Method for training a subsequent journey trajectory on the basis of route sections driven through manually and at least semi-autonomously, and electronic systems

Abstract

The invention relates to a method for training a subsequent journey trajectory (19) for a vehicle (9), comprising the following steps: - driving, with the vehicle (9), through a first route section (17) with a first degree of autonomy and recording the first route section (17); - driving, with the vehicle (9), through a second route section (18), following on from the first route section (17) with a second degree of autonomy which differs from the first degree of autonomy and recording the second route section (18); - determining the subsequent driving trajectory (19) from the recorded first route section (17) and the recorded second route section (18), wherein the subsequent driving trajectory (19) is created for the at least semi-autonomous driving of a subsequent journey. The invention also relates to methods and systems (12, 13).

Description

[0001] Method for training a tracking trajectory based on manually driven and at least semi-autonomously driven route sections, as well as electronic systems

[0002] One aspect of the invention relates to a method for training a tracking trajectory for a vehicle. Another aspect of the invention relates to a method for performing a driving maneuver with a vehicle. Yet another aspect relates to an electronic analysis system. Furthermore, one aspect of the invention relates to an electronic vehicle guidance system. Yet another independent aspect of the invention relates to a computer-implemented method for generating a tracking trajectory for a vehicle.

[0003] It is known from the prior art that for specific driving maneuvers, such as shunting maneuvers, follow-up trajectories are trained which then form the basis for the shunting maneuver and are then followed. Such a shunting maneuver can, for example, be driving through a specific area where, for example, parking in a predetermined parking zone then takes place. This is known, for example, from EP 3 401 190 B1. There, manual driving maneuvers are carried out from a start position to a target position, which are intended to serve as the basis for a follow-up trajectory to be trained. In the scenario there, these manually driven route sections are carried out in a specific area, for example on private property. The vehicle is thus first moved from, for example, a property boundary on the property towards a parking zone and is then finally parked in the parking zone.

[0004] In these systems, the training of the tracking trajectory with regard to the starting position and the target position is defined by the user. This training is thus assigned to very specific situations. This limits the flexibility with regard to generating a tracking trajectory, as well as the recognition of specific situations that make generating a tracking trajectory possible or seem sensible.

[0005] The object of the present invention is to provide methods and systems with which tracking trajectories can be designed and provided more variably and more flexibly with regard to specific situations. This object is achieved by methods and systems according to the independent claims.

[0006] One aspect of the invention relates to a method for training a tracking trajectory for a vehicle, in particular comprising the following steps:

[0007] Driving through a first section of the route with a first degree of autonomy with the vehicle and recording the first section of the route;

[0008] Driving the vehicle through a second route section following the first route section with a second level of autonomy different from the first level of autonomy and recording the second route section;

[0009] Determining the follow-up trajectory from the recorded first route section and the recorded second route section, wherein the follow-up trajectory is generated for at least semi-autonomous traversing of a follow-up journey.

[0010] Such a method makes it possible to train different tracking trajectories in a more diverse and flexible manner. In particular, it allows for a wider range of situations suitable for driving with a tracking trajectory, especially since they are repeated several times. Precisely because a tracking trajectory can be formed from route sections traversed with varying degrees of vehicle autonomy, longer or more complex tracking trajectories can also be generated. This, in turn, allows highly customized environments to be traversed with such a tracking trajectory. This significantly expands the possibilities for navigating a wide variety of environments with individual, trained tracking trajectories.It is particularly advantageous that the tracking trajectory is generated and configured on the basis of the information from the recorded route sections in such a way that it can be driven through at least semi-autonomously, in particular fully autonomously. This is particularly advantageous because it also allows route sections that are, for example, manually driven by the vehicle and used to train the tracking trajectory to be driven through with a higher degree of autonomy, namely at least semi-autonomously, during the tracking trajectory. The composition of a tracking trajectory from route sections that are driven through with different degrees of autonomy significantly increases the diversity of the generation of the tracking trajectory. Last but not least, it is particularly advantageous to change, in particular to align, the degree of autonomy of route sections driven through with different degrees of autonomy during the tracking journey.This allows the tracking journey, based on the tracking trajectory, to be completed with a uniform degree of autonomy, namely at least completely semi-autonomously, over the entire route. For example, the first section of the route is completed manually, and the second section is completed with the assistance of a parking assistant.

[0011] In particular, this also makes it possible to expand environmental situations that make subsequent repeated driving based on a tracking trajectory seem sensible, or that are specifically enabled by the proposed method. This not only allows a user to specify the route sections that are considered for the tracking trajectory, but also, in one exemplary embodiment, allows for increased flexibility with regard to deciding which route sections are used for a tracking trajectory.This method now also makes it possible, based on the combination of route sections traveled with varying degrees of autonomy, to decide, depending on the situation, in particular based on the recognized environmental situation, whether it is sensible and / or suitable to continue to travel through this environment in the future and upon repeated use based on an at least semi-autonomously traversable tracking trajectory. This recognition scenario, which is also possible during training, can also be implemented in such a way that the route sections are recorded in a blanket manner and without any prior triggering criteria, with the relevant information then being provided anyway. This means that the decision to train a tracking trajectory can only be made once recording of at least one route section has already begun.Thus, in one embodiment of the method, it is possible to decide in an intelligent manner, even after the start of recording at least one route section, whether a follow-up trajectory should be trained on the basis of these route sections or whether a follow-up trajectory that can be driven semi-autonomously should be trained or generated on the basis of these route sections.

[0012] In one embodiment, a third route section is traversed and recorded with the first degree of autonomy, wherein the follow-up trajectory is determined from the recorded first route section, the recorded second route section, and the recorded third route section, and wherein the follow-up trajectory is generated for at least semi-autonomous traversal of a follow-up journey. For example, the first route section is traversed with the aid of a rear parking assistant, the second route section is traversed manually, and the third route section is traversed with the aid of a parking assistant.

[0013] In one embodiment, the route length and / or the course of the tracking trajectory are determined based on the summed two route lengths. This is a further advantageous embodiment, since this combination of the two route sections allows for the determination of a coherent tracking trajectory, which may also be relatively long and / or have a very complex course. Nevertheless, it allows the entire tracking trajectory to be traversed at least semi-autonomously.

[0014] In one embodiment, the follow-up trajectory is determined such that its route length and its course correspond to the route length and / or the course of the summed two route lengths. This ensures that the available information about the route sections is utilized to the maximum and a particularly advantageous follow-up trajectory can be generated.

[0015] In one embodiment, a section of road is traversed by a driver manually driving the vehicle as an example of a degree of autonomy. In this regard, the degree of autonomy can thus be set to zero. This means that steering and / or acceleration and braking are carried out by the driver. Such a specification allows, on the one hand, specific route lengths and / or courses of the route section to be specified in different environmental conditions and based on the individual wishes of a driver, which are to be taken into account in the tracking trajectory. This allows the individual wishes of a driver to be met in a particularly advantageous manner. On the other hand, it also makes it possible for a recording of such a manually traversed section of road to be carried out individually, in particular to be started.Thus, in one embodiment, it is possible for the driver or another vehicle occupant to initiate recording of such a manually driven section of the route. However, it is also possible for an electronic vehicle guidance system, particularly based on acquired environmental information, to detect an environment that is assessed as suitable or appropriate for subsequent repeated driving based on a follow-up trajectory. It is then possible for the vehicle guidance system itself to automatically initiate recording of this manually driven section of the route. Thus, recording can then also begin automatically, and the information recorded can be used for determining, in particular subsequently, the follow-up trajectory.This allows for a particularly advantageous extension of the training scenario, especially when a section of road is driven through manually. This is because, in such maneuvers, the driver is usually the highest authority, who is therefore also responsible for driving the vehicle safely and must have a comprehensive view of the surroundings. Since, in such situations, the system does not actually know whether the driver is deliberately driving through this section of road with the intention of using it for a follow-up trajectory, the above-mentioned scenario is advantageous. This allows the manually driven section of road to be recorded as a precautionary measure. This creates an information base or database that can also be subsequently made available and used to determine a follow-up trajectory.

[0016] In one embodiment, a route section is traversed at least semi-autonomously using a vehicle guidance system of the vehicle as an example of a degree of autonomy. This now also makes it possible to consider semi-autonomous or even fully autonomous journeys along a route section for generating and training a tracking trajectory for a tracking journey. For this purpose, the steering and / or acceleration and / or braking of the vehicle is taken over by the electronic vehicle guidance system. This now also makes it possible to incorporate route sections for training a tracking trajectory that are at least partially decoupled from a vehicle driver. In particular, it is therefore also possible to design a tracking trajectory for more diverse and individual environmental scenarios. This is because such a route section traversed at least semi-autonomously can then also be used as a basis, as needed.In particular, it is therefore also advantageous if an environmental scenario exists in which such an at least semi-autonomously traversed route section is more suitable or is desired by a vehicle driver, is recorded and taken into account for the follow-up trajectory. Last but not least, this also makes it possible to combine a follow-up trajectory in a variety of ways using at least two such route sections traversed with different degrees of autonomy. Thus, for example, in one exemplary embodiment, the first route section along the route can be manual or at least semi-autonomous. The second route section following the first route section along the route, in particular immediately adjacent to it, can then be traversed with a different degree of autonomy in comparison.Depending on the first route section, the second route section can then also be traveled manually or at least semi-autonomously, for example. Thus, it is possible for the first route section to be traveled manually, for example, and the second route section to be traveled at least semi-autonomously, in particular fully autonomously. In another exemplary embodiment, the first route section can be traveled semi-autonomously, and the second route section can be traveled manually or fully autonomously, for example. If the first route section is traveled fully autonomously, the second route section can be traveled manually or semi-autonomously, for example.

[0017] The proposed method, which involves driving through at least two route sections with different levels of autonomy, makes it possible, especially in such constellations, to generate a tracking trajectory that exhibits a seamless and thus continuous trajectory progression without any jumps or offsets. A very precise transition between the route sections driven with different levels of autonomy can then be generated in the tracking trajectory. In particular, the tracking trajectory is driven with the same level of autonomy over the entire route length, namely at least semi-autonomously or even fully autonomously.

[0018] In one embodiment, the recording of at least the section of the route with which the tracking trajectory is to begin is initiated by a vehicle occupant, as mentioned above. This allows for user-specific recording.

[0019] In one embodiment, it is also possible for such a start of recording of at least the route section with which the tracking trajectory is to begin, as mentioned, to be suggested by a vehicle guidance system. In such an embodiment, a vehicle occupant can then confirm this suggestion and thus initiate the actual start. As already explained above, such an automatic and thus system-based suggestion of a starting point or a start time for recording such a route section can be made depending on at least one decision criterion. In this context, a decision criterion can be, for example, environmental information and / or a vehicle parameter.In this context, such a system-side suggestion for starting recording can be made, for example, based on objects detected in the vehicle's surroundings and / or on a position of the vehicle in the surroundings. In particular, if a specific environmental scenario is recognized and assessed on the basis of this environmental information, such a system-side suggestion can be made. In particular, the environmental information is assessed to determine whether the environmental scenario is suitable or appropriate for repeated driving through it based on an at least semi-autonomous tracking trajectory. This decision can preferably also be made by the vehicle guidance system. For example, this can be assessed based on previously recurring driving through this environmental scenario and / or depending on specific infrastructure in the surroundings.For example, such infrastructure could be a specific parking garage. This could be a parking garage at a workplace or for shopping. It is also possible that the environmental scenario represents a specific property area, such as private property, and the corresponding detection is performed. In particular, the detection of a garage or parking space in this environmental area can also contribute to the assessment of such an environmental scenario as one suitable for subsequent repeated driving through this area with a tracking trajectory.

[0020] However, other environmental scenarios can also include driving maneuvers, for example, involving the attachment or coupling of a trailer to the vehicle. Approaching an energy delivery unit, such as a gas pump at a petrol station or a charging station for electrically charging the vehicle, can also be assessed accordingly. Driving through a car wash can also be an environmental scenario that is assessed as suitable or useful for semi-autonomous follow-up driving with a follow-up trajectory. In one embodiment, data recording of at least the route section with which the follow-up trajectory is to begin is proposed using a trajectory determination system. In particular, starting is proposed for selection by a vehicle occupant. This can be done acoustically and / or visually.The trajectory determination system can be part of the vehicle.

[0021] In one embodiment, the recording of the route section with which the tracking trajectory is to begin is initiated, in particular automatically, by a trajectory determination system. The advantages of the aforementioned embodiments have already been explained above.

[0022] In one exemplary embodiment, this automatic start of recording at least the route section with which the tracking trajectory is to begin is carried out depending on an environmental scenario and / or depending on at least one vehicle parameter of the vehicle. An environmental scenario can be detected, for example, using at least one detection unit of the vehicle. For example, a detection unit can be a camera that is sensitive in the spectral range visible to humans. However, a detection unit can additionally or instead be, for example, a LIDAR sensor and / or an ultrasonic sensor and / or a radar sensor or the like. Preferably, the detection of the environment can be carried out using a front camera that can detect the surrounding area in front of the vehicle. This front camera can be arranged, for example, in the passenger compartment.For example, it can be located behind a vehicle's windshield and detect through the windshield.

[0023] In one embodiment, when driving through the first route section, the surroundings are recorded using at least one detection unit of the vehicle, and this recorded information is stored. Additionally or instead of this, when driving through the second route section, the surroundings can be recorded using at least one detection unit of the vehicle, and this recorded information is stored. It is particularly advantageous if a digital map of the surroundings is generated based on this recorded information.

[0024] In one embodiment, the localization of the vehicle and the generation of a digital map of the surroundings while traveling through the first route section and / or while traveling through the second route section are performed using a SLAM (Simultaneous Localization and Mapping) method or a VSLAM (Visual Simultaneous Localization and Mapping) method. This allows very precise information about the vehicle to be generated, even in more complex environmental scenarios and / or more complex route sections. In particular, this also enables the vehicle's position in the surroundings to be determined very precisely.

[0025] In one embodiment, the surroundings are captured using at least one camera as one embodiment of a capture unit. In particular, the surroundings are captured using multiple cameras. In one embodiment, the camera has a capture range with an aperture angle greater than or equal to 150°, in particular between 160° and 180°. In one embodiment, in addition to or instead of this, the surroundings can be captured using at least one front camera of the vehicle that detects to the front. This front camera preferably has a capture range with an aperture angle less than or equal to 150°, in particular between 150° and 100°.

[0026] It's possible that two, three, four, or six cameras are installed on the vehicle to capture the surroundings. In one embodiment, these cameras can be surround-view cameras. Such a surround-view camera can, for example, be a fisheye camera.

[0027] Especially when a VSLAM method is used, the vehicle's surroundings are recorded using at least one camera, for example, when manually driving through a section of road. Camera images can be analyzed at time intervals or route intervals. Specific features of the surroundings can be determined from these camera images. For example, pixels of an image can be combined into features. Alternatively, and thus also in addition to or instead of, a location parameter or position parameter can be assigned to the features. A digital map is generated, in particular, from a multitude of such features.

[0028] Furthermore, for example, the position of the vehicle can be determined in relation to these features, so that a trajectory is formed from a plurality of such positions. This trajectory, which is generated when driving through a section of road, is stored. In particular, this occurs when driving through a section of road manually. A further aspect of the invention relates to a method for performing a driving maneuver with a vehicle, in which the vehicle is provided with a tracking trajectory, as determined using a method according to the above-mentioned aspect or an advantageous embodiment thereof. As a driving maneuver, an at least semi-autonomous, in particular fully autonomous, tracking drive along the tracking trajectory is performed.

[0029] A further aspect of the invention relates to an electronic analysis system for generating a tracking trajectory. The analysis system has at least one evaluation unit. The analysis system is designed to perform a method according to the above-mentioned aspect or an advantageous embodiment thereof. In particular, this method is performed using this analysis system.

[0030] A further aspect of the invention relates to an electronic vehicle guidance system with a control unit. The vehicle guidance system is designed to carry out a method according to the above-mentioned aspect relating to performing a driving maneuver with a vehicle. In particular, this method is carried out with the vehicle guidance system.

[0031] A further independent aspect of the invention relates to a computer-implemented method for generating a tracking trajectory for a vehicle, comprising the following steps:

[0032] Providing recorded environmental information of an environment and / or vehicle information of the vehicle to an evaluation unit, wherein the information was recorded when driving through a first route section with a first degree of autonomy with the vehicle;

[0033] Providing recorded environmental information of the environment and / or vehicle information of the vehicle to an evaluation unit, wherein the information was recorded when driving through a second route section following the first route section with a second level of autonomy different from the first level of autonomy with the vehicle;

[0034] Determining the follow-up trajectory from the information provided on the recorded first route section and on the recorded second route section with the evaluation unit, wherein the follow-up trajectory is generated for at least semi-autonomous traversing of a follow-up journey.

[0035] In a quick and easy way, the evaluation unit can thus generate a tracking trajectory based on the information provided.

[0036] Advantageous embodiments of the above-mentioned method are to be regarded as advantageous embodiments of the computer-implemented method. In particular, it is provided that acquired and / or evaluated information is provided to the evaluation unit. Providing information in this context may not only mean receiving the information by the evaluation unit. The evaluation unit may be a single evaluation unit. However, it may also comprise multiple evaluation units. In this case, it may be referred to as an evaluation system.

[0037] A further aspect of the invention relates to a computer program product comprising instructions which, when executed on a computer or on a computing unit, cause the computer or the computing unit to carry out a method according to the above-mentioned aspect or an advantageous embodiment thereof.

[0038] A further aspect of the invention relates to a computer-readable storage medium comprising instructions that cause a computer or a computing unit to execute the above-mentioned method or an advantageous embodiment thereof when the instructions are loaded onto or run on the computer.

[0039] A further aspect of the invention relates to a vehicle with an electronic analysis system and / or an electronic vehicle guidance system according to the above-mentioned aspect or an advantageous embodiment thereof.

[0040] The vehicle guidance system may also have at least one detection unit.

[0041] In situations where there is no information and / or only insufficient information about route sections and / or the surroundings of the vehicle and / or vehicle parameters, which result in a tracking trajectory not being able to be trained and / or a driving maneuver based on a trained tracking trajectory not being able to be carried out, an error message can be output on an output unit, for example, and / or the training and / or the execution of the driving maneuver can be aborted.

[0042] An exemplary embodiment of the invention is explained in more detail below with reference to a schematic drawing. The sole Figure 1 shows a schematic plan view of a scenario 1.

[0043] Scenario 1 depicts a specific environment. This could, for example, be a vehicle user's private property. In another exemplary embodiment, however, environment 2 could also be the area of ​​a car wash or an energy supply station, such as a gas station or a charging station with electric charging stations.

[0044] In the exemplary embodiment, the environment 2 is limited by boundaries 3. However, this does not have to be the case.

[0045] These boundaries 3 represent objects in the environment 2. The environment 2 is separated, in particular, by these boundaries 3 from a roadway 4 that does not belong to the environment 2. The environment 2 has an entrance 5 and / or exit 6. In the environment 2, there are also further objects 7, the number and position of which are merely examples and are by no means exhaustive. These can be static objects or dynamically movable objects. They can be plants or trees or the like. However, they can also represent pillars or other objects. In the exemplary embodiment, a fixed parking zone 8 for parking a vehicle is formed in the environment 2. The parking zone 8 can be a garage or a parking space or the like. It is intended to be used to park a vehicle there. In another exemplary embodiment, the parking zone 8 can be, for example, a gas pump or a charging station.However, it is also possible that parking zone 8 is a cleaning chamber of a car wash.

[0046] In another exemplary embodiment, however, the parking zone 8 can also be a terminal at which an interaction of the vehicle and / or a vehicle occupant with this terminal can take place. In the exemplary embodiment, a vehicle 9 is shown. The vehicle 9 is shown here along a route that extends, for example, from a starting position 10 to an end position 11. The vehicle 9 is also shown schematically at the starting position 10 and the end position 11.

[0047] The vehicle 9 has an electronic vehicle guidance system 12. It is also possible for the vehicle 9 to have an electronic analysis system 13.

[0048] In another embodiment, however, the electronic analysis system 13 can also be arranged externally to the vehicle 9. For example, this electronic analysis system 13 can be arranged in a traffic control center or a computing unit. It can be configured for wireless communication with the vehicle 9.

[0049] The vehicle 9 preferably has at least one evaluation unit 14. This can be a component of the vehicle guidance system 12 and / or the analysis system 13. Furthermore, the vehicle 9 can also have a control unit 15. This can have the evaluation unit 14. The control unit 15 can be a component of the vehicle guidance system 12 and / or the analysis system 13. In one exemplary embodiment, the vehicle 9 has at least one detection unit 16. It preferably has a plurality of detection units 16, for example two or three or four or five or six such detection units 16. For the sake of clarity, only two such detection units 16 are shown here. The detection units 16 can be cameras and / or LIDAR sensors and / or radar sensors. If at least one detection unit 16 is a camera, it can be an all-round view camera.In another embodiment, a camera may be designed as a front camera in addition to or instead of this.

[0050] This electronic vehicle guidance system 12 and / or the analysis system 13 can comprise a computer program product. This computer program product can be used to perform a computer-implemented method for generating a tracking trajectory for the vehicle 9, particularly if this computer program product is present on the evaluation unit 14 and / or the control unit 15. This computer-implemented method preferably performs the following steps:

[0051] Providing captured environmental information of environment 2 and / or

[0052] Vehicle information of the vehicle 9 to the evaluation unit 14, wherein the information was recorded when driving through a first route section 17 with a first degree of autonomy with the vehicle 9;

[0053] Providing recorded environmental information of the environment 2 and / or vehicle information of the vehicle 9 to the evaluation unit 14, wherein the information was recorded when driving through a second route section 18 following the first route section 17 with the vehicle 9 having a second degree of autonomy different from the first degree of autonomy;

[0054] Determining the follow-up trajectory 19 from the information provided on the recorded first route section 17 and on the recorded second route section 18 with the evaluation unit 14. Preferably, the follow-up trajectory 19 is generated for at least semi-autonomous, in particular fully autonomous, driving through.

[0055] Furthermore, a method for training a tracking trajectory 19 is carried out with the vehicle 9. For this purpose, in one exemplary embodiment, a first route section 17 in the environment 2 is traversed with a first degree of autonomy of the vehicle 9, and this first route section 17 is recorded. In the exemplary embodiment, the first route section 17 is traversed manually. This means that the vehicle 9 is driven from the starting position 10 by manual guidance by a driver. The starting position 10 and thus, in particular, also the starting of the recording of this first route section 17 can be carried out by a vehicle occupant in one exemplary embodiment.In another exemplary embodiment, it is preferably possible for this starting of the recording of at least the first route section 17 to be suggested by the electronic vehicle guidance system 12 and for this suggestion to be accepted or confirmed, if appropriate, by a vehicle occupant of the vehicle 9. In another exemplary embodiment, the starting of the recording of at least the first route section 17 can also be carried out fully automatically by the electronic vehicle guidance system 12 and / or the electronic analysis system 13. In this exemplary embodiment, the environment 2 can be assessed depending on environmental information and / or vehicle parameters. Depending on such an assessment result, the respective system 12 and / or 13 can then automatically start the recording.If, in one embodiment, a specific environmental scenario is recognized as environment 2, the system 12 and / or 13 can decide where and / or when the recording of the first route section 17 traveled manually will start.

[0056] In the exemplary embodiment, a second route section 18 is traversed by the vehicle 9 along the route traveled, directly following the first route section 17. During this second route section 18, the vehicle 9 is not guided purely manually by the driver, but rather is guided at least semi-autonomously by the electronic vehicle guidance system 12. This means that along the entire route, which is formed by the at least two consecutive route sections 17 and 18, the vehicle 9 is operated in sections with different degrees of autonomy.

[0057] In the exemplary embodiment, the vehicle 9 is moved at least semi-autonomously, in particular fully autonomously, on the second route section 18 to the target position 11, as shown in Fig. 1. This second route section 18 is also recorded in full. During recording, in particular environmental information and / or vehicle information and / or the route length and / or the course of the route traversed by the two route sections 17 and 18 are recorded. Preferably, a digital map of the environment 2 is generated depending on the recorded information. In particular, the localization of the vehicle 9 when traveling through the route sections 17 and 18 and the generation of a digital map of the environment 2 when traveling through the first route section 17 and / or when traveling through the second route section 18 are carried out using a SLAM or VSLAM method.By driving through the two track sections 17 and 18, a resulting follow-up trajectory 19 is trained.

[0058] In particular, this tracking trajectory 19 is generated in the training scenario described so far from the recorded information on the first route section 17 and the recorded information on the second route section 18. The tracking trajectory 19 is determined or generated, in particular, for at least semi-autonomous driving during a subsequent and repeated tracking journey. In particular, the tracking trajectory 19 and thus its route length and / or its course are determined depending on the summed two route sections 17 and 18. In particular, the tracking trajectory 19 is determined such that it corresponds to the route length and / or the course of the summed two route sections 17 and 18.Thus, after the training scenario, during at least one further repeated travel through the environment 2, in particular starting from reaching the starting position 10, the route between the starting position 10 and the target position 11 is traveled completely at least semi-autonomously on the basis of the trained and stored follow-up trajectory 19.

[0059] In the exemplary embodiment, it is possible for the vehicle guidance system 12 to detect that the starting position 10 or a tolerance range thereto has been reached during such a follow-up journey, and then to automatically continue driving to the target position 11 based on the follow-up trajectory 19, at least semi-autonomously, in particular fully autonomously. In particular, upon detection of the starting position 10 in this way, the vehicle occupant can be given an indication that the electronic vehicle guidance system 12 will take over further vehicle guidance to the target position 11 and, in this regard, the follow-up trajectory 19 will be traversed at least semi-autonomously, in particular fully autonomously.

[0060] The same scenario as shown and explained in Fig. 1 can also apply to another environment that is not private property with a garage or parking space as parking zone 8, in particular to a public car wash or a petrol station or a charging park or the like.

Claims

Patent claims 1. A method for training a tracking trajectory (19) for a vehicle (9), comprising the following steps: Driving through a first route section (17) with a first degree of autonomy with the vehicle (9) and recording the first route section (17); Driving through a second route section (18) following the first route section (17) with a second degree of autonomy different from the first degree of autonomy with the vehicle (9) and recording the second route section (18); Determining the follow-up trajectory (19) from the recorded first route section (17) and the recorded second route section (18), wherein the follow-up trajectory (19) is generated for at least semi-autonomous traversing of a follow-up journey.

2. Method according to claim 1, wherein the route length and / or the course of the follow-up trajectory (19) is determined as a function of the two route sections (17, 18).

3. Method according to claim 1 or 2, wherein the follow-up trajectory (19) is determined such that it corresponds to the route length and / or the course of the summed two route lengths (17, 18).

4. Method according to one of the preceding claims, wherein a route section (17, 18) is traversed by manually driving the vehicle (9) with a driver as the degree of autonomy.

5. Method according to one of the preceding claims, wherein a route section (17, 18) is traversed at least semi-autonomously with a vehicle guidance system (12) of the vehicle (9) as the level of autonomy.

6. Method according to one of the preceding claims, wherein the recording of at least that section of the route (17, 18) with which the follow-up trajectory (19) is to begin is started by a vehicle occupant.

7. The method according to claim 6, wherein the starting of the recording of at least that section of the route (17, 18) with which the follow-up trajectory (19) is to begin is proposed by a trajectory determination system (12, 13), in particular the starting is proposed for selection by a vehicle occupant.

8. Method according to one of the preceding claims 1 to 6, wherein the starting of the recording of the route section (17, 18) with which the follow-up trajectory (19) is to begin is started by a trajectory determination system (12, 13).

9. The method according to claim 8, wherein the automatic starting takes place depending on an environmental scenario of the environment (2) and / or depending on at least one vehicle parameter of the vehicle (9).

10. Method according to one of the preceding claims, wherein when driving through the first route section (17) the surroundings (2) are detected with at least one detection unit (16) of the vehicle (9) and this detected information is stored and / or when driving through the second route section (18) the surroundings (2) are detected with at least one detection unit (16) of the vehicle (9) and this detected information is stored.

11. Method according to claim 10, wherein a digital map of the environment (2) is generated depending on the acquired information.

12. Method according to claim 10 or 11, wherein the localization of the vehicle (9) and the generation of a digital map of the surroundings (2) when driving through the first route section (17) and / or when driving through the second section (18) using a SLAM or VSLAM method.

13. Method according to one of the preceding claims, wherein the environment (2) is detected with at least one camera and / or at least one radar sensor and / or at least one LIDAR sensor as detection unit (16).

14. The method according to claim 13, wherein the environment (2) is detected with at least one camera, in particular a plurality of cameras, which has a detection range with an opening angle greater than or equal to 150°, in particular between 160° and 180° or 200°, and / or the environment (2) is detected with at least one forward-detecting front camera of the vehicle (9) which has a detection range with an opening angle less than or equal to 150°, in particular between 150° and 100°.

15. Method for carrying out a driving maneuver with a vehicle (9), in which the vehicle (9) is provided with a follow-up trajectory (19) as determined by a method according to one of the preceding claims, and as a driving maneuver an at least semi-autonomous, in particular fully autonomous, follow-up drive is carried out as a driving maneuver along the follow-up trajectory (19).

16. Electronic analysis system (13) for generating a tracking trajectory (19), with at least one evaluation unit (14), wherein the analysis system (13) is designed to carry out a method according to one of the preceding claims 1 to 14.

17. Electronic vehicle guidance system (12) with a control unit (15), wherein the vehicle guidance system (12) is designed to carry out a method according to claim 15.

18. Computer-implemented method for generating a tracking trajectory (19) for a vehicle (9), comprising the following steps: Providing recorded environmental information of an environment (2) and / or vehicle information of the vehicle (9) to an evaluation unit (14), wherein the information was recorded when driving through a first route section (17) with a first degree of autonomy with the vehicle (9); Providing recorded environmental information of the environment (2) and / or vehicle information of the vehicle (9) to an evaluation unit (14), wherein the information was recorded when driving through a second route section (18) following the first route section (17) with the vehicle () with a second degree of autonomy different from the first degree of autonomy; Determining the follow-up trajectory (19) from the information provided on the recorded first route section (17) and on the recorded second route section (18) with the evaluation unit (14), wherein the follow-up trajectory (19) is generated for at least semi-autonomous traversing of a follow-up journey.

Images