Method for at least semi-autonomous maneuvering of a vehicle to an intermediate position and / or an endpoint along a trained driving trajectory, electronic maneuvering system, and control unit

Abstract

One aspect of the invention relates to a method for at least semi-autonomous maneuvering of a vehicle (1), comprising the following steps: - Moving the vehicle (1) and recording a driving trajectory (17) of the vehicle (1) during this movement; - Providing environmental information about the environment (6) of the vehicle (1) along the driving trajectory (17); - Planning a maneuver of the vehicle (1) by selecting specific environmental information and / or at least a zone around a specific trajectory point (19) of the driving trajectory (17), wherein the specific environmental information and / or the specific trajectory point (19) characterizes a specific sub-section (23) of the driving trajectory (17); - Maneuvering the vehicle (1) by at least semi-autonomously driving the vehicle (1) over the section (23).

Description

[0001] One aspect of the invention relates to a method for at least semi-autonomous maneuvering of a vehicle. Another aspect of the invention relates to a computer-implemented method for determining a driving path for a vehicle. Yet another aspect of the invention relates to an electronic maneuvering system. Another aspect of the invention relates to a control unit for a vehicle. Furthermore, another aspect of the invention relates to a vehicle.

[0002] It is known that electronic maneuvering systems can be, for example, an electronic parking assistance system or a driver assistance system. With electronic maneuvering systems, it is also possible for driving trajectories to be recorded or stored. Such trained driving trajectories can then be used for subsequent maneuvering operations. However, this limits flexibility.

[0003] While recording a driving trajectory, a driver can travel along a trajectory being recorded and simultaneously record any maneuver (e.g., entering their garage). The recorded driving trajectory is then stored by the vehicle, for example, by a control unit of the appropriately trained driver assistance system. When the driver approaches the recorded and stored driving trajectory, the driver assistance system can recognize it and offer a fully or at least partially autonomous maneuver to follow the stored trajectory.

[0004] With current systems, the driver must always record the entire trajectory from start to finish. In other words, the vehicle must travel the entire trajectory to reach the endpoint in order to record it completely. The driver can only store one endpoint per trajectory, which means they have to record multiple trajectories for multiple endpoints.

[0005] The problem for the driver is that, while recording the trajectory, they must be able to execute the entire maneuver, which can be difficult, for example, if the road is blocked. Furthermore, to record the trajectory, including the endpoint, both the starting and ending points must be clear so that the trajectory can be saved.

[0006] The object of the present invention is to provide a method, an electronic shunting system, a control unit and a vehicle in which the flexibility of at least semi-autonomous driving maneuvers based on a known driving trajectory is increased.

[0007] This task is solved by a method, an electronic shunting system, a control unit, and a vehicle according to the independent claims.

[0008] One aspect of the invention relates to a method for at least semi-autonomous maneuvering of a vehicle. In particular, this method comprises the following steps: - In particular, moving the vehicle and recording the vehicle's trajectory during this movement; - In particular, providing environmental information about the vehicle's surroundings along its trajectory; - In particular, planning a maneuver of the vehicle by selecting specific environmental information and / or at least a zone around a specific trajectory point of the driving trajectory and / or a specific, in particular predetermined and discrete, trajectory point, a specific sub-section, in particular at least one of its ends and / or its length, of the driving trajectory is characterized, wherein a specific sub-section of the driving trajectory is characterized by the specific environmental information and / or the specific zone and / or the specific trajectory point; - In particular, maneuvering the vehicle by at least semi-autonomously driving the vehicle along the section.

[0009] This method thus increases the flexibility of a semi-autonomous vehicle maneuvering process based on a known trajectory. It now allows for the specific use of this previously traveled and, in particular, stored trajectory. Specifically, it enables the use of only specific segments of this trajectory, which can be predefined by the user, to reach intermediate positions along the route in a targeted and desired manner. The vehicle can therefore travel along this known trajectory with user-specific precision, without having to repeat the entire route. Instead, intermediate destinations along this trajectory can be individually selected and reached through at least semi-autonomous, and in particular, fully autonomous, maneuvers.A particular advantage is that this planning of the vehicle's maneuvering, based on the already known trajectory, can be carried out according to various specifications, especially those of a user. This makes it possible, on the one hand, to select a very specific point along the trajectory, which then defines the segment of the trajectory to be traveled. In this context, a segment is defined as a section of the trajectory that is not the entire trajectory.

[0010] In this context, it is particularly advantageous that, optionally, a user can determine a specific segment of the trajectory to be traveled based on the provided environmental information. This means that it is not mandatory to define, or even select, a precise point along the trajectory to determine the segment. This also allows for the selection of infrastructure in the surrounding area that can be reached along the segment. This, in turn, enables precise maneuvering of the vehicle based on the trajectory, but only along the selected segment.

[0011] In another embodiment, however, the desired segment can be determined either based on a local point along the trajectory or by selecting environmental information. This method advantageously allows for the selective characterization of this segment of the trajectory. In particular, this enables users to decide, in a highly individualized and situation-dependent manner, which approach to use to determine the segment and plan the maneuvering.

[0012] In general, trajectory points are fixed, predetermined locations along the driving trajectory. They are generated when the driving trajectory is recorded. They serve as reference points along the driving trajectory.

[0013] In particular, the determination of the subsection of the travel trajectory as a route depends on at least the input information, especially with a computing unit.

[0014] In particular, it is provided that a vehicle control unit is configured to generate control signals for at least one functional unit of the vehicle, depending on information, especially the section of the route, in order to enable driving on that section. Specifically, these control signals are generated by the control unit. The processing unit can be part of the vehicle. However, it can also be located externally. For example, it can be located in a data center or in another vehicle.

[0015] The maneuver is specifically a shunting maneuver. In particular, this means that the vehicle's speed over the entire distance of the maneuver is less than or equal to 40 km / h, and especially less than or equal to 30 km / h.

[0016] A functional unit can be a vehicle's braking system, steering system, assistance system, drive system, etc.

[0017] In one embodiment, the selected environmental information and / or the selected zone and / or a specific trajectory point of the vehicle's trajectory define an endpoint of the segment to which the vehicle is to move, at least semi-autonomously, and in particular fully autonomously, from its current position. This selection and thus definition of the endpoint precisely defines the length of the segment. The vehicle's maneuvering is therefore also specified with exceptional accuracy. Reaching this desired location during at least semi-autonomous maneuvering is thus possible with remarkable precision. Especially during fully autonomous maneuvering, the local position can therefore be approached with high accuracy.This is particularly advantageous when a specific position relative to an infrastructure in the vicinity needs to be reached, and / or interaction with the infrastructure needs to be enabled, and / or entry into the infrastructure is required. This is because the precision of the maneuvering process is significantly enhanced by this specific endpoint definition.

[0018] In one embodiment, when selecting a zone, a trajectory segment of the driving trajectory between two, particularly adjacent, trajectory points is selected as the endpoint. This allows for direct localization on the driving trajectory, characterizing this segment. In this context, for example, a trajectory point can be selected directly, or the trajectory segment between two trajectory points can be selected. Such a selection can be made, for example, by a voice signal and / or by touching the driving trajectory displayed on a screen. It is possible that the trajectory points, which represent very discrete and local points on the driving trajectory, are numbered and / or color-coded. They can also be geometrically individualized, either additionally or instead of this.This allows, for example, the selection of such a trajectory point by a simple voice signal, such as mentioning the number and / or the geometry and / or the color.

[0019] It is also possible to select a point in the zone that is not on the vehicle's trajectory as the endpoint. This allows, to a particularly advantageous extent, the vehicle to be directed to a point that is not precisely on the trajectory. This can be beneficial when specific environmental conditions and / or situations, such as those dependent on weather, traffic conditions, or traffic volume, need to be taken into account. Furthermore, this embodiment also allows for a certain degree of tolerance for the vehicle's endpoint when a specific exactness of a point or position on the trajectory is not strictly necessary.

[0020] In another embodiment, a trajectory point on the driving trajectory can be selected directly and exactly as the endpoint.

[0021] This also provides different alternatives, particularly as options. This allows users to choose a different endpoint scenario depending on the specific situation and their needs.

[0022] In one embodiment, trajectory points of the driving trajectory can be individualized geometrically, in terms of color, and / or in terms of size. These individualizations can be visually displayed on a screen. Such individualization of trajectory points can provide the user with additional visual information. It is also possible for a trajectory point to dynamically change its color, size, and / or geometry. This, too, can provide the user with personalized information. For example, such individualization of trajectory points can signal critical sections of the driving trajectory. It is also possible that, for example, a particularly difficult maneuver or particularly tight conditions exist in a section of the driving trajectory.This allows users to be notified that, for example, selecting a specific section might involve a more difficult driving maneuver, which could result in the maneuver taking longer and / or requiring manual intervention from the user to control the vehicle. This also enables users to receive information on whether they should adjust their choice of destination and thus the section, and perhaps plan differently.For example, this can also achieve a situation where, if a trajectory point is selected, or a trajectory segment containing this trajectory point is selected, and / or an endpoint adjacent to the driving trajectory in a zone is located near such a trajectory point, maneuvering may be simpler and / or safer if an adjacent trajectory point is selected as the endpoint. In one embodiment, this allows for a specific classification of trajectory points, according to which specific sub-segments can be categorized with regard to safe and / or fast driving maneuvers, should the endpoint need to be selected at that corresponding location.

[0023] In one embodiment, it is possible to select the endpoint, particularly if a point not on the trajectory is chosen as the endpoint, while the trajectory is being recorded. The maneuvering of the vehicle along the segment of the trajectory leading to the selected endpoint is planned during the recording process. In other words, a user or driver of the vehicle does not actually have to travel to the endpoint while recording the trajectory, but can save it as the endpoint of the currently recorded trajectory before reaching it.This allows the vehicle to be maneuvered to the endpoint during the current journey or at a later time using the recorded trajectory, without the driver actually having driven to the endpoint during the initial trajectory run to record it as the endpoint of the trajectory.

[0024] Preferably, the endpoint is selected within the detection range of a sensor unit, such as an environmental sensor in the vehicle. For example, the detection range of a front camera can be displayed graphically on a display unit inside the vehicle. Within the displayed detection range, the user or driver can select the endpoint, which is not on the vehicle's trajectory, for example, by touch input on the display unit.

[0025] In one embodiment, a point beyond the most recent trajectory point is selected as the endpoint. In other words, a point adjacent to the end of the recorded trajectory can be selected as the endpoint. The path from the most recent trajectory point to the endpoint is defined as the segment along which the vehicle is maneuvered.

[0026] Preferably, a known parking space located in the vicinity and / or within the zone and / or detection range of the vehicle's sensor unit is provided as a selectable endpoint. In other words, the user or driver of the vehicle can be presented with one or more known parking spaces as potential endpoints for the driving trajectory, for example, on the display unit. Thus, if the recorded driving trajectory ends near one or more known parking spaces, one, several, or all of these parking spaces can be suggested as endpoints for the trajectory, without the vehicle ever having to actually visit them to train these endpoints. Depending on current environmental conditions, a currently available and accessible parking space can be automatically selected as the endpoint for the current driving maneuver.If several of the known parking spaces are currently free and accessible, the parking space that meets predetermined criteria can be selected as the endpoint, for example, because it is within a maximum acceptable distance to an exit of a parking garage.

[0027] In other words, at least one known parking space can be automatically selected as the endpoint from a large number of known parking spaces based on predefined selection criteria. Each parking space from the large number of known parking spaces can be checked for its current availability and / or accessibility before being selected as the endpoint. If a parking space is currently unavailable, it can still be saved as the endpoint for a future maneuver. In this way, multiple endpoints for a single driving trajectory can be saved and selected and navigated to as needed.

[0028] In other words, during at least one further maneuver of the vehicle, in addition to the selected endpoint, further parking spaces from the multitude of known locations can be considered as potential endpoints for the driving trajectory, depending on their current availability and / or accessibility. The trajectory can therefore be extended, for example, to include known parking positions in the vicinity as endpoints.

[0029] In another embodiment, the endpoint can be selected by the vehicle user via a user interface on a display unit in the vehicle. For example, known parking spaces can be displayed as potential endpoints on a screen of the display unit, perhaps in a map view, and the vehicle user can select one of the known parking spaces by touch input on the display unit's screen.

[0030] Preferably, the selected endpoint of the driving trajectory is added as a new trajectory point. In other words, the driving trajectory is supplemented or extended by the selected endpoint. The driving trajectory thus supplemented can be made available for further, at least semi-autonomous, maneuvers of the vehicle.

[0031] Alternatively or additionally, several potential endpoints can be stored for the driving trajectory, whereby one of the endpoints is selected as the actual endpoint for the driving trajectory currently being traveled by the vehicle.

[0032] The described embodiments show how the driver can select the endpoint for an active recording of a driving trajectory without actually having to drive the vehicle to the endpoint.

[0033] As described, a driver assistance system, for example, can give the driver the option of choosing their own destination. To avoid the problem of the driver having to drive the entire route themselves, the driver could be able to select the destination via a human-machine interface within a defined area around the vehicle. This area can be selected based on the detection range of a sensor on the vehicle to check whether the chosen destination is accessible by car and / or whether a collision-free approach to the destination is possible.

[0034] Several options are available for the playback phase of the recorded driving trajectory: The route planned by the driver during recording when selecting the endpoint (the recorded driving trajectory) can be saved at the end of the recording, so that it is always played back exactly as planned during the recording. Alternatively or additionally, the driver assistance system can plan different paths during playback of the recorded trajectory, with the respective endpoint of the trajectory or path being spontaneously selected, for example, depending on current environmental conditions.

[0035] As described, available parking spaces that are within reach can be identified and offered as endpoints to be saved in the recorded trajectory. Alternatively, the system can generally search for potential, unoccupied endpoints that can be offered to the driver for saving, provided they are unoccupied and reachable without collisions. In addition, currently unoccupied parking spaces can be searched for as potential endpoints.

[0036] Occupied but reachable endpoints can be identified and offered to the driver so they can be saved as endpoints for the recorded trajectory. To avoid the issue of requiring an endpoint to be reachable without collisions for trajectory recording, occupied endpoints, such as occupied parking spaces, are also preferentially identified and offered as selectable endpoints. In other words, if a driver is recording a trajectory and sees that the desired endpoint is already blocked, they can still save the recording with the occupied endpoint. Therefore, occupied endpoints can be searched for and offered to the driver for saving with the currently recorded trajectory if a drivable and collision-free route to such an endpoint can be planned.The driver can then also select the occupied endpoint that should be saved in the trajectory recording.

[0037] As described, multiple endpoints can be saved for a currently recorded trajectory. If several of these endpoints are free, the user can select one of them for an autonomous or at least partially autonomous driving maneuver currently being performed by the vehicle, depending on the user's priority.

[0038] In a concrete example, the user or driver of the vehicle can first record the initial part of a currently traveled trajectory. A user-selected endpoint can be added to this first part. The user can then continue recording a second part or segment of the trajectory. During the recording of the second part, further endpoints can be offered, such as available or occupied parking spaces within the detection range of the vehicle's sensor. The user can also select these endpoints and add them to the recorded trajectory. In a subsequent autonomous or at least partially autonomous driving maneuver, the vehicle can then follow the previously recorded trajectory, completing the maneuver at one of the added endpoints, for example, depending on the user's priority.The priority of the endpoints can follow the order in which they are added to the recorded trajectory. As described, the driver can also record trajectories with endpoints that are currently occupied.

[0039] In another embodiment, environmental information values ​​are captured during the recording of the vehicle's trajectory using at least one of the vehicle's sensors. This provides particularly up-to-date information about the environment as the trajectory is driven and thus trained and learned. This enables a particularly reliable and time-coupled correlation between the recorded trajectory and the environmental information prevailing at that time.

[0040] It is also possible to capture landmarks, particularly objects, as environmental information. This provides particularly useful anchor information that correlates with the vehicle's trajectory. This allows for a very advantageous and precise selection of a section of the route based on such environmental information in the form of landmarks. Since very accurate positional information for these landmarks can also be generated and provided, maneuvering along a section of the route, especially along or adjacent to such a landmark, can be planned with exceptional precision.

[0041] In one embodiment, the vehicle's position is determined during movement, particularly based on geometric data. Specifically, this position information is correlated with environmental information along the vehicle's trajectory. This ensures that environmental information is uniquely assigned to the vehicle's position along this known and stored trajectory. In particular, this provides a highly accurate representation of the environment, especially based on a digital map.

[0042] In one embodiment, several discrete trajectory points are generated as the vehicle moves. Each trajectory point is characterized, in this embodiment, by at least one mapping cluster consisting of environmental information and the vehicle's position. This increases accuracy and, particularly advantageously, provides a local selection point via the trajectory points, enabling a highly precise determination and traversal of the segment. In particular, this also allows for the precise determination and specification of the end of the segment.

[0043] In one embodiment, the trajectory points are generated at intervals between 0.40 m and 1.50 m, particularly between 0.70 m and 1.20 m. This advantageously creates a chain of points. These intervals are particularly beneficial because, on the one hand, the number of trajectory points and their positional assignment allow for a very fine-grained selection of sub-segments, thus enabling a very precise determination of an endpoint for each sub-segment. This allows for at least semi-autonomous, and especially fully autonomous, movement or maneuvering of the vehicle, particularly in correlation with a desired infrastructure facility in the vicinity of the trajectory, so that this infrastructure facility can also be approached very precisely.Even when multiple infrastructure units are located close to each other in the local area, determining the distances between trajectory points allows for a selection process that enables precise maneuvering towards a desired infrastructure facility within such an infrastructure cluster. Furthermore, the density of trajectory points, due to these specific distances, is not so high as to make selection difficult or increasingly inaccurate. This effectively prevents input errors and the need for deletions or replanning maneuvers.

[0044] Furthermore, these distances do not undesirably increase the effort required for creation and thus the computing power for a control unit or computing unit.

[0045] In one embodiment, the environmental information and / or the zone during planning and / or a specific, predefined trajectory point are selected by the user. This allows a person to perform these specific steps themselves, as an alternative to system-based selection, in order to define the respective segment according to their individual needs and preferences. This user-specific selection option offers a high degree of flexibility and numerous variations, enabling the user to drive the vehicle semi-autonomously or fully autonomously along the known trajectory to various endpoints of a segment, depending on the situation. In particular, this also enables a highly advantageous and versatile remote control of the vehicle when the user is outside the vehicle and wishes to drive it to a specific endpoint along the known and learned trajectory.

[0046] In one embodiment, the vehicle trajectory and environmental information, and / or zones and / or trajectory points, are displayed on a display unit. Selection can be made by input, particularly manual input, such as via touch-sensitive areas of the display unit, and / or by voice commands. It is also possible for the vehicle itself, with its current position, to be displayed on this unit. This display can be on an internal vehicle display unit. Additionally, or instead of this, and particularly advantageous, it is possible for this information to be displayed on an external display unit. For example, this could be a portable display unit. This is especially advantageous if the portable device is a communication device in conjunction with the display unit.This could be, for example, a mobile device such as a tablet or smartphone. A portable remote control unit is also possible. Furthermore, the display unit could be located, for example, in a teleoperations center. This would allow, for instance, a person acting as a teleoperator, who is located remotely from the vehicle and cannot directly inspect or see it, to maneuver a section of the vehicle along a known trajectory, at least semi-autonomously.

[0047] In one embodiment, the vehicle's movement along a section of the route is displayed in real time on a screen. Specifically, the vehicle's trajectory and the dynamic changes in the already traversed portion of the section are shown. This allows the vehicle's movement to be visually observed at any time as it travels along this selected section. This enables particularly safe and planned maneuvers, especially shunting. This is especially beneficial when the user is not directly inside the vehicle but is located externally.

[0048] In an advantageous embodiment, when driving along this selected section, the vehicle's localization is performed based on environmental information and a SLAM (Simultaneous Localization and Mapping) method. This combination enables highly accurate localization, even during the vehicle's dynamic movement. As a result, the safety of maneuvering the vehicle while driving along this section is particularly high. In particular, this also makes it possible to achieve highly accurate repeated driving along the already known trajectory.

[0049] In one embodiment, the segment is traversed in the reverse direction of the driving trajectory. This means, for example, that if the driving trajectory has already been traversed forwards, the subsequently selected segment of this now-known driving trajectory is traversed in reverse. If the vehicle is moving in reverse during the recording and learning of the driving trajectory, another embodiment allows the selected segment to be traversed in the vehicle's forward direction. This eliminates the need to turn the vehicle around to traverse the segment.

[0050] In another embodiment, however, it is also possible for the section to be traversed in the same direction as the driving trajectory. In particular, it is then provided that the vehicle is turned around after traversing the driving trajectory and / or before beginning to traverse the selected section. This can also be carried out at least semi-autonomously, and in particular fully autonomously.

[0051] In one embodiment, the environmental information and / or the vehicle trajectory are stored in an external unit. For example, they could be stored in a cloud. This is advantageous because the vehicle itself does not require such data storage. Furthermore, it also allows the relevant information to be made available to other vehicles. Finally, it also enables corresponding processing to be carried out in the cloud.

[0052] In one embodiment, the driving trajectory can preferably be up to 250 m long. In particular, it can be at least 10 m, more specifically at least 50 m, and more specifically at least 100 m long. In this context, all values ​​of the driving trajectory length between these aforementioned interval limits, resulting in one-meter increments, are also considered disclosed. It is also possible for the driving trajectory to be longer than 250 m. Such long distances are particularly advantageous in that they allow for the selection and approach of a wide variety of sub-sections, and thus also enable the selection of a wide variety of intermediate positions along the driving trajectory as endpoints for the sub-section.This makes it particularly advantageous that, with such a stored, especially single, driving trajectory, a wide variety of infrastructure facilities located along the trajectory can be approached at least semi-autonomously, and in particular fully autonomously, by selecting such a segment. This scenario is also particularly advantageous in that a vehicle's reversing assistance system can be used, especially during maneuvering, which preferably takes place at speeds of 40 km / h or less. Such a reversing assistance system enables the vehicle to move along a known and trained driving trajectory over a longer distance, at least semi-autonomously, and in particular fully autonomously.In particular, this learned driving trajectory can thus be repeated very precisely over a longer stretch of road and, especially, driven again in the opposite direction. This is a further advantage, as this approach is limited with conventional parking assistance systems, which, unlike reversing assistance systems, have a significantly shorter range in terms of their operating zone, and are intended to have such a shorter range.

[0053] In one embodiment, the user can also be positioned outside the vehicle when planning the maneuver and moving the vehicle along the section.

[0054] In one embodiment, the zone can be defined by a maximum distance from the driving trajectory and / or by a maximum radius around a specific trajectory point, in particular by a maximum distance and / or radius of preferably a maximum of 2.00 m, more specifically 1.50 m, from the trajectory point. Such zone dimensioning allows for a wide range of selection flexibility, as already mentioned above. Nevertheless, the distance from the driving trajectory is not so large that undesirable inaccuracies in reaching the endpoint of the sub-segment would occur. This makes it particularly advantageous to avoid unsafe and critical driving maneuvers along the sub-segment. On the other hand, this possible zone also makes it particularly advantageous to be able to reach points in the vicinity that deviate from the driving trajectory itself using a simple tolerance maneuver.If endpoints are selected for the segment that are not on the driving trajectory itself, for example a free parking space within the detection range of the vehicle's detection unit, this maximum distance and / or this maximum radius can be extended, for example to 10 or even to 15 m.

[0055] Another aspect of the invention relates to a method, in particular a computer-implemented method, for determining a route for a vehicle. This method preferably comprises the following steps: - Providing a driving trajectory, driven in an environment where the driving route is to be determined, to a computing unit; - Providing environmental information about the vehicle's surroundings along the driving trajectory to the processing unit; - Providing input information from a user regarding the selection of specific environmental information and / or at least a zone around a specific trajectory point and / or a trajectory point of the driving trajectory and / or an endpoint of the driving trajectory, wherein the specific environmental information and / or the specific trajectory point and / or the specific zone and / or the endpoint characterizes a specific subsection of the driving trajectory, in particular measured from a starting point or an endpoint of the driving trajectory, to the computing unit; - Determining the segment of the travel trajectory as the route depending on at least the input information using the computing unit.

[0056] This method makes it particularly easy to determine, precisely and according to need, even a specific segment of the vehicle's trajectory by the processing unit, by providing highly specific information. Providing information can involve receiving it. However, it is also possible for the processing unit to retrieve the information from memory.

[0057] Another aspect of the invention relates to a computer program product comprising instructions that cause a computer to execute the steps according to the above-mentioned method or an advantageous embodiment thereof. In particular, at least the determination of the segment of the travel trajectory as the route is carried out by the computer program depending on at least the input information. Another aspect of the invention relates to a computer-readable data carrier, in particular one comprising the above-mentioned computer program.

[0058] Another aspect of the invention relates to an electronic maneuvering system for a vehicle. The maneuvering system comprises, in particular, at least one processing unit. It preferably also comprises an input unit. The maneuvering system is configured to carry out a method according to one of the aspects mentioned above or an advantageous embodiment thereof. In particular, this method is carried out using the electronic maneuvering system.

[0059] Another aspect of the invention relates to a control unit for a vehicle which is configured to generate control signals for at least one functional unit of a vehicle, depending on information such as that generated in particular by a method according to one of the aspects mentioned above or an advantageous embodiment thereof, in order to carry out the traversing of the subsection.

[0060] Another aspect of the invention relates to a vehicle with an electronic maneuvering system and / or a control unit according to the aspects mentioned above.

[0061] An embodiment of the invention is explained in more detail below with reference to schematic drawings. These show: Fig. 1 a schematic representation of an embodiment of a shunting system according to the invention and an embodiment of a vehicle according to the invention; Fig. 2 a schematic top view of a representation with an environment, a vehicle and a learned driving trajectory; Fig. 3 a representation accordingly Fig. 2, wherein in Fig. 3 a selected section was traversed by the vehicle; Fig. 4 a schematic top view of a representation with a vehicle, a learned driving trajectory and a selected endpoint marking the end of a sub-section along which the vehicle is maneuvered; Fig. 5 another schematic top view of a representation according to Fig. 4, where the endpoint was selected by a user within a sensor detection range; and Fig. 6 another schematic top view of a representation according to Fig. 4 and / or 5, wherein in Fig. Six endpoints are offered, both inside and outside the detection range of the vehicle sensor.

[0062] In the figures, identical or functionally equivalent elements are given the same reference symbols.

[0063] In Fig. Figure 1 shows a schematic representation of an exemplary embodiment of a vehicle 1. The vehicle 1 can be, for example, a passenger car or a truck. The vehicle 1 has an electronic maneuvering system 2. The vehicle 1 has a control unit 3. The control unit 3 can be part of the electronic maneuvering system 2. In particular, the vehicle 1 has an electronic reversing assistance system 2a. This can be part of the electronic maneuvering system 2. Furthermore, in one exemplary embodiment, the vehicle 1 can have an electronic parking assistance system 2b. This can also be part of the electronic maneuvering system 2. The functional range of the electronic reversing assistance system 2a is greater than that of the parking assistance system 2b. In particular, the electronic reversing assistance system 2a is configured to be able to drive along a known and stored driving trajectory.In particular, this is also possible fully autonomously over a distance greater than 50 m, especially greater than 100 m, and especially up to 300 m. The parking assistance system 2b is specifically designed such that its functional range is less than 50 m. This means, in particular, that maneuvering the vehicle 1 based on the parking assistance system 2b can only be carried out semi-autonomously, and in particular fully autonomously, over a maximum distance of 50 m.

[0064] It is possible that the shunting system 2 includes a computing unit 2c. This can be part of a control unit 2d of the vehicle 1. It is also possible that the control unit 2d and the computing unit 2c are separate components.

[0065] Furthermore, in Fig. Figure 1 also shows a cloud 4 arranged externally to the vehicle 1. This can be part of the shunting system 2 if this shunting system 2 is not formed exclusively by components internal to the vehicle. In this context, the exemplary embodiment then shows a system 5 which comprises the vehicle 1 and the cloud 4. Furthermore, in an exemplary embodiment according to Fig. 1. A display unit 6a is also provided. This can be located inside the vehicle 1 or externally to the vehicle 1. It can be part of a portable communication device, such as a smartphone or tablet, or another remote control unit.

[0066] In Fig. Figure 2 shows a schematic representation, in particular a top view, of an environment 6. This environment 6 could be, for example, a private property, a public parking garage, or another type of environment. In the illustrated embodiment, the environment 6 is a private property. Infrastructure facilities include a parking area, for example, a garage 7. Furthermore, a building 8, which could be, for example, a residential building, is also present. In addition, further parking zones 9 and 10 are provided in this environment 6 as examples of infrastructure areas or facilities. Other infrastructure facilities or areas, such as a loading zone 11, are also shown, but only as examples. This loading zone 11 is designed as a facility where the vehicle 1 can be loaded or unloaded, or where specific goods can be loaded.It is also possible that an infrastructure facility is a communication facility, such as a terminal or a drive-through window, or the like. Furthermore, additional objects 12, 13, and 14 are provided, which could be, for example, plantings and / or other objects and / or traffic signs. In addition, a garden area 15 is present in the exemplary embodiment. This can be bordered by an edge zone, such as kerbstones or the like. It can be a lawn area or a planting area. These can also be infrastructure facilities. Unlike infrastructure facilities 7, 9, and 10, these are not infrastructure facilities that are intended for use by vehicle 1, for example, for parking.

[0067] The infrastructure facilities and / or their specific boundaries can serve as landmarks. Fig. Figure 2 shows a representation that can be displayed, for example, on display unit 6a. Vehicle 1 is also shown here in its current position.

[0068] Based on this exemplary and by no means exhaustive scenario, a method for at least semi-autonomous maneuvering of a vehicle, here vehicle 1, is explained. Additionally, it is possible for vehicle 1 to start from the position shown in the diagram. Fig. The vehicle 1 moves to the position shown in point 2 and, in doing so, covers the distance 16 already shown. During this movement, the generated trajectory 17 is recorded. In particular, the position of the vehicle 1 is preferably determined based on at least odometry data. Furthermore, it is specifically provided that the environment is detected by at least one detection unit 18 of the vehicle 1. In this embodiment, environmental information is generated. This environmental information of the environment 6 is made available. In particular, it is made available to the processing unit 2c. Thus, environmental information of the environment 6 of the vehicle 1, which was acquired during the recording of the trajectory 17, is made available. This also makes it possible, in particular, to assign this environmental information to the trajectory 17.Environmental information is also assigned to the positions of vehicle 1 along the driving trajectory 17. In the example, driving trajectory 17 extends from the position of vehicle 1 shown at the property entrance of environment 6 to the final position of vehicle 1 in garage 7. However, this is merely an example, and many other driving trajectories can be generated.

[0069] In this embodiment, several discrete trajectory points 19 are generated during movement and thus during the recording of the vehicle's trajectory 17. For clarity, only some of these trajectory points 19 are shown here with the corresponding reference symbol. Each trajectory point 19 is preferably assigned at least one mapping cluster, or each trajectory point 19 is characterized by such a mapping cluster. A mapping cluster is characterized in particular by environmental information and the position of the vehicle 1.

[0070] Preferably, the trajectory points 19 are generated at a distance between 0.40 m and 1.50 m and in particular between 0.70 m and 1.20 m.

[0071] In the embodiment shown here, vehicle 1 has driven into garage 7 and parked there.

[0072] Preferably, it is provided that this information concerning at least one known travel trajectory 17, preferably also the trajectory points 19 and the recorded environmental information, is transmitted to or provided by Cloud 4.

[0073] In one embodiment, distances a, b, c from the travel trajectory 17 to the infrastructure facilities 7, 8, 9, 10, 11, 12, 13, 14, 15, and 20 are determined. These distances a, b, c can also be provided by Cloud 4 in another embodiment. It is possible for these distances a, b, c to be determined based on the available information.

[0074] In Fig. Figure 2 shows, for clarity only, a representation of the distances a, b, c. Distances to the other example infrastructure facilities mentioned can also be determined and provided.

[0075] In Fig. 3 is a corresponding representation as in Fig. Figure 2 shows a scenario in which vehicle 1 is parked, for example, in garage 7. It is intended, for instance, that vehicle 1 should travel along the known and stored trajectory 17 to a specific point, but not to a starting point 17a of the trajectory 17, at least semi-autonomously, and in particular fully autonomously. Rather, it is intended that maneuvering of vehicle 1 is planned, in particular taking into account the trajectory 17 and preferably also the acquired environmental information. This planning of the maneuvering of vehicle 1, starting from the current position of vehicle 1, here the position in garage 7, is preferably carried out by a user 21. The current position of vehicle 1 is, in particular, on the trajectory 17, here, for example, at the end of the trajectory 17.In particular, this planning is carried out by selecting specific environmental information and / or at least a zone around a specific trajectory point 19 of the driving trajectory 17 and / or a specific trajectory point 19 and / or by selecting an endpoint 19a. This specific environmental information and / or at least a specific trajectory point 19 and / or the endpoint 19a characterizes a specific sub-section 23 (shown as a dashed line directly next to the driving trajectory 17 for identification purposes) of the driving trajectory 17. In this context, it is possible that when selecting a zone, a trajectory section of the driving trajectory 17 between two, in particular adjacent, trajectory points 19 is selected as the endpoint of this sub-section 23. It is also possible that a point not lying on the driving trajectory 17 within such a zone is selected as the endpoint 19a of the sub-section 23.

[0076] In another embodiment, a trajectory point 19 on the driving trajectory 17 can be selected as the endpoint 19a of this subsection 23. It is possible that this representation is shown according to Fig. 3 is displayed on the display unit 6a. In particular, this selection of the subsection 23 can be made by user 21, for example by voice signals and / or by touching the display unit 6a on the display surface. Therefore, the length of the subsection 23, and thus also its beginning, which is defined in particular by the current position of the vehicle 1 on the travel trajectory 17, and its end, can be determined, in particular according to user preferences. Fig. In the embodiment shown in Figure 3, the vehicle 1 is intended to travel semi-autonomously, and in particular fully autonomously, from its position in garage 7 to a location 22. It is possible that an endpoint of this section 23 is defined by the trajectory point 19a. As already explained above, this point can be selected in one embodiment by touching the display unit 6a at a specific location when the driving trajectory 17 and the discrete trajectory points 19 are displayed. This characterizes the section 23. Additionally or alternatively, it is also possible that such a section 23, and in particular an endpoint of this section 23, is selected by selecting environmental information, for example, a landmark, in particular a boundary of an infrastructure facility 7, 8, 9, 10, 11, 12, 13, 20.This can also be done, for example, by voice signals and / or by touching the display unit 6a at the corresponding location where this infrastructure facility 7, 8, 9, 10, 11, 12, 13, 20 is shown.

[0077] If the vehicle 1 is subsequently driven from the position in garage 7, as shown in this example, to this trajectory point 19a, which was selected as the preferably endpoint of the subsection 23, the localization of the vehicle 1 during this traversal of the subsection 23 is preferably based on the environmental information, in particular the environmental information acquired when recording the driving trajectory 17 and / or currently acquired environmental information, as well as a SLAM procedure.

[0078] As can be seen, when traversing section 23, vehicle 1 preferably travels along, or substantially along, the already known trajectory 17. This traversal of section 23 can be in the opposite direction or in the same direction as the vehicle 1 traveled along trajectory 17. The trajectory 17 can preferably have a length of up to, for example, 250 m. When planning the maneuver based on section 23 and / or when moving vehicle 1 along section 23, the user 21 is preferably positioned outside of vehicle 1.

[0079] The Fig. Figures 4 to 6 show exemplary use cases where a respective endpoint 19a is selected for the subsection 23 that does not lie on the travel trajectory 17. In other words, the Fig. 4 to 6 are therefore use cases in which the vehicle 1 deviates from the previously planned trajectory 17 when driving at least semi-autonomously along section 23. The environment 6 in which the shown use cases take place can include a private property on which, for example, several known parking spaces and / or a garage structure 8 and / or several trees 14 are located.

[0080] In Fig. Figure 4 illustrates a situation where vehicle 1 initially records the driving trajectory 17. During recording, an endpoint 19a can be selected, which in this case is located in a known parking lot. This selected endpoint 19a marks the end of subsection 23 and characterizes it accordingly. The special feature here is that vehicle 1 can add the selected endpoint 19a to the driving trajectory 17 without actually having driven to endpoint 19a during the recording of the driving trajectory 17, i.e., during the training phase. The parking lot where endpoint 19a is located can be displayed to the driver of vehicle 1, for example, on a map view of the surroundings 6 in vehicle 1 during the recording of the driving trajectory 17. This display can, for example, be shown on the display unit 6a in vehicle 1.If the driver or user 21 is outside the vehicle 1 during this time, the display can also be shown on an external display unit 6a, as in connection with . Fig. 3 described. If the display unit 6a has a touch-sensitive screen, the selection of endpoint 19a in the parking lot can be made directly by touching the screen.

[0081] Once endpoint 19a is selected, section 23 can be planned towards endpoint 19a and driven at least semi-autonomously by vehicle 1 until endpoint 19a is reached.

[0082] In Fig. Figure 5 schematically shows a similar scenario to that in Fig. 4, where the selection of endpoint 19a takes place within a detection range 24 of the detection unit 18 of vehicle 1. This means that, according to the example shown here, the user 21 can select an endpoint 19a that is located, for example, within the field of view (FoV) of a camera sensor of vehicle 1. This has the advantage that the camera sensor images can also be used to determine whether the selected endpoint 19a is currently accessible. This means that it can be determined whether another vehicle is already there and / or whether the path to endpoint 19a is accessible or blocked. Once endpoint 19a is selected, it can be added to the driving trajectory 17 and navigated to. In the example of the Fig. 5. Vehicle 1 performs a multi-stage reversing parking maneuver, as indicated by the dashed section 23. If endpoint 19a is currently occupied, for example by another vehicle, it can still be saved as a possible endpoint 19a for the driving trajectory 17. If, during a subsequent drive, vehicle 1 approaches driving trajectory 17, in particular within 2 m of any trajectory point 19 of driving trajectory 17, the system can offer to follow driving trajectory 17. Along with this offer, possible endpoints 19a for driving trajectory 17 can be suggested, for example, the endpoint 19a described here in the parking space that was occupied during the previous drive.

[0083] Fig. Figure 6 shows a schematic top view of a further development of the [product / service] in connection with Fig. 5 described scenarios. In the example of the Fig. 6. Endpoints 19a outside the detection range 24 of the driving trajectory 17 are also stored. These endpoints 19a outside the detection range 24 may, for example, have been recorded and stored during a recording run of vehicle 1 that occurred even further in the past. According to the exemplary embodiment of the Fig. Each segment 23 of the driving trajectory 17 leads to a respective endpoint 19a. From the stored endpoints 19a, the user 21 or a parking assistance system of the vehicle 1 can select a currently desired endpoint 19a. This selection can be made, for example, based on predefined criteria, such as the distance of a respective endpoint 19a from an exit of the environment 6.

[0084] Overall, the examples show how flexibility can be increased in driving maneuvers that can be performed at least semi-autonomously based on a known driving trajectory.

Claims

[1] Method for at least semi-autonomous maneuvering of a vehicle (1) comprising the following steps: - Moving the vehicle (1) and recording a driving trajectory (17) of the vehicle (1) during this movement; - Providing environmental information about the environment (6) of the vehicle (1) along the driving trajectory (17); - Planning a maneuver of the vehicle (1) by selecting specific environmental information and / or at least a zone around a specific trajectory point (19) of the driving trajectory (17) and / or a specific trajectory point (19), wherein the specific environmental information and / or the specific zone and / or the specific trajectory point (19) characterizes a specific sub-section (23) of the driving trajectory (17); - Maneuvering the vehicle (1) by at least semi-autonomously driving the vehicle (1) over the section (23). [2] Method according to claim 1, wherein the selected environmental information and / or the selected zone defines an endpoint (19a) of the subsection (23) to which the vehicle (1) is to move at least semi-autonomously from its current position. [3] Method according to claim 2, wherein when selecting a zone a trajectory section of the driving trajectory (17) between two, in particular adjacent, trajectory points (19) is selected as the endpoint (19a), or a point in the zone not lying on the driving trajectory (17) is selected as the endpoint, or a trajectory point (19) on the driving trajectory (17) is selected as the endpoint (19a). [4] Method according to claim 3, wherein the endpoint (19a), in particular in the case that a point not lying on the driving trajectory (17) is selected as the endpoint, is selected during the recording of the driving trajectory (17), wherein the maneuvering of the vehicle (1) along the subsection (23) of the driving trajectory (17) which leads to the selected endpoint (19a) is planned during the recording of the driving trajectory (17). [5] Method according to claim 3 or 4, wherein the endpoint (19a) is selected within a detection range (24) of a detection unit (18) of the vehicle (1). [6] Method according to any one of claims 3 to 5, wherein a point located beyond a currently last trajectory point (19) is selected as the endpoint (19a). [7] Method according to any one of claims 3 to 6, wherein at least one known parking space is available in the vicinity (6) and / or in the zone and / or in the detection area (24) of the detection unit (18) of the vehicle (1), which is provided as an endpoint (19a) for selection. [8] Method according to claim 7, wherein the at least one known parking space is automatically selected as the endpoint (19a) from a plurality of known parking spaces based on predetermined selection criteria. [9] Method according to claim 8, wherein each of the parking spaces of the plurality of known parking spaces is checked for its current availability and / or accessibility before its selection as endpoint (19a). [10] Method according to one of claims 8 or 9, wherein during at least one further maneuver of the vehicle (1) in addition to the selected endpoint (19a) further parking spaces of the plurality of known parking spaces are considered as potential endpoints (19a) for the driving trajectory (17) depending on their current availability and / or accessibility. [11] Method according to any one of claims 3 to 10, wherein the endpoint (19a) is selected by an operating input of a user (21) of the vehicle (1) on a user operating interface of a display unit (6a) of the vehicle (1). [12] Method according to any one of claims 3 to 11, wherein the selected endpoint (19a) of the driving trajectory (17) is added as a new trajectory point (19). [13] Method according to claim 12, wherein the driving trajectory (17) is provided with the new trajectory point (19) for further, at least semi-autonomous maneuvers of the vehicle (1). [14] Method according to any one of claims 3 to 13, wherein several potential endpoints (19a) are stored for the driving trajectory (17), wherein one of the endpoints (19a) is selected as the actual endpoint (19a) for the driving trajectory (17). [15] Method according to one of the preceding claims, wherein the environmental information is acquired during the recording of the driving trajectory (17) with the at least one detection unit (18) of the vehicle (1). [16] Method according to one of the preceding claims, wherein landmarks, in particular objects, in the surroundings (6) are recorded as environmental information. [17] Method according to one of the preceding claims, wherein, during movement, the position of the vehicle (1) is determined, in particular on the basis of odometry data, and in particular environmental information is assigned to positions of the vehicle (1) along the driving trajectory (17). [18] Method according to claim 17, wherein several discrete trajectory points (19) are generated during movement, and each trajectory point (19) is characterized at least by an assignment cluster of environmental information and position of the vehicle (1). [19] Method according to claim 18, wherein the trajectory points (19) are generated at a distance between 0.40 m and 1.50 m, in particular between 0.70 m and 1.20 m from each other. [20] Method according to any of the preceding claims, wherein the environmental information and / or the zone is selected by the user (21) during planning. [21] Method according to one of the preceding claims, wherein the driving trajectory (17) and the environmental information and / or the zones and / or the trajectory points (19) and / or the endpoints (19a) are displayed on the display unit (6a) and are selected by manual input, in particular by touch-sensitive areas of the display unit (6a), and / or by voice signals. [22] Method according to one of the preceding claims, wherein the traversing of the subsection (23) is displayed in real time on a display unit (6a), in particular the trajectory (17) and the dynamic change of the already traversed section of the subsection (23) is displayed. [23] Method according to one of the preceding claims, wherein when driving through the subsection (23) the localization of the vehicle (1) is carried out on the basis of the environmental information and a SLAM method. [24] Method according to one of the preceding claims, wherein the subsection (23) is traversed in the opposite direction to the travel trajectory (17), or in the same direction as the travel trajectory (17). [25] Method according to one of the preceding claims, wherein the environmental information and / or the driving trajectory (17) is stored in a vehicle-external unit, in particular in a cloud (4). [26] Method according to one of the preceding claims, wherein the user (21) is positioned outside the vehicle (1) when planning the maneuver and when moving the vehicle (1) along the subsection (23). [27] Method, in particular a computer-implemented method, for determining a route for a vehicle (1) comprising the following steps: - Providing a driving trajectory (17) driven in an environment (6) in which the driving path is to be determined to a computing unit (2c); - Providing environmental information of the environment (6) of the vehicle (1) along the driving trajectory (17) to the computing unit (2c); - Providing input information from a user (21) relating to the selection of specific environmental information and / or at least a zone around a specific trajectory point (23) and / or a trajectory point (19) of the driving trajectory (17) and / or an endpoint (19a) of the driving trajectory (17), wherein the specific environmental information and / or the specific trajectory point (19) and / or the endpoint (19a) and / or the specific zone characterizes a specific sub-section (23) of the driving trajectory (17), in particular measured from a starting point or an endpoint of the driving trajectory (17), to the computing unit (2c); - Determining the subsection (23) of the travel trajectory (17) as the route depending on at least the input information using the computing unit (2c). [28] Electronic shunting system (2) for a vehicle (1) comprising a computing unit (2c) and an input unit, wherein the shunting system (2) is configured to perform a method according to one of the preceding claims. [29] Control unit (2d) for a vehicle (1) which is configured to generate control signals for at least one functional unit of a vehicle (1) depending on information such as that generated by a method according to any one of the preceding claims 1 to 18, in order to carry out the driving of the subsection (23).

Images