Method and apparatus for vehicle control

Abstract

The invention relates to a method (200) for controlling a vehicle, wherein the method (200) comprises the following steps: reading in (205) a camera signal (125) representing an optically sensed image of a lane to be traveled by a vehicle (100); performing (210) semantic segmentation on the image represented by the camera signal (125) in order to identify a free area in front of the vehicle (100) as a drivable lane section (305) from the image, and providing a lane signal (135) representing the identified drivable lane section (305); and determining (215) a control signal (140) for actuating at least one vehicle component (145) using the lane signal (135). The invention also relates to a device and a machine-readable storage medium.

Description

Technical Field

[0001] The present invention is based on a method and a device for vehicle control. The present invention also provides a computer program. Existing technology

[0002] Vehicles can include a single-camera sensor system for automated or partially automated driving systems, such as lane keeping assist or lane following. Object recognition can be used to identify other vehicles and lane markings, for example. Object recognition can also depend on the vehicle's surroundings, such as heavy rain or backlighting. Summary of the Invention

[0003] Against this background, the present invention provides a method for controlling a vehicle, a device using the method, and finally a corresponding computer program. The measures listed in the preferred embodiments enable advantageous developments and improvements of the present invention.

[0004] The proposed approach is based on the recognition that vehicle control based on the evaluation of optically sensed images can be improved by evaluating the images using semantic segmentation. Advantageously, semantic segmentation analysis allows for the determination of regions consisting of content-related image points from the image in order to identify drivable lane sections. Advantageously, this can be achieved quickly and reliably even under adverse environmental conditions, as conventional object recognition, which can be limited by environmental conditions such as heavy rain or backlighting, can be omitted. Based on the lane sections identified using semantic segmentation, control signals can be determined to guide the vehicle along the drivable lane sections.

[0005] A method for vehicle control is proposed. The method includes at least one step of reading in a camera signal, performing semantic segmentation, and determining a control signal. The camera signal read in the reading step represents an optically sensed image of the lane the vehicle is to travel. In the performing step, semantic segmentation is performed on the image represented by the camera signal to identify a free area in front of the vehicle as a drivable lane section from the image. Furthermore, in the performing step, a lane signal is provided. This lane signal represents the result of the semantic segmentation, i.e., the identified drivable lane section. In the determining step, a control signal is determined using the lane signal to actuate at least one vehicle component.

[0006] The vehicle can be a motor vehicle, a truck, a bus, a motorcycle, or a ground vehicle. The vehicle can have partially automated or fully automated driving operations. For example, the camera signal can be provided by the vehicle's optical environment sensing device. When implementing semantic segmentation, content-related areas can be formed from a collection of adjacent image points, such as pixels, so that the drivable area of ​​the lane can be divided into closed areas and the closed areas are provided as identified drivable lane sections. In this case, the area is an idle area because there are no other objects that could potentially endanger the safe driving of the vehicle, such as other vehicles or other objects located in the lane. The control signal for operating at least one vehicle component can be provided, for example, to a controller or vehicle component of the vehicle. For example, the vehicle component is a driving assistance system, such as lane keeping assistance or distance recognition, or a braking device, an accelerator, or a steering device. For example, the control signal is configured to change the settings of a vehicle component or to activate or deactivate a driving system.

[0007] According to one specific embodiment, the control signal is additionally determined in the ascertainment step using a marking signal. The marking signal represents the recognized lane markings. The recognized lane markings are based, for example, on an evaluation of the camera signal using conventional object recognition. Advantageously, the information about the recognized lane markings can be combined with the information about the drivable lane section identified using semantic segmentation to control vehicle components.

[0008] According to one embodiment, the control signal is additionally determined in the ascertainment step using a vehicle identification signal. The vehicle identification signal represents the identified preceding vehicle. Advantageously, this allows objects located in the optical image area that limit the identified drivable lane section, for example in length or width, to be identified as preceding vehicles, for example to implement a follow-up function.

[0009] According to one embodiment, the semantic segmentation step can be repeated to identify changes in the extension of the drivable lane segment. For example, the lengths of at least two identified drivable lane segments can be compared continuously or at specific time intervals, and additionally or alternatively, the widths can be compared to identify changes that occur, such as sudden changes, such as a sudden shortening of the lane segment that is dependent on the vehicle's speed and the preceding vehicle, when another vehicle enters the lane or when a preceding vehicle exits. The identified change in the extension of the identified drivable lane segment can be provided in the implementation step as an extension signal. In the determination step, the control signal can be determined using the extension signal.

[0010] According to one embodiment, in the determining step, a control signal may be determined for controlling the lateral guidance of the vehicle within the identified drivable lane section. For example, the control signal may be configured to adjust the orientation of the vehicle within the identified drivable lane section, for example by controlling the vehicle's steering system, when traveling through the identified drivable lane section.

[0011] Furthermore, according to one embodiment, in the ascertainment step, a control signal is ascertained for controlling the longitudinal guidance of the vehicle relative to the identified end of the identified drivable lane section. This allows for advantageous use of the control signal, for example, for a driver assistance system for automatically maintaining distance or for cruise control. The control signal can be designed, for example, to control the speed or acceleration of the vehicle.

[0012] Furthermore, according to one embodiment, the control signal can be determined in the determination step based on a threshold value. For example, the threshold value can represent a minimum width or length required for the vehicle to safely travel within the identified drivable lane section. To this end, the threshold value can be preset or read in, so that the threshold value is determined, for example, based on the vehicle type, the nature of the surface to be traveled on, the vehicle's surrounding conditions, or the vehicle's speed.

[0013] According to one specific embodiment, in the ascertainment step, a warning signal can also be ascertained based on a threshold value to output an acoustic and, in addition or as an alternative, visual warning to the driver. For example, a warning signal can be provided if a shortening of the identified drivable lane section occurs, for example, if another vehicle suddenly enters the identified drivable lane section from an adjacent lane.

[0014] According to one embodiment, in the determination step, control signals can also be determined using the read-in driving parameters in order to adjust the driving parameters. The read-in driving parameters can represent, for example, the vehicle speed or the steering wheel position. The determined control signals can be designed to control a change in the adjusted driving parameters, for example, also to control the adjustment of the longitudinal or transverse guidance of the vehicle as a function of the read-in driving parameters.

[0015] According to one embodiment, the method may further include the step of optically sensing the lane the vehicle is traveling through using a monocular camera sensor device to provide a camera signal. The monocular camera sensor device may, for example, comprise a monocular camera system. To sense the lane the vehicle is traveling through, the monocular camera sensor device may, for example, have a sensing area oriented in the direction of travel.

[0016] For example, the method can be implemented in the form of software or hardware or in the form of a mixture of software and hardware, for example in a controller.

[0017] The present invention also proposes a device, which is configured to implement, control or realize the steps of the variant of the method proposed here in a corresponding device. By this embodiment variant of the present invention in the form of a device, the task on which the present invention is based can also be solved quickly and effectively.

[0018] To this end, the device can include at least one computing unit for processing signals or data, at least one memory unit for storing signals or data, at least one interface to a sensor for reading in sensor signals from the sensor or an interface to an actuator for outputting data or control signals to the actuator, and / or at least one communication interface for reading in or outputting data embedded in a communication protocol. The computing unit can be, for example, a signal processor, a microcontroller, or the like, while the memory unit can be a flash memory, an EEPROM, or a magnetic memory unit. The communication interface can be designed for wireless and / or wired data reading in or outputting data, wherein a communication interface capable of reading in or outputting wired data can, for example, electrically or optically read in this data from a corresponding data transmission line or output this data to a corresponding data transmission line.

[0019] In the present context, a device is understood to be an electrical device that processes sensor signals and outputs control and / or data signals accordingly. The device may have an interface, which can be implemented in hardware and / or software. In the case of a hardware implementation, the interface may, for example, be part of a so-called system ASIC that contains the various functions of the device. However, it is also possible for the interface to be its own integrated circuit or to be composed at least partially of discrete components. In the case of a software implementation, the interface may be, for example, a software module that exists alongside other software modules on a microcontroller.

[0020] Advantageously, a computer program product or a computer program with a program code is also proposed, which program code can be stored on a machine-readable carrier or storage medium, such as a semiconductor memory, a hard disk memory or an optical memory, and is used to implement, realize and / or control the steps of the method according to any of the aforementioned embodiments, in particular when the program product or program is executed on a computer or device. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] An exemplary embodiment of the present invention is illustrated in the drawings and explained in more detail in the following description.

[0022] Figure 1 A schematic diagram of a vehicle having an apparatus for vehicle control according to one embodiment;

[0023] Figure 2 A flow chart of a method for vehicle control according to one embodiment;

[0024] Figure 3 Illustration of drivable lane segments identified using semantic segmentation;

[0025] Figure 4 Illustration of drivable lane segments identified using semantic segmentation;

[0026] Figure 5 Illustration of drivable lane segments identified using semantic segmentation;

[0027] Figure 6 A graphic representation of the drivable lane segments identified using semantic segmentation; and

[0028] Figure 7 Signal flow diagram of a method for vehicle control according to one embodiment. DETAILED DESCRIPTION

[0029] In the following description of advantageous exemplary embodiments of the present invention, identical or similar reference numerals are used for similarly acting elements that are shown in the various figures, wherein a repeated description of these elements is omitted.

[0030] Figure 1 According to one exemplary embodiment, a schematic diagram of a vehicle 100 is shown having a device 105 for vehicle control. Device 105 includes a read-in interface 110, an implementation device 115, and an ascertainment device 120. Device 105 is configured to read in a camera signal 125 via read-in interface 110. Camera signal 125 represents an optically captured image of a lane that vehicle 100 is traveling through. Camera signal 125 is provided, for example, by a monocular camera sensor device 130 of vehicle 100. Implementation device 115 is configured to perform semantic segmentation on the image represented by camera signal 125 in order to identify a free area in front of vehicle 100 as a drivable lane section from the image. Implementation device 115 is also configured to provide a lane signal 135 representing the identified drivable lane section. Lane signal 135 is provided to ascertainment device 120. Ascertainment device 120 is configured to use lane signal 135 to ascertain a control signal 140 for actuating at least one vehicle component 145 of vehicle 100. Vehicle components 145 of the vehicle are, for example, control units, steering systems, brake systems, or accelerators of vehicle 100. Thus, using control signal 140, for example, the driving speed of vehicle 100 in the identified lane section to be driven and, in addition or as an alternative, the driving direction can be set.

[0031] According to the exemplary embodiment shown here, ascertainment device 120 is designed to ascertain control signal 135 while additionally using marking signal 150. Marking signal 150 represents the recognized lane markings. Marking signal 150 is provided, for example, by lane marking recognition device 155 of vehicle 100. The recognized lane markings are, for example, markings recognized by conventional object recognition using camera signal 125. Advantageously, lane recognition, for example, based on object recognition, can thus be used in addition to recognizing lane sections using semantic segmentation.

[0032] According to the exemplary embodiment shown here, ascertainment device 120 is further configured to ascertain control signal 140 using a vehicle identification signal 160. Vehicle identification signal 160 represents a detected preceding vehicle. Vehicle identification signal 160 is provided, for example, by a detection device 165 of vehicle 100. Detection device 165 may include, for example, a lidar sensor device or a radar sensor device for detecting preceding vehicles, or may be configured to detect preceding vehicles using camera signal 125 and object detection.

[0033] According to one exemplary embodiment, implementation device 115 is configured to repeatedly perform semantic segmentation in order to detect changes in the extension of the drivable lane segment. In this case, implementation device 115 is also configured to provide an extension signal 170 that represents the detected change in the extension of the identified drivable lane segment. Ascertainment device 120 is configured to determine control signal 140 using extension signal 170. Advantageously, this allows for rapid and reliable detection of the entry or exit of a preceding vehicle in order to detect changes in the drivable lane segment.

[0034] Furthermore, according to one exemplary embodiment, ascertainment device 120 is configured to ascertain a control signal 140 for controlling the lateral guidance of vehicle 100 within the identified drivable lane section. Lane signal 135 includes, for example, information about the length and width of the identified drivable lane section. To control the lateral guidance of vehicle 100, control signal 140 is configured, for example, to control the setting of a steering angle of a steering system of vehicle 100.

[0035] Furthermore, according to one exemplary embodiment, ascertainment device 120 is further configured to ascertain a control signal 140 for controlling the longitudinal guidance of the vehicle with respect to the identified end of the identified drivable lane section. To control the longitudinal guidance of vehicle 100, control signal 140 is configured, for example, to control the setting of the speed or acceleration of vehicle 10. This is advantageous so that, using control signal 140, an assistance function of vehicle 100 can be implemented, for example, to automatically maintain a distance to a preceding vehicle.

[0036] According to one exemplary embodiment, ascertainment device 120 is also configured to ascertain control signal 140 as a function of a threshold value for the minimum width or length of the identified drivable lane section. This threshold value is, for example, predefined or read in via read-in interface 110, for example, in order to use the threshold value as a function of a detected speed limit on the lane section to be driven.

[0037] According to the exemplary embodiment shown here, ascertainment device 120 is also configured to additionally provide a warning signal 175 as a function of a threshold value for outputting an acoustic and, in addition or alternatively, a visually perceptible warning to the driver. Warning signal 175 is provided, for example, to an output device 180 of vehicle 100.

[0038] In order to adjust at least one driving parameter, such as the speed of vehicle 100 or the driving direction or trajectory of vehicle 100, according to the exemplary embodiment shown here, ascertainment device 120 is designed to ascertain control signal 140 using read-in driving parameter 185 in order to adjust the driving parameter. For this purpose, at least one adjusted driving parameter 185 is read in, for example, in the form of a driving parameter signal via read-in interface 110, and control signal 140 is ascertained using this driving parameter signal in order to control a change in driving parameter 185 and, in addition or as an alternative, a change in another driving parameter.

[0039] Camera signal 125 is provided here by monocular camera sensor device 130 of vehicle 100 . Alternatively, device 100 may also include a sensing device, for example a monocular optical sensing device, which is designed to optically sense the lane to be traveled by vehicle 100 in order to provide camera signal 125 .

[0040] Figure 2 According to one embodiment, a flow chart of a method 200 for vehicle control is shown. The method 200 can be implemented, for example, using one embodiment of the above-described device for vehicle control. The method 200 includes at least one reading step 205, an implementation step 210, and a determination step 215. In the reading step 205, a camera signal is read in, which represents an optically sensed image of the lane through which the vehicle is to travel. In the implementation step 210, semantic segmentation is performed on the image represented by the camera signal in order to identify a free area in front of the vehicle as a drivable lane section from the image. In addition, in the implementation step 210, a lane signal is provided, which represents the identified drivable lane section. In the determination step 215, a control signal is determined using the lane signal to operate at least one vehicle component.

[0041] According to the embodiment shown here, method 200 further includes an optional optical sensing step 220. In optical sensing step 220, the lane to be traveled by the vehicle is optically sensed using a monocular camera sensor device to provide a camera signal. Step 220 is optionally performed before step 205.

[0042] Figure 3 A diagram shows a drivable lane section 305 identified using semantic segmentation. This shows lane section 305 identified by semantic segmentation for the vehicle's own lane in a rainy weather sequence, with poor visibility of the lane markings. In the driving situation illustrated here, it is advantageous to use an embodiment of a device for vehicle control to supplement or replace situation-dependent lateral guidance of a vehicle that encounters a system boundary when lane markings are poorly visible. The approach described with reference to the preceding figures advantageously enables the identification of the own lane identified using semantic segmentation as drivable lane section 305 to improve the longitudinal and lateral guidance of the vehicle.

[0043] Using semantic segmentation to identify drivable lane sections 305 and to control vehicle components based on the identified drivable lane sections 305 is particularly advantageous when a vehicle utilizes a monocular camera-based sensor solution for vehicle control. For example, this allows for incorporating machine learning-based identification of drivable lane sections 305, i.e., the "holistic egolane," into the control of a (partially) automated driving system. For example, in situations where the road surface is glistening due to backlighting or rain, conventional object recognition methods for identifying lane markings and vehicles cannot be used, or can only be used to a limited extent, with a single-sensor system, such as a single-video-based lane keeping assistance system. In such situations, it is advantageous to use semantic segmentation in addition to or as an alternative to conventional object recognition to provide automated assistance to the vehicle driver even in such challenging situations.

[0044] For example, semantic segmentation as a machine learning method can be used as a complementary approach to conventional model-assisted methods, for example for the purpose of increasing robustness. During semantic segmentation, the drivable surface or lane occupied by the vehicle is segmented into closed areas in the pixel image in order to detect these areas as drivable lane segments 305.

[0045] In particular in the case of backlighting and surface reflections, the use of semantic segmentation is advantageous, since such a network also always utilizes the entire scene content for the perception task. Therefore, information about the vehicle traveling in front is always utilized when segmenting the own lane, rather than just pure lane markings as in conventional methods. The longitudinal and lateral guidance of the vehicle can be controlled based on the drivable lane section 305 identified in this way. Therefore, conventional lane marking recognition and conventional vehicle recognition may be limited in the case of strong backlighting and lane reflections, even though in this case, for example, the edges of the drivable lane section 305 identified via semantic segmentation can be used by means of the overall content of the scene (including the driver in front) in order to thereby achieve lateral adjustment of the vehicle. Under favorable conditions, it is advantageous to use the semantic segmentation method based on machine learning proposed here at least complementary, or it is advantageous to use conventional object recognition and semantic segmentation in a fusion, as described below with reference to. Figure 7 As described.

[0046] Figure 4 A diagram shows a drivable lane section 305 identified using semantic segmentation. An exemplary scenario is shown in which semantic segmentation is advantageous for identifying drivable lane section 305, for example, in order to control or assist in vehicle following guidance using identified drivable lane section 305. In the illustrated scenario, reflections may occur at vehicle edges in strong sunlight, which may result in large intensity differences. In principle, intensity differences are typical features used in conventional methods for object classification. However, reflections may cause irregularities in the position and intensity of these features and thus lead to features that differ from object features under normal lighting. Typically, intensity normalization of features is also used, which may cause typical object features to be more likely to be suppressed in strong sunlight, as a result of which object plausibility checks may be delayed or impossible. Therefore, in the scenario shown here, the vehicle control method presented here is advantageous when using drivable lane section 305 identified by means of semantic segmentation: If drivable lane section 305 ends abruptly very close to the vehicle and no preceding driver 405 is identified, it is assumed that the vehicle, i.e., preceding driver 405, is at the end of the identified drivable lane section 305. In this case, it is advantageous to adjust the longitudinal control of the vehicle to the end of the own lane, i.e., the end of the identified drivable lane section 305.

[0047] Figure 5A diagram shows a drivable lane section 305 identified using semantic segmentation. This diagram illustrates an exemplary scenario in which semantic segmentation is advantageous for identifying drivable lane section 305, in this case a situation with rain and an object, a preceding vehicle 405, about to enter the vehicle's lane. Advantageously, the method for implementing semantic segmentation proposed herein allows for reliable and early identification of objects, such as, in this case, an approaching preceding vehicle 405. If the identified drivable lane section 305 suddenly shortens, an approach can be assumed, and the automatic longitudinal control can be adapted using control signals. The same method can also be used for early identification of exiting maneuvers. In this case, the vehicle's automatic longitudinal control can be activated to accelerate early, resulting in increased dynamic perceptibility and perceived by the driver as human-like behavior.

[0048] Furthermore, it is possible to use machine learning-based changes or deformations of the lane, i.e., the identified drivable lane section 305, for anomaly detection, for example in the case of a vehicle protruding into the lane, which vehicle could not be identified as an object relevant for longitudinal adjustment using conventional methods; or in the case of an object in the area of ​​the identified drivable lane section 305, for example, in the case of lost cargo on the lane.

[0049] In these situations, driver warnings can be requested using control signals, and automatic longitudinal or lateral interventions can be performed in addition or as an alternative. Even in vehicles with multiple sensors, such as those with a multi-sensor assistance system, the vehicle control method presented here, which uses drivable lane segments 305 identified by semantic segmentation, is advantageous, for example as an additional safeguard or in the event of damage or failure of other sensor systems, such as radar sensor systems and / or lidar sensor systems, or as a so-called "fallback solution." In this case, for example, driving modes with a limited range of functions (e.g., no lane changes, at a limited speed) can also be controlled or assisted. Furthermore, in highly automated systems, a period of time can be allowed until the driver can resume driving after detecting a sensor failure.

[0050] Figure 6 A diagram showing the drivable lane sections identified using semantic segmentation is shown. Figure 5 A subsequent scenario of the described application situation: preceding vehicle 405 is shown here while entering the area previously identified as drivable lane section 305. Due to the entry of preceding vehicle 405, lane section 305 identified as drivable is shortened, and the object is identified as preceding vehicle 405.

[0051] Figure 7A signal flow diagram of a method for vehicle control according to one embodiment is shown. The diagram shows an exemplary incorporation of this path based on machine learning, i.e., an exemplary application of the method presented herein for identifying drivable lane sections using semantic segmentation and controlling the vehicle based on the identified drivable lane sections.

[0052] In the signal flow diagram shown here, conventional perception methods for lateral guidance, namely model-based lane marking recognition, and conventional perception methods for longitudinal guidance, namely object classification using pattern recognition, such as Viola Jones or "support vector machines" (SVM), are supplemented by a largely complementary, machine learning-based approach, namely semantic segmentation of the own lane.

[0053] In the exemplary embodiment shown here, camera signal 125 is provided by a monocular video sensor, namely, a monocular camera sensor device 130. Camera signal 125 is provided, by way of example, to a lane marking recognition device 155, an implementation device 115, a detection device 165, and a checking device 705. Lane marking recognition device 155 performs conventional lane marking recognition and provides a marking signal 150 to an ascertainment device 120. Implementation device 115 performs semantic segmentation of the lane and provides a lane signal 135. Detection device 165 performs conventional vehicle recognition and provides a vehicle recognition signal 160. Checking device 705 performs a feasibility check of a conventional method and provides a modulation signal 710. Ascertainment device 120 includes a lateral guidance data fusion device 715 and a longitudinal guidance data fusion device 720. Modulation signal 710 is provided to lateral guidance data fusion device 715 and longitudinal guidance data fusion device 720 to produce a modulating effect on the data fusion. Furthermore, marking signal 150 and lane signal 135 are provided to lateral guidance data fusion device 715. According to the exemplary embodiment shown here, lateral guidance data fusion device 715 is designed to determine control signal 140 for lateral guidance using marking signal 150, lane signal 135, and modulation signal 710, and to provide control signal 140 for lateral guidance to a vehicle controller as vehicle component 145. Vehicle identification signal 160 and lane signal 135 are provided to longitudinal guidance data fusion device 720. According to the exemplary embodiment shown here, longitudinal guidance data fusion device 720 is designed to determine control signal 140 for longitudinal guidance using vehicle identification signal 160, lane signal 135, and modulation signal 710, and to provide control signal 140 for longitudinal guidance to a vehicle controller as vehicle component 145.

[0054] For example, it is possible to rely more heavily on machine learning methods in data fusion in scenarios that are challenging for conventional methods. In the best case, conventional methods are preferred in data fusion if they can already describe the driver's expectations well.

[0055] If an embodiment includes an “and / or” connection between a first feature and a second feature, this should be interpreted as meaning that the embodiment has both the first and the second feature according to one embodiment and either only the first or only the second feature according to another embodiment.

Claims

1. A method (200) for vehicle control, wherein: The method (200) comprises the following steps: - reading in a camera signal (125) representing an optically sensed image of the lane to be traveled by the vehicle (100); - performing semantic segmentation on an image represented by the camera signal (125) in order to identify a free area in front of the vehicle (100) as a drivable lane section (305) from the image and to provide a lane signal (135) representing the identified drivable lane section (305), wherein the image is a pixel image and wherein, when performing semantic segmentation, content-related regions are formed from a collection of adjacent image points in order to segment the drivable area of ​​a lane or a lane used by the vehicle (100) into closed regions in the pixel image and to provide the closed regions as the identified drivable lane section (305); and - using the lane signal (135), determining a control signal (140) for actuating at least one vehicle component (145), The step of performing semantic segmentation is repeatedly performed to identify a change in the extent and / or width of the drivable lane section (305), and an extension signal (170) is provided, which represents the identified change in the extent and / or width of the identified drivable lane section (305), wherein in the step of determining, the control signal (140) is determined using the extension signal (170).

2. The method (200) according to claim 1, wherein: In the ascertaining step, the control signal (140) is additionally ascertained using a marking signal (150), wherein the marking signal (150) represents the recognized lane marking.

3. The method (200) according to claim 1 or 2, wherein: In the ascertaining step, the control signal (140) is additionally ascertained using a vehicle identification signal (160), wherein the vehicle identification signal (160) represents the identified preceding vehicle (405).

4. The method (200) according to claim 1 or 2, wherein: In the ascertaining step, the control signal (140) is ascertained for controlling the lateral guidance of the vehicle (100) within the identified drivable lane section (305).

5. The method (200) according to claim 1 or 2, wherein: In the ascertaining step, the control signal (140) is ascertained for controlling the longitudinal guidance of the vehicle (100) relative to the identified end of the identified drivable lane section (305).

6. The method (200) according to claim 1 or 2, wherein: In the ascertaining step, the control signal (140) is ascertained as a function of a threshold value with respect to a minimum width or length of the identified drivable lane section (305).

7. The method (200) of claim 6, wherein: In the ascertaining step, a warning signal (175) is additionally ascertained as a function of the threshold value for outputting an acoustically and / or visually perceptible warning to the driver.

8. The method (200) according to any one of claims 1, 2 and 7, wherein: In the ascertaining step, the control signal (140) is ascertained using the read-in driving parameters (185) in order to adjust the driving parameters.

9. The method (200) according to any one of claims 1, 2 and 7, wherein: The invention comprises a step of optically sensing a lane to be traveled by the vehicle (100) using a monocular camera sensor device (130) in order to provide the camera signal (125).

10. A device (105) configured to carry out and / or control the steps of the method (200) according to any one of claims 1 to 9 in corresponding units (110, 115, 120).

11. A computer program product comprising a computer program, which is configured to carry out and / or control the steps of the method (200) according to any one of claims 1 to 9.

12. A machine-readable storage medium having a computer program stored thereon, the computer program being configured to execute and / or control the steps of the method (200) according to any one of claims 1 to 9.

Images