention shows how a method for stabilizing a live track detection of a front camera using swarm data 18 can be implemented. Reference symbol list 10 Street 12 lanes 14 Motor vehicle 16 road users 18 swarm data 20 Server setup 22 Camera setup 24 viewing area 26 Road marking 28 first lane marking 30 second lane marking 32 sensor data 34 Characteristic map 36 Situation V1a to V4 respective procedural steps

Claims

[1] Method for guiding a motor vehicle (14) across a road (10), in which - a lane marking (28) assigned to a driver's side of the motor vehicle (14) is assigned to a lane type (V1b) based on sensor data (32) representing an environment of the motor vehicle (14), wherein the lane type characterizes whether the lane marking (28) is assigned to an ego lane on which the motor vehicle (14) is to be driven, or to an adjacent lane near the ego lane, - the lane marking (28) assigned to the driver's side of the motor vehicle (14) is assigned to the lane type (V1a) based on swarm data (18) received from other vehicles, - the lane type assigned to the lane marking (28) is determined as the lane type determined based on the swarm data (18) if the lane type determined on the basis of the sensor data (32) and the lane type determined on the basis of the swarm data (18) differ (V4), and - the motor vehicle (14) is assisted by means of a control device depending on the specified lane type of the lane marking (28) assigned to the driver's side of the motor vehicle (14) by means of a control device, is steered laterally (V4). [2] Method according to claim 1, wherein the motor vehicle (14) is laterally controlled based on the sensor data (32). [3] Method according to claim 1, wherein the motor vehicle (14) is laterally controlled based on the swarm data (18). [4] Method according to one of the preceding claims, wherein the lane type assigned to the road marking (28) is determined as the lane type determined as a function of the sensor data (32) when, for a road marking (30) assigned to a passenger side of the motor vehicle (14), a lane type determined on the basis of the sensor data (32) and a lane type determined on the basis of the swarm data (18) differ. [5] Method according to one of the preceding claims, wherein the sensor data (32) comprise camera data, on the basis of which the lane marking (28) assigned to the driver's side of the motor vehicle (14) is analyzed by means of an image recognition method. [6] Method according to one of the preceding claims, wherein the lane type assigned to the road marking (28) is determined as the lane type determined depending on the swarm data (18) if the swarm data (18) meet at least one predetermined quality criterion. [7] Method according to claim 6, wherein a quality criterion is a correspondence rate of characteristic values ​​determined from the swarm data (18) with characteristic values ​​determined from the sensor data (32) for a defined section of the route and / or a localization quality of a localization of the motor vehicle (14) on the road (10) based on the swarm data (18) is specified as a quality criterion. [8] Method according to one of the preceding claims, wherein the lane type assigned to the road marking (28) is determined as the lane type determined depending on the swarm data (18). - while a predetermined time interval is running, and / or - while the motor vehicle (14) travels a predetermined distance, and / or - while the motor vehicle (14) is moving in the direction of travel at a speed within a specified speed range, and / or - if the road (10) has a curvature above a specified curvature limit, and / or - if the road (10) bearing the road marking (28) has a width within a specified width range, and / or - if the motor vehicle (14) is within a specified distance range from a further lane marking (30) assigned to a passenger side of the motor vehicle. [9] Method according to one of the preceding claims, wherein the lane type assigned to the driver's side of the motor vehicle (14) is determined as the lane type determined based on the sensor data (32) if the lane type determined on the basis of the sensor data (32) and the lane type determined on the basis of the swarm data (18) are identical. [10] Motor vehicle (14) which is equipped to carry out a method according to any of the preceding claims.

Images