Trajectory planning for a vehicle

Abstract

The present invention relates to trajectory planning for a vehicle (10) in a predefined surrounding area (U) in which the vehicle (10) can assume vehicle positions (P) which are at different heights (H) from one another. First an entry position (E) of the vehicle (10) in the surrounding area (U) is determined and, proceeding from the entry position (E), a height (H) of the vehicle (10) relative to the entry position (E) is determined for, in each case, a current vehicle position (P) of the vehicle (10) by means of inclination measurement, as the surrounding area (U) is traveled. For the trajectory planning, the determined height (H) for the current vehicle position (P) is then used as a starting height and, proceeding therefrom, a destination position (Z) having an associated destination height is reached.

Description

[0001] 2023PF00700

[0002] 1

[0003] Trajectory planning for a vehicle

[0004] The present invention relates to a method for trajectory planning for a vehicle, particularly in a multi-level parking garage or multi-level parking structure. The invention also relates to a computer program, a control unit, and a vehicle for implementing such trajectory planning.

[0005] Modern vehicles offer the possibility of autonomous or automatic route planning using a driver assistance system. This system utilizes a trajectory that recreates a continuous path formed by vehicle positions, which the vehicle can then follow automatically.

[0006] In a three-dimensional environment, such as a multi-level or multi-story parking garage, vehicle positions can have different altitudes but a similar or identical horizontal position (azimuth). Therefore, a trajectory spanning multiple floors should be able to differentiate between the altitudes of the vehicle positions.

[0007] DE 10 2013 215 960 A1 discloses a method for determining the position information of a vehicle, comprising capturing movement information of the vehicle, determining a movement path of the vehicle based on the movement information and automatically determining floor information of a parking level or parking ramp in a parking garage on which the vehicle is located, based on the movement path.

[0008] US Patent 9,026,263 B2 discloses a method in which the internal geometry of a sensor position relative to a vehicle's rear axle is used to obtain more accurate results when determining position using GPS signals, even when GPS signal reception is disrupted. This allows, for example, the storage of elevation information for a vehicle's trajectory in the vicinity of a parking garage.

[0009] One disadvantage of the known state of the art is that height detection is often very inaccurate. The resolution of the described systems, especially 2023PF00700, is not sufficient.

[0010] 2

[0011] GPS systems typically have an error rate of + / - 10 meters, which in a parking garage, for example, means a difference of three or four floors.

[0012] The object of the present invention is to provide the most precise altitude information possible when planning a trajectory for a vehicle in an environment with different altitudes.

[0013] The problem is solved by the subject matter of the independent patent claims. Advantageous embodiments of the invention are disclosed by the dependent patent claims, the description, and the figures.

[0014] Accordingly, one aspect of the invention proposes a method for trajectory planning for a vehicle in a given environment, wherein the vehicle can assume vehicle positions in the environment that are at different altitudes relative to one another. In this context, "trajectory planning" refers specifically to recording or training a trajectory, and / or replaying or automatically tracking or following a previously stored or recorded trajectory. The aim is therefore to capture and / or reproduce the trajectory in order to use it for automated tracking. A trajectory to be recorded is hereinafter also referred to as a trajectory to be planned. A trajectory that has been recorded and is to be tracked is hereinafter also referred to as a pre-planned trajectory.

[0015] In the method according to the invention, an entry position of the vehicle into the surrounding area is first determined for trajectory planning. As the vehicle travels through the surrounding area, its altitude relative to the entry position is then determined for each current vehicle position within the surrounding area by means of inclination measurement. That is, the geographical altitude of each vehicle position is measured, with the geographical altitude of the entry position being used as the reference altitude or "zero point". For inclination measurement, the vehicle's inclination is preferably measured or determined internally by the vehicle itself. Therefore, no external signals, such as GPS signals, are necessary.

[0016] For trajectory planning, the inventive method now provides that for the automated tracking of a pre-planned trajectory or for recording 2023PF00700

[0017] 3. In the planning of a trajectory for the vehicle in the surrounding area to a target position described by a target elevation, the determined elevation for the current vehicle position is used as the starting elevation for reaching the target position with the target elevation. That is, the trajectory is planned or used taking into account the current elevation of the vehicle in the surrounding area.

[0018] The described method offers the advantage that estimating a vehicle's altitude for trajectory planning using inclination measurement can be particularly precise. In particular, this prevents a driver assistance system from suggesting or outputting incorrect or invalid trajectories for trajectory planning. Offering undesirable or invalid trajectories can occur, especially in symmetrical environments, when there are vehicle positions that are horizontally similar or identical but differ in altitude. If the vehicle's localization or position determination in such environments is based solely on objects or their characteristics in the surrounding area, it can happen that a trajectory is offered that is assigned to a different altitude than the vehicle's actual altitude.In a multi-story or multi-level parking garage, this can mean that the vehicle is directed, for example, to a parking space that is not intended for it or for which it is not authorized. Overall, the described procedure can thus stabilize and / or refine the vehicle's location.

[0019] The term "surrounding area" refers specifically to a spatially limited area. It is preferably a three-dimensional space traversable by the vehicle. Examples include a building or infrastructure with multiple levels or floors, such as a garage, a parking garage, a road bridge over a street, or a winding mountain pass. Preferably, satellite reception in the surrounding area is at least limited, so that vehicle localization using GNSS (Global Navigation Satellite System) is either not possible or only possible to a limited extent.

[0020] The entry position into the surrounding area describes, in this context, a vehicle position that relates to the vehicle's entry into the surrounding area. The entry position is therefore the starting point from which the vehicle travels through the surrounding area. "Traveling" 2023PF00700

[0021] The number 4 here means that the vehicle is moving or has moved into the surrounding area from its respective entry position. This means the vehicle is within the surrounding area. Driving within the surrounding area can be done manually or automatically. This means that a route or path the vehicle follows can be freely or manually selected by a driver, or the vehicle can be automatically controlled along a trajectory, for example, by a driver guidance system.

[0022] As the vehicle travels through or traverses the surrounding area, it assumes different vehicle positions or waypoints. Preferably, these vehicle positions form a continuous route. Position data is recorded for altitude measurement at each vehicle position. This position data is specified, for example, as geocoordinates, particularly three-dimensional geocoordinates. A spherical, Cartesian, or any other coordinate system can be used to describe the position data, especially the geocoordinates. The position data thus includes, in particular, a geographical description of the vehicle's altitude, which is referred to as elevation or altitude in spherical coordinate systems and as position in the z-direction in Cartesian coordinate systems.Additionally, the position data can also describe the vehicle's horizontal orientation, which is referred to as azimuth in spherical coordinate systems and as position in the xy direction in Cartesian coordinate systems. With respect to the vehicle's dimensions, the vertical orientation corresponds to a coordinate in the vehicle's vertical direction relative to the vehicle's position at the entry point. The horizontal orientation corresponds, for example, to coordinates in the vehicle's longitudinal and transverse directions relative to the vehicle's position at the entry point.

[0023] The vehicle's altitude is determined using tilt measurement. For this purpose, the vehicle can be equipped with at least one tilt sensor, which can, for example, determine the vehicle's pitch angle. The tilt sensor can be installed or integrated into an Inertial Measurement Unit (IMU). An IMU is a measurement unit that uses a combination of several inertial or tilt sensors to measure inclination. Accelerometers and / or yaw rate sensors, for example, can be used as inertial sensors. In contrast to altitude measurement using GPS, such tilt measurement allows for more precise results. Altitude can be determined with centimeter accuracy. 2023PF00700

[0024] 5

[0025] For trajectory planning, i.e., for the automated tracking or recording of the planned trajectory, the vehicle can, for example, include an electronic vehicle guidance system. An electronic vehicle guidance system can be understood, in particular, as an electronic system designed to guide a vehicle fully automatically or autonomously, especially without requiring driver intervention. To this end, the electronic vehicle guidance system uses the pre-planned or stored trajectory as a guide or route plan. In tracking mode, the vehicle performs all necessary functions along the pre-planned trajectory, such as steering, braking, and / or acceleration maneuvers, monitoring and recording road traffic, and reacting accordingly.In a training mode, the vehicle automatically performs all necessary functions, such as observing and recording road traffic, the environment, and driving maneuvers, as well as locating the vehicle within its surroundings, and generates the trajectory from the collected data. In this context, an electronic vehicle guidance system can also be understood as an advanced driver assistance system (ADAS), which supports the driver during partially automated or semi-autonomous driving.

[0026] Preferably, the method allows switching from a free-ride mode, in which the vehicle is manually controlled, to a planning mode, which concerns recording or replay. When switching from free-ride mode to planning mode, the current altitude is taken into account as the starting altitude. For replay, the vehicle can therefore only be offered a trajectory assigned to the current altitude. For recording, the vehicle's current altitude is specifically recorded as a reference point and stored for the trajectory.

[0027] The invention includes embodiments that offer additional advantages.

[0028] According to one embodiment, position data is acquired from the current vehicle position to record the trajectory to be planned. For each vehicle position along the trajectory until the target position is reached, the position data specifies an altitude relative to the entry position. This includes or 2023PF00700

[0029] 6. The position data includes the starting altitude in a starting position and the target altitude in a target position for the vehicle.

[0030] In this embodiment, the training mode or recording process is described in more detail. The trajectory being recorded is limited by the starting position and the target position as respective waypoints. The starting elevation is stored in the position data for the starting position, while the target elevation is stored for the target position. Preferably, the respective elevation relative to the entry position is also stored in the position data for all other waypoints. This results in a continuous driving path where the elevation is known at every position.

[0031] By recording or capturing the planned trajectory, it can be saved and thus made available as a pre-planned trajectory for replay. When using the replay, it is therefore possible to utilize the altitude information associated with the trajectory to suggest or track only trajectories that are relevant to the vehicle's current altitude at its position.

[0032] According to one embodiment, for the automated tracking of the pre-planned trajectory, a stored trajectory or, for example, a stored trajectory segment is selected or, preferably, suggested to the driver. The trajectory is described by previously known position data that specifies the altitude relative to the entry position for each vehicle position along the trajectory. This previously known position data can therefore be the position data acquired during training mode. Only a trajectory whose position data includes the starting altitude at a starting position and the target altitude at a target position for the vehicle is selected for automated tracking.

[0033] This embodiment describes the tracking mode or replay of the previously recorded trajectory. In tracking mode, a selection can be made from several trajectories. The selection process includes discarding trajectories whose position data does not contain the starting altitude, i.e., the altitude at the current vehicle position. Only if the starting altitude is present in the position data, and in particular if the starting position of the trajectory matches the current vehicle position, is the trajectory used for vehicle guidance. 2023PF00700

[0034] 7

[0035] According to one embodiment, it is provided that during automated tracking of the trajectory, i.e. while driving along the trajectory, the altitude for the current vehicle position on the trajectory is determined from the position data and a remaining route information from the current vehicle position to the target position is updated depending on the altitude and the target altitude.

[0036] In other words, the remaining route information is updated at predetermined intervals until the destination is reached. This involves a recalculation of the route information. Preferably, this route information can be displayed to the driver. The route information could include, for example, an arrival time or the remaining distance from the current vehicle position to the destination.

[0037] According to one embodiment, for measuring the vehicle's inclination while driving through the surrounding area, and in particular along the route travelled within that area, an inclination profile of the vehicle is recorded. This profile indicates at least one pitch angle of the vehicle for its current position. The vehicle's altitude is then determined from this inclination profile, specifically based on the pitch angle.

[0038] The tilt profile thus allows the creation of a characteristic map or function in which the pitch angle is assigned or specified for each vehicle position along the traveled route. The pitch angle is a positional angle along a pitch or lateral axis of the vehicle. The pitch angle therefore describes the change in position or angle of the vehicle along the vehicle's lateral direction as the axis of rotation. Preferably, the tilt profile can also include information about other positional angles, such as yaw and / or roll angles.

[0039] According to one embodiment, the number of pitch angles recorded up to that point is calculated to determine the altitude for the current vehicle position. For example, the recorded pitch angles from the entry position to the current vehicle position can be summed or integrated. In this way, information about the altitude, i.e., angular information, can be used. 2023PF00700

[0040] 8 to estimate the height of the vehicle, i.e. the floor or level on which the vehicle is located.

[0041] According to one embodiment, individual track segments are identified in this context to determine the elevation within the slope profile, depending on the magnitude of the pitch angles for the respective vehicle position. These track segments differ in their respective inclinations along a pitch axis of the vehicle according to a predetermined distinguishing criterion. Based on the track segments in the surrounding area, planes with different elevations relative to each other are then determined within that area.

[0042] In other words, the levels or levels of the surrounding area can be counted by alternating between level and inclined track sections. According to the distinguishing criterion, the track sections can be classified into different inclination categories. These inclination categories are defined by the distinguishing criterion. The categories are preferably described by two or more distinct, delimited ranges of values ​​for the pitch angle. Different pitch angle values ​​are assigned to each value range and are classified as identical or corresponding. Based on the inclination categories, a distinction can thus be made, for example, between level and inclined track sections.

[0043] According to one embodiment, an object recognition algorithm is executed and / or navigation satellite information is evaluated and / or map data is evaluated to determine the entry position.

[0044] To execute the object recognition algorithm, the vehicle may, for example, have a device for environmental sensing or detection. This environmental sensing device may be designed as a sensor system capable of generating sensor data or signals that map, display, or reproduce the device's surroundings. Examples include cameras, radar systems, lidar systems, or ultrasonic systems.

[0045] In this context, an object recognition algorithm can be understood as a computer algorithm that is capable of recognizing one or more objects within a provided input data set, for example, an input image. 2023PF00700

[0046] 9. To identify and locate objects, for example, corresponding boundary boxes or regions of interest (ROIs) can be defined, and in particular, each boundary box can be assigned a corresponding object class. The object classes can be selected from a predefined set of object classes. For the assignment, a corresponding confidence value or probability value can be determined, indicating that the identified object within the boundary box belongs to the corresponding object class. The algorithm can, for example, provide such a confidence value or probability for each object class for a given boundary box. The object class assignment can, for example, involve selecting or providing the object class with the highest confidence value or probability.Alternatively, the algorithm can simply define the boundary boxes without assigning a corresponding object class. The acquisition of the input data set for object recognition can be achieved, for example, using an environmental sensing device, particularly an environmental sensor system.

[0047] The object recognition algorithm can be implemented particularly favorably as a computer vision algorithm or using machine learning. This is an image processing or image analysis algorithm that is or has been trained through machine learning and can, for example, be based on an artificial neural network, especially a convolutional neural network.

[0048] To receive and process navigation satellite information, the vehicle can, for example, include a GNSS positioning device. This device can include one or more GNSS-enabled antennas. Information from a global navigation satellite system, such as GPS or Galileo, is used to determine that the vehicle is in the entry position. This can involve, for example, using a satellite signal that contains not only the vehicle's position but also positional information about the surrounding area, thus providing a possible entry point.

[0049] Map data, in this context, refers specifically to data describing a road map with a network of drivable roads or paths for vehicles. Preferably, the map data can be stored in the form of an HD map (HD: High Definition). The map data can be, for example, 2023PF00700

[0050] 10

[0051] This example involves navigation data provided by a device for navigating the vehicle, such as a navigation system. The map data can, for example, contain information, particularly the local geographical boundaries of the surrounding area, so that if the current vehicle position is known, it can be identified as the entry point into that area.

[0052] According to one embodiment, for the automated tracking or recording of the planned trajectory, position data is considered in addition to the altitude relative to the current vehicle position. This data is acquired or used to specify environmental features for each vehicle position along the trajectory until the target position is reached. In other words, the position data also specifies or stores environmental features or markers for each vehicle position. These environmental features can be, for example, object classes that uniquely identify an object in the environment. The environmental features can be acquired, for example, using environmental recognition, such as the object recognition algorithm described earlier.The detected or stored environmental features can be used, for example, in a feature matching algorithm, also called a feature matching algorithm, for example for replay.

[0053] Preferably, the recognition of environmental features is based on the so-called vSLAM algorithm (visual simultaneous localization and mapping). This is a well-known system, or rather a known algorithm, for visual environment recognition using a camera. Using image information provided by the camera during environmental observation, software constructs a constantly updated three-dimensional map of the space, i.e., the environment. Objects in the space are classified in training mode and can be recognized again in tracking mode (feature matching).

[0054] According to one aspect, the invention relates to a computer program product comprising instructions which, when the program is executed by a computer, cause it to execute or carry out the trajectory planning method for a vehicle described above.

[0055] The computer program product can, for example, be provided or supplied as a storage medium or storage unit on which the computer program 2023PF00700 is stored.

[0056] 11 is stored or saved. A storage unit can be volatile data storage, for example as dynamic random access memory (DRAM) or static random access memory (SRAM), or as non-volatile data storage, for example as read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory or flash EEPROM, ferroelectric random access memory (FRAM), or magnetoresistive random access memory.It can be designed as MRAM (magnetoresistive random access memory) or as phase-change random access memory, PCRAM (phase-change random access memory). Alternatively, the memory unit can also be provided or delivered as a downloadable file from a server on a network.

[0057] According to one aspect, the invention relates to a computing unit for a vehicle, which is configured to execute or perform the previously described method for trajectory planning for a vehicle. The computing unit can, for example, be comprised of a control device, such as an ECU (electronic control unit) of the vehicle. The computing unit can, for example, be functionally assigned to an electronic vehicle guidance system.

[0058] In this context, a computing unit can be understood, for example, as a data processing device with processing circuits. A computing unit can therefore perform arithmetic operations to process data. These operations can also include indexed access to a data structure, such as a lookup table (LUT). A computing unit can, in particular, comprise one or more computers, one or more microcontrollers, and / or one or more integrated circuits, such as one or more application-specific integrated circuits (ASICs), one or more field-programmable gate arrays (FPGAs), and / or one or more systems on a chip (SoCs). The computing unit can also contain one or more processors, such as one or more microprocessors or one or more central processing units (CPUs).

[0059] 12

[0060] (English: central processing unit), one or more graphics processing units (GPUs), and / or one or more signal processors, in particular one or more digital signal processors (DSPs). The processing unit may also include a physical or virtual cluster of computers or other units of the aforementioned type. A processing unit may also include one or more hardware and / or software interfaces and / or one or more storage units.

[0061] According to one aspect, the invention also relates to a vehicle with a computing unit as previously described. Preferably, the vehicle is designed as a motor vehicle, in particular as a motor car, such as a passenger car, truck, bus, or motorcycle. Optionally, in addition to the computing unit, the vehicle may also include the previously described device for determining the entry position, a device for measuring inclination, such as an IMU, and / or an electronic vehicle guidance system for trajectory planning.

[0062] Further embodiments of the computer program product, the computing unit, and / or the vehicle according to the invention follow directly from the various configurations of the method according to the invention, and vice versa. In particular, individual features and corresponding explanations as well as advantages relating to the various embodiments of the method according to the invention can be transferred analogously to corresponding embodiments of the computer program product, the computing unit, and / or the vehicle according to the invention.

[0063] The invention is explained in more detail below with reference to specific embodiments and associated schematic drawings. In the figures, identical or functionally equivalent elements may be designated with the same reference numerals. The description of identical or functionally equivalent elements is not necessarily repeated with respect to different figures.

[0064] The figures show:

[0065] Fig. 1 a schematic representation of a vehicle with a driver assistance system for trajectory planning; and 2023PF00700

[0066] 13

[0067] Fig. 2 shows a schematic representation of a multi-story car park, in which trajectories are available for trajectory planning for a vehicle on the different floors;

[0068] Fig. 3 shows a schematic process flow diagram with individual process steps for trajectory planning for the vehicle, for example in the parking garage.

[0069] Fig. 1 shows a schematic representation of a vehicle 10, for example, a passenger car. The vehicle 10 includes a driver assistance system 11. The driver assistance system 11 is, for example, a parking assistant that can automatically guide or steer the vehicle 10 to a desired parking space. For this purpose, the driver assistance system 11 performs trajectory planning. In this case, the driver assistance system 11 uses a processing unit 12, environmental sensors 13, and tilt sensors 14 for trajectory planning. The processing unit 12 is, for example, comprised of an electronic control unit of the driver assistance system 11. In the embodiment shown in Fig. 1, the environmental sensors are, for example, represented as a camera or camera system with one or more cameras for environmental detection.

[0070] Trajectory planning, in this context, refers specifically to the recording or training and / or the playback or tracking of a recorded trajectory T. This means that the driver assistance system 11 can record driving maneuvers related to a trajectory T, i.e., a route that the vehicle 10 is to follow. The next time the vehicle is, for example, near the recorded trajectory T, and especially when a user, such as a driver, accepts the playback of the trajectory T or the driving maneuver (replay offer), the maneuver can be played back or repeated. The driver assistance system 11 then performs all necessary driving maneuvers, such as braking, accelerating, and / or steering the vehicle, preferably independently or automatically, without the driver needing to intervene in the vehicle controls.Typically, a parking assistant can store more than one, i.e., two or more trajectories T, which can be used to reach the same or different target positions. 2023PF00700.

[0071] 14

[0072] For trajectory planning, the driver assistance system 11 can, for example, use so-called SLAM technology. SLAM technology describes a system and / or an algorithm for environmental perception, in which suitable environmental sensors 13, such as cameras, lidar, or radar sensors, are used to capture the surroundings and create a real-time three-dimensional map of the space. Objects in the map are classified during training, i.e., during recording, and can be recognized during replay (feature matching). This feature matching makes it possible to locate the vehicle 10 within its surroundings, i.e., an area U in which the vehicle is currently located.

[0073] In environments U that are symmetrical or at least partially identical, the use of SLAM technology can lead to undesirable or incorrect localization results. This is because, in a symmetrical environment U, there are areas where the environmental features are identical or at least so similar to those of other areas that the driver assistance system 11 cannot distinguish between them. As a result, during trajectory planning, the driver may be offered a trajectory T that was not originally intended for the environment or for a different vehicle position than the one in which the vehicle 10 is currently located.

[0074] An example of a symmetrical environment U is shown in Fig. 2. The environment U is, for example, a parking garage G with several different floors or levels L0, L1, L2, L3, and L4. In such a multi-level parking garage G, the structure of the individual floors, as shown in Fig. 1, is essentially symmetrical, and the access areas or ramps to the individual levels L0 to L4 are generally indistinguishable to the driver assistance system.

[0075] In the scenario shown in Fig. 2, for example, several different trajectories T1, T2, T3, T4, and T5 are recorded and / or stored for tracking by vehicle 10. Each of the trajectories T1 to T5 is described by individual position data that define a continuous path of vehicle positions for vehicle 10. Each of the trajectories T is bounded by a starting position S and a destination position Z. Trajectory T1, for example, shows a route for vehicle 10 from level L0 (ground floor) to a destination position Z, such as a parking space, on level L1. Trajectory T2 shows a route for vehicle 10 from a starting position S.

[0076] Trajectory 15 begins at the entrance to level L1 and ends at the same destination position Z as trajectory T1, namely the parking lot on level L1. Similarly, trajectories T3 and T4 show a route with a starting position S in the entrance area to level L2 and level L3, respectively, and an associated destination position Z in level L2 and level L3, respectively. Trajectory T5 shows a route with a starting position S in level L4, ending at the same destination position Z as trajectory T4, namely a parking lot in level L3.

[0077] In the schematic representation according to Fig. 2, it is now clearly evident that the trajectories T are specified by position data that overlap in azimuth or horizontal position (xy-direction) or are at least similar in sections, but differ in their altitude H (z-direction). Thus, when traversing the surrounding area U, the vehicle 10 can assume positions P that are at different altitudes H relative to each other. In the exemplary embodiment shown in Fig. 2, three different positions P for the vehicle 10 are depicted, in which the vehicle 10 is at different altitudes H. Here, altitude H refers in particular to a height or elevation angle relative to a predetermined reference altitude.The reference elevation, which in this case is used, for example, as the zero point or starting point, is given, for example, at an entry position E of the vehicle 10 into the surrounding area U on the lowest floor, i.e. level L0.

[0078] If, for example, only azimuth information is considered for trajectory planning based on SLAM technology, such as through feature matching, this can lead to incorrect results or suggestions. Localization using feature matching can therefore lead to limitations in the localization of vehicle 10. For example, if vehicle 10 is located in the approach area to level L1, the symmetrical layout of levels L0 to L4 might result in the driver being automatically suggested trajectories T1, T2, T3, and T4. However, of the four trajectories T, only the first two are correct with respect to the vehicle's position P if the altitude H information is disregarded.

[0079] Therefore, in symmetrical environments U, such as parking garage G, it can be useful to record or consider information about the altitude of vehicle 10. One approach for this is GPS-2023PF00700.

[0080] 16

[0081] to use information or another type of satellite information for satellite-based navigation. However, the disadvantage of using GPS information is that GPS altitude resolution is imprecise and typically ranges from several meters to approximately + / - 10 meters. This is insufficient for accurate altitude determination in a scenario like a parking garage G, where the levels are usually less than 5 meters apart in height H. Furthermore, there are surrounding areas U, such as the parking garage shown in Fig. 2, where GPS data is no longer available, for example, due to a lack of satellite reception.

[0082] Therefore, in such scenarios, it is necessary to consider an alternative method for determining the vehicle's altitude (10) for trajectory planning. In this case, altitude and / or angle information for the recorded or to-be-recorded trajectory T, starting from the entry position E, is used to estimate or specify the altitude, and in particular the relative altitude H, of the vehicle (10) for each vehicle position P, in addition to feature matching for locating the vehicle (10). In the context of the parking garage shown in Fig. 2, this essentially means that the altitude H for the recorded or to-be-recorded trajectory T should be included, and the entry position E, i.e., the entrance to the parking garage (G), should be recorded.This combined information from altitude and entry position E can be used to assign the floor on which vehicle 10 is currently located to the current vehicle position.

[0083] Information about the elevation H is obtained by means of inclination measurement. That is, the inclination of the vehicle 10 relative to the inclination of the road surface at the entry position E is measured or recorded. For this purpose, the vehicle 10 includes the inclination sensor 14 as a device for inclination measurement. In Fig. 1, this is shown, for example, as an inertial measurement unit (IMU). For inclination measurement, the IMU preferably comprises one or more inclination sensors or inertial sensors, such as an accelerometer and / or a yaw rate sensor.

[0084] The consideration of the elevation H for trajectory planning will be described in more detail using the exemplary embodiment in Fig. 3. Fig. 3 shows a process flow diagram for an exemplary trajectory planning method. The 2023PF00700

[0085] 17

[0086] The procedure according to Fig. 3 can, for example, be carried out or performed using the computing unit 12.

[0087] For trajectory planning, the procedure first determines the entry position of each vehicle 10 in the surrounding area U in step S1. To identify the entry position E, the processing unit 12 can, for example, execute a computer vision algorithm or a machine learning algorithm. This means that the processing unit 12 can, for example, apply an image analysis algorithm or an object recognition algorithm to image data provided by the environmental sensors 13, such as the associated camera, when capturing the environment. Alternatively, it is conceivable, for example, that GPS information or information from a navigation map or map data could be used to identify the entry position E. The procedure then continues in step S2.

[0088] In step S2, starting from the entry position during or while traversing the surrounding area U, the elevation H relative to the entry position E is determined for the current vehicle position P of vehicle 10 by means of tilt measurement. The elevation H is determined, for example, using the tilt sensor 14.

[0089] This allows, for example, the recording of a tilt profile of vehicle 10, which specifies at least one pitch angle of vehicle 10 for the current vehicle position P. This means that, starting from the entry position E to the current vehicle position P, a map is created in which the pitch angles are recorded. To determine the altitude H, the number of pitch angles recorded up to that point can now be calculated together for the current vehicle position. That is, the pitch angles can, for example, be summed or integrated.

[0090] Alternatively or additionally, it is conceivable, for example, to identify sections in the gradient profile whose pitch angles differ according to a predetermined distinguishing criterion. This distinguishing criterion specifies, for example, a respective range of values ​​for the pitch angle magnitude within which the pitch angles are considered identical or congruent. Thus, the gradient profile contains alternating sections, and therefore travel distances or trajectory areas, with pitch angles that differ from one another. These value ranges are preferably selected such that, between approach areas to the individual 2023PF00700

[0091] 18

[0092] Levels L0 to L4, which are inclined relative to the road surface at entry position E, can be distinguished from level parking areas, i.e., those areas in the parking garage that are parallel or level relative to the road surface at entry position E. By analyzing the changes between level and inclined sections in the gradient profile, the individual levels or floors of the parking garage can be counted, and thus the vehicle's height H at the respective vehicle position can be determined. The process then continues in step S3.

[0093] In step S3, trajectory planning takes place. This means that a pre-planned trajectory T is automatically followed, or a trajectory T to be planned is recorded. For this purpose, the system can switch from a free-travel mode B1, to which steps S1 and S2 are assigned, for example, to a planning mode B2, to which step S3 is assigned. During trajectory planning according to step S3, the determined altitude H for the current vehicle position P is used as the starting altitude for reaching a target position Z assigned to trajectory T at the target altitude. As can be seen in Fig. 2, for example, trajectories T1 and T2 are assigned the target altitude H1. Trajectory T3 is assigned the target altitude H2, and trajectories T4 and T5 are assigned the target altitude H3.

[0094] To record the planned trajectory T, the vehicle 10, in particular the driver assistance system 11, is operated in a recording mode or training mode. Starting from the current vehicle position P, position data is recorded that specifies an altitude H relative to the entry position E for each vehicle position P on the trajectory until the target position Z is reached. The position data includes the starting altitude at a starting position S and the target altitude at the target position Z for the vehicle 10.

[0095] For the automated tracking of the pre-planned trajectory T, the vehicle 10, in particular the driver assistance system 11, is operated in a replay mode or tracking mode. Here, a stored or saved trajectory or a segment thereof is selected as the pre-planned trajectory T, and in particular, suggested to the driver for selection. The trajectory T is described or specified by previously known position data, as described above. Preferably, only a trajectory T is selected whose position data includes the starting altitude at the starting position S and the target altitude at the target position Z for the vehicle. That is, only such trajectories T are available for selection.

[0096] 19 are taken into account, whose assigned altitudes H correspond to the altitude H of vehicle 10 at the current vehicle position P.

[0097] When traveling along the selected trajectory T, remaining route information is cyclically updated, preferably based on the known altitude H at the current vehicle position P. Depending on the altitude H, an arrival time or, for example, the remaining distance to reach the destination is displayed to the driver while following the trajectory T.

[0098] Overall, the described method allows elevation data to be used to identify the floor on which vehicle 10 is located, even without GPS information, and thus to pinpoint its altitude. This elevation information can be used in conjunction with the feature matching capabilities of SLAM technology to prevent, for example, the provision or use of unwanted or incorrect trajectories that actually belong to a different floor. As a result, the vehicle localization results can be more reliable and stable.

[0099] Overall, the examples demonstrate how to avoid incorrect route suggestions in a multilevel parking garage scenario, particularly when no GPS signals are available.

Claims

2023PF00700 20 Patent claims 1. A method for trajectory planning for a vehicle (10) in a given surrounding area (U), wherein the vehicle (10) can assume vehicle positions in the surrounding area that are at different elevations (H) relative to each other, characterized in that an entry position (E) of the vehicle (10) into the surrounding area (U) is determined, and starting from the entry position (E), when traversing the surrounding area (U), an elevation (H) of the vehicle (10) relative to the entry position (E) is determined for each current vehicle position (P) of the vehicle (10) by means of an elevation measurement, wherein for trajectory planning for the automated following of a pre-planned or for recording a trajectory (T) to be planned by the vehicle in the surrounding area (U) to a target position (Z) which is described by a target elevation,The determined altitude (H) for the current vehicle position (P) is used as the starting altitude to reach the target position (Z) at the target altitude.

2. Method according to claim 1, wherein for recording the trajectory (T) to be planned starting from the respective current vehicle position (P), position data are recorded which specify for each vehicle position (P) up to reaching the target position (Z) an altitude (H) relative to the entry position (E), wherein the position data include the starting altitude in a starting position (S) and the target altitude in the target position (Z) for the vehicle (10).

3. A method according to one of the preceding claims, wherein for the automated tracking of the pre-planned trajectory (T) a stored trajectory (T) is selected which is described by previously known position data which specifies the altitude (H) relative to the for each vehicle position (P). 2023PF00700 21 Specify entry position (E), selecting only one trajectory (T) whose position data includes the starting altitude in a starting position (S) and the target altitude in the target position (Z) for the vehicle (10).

4. Method according to claim 3, wherein, in the automated tracking of the trajectory (T), the altitude (H) for the current vehicle position (P) on the trajectory (T) is determined from the position data and a remaining route information from the current vehicle position (P) to the target position (Z) is updated depending on the altitude (H) and the target altitude.

5. Method according to one of the preceding claims, wherein for the inclination measurement when driving through the surrounding area (U) an inclination profile of the vehicle (10) is recorded, which indicates at least one pitch angle of the vehicle (10) for the respective current vehicle position (P), and the respective altitude (H) is determined from the inclination profile.

6. Method according to claim 5, wherein the quantity of pitch angles recorded up to that point are calculated together to determine the altitude (H) for the respective current vehicle position (P).

7. Method according to claim 5 or 6, wherein, in order to determine the elevation (H) in the inclination profile, individual track sections are identified as a function of the size of the pitch angles for the respective vehicle position (P), which differ in their respective inclination along a pitch axis of the vehicle (10) according to a predetermined distinction criterion, and planes with different elevations (H) relative to each other are determined as a function of the track sections in the surrounding area (U).

8. Method according to one of the preceding claims, wherein an object recognition algorithm is executed to determine the entry position (E), and / or navigation satellite information is evaluated, and / or map data is evaluated. 2023PF00700 22 9. Method according to one of the preceding claims, wherein, in addition to the altitude (H) starting from the respective current vehicle position (P), position data are taken into account for the automated tracking of the pre-planned or the recording of the trajectory to be planned (T), which are available for each vehicle position (P) until the target position (Z) is reached. Specify environmental characteristics.

10. Computer program product comprising instructions which, when the program is executed by a computer, cause it to execute the method for trajectory planning for a vehicle (10) according to one of claims 1 - 9.

11. Computing unit (12) for a vehicle (10) which is configured to execute the trajectory planning method for a vehicle (10) according to one of claims 1 - 9.

12. Vehicle (10) with a computing unit (12) according to claim 11 .

Images