Motor vehicle and method for motion planning of an autonomous motor vehicle for a parking maneuver

Abstract

The invention relates to a motor vehicle (10) and a method for motion planning of an autonomous motor vehicle (10) for a parking maneuver. The method comprises determining (S1) an initial state and environment of the motor vehicle (10); determining (S2) parking options, wherein a target state is determined for each parking option; and determining (S3) configuration samples (20, 20', 20") in a configuration space (18) depending on a motion model, wherein respective sequences of configuration samples (20, 20', 20") define possible paths (22, 22') of the motor vehicle (10) through the configuration space (18) to reach the respective target state, wherein the paths (22, 22') for the respective parking options are determined simultaneously.Furthermore, the paths (22, 22') are evaluated (S4); a respective path (22') is selected (S5) based on the evaluation and a respective movement trajectory (24) is generated for the selected paths (22'); a ranking of the generated movement trajectories (24) is provided; and a movement trajectory (24) is selected (S7) from the ranking.

Description

[0001] The invention relates to a method for planning the movement of an autonomous motor vehicle for a parking maneuver. Furthermore, the invention relates to a motor vehicle configured to carry out the method.

[0002] In autonomous vehicles, parking options are often optimized for specific use cases or are very computationally intensive. Thus, complete spatial coverage of all parking options for any given situation is often not possible, and / or methods for motion planning during a parking maneuver are not suitable for real-time use due to the computational intensity.

[0003] The object of the invention is to improve motion planning for a parking process.

[0004] This problem is solved by the independent patent claims. Advantageous further developments are disclosed in the dependent patent claims, the following description, and the figures.

[0005] The invention is based on the idea of ​​using a sample-based algorithm for motion planning. This allows paths to each identified parking space in the surrounding area to be determined simultaneously, allowing for rapid response to changes in the environment if an originally planned trajectory is no longer feasible. A motion model of the motor vehicle is provided for motion planning, which generates only sample values ​​that are possible based on an initial orientation and position of the motor vehicle.

[0006] One aspect of the invention relates to a method for planning the movement of an autonomous motor vehicle for a parking maneuver. The method comprises the following steps: determining an initial state of the motor vehicle and the surroundings of the motor vehicle by a sensor device of the motor vehicle; determining parking options from the determined surroundings of the motor vehicle by a computing device of the motor vehicle, wherein a target state is determined for each parking option; and determining configuration samples in a configuration space by the computing device. The configuration samples are generated depending on a movement model of the motor vehicle, thus excluding impermissible movements of the motor vehicle.wherein the configuration space represents the environment of the motor vehicle and the configuration samples comprise possible positions of the motor vehicle with a respective associated state of the motor vehicle, wherein the configuration samples are determined using a sampling-based algorithm in predefined sampling steps, wherein in each sampling step a plurality of configuration samples are determined in parallel using at least one graphics processing unit of the computing device, wherein the configuration samples generated in each sampling step are based on the state and position of the configuration samples of the previous sampling step, wherein respective sequences of configuration samples define possible paths of the motor vehicle through the configuration space to the respective parking options for reaching the respective target state,wherein the possible paths for the respective parking spaces are determined simultaneously by the computing device. Furthermore, the possible paths of the motor vehicle to the respective parking space are evaluated after each scanning step by the computing device depending on a path criterion, a respective path to the respective parking space is selected from the possible paths based on the evaluation by the computing device, wherein a respective movement trajectory of the motor vehicle, comprising a speed profile of the motor vehicle to the respective parking space, is generated from the respectively selected paths,The computing device evaluates the generated movement trajectories based on a maneuver criterion and provides a ranked list of the movement trajectories based on the evaluation. Furthermore, a movement trajectory is selected from the ranked list and the selected movement trajectory is provided to a control device of the motor vehicle for moving the autonomous motor vehicle to the target state in the parking space.

[0007] In other words, a sensor device of the motor vehicle can initially capture sensor data that describe the surroundings of the motor vehicle and an initial state of the motor vehicle. For example, the surroundings of the motor vehicle can be defined as static objects, in particular road markings and / or other vehicles; dynamic objects, in particular moving vehicles and / or pedestrians; and parking options located in the surroundings of the motor vehicle. The initial state of the motor vehicle can be defined as, for example, an orientation of the motor vehicle in the surroundings, a steering angle, and / or a speed of the motor vehicle. The sensor device can, in particular, comprise steering angle sensors, speed sensors, GPS sensors, radar sensors, laser sensors, in particular lidar sensors, ultrasonic sensors, and / or camera sensors.

[0008] The parking options in the identified surroundings can be determined, for example, using an image recognition algorithm. For each parking option, a target state can be determined that the motor vehicle should assume when parking in the respective parking space. This means that the target state includes, for example, a position and / or orientation of the motor vehicle in the respective parking space. Preferably, a plurality of parking options is detected, i.e., two or more. A parking option refers to a possible parking position for the motor vehicle, for example, a parking space or a parking space.

[0009] From the acquired sensor data, a computing device of the motor vehicle, which can be provided, for example, as a vehicle computer, can then generate a configuration space that virtually represents the surroundings of the motor vehicle and in which possible paths of the motor vehicle to the respective parking spaces can be determined. The paths are generated using configuration samples that include a respective position of the motor vehicle in the configuration space and an associated state. This means that the configuration samples can include the respective position in the configuration space and the associated configuration or the associated state of the motor vehicle. In particular, the state includes the respective steering angles required at the corresponding positions.For this purpose, a sampling-based algorithm can be used, which does not generate complete trajectories through space, but rather individual movement points in space, which are based in particular on the previous sampling points, i.e., the previous states and positions of the motor vehicle. This generation of configuration samples is preferably carried out by means of a graphics processing unit, in particular one or more graphics processors (GPUs), which enables massively parallel generation of the configuration samples. Thus, a multitude of possible movements through the configuration space can be determined, in particular at least one, preferably several, for each parking option. In particular, this allows for an improved determination of the parking process for a variety of situations.Preferably, multiple sampling steps can be performed, whereby subsequent configuration samples can be determined for each configuration sample depending on the previously determined state of the motor vehicle and the position. For example, such a sampling-based method with a configuration space and configuration samples is known from WO 2022 / 268323 A1.

[0010] To limit the computational effort for the sampling algorithm and provide improved motion planning, the configuration samples can be generated using a motion model of the motor vehicle, thus limiting the number of configuration samples in the configuration space. In particular, only permissible movements of the motor vehicle can be defined and permitted in the motion model, so that configuration samples that contradict the motion model are not generated or considered. In particular, only configuration samples that are actually achievable by a steering movement of the motor vehicle can be generated, thus excluding, for example, movements of the motor vehicle in the transverse direction of the vehicle.The motion model can use the previous state and position as input variables and thus determine the configuration samples based on them in the subsequent sampling step.

[0011] A respective sequence of configuration samples forms a possible geometric path of the motor vehicle through the configuration space, whereby the configuration samples for the respective target state of each parking space, and thus all possible paths to each parking space, are determined simultaneously. The "geometric path" here refers to the respective spatial distance to the parking spaces, which, however, does not yet include a speed profile. However, the individual configuration samples can include speed information, which is required, for example, to move from one configuration sample to the next, for example, if such a movement requires steering while the motor vehicle is stationary.The simultaneous determination of the configuration samples or the possible paths to the respective parking spaces means that the paths for all parking spaces are determined in a single computational sequence of sampling steps, rather than in separate individual calculations for each parking space. Since the configuration samples determine all possible movements through the configuration space and the configuration samples are based on the state of the previous sampling step, computational intensity and / or time can be reduced, as movements through the configuration space to different parking spaces can at least partially use the same configuration samples, which thus do not need to be generated again.

[0012] After all possible paths have been generated in a scanning step, they can be evaluated for each scanning step depending on a path criterion. The path criterion can, for example, be a condition and / or a rule that the respective path should or must fulfill. For example, the paths can be evaluated depending on a distance, a number of required steering movements or changes of direction of the vehicle and / or based on a maintained distance from static and / or dynamic objects. Paths that do not fulfill the path criterion can preferably be truncated or not pursued further. Particularly preferably, the determination of the configuration sampling values ​​and a subsequent evaluation of the resulting paths can thus be carried out in each scanning step, wherein in particular the best-rated paths can be pursued iteratively.This can be done until the respective paths lead to the target state of the respective parking option.

[0013] Once paths to the respective parking options have been provided, a path can be selected for each parking option based on the last rating, preferably the best-rated path. This can be determined, in particular, using a loss function (“loss function” or “cost function”).

[0014] Once a path to each parking space has been selected, a movement trajectory to each parking space can be created using a velocity profile. In other words, the movement trajectory represents the geometric path through space, including the associated velocity at each position along the path. This means that a movement trajectory can be provided for each parking space.

[0015] The provided movement trajectories can then be evaluated by the computing device based on a maneuver criterion. The maneuver criterion can be a further condition and / or rule by which the provided movement trajectories can be compared with one another and thus evaluated. In particular, the maneuver criterion can include the duration with which a movement trajectory can be executed, whereby a short duration can be evaluated more favorably than a long duration. Alternatively or additionally, for example, a required fuel consumption for the respective movement trajectories can also be determined using the maneuver criterion, whereby low fuel consumption can be evaluated more favorably than high fuel consumption.It is also possible to evaluate, for example, the target state of the vehicle after completing each movement trajectory and thus, for example, how much space is available around the vehicle to exit. In other words, the maneuver criterion can be used to evaluate the movement planning for each parking option to determine which parking option is most suitable.

[0016] The movement trajectories evaluated using the maneuver criterion can then be ranked based on the rating, from which a movement trajectory and thus a parking option can be selected. The ranking can, for example, be displayed to a user, who can then select a movement trajectory or a parking option from the ranking. Alternatively or additionally, the selection can also be performed automatically by the computing device, which can, for example, use the best-rated movement trajectory from the ranking.

[0017] Finally, the selected movement trajectory can be provided to a control device of the motor vehicle, which can be designed to steer the motor vehicle autonomously into the parking space by means of the selected movement trajectory.

[0018] The invention offers the advantage that a decision for the best parking option from among many available parking options can be made automatically. Since the determination of the movement trajectories for each parking option is based on the same or simultaneous calculation, which maps all permissible movements of the vehicle in space, the computational effort can be reduced compared to separate calculations for movement planning, i.e., for each parking option individually, thus enabling real-time capability. Furthermore, if the environment changes, a quick switch to another movement trajectory is possible, since all possible movements in space have been determined.

[0019] The invention also includes further developments which result in additional advantages.

[0020] A further development provides that the respective state of the motor vehicle comprises at least an orientation of the motor vehicle in the environment and a steering angle, wherein the motion model does not generate configuration samples that are excluded by the orientation of the motor vehicle and the steering angle. However, multiple movements in the forward and reverse directions, a change of direction at any point in the environment, and even a change in the steering angle while the motor vehicle is stationary can be taken into account. In particular, this prevents the generation of multiple, successive configuration samples that provide paths that are completely impossible due to the orientation and the current steering angle of the motor vehicle.The configuration samples comprising the state can build on one another, in particular in several consecutive sampling steps, whereby it can thus be determined which configuration samples are possible based on the previous configuration sample. The respective previous state thus forms the basis for a new exploration. The first state is the initially determined state of the motor vehicle, and the subsequent states are determined using the configuration samples. Preferably, the respective configuration sample and / or state can also include a possible speed of the motor vehicle, a distance or a position relative to previous configuration samples, and / or a steering angle rate or a steering angle change rate, by means of which it can be determined whether a deceleration and / or a change in direction of the motor vehicle's movement is necessary.For example, it can be determined whether steering while the vehicle is stationary is necessary for the motion model to generate subsequent configuration samples. This refinement offers the advantage of saving computing capacity, which contributes to improved real-time capability.

[0021] A further development provides that the motion model is a single-track model. The single-track model provides a model of a motor vehicle for lateral dynamics, by means of which movements of the motor vehicle can be described. Preferably, the single-track model takes into account a front axle control or steering of the motor vehicle in a forward and / or reverse direction. This means that, for the provision of the configuration sample values, movements of the motor vehicle whose front axle is movable are permitted, and thus movements in the forward and / or reverse directions can be determined based on the single-track model. This makes it possible to provide a preferred embodiment of a motion model.

[0022] A further development provides that the parking options in the detected surroundings of the motor vehicle are determined using an image recognition algorithm, whereby only parking options are provided that match a specified dimension of the motor vehicle. In other words, the sensor device can have at least one camera through which the surroundings are detected, whereby the detected surroundings can be checked for available parking options using an image recognition algorithm. For example, lane markings and / or street signs and / or gaps between parked vehicles can indicate the presence of a parking option. Preferably, only parking options whose dimensions match the motor vehicle can be tracked.This means, for example, that the dimensions of the vehicle can be specified, and the image recognition algorithm can be used to check whether the detected parking space at least matches the dimensions. This development offers the advantage of automatically recognizing parking spaces.

[0023] A further development provides that the evaluation of the possible paths by the path criterion is carried out based on a distance of the respective path to objects in the environment and / or based on a number of required steering maneuvers on the respective path and / or based on a number of required changes of direction and / or based on a path length of the respective path and / or based on a number of steering maneuvers when the motor vehicle is stationary. In other words, paths with a large distance to objects in the environment can be rated better than paths with a small distance. Alternatively or additionally, paths requiring many steering maneuvers can receive a worse rating than paths requiring few steering maneuvers, and / or a short path length of the respective path can receive a better rating than a long path length.Alternatively or additionally, an increasing number of required direction changes and / or an increasing number of steering maneuvers while the vehicle is stationary can negatively impact the evaluation, meaning that paths with a high number of direction changes and / or steering maneuvers while stationary are rated lower than those with a low number. This has the advantage that a trajectory is only created for the best-rated path, thus improving the parking process.

[0024] A further development provides that the generated movement trajectories are evaluated by the maneuver criterion based on the respective target state and / or based on an estimated fuel consumption and / or based on the duration of the parking maneuver. In other words, to evaluate the movement trajectories for the ranking list, it can be checked which fuel consumption is required for the respective movement trajectory and / or how long it takes to drive through the respective movement trajectories and / or which target state the motor vehicle is in after driving through the movement trajectory. With regard to fuel consumption, low fuel consumption can preferably be rated better than high fuel consumption. With regard to duration, a short duration can preferably be rated better than a long duration.When assessing the target state, for example, the distance between the vehicle and neighboring objects in the parking space can be checked, with a larger distance being rated more favorably than a smaller distance. This can be used, for example, to check how easy it is for the vehicle to get in or out.

[0025] In a further development, it is provided that the ranking list with the movement trajectories is provided to a user interface of the motor vehicle, wherein the selection of the movement trajectory from the ranking list is carried out by means of the user interface. In other words, the ranking list can be provided to a user who can select the movement trajectory desired for the parking process via the user interface. The user interface can preferably comprise a display or a screen, in particular a touch-sensitive screen, on which the ranking list can be displayed. Particularly preferably, an overview of all parking options with the determined movement trajectories can also be displayed, so that the user receives a visual overview from which the movement trajectory and thus the parking option can be selected.In addition to a touchscreen, the selection can also be made using other input methods, such as voice input. This offers the advantage of assisting the user in selecting the best parking options.

[0026] In an advantageous further development, the computing device automatically selects the best-rated movement trajectory from the ranking list. This means that no further user input is necessary, and thus the best-rated movement trajectory can be automatically used for the parking process.

[0027] A further development provides that, if the sensor device detects a change in the environment of the motor vehicle, a new movement trajectory is selected from the ranked list of movement trajectories that matches the changed environment. In other words, a change in the environment may occur, for example due to dynamic objects, in particular other road users, which restricts or makes impossible the execution of the previously selected movement trajectory. However, since the movement trajectories for the other parking options have already been determined, a new movement trajectory can be quickly selected from these that is not affected by the change in the environment. Preferably, the next best movement trajectory from the ranked list that is not influenced by the change in the environment can be used.This training offers the advantage of being able to react quickly to changes without having to carry out additional calculations.

[0028] A further aspect of the invention relates to a motor vehicle comprising a sensor device, a computing device, and a control device. The sensor device is configured to determine an initial state of the motor vehicle and an environment of the motor vehicle. The computing device is configured to determine parking options from the determined environment of the motor vehicle, including a target state for the respective parking option, to determine configuration samples in a configuration space using a sampling-based algorithm in predefined sampling steps, and to generate the configuration samples depending on a movement model of the motor vehicle, thus excluding impermissible movements of the motor vehicle.wherein the configuration space represents the environment of the motor vehicle, and the configuration samples comprise possible positions of the motor vehicle with a respective associated state of the motor vehicle. In each sampling step, a plurality of configuration samples can be generated in parallel using a graphics processing unit of the computing device. The configuration samples generated in each sampling step are based on the state and position of the configuration samples of the previous sampling step. The respective sequences of configuration samples define possible paths of the motor vehicle to the respective parking options for reaching the respective target state. Furthermore, the computing device is configured to determine the possible paths for the respective parking options simultaneously.to evaluate the possible paths of the motor vehicle to the respective parking space depending on a path criterion, to select a respective path to the respective parking space from the possible paths based on the evaluation, and to generate a respective movement trajectory of the motor vehicle, comprising a speed profile of the motor vehicle, to the respective parking space from the respectively selected paths, to evaluate the generated movement trajectories based on a maneuver criterion and to provide a ranked list of the movement trajectories depending on the evaluation, and to provide a movement trajectory selected from the ranked list to the control device. The control device is designed to steer the motor vehicle autonomously to the target state in the parking space based on the provided movement trajectory. In other words, the motor vehicle is designed toto carry out the method according to the above-mentioned aspect of the invention. This results in the same advantages and possible variations as in the method.

[0029] For use cases or application situations that may arise during the method and which are not explicitly described here, it may be provided that, in accordance with the method, an error message and / or a request to enter user feedback is issued and / or a default setting and / or a predetermined initial state is set.

[0030] The invention also includes the control device for the motor vehicle. The control device can have a data processing device or a processor device that is configured to carry out an embodiment of the method according to the invention. For this purpose, the processor device can have at least one microprocessor and / or at least one microcontroller and / or at least one FPGA (Field Programmable Gate Array) and / or at least one DSP (Digital Signal Processor). In particular, a CPU (Central Processing Unit), a GPU (Graphical Processing Unit), or an NPU (Neural Processing Unit) can be used as the microprocessor. Furthermore, the processor device can have program code that is configured to carry out the embodiment of the method according to the invention when executed by the processor device. The program code can be stored in a data memory of the processor device.The processor device can be based, for example, on at least one circuit board and / or on at least one SoC (System on Chip).

[0031] The invention also includes further developments of the motor vehicle according to the invention that have features already described in connection with the further developments of the method according to the invention. For this reason, the corresponding further developments of the motor vehicle according to the invention are not described again here.

[0032] The motor vehicle according to the invention is preferably designed as a motor vehicle, in particular as a passenger car or truck, or as a passenger bus or motorcycle.

[0033] As a further solution, the invention also encompasses a computer-readable storage medium comprising program code which, when executed by a computer or computer network, causes the computer to carry out an embodiment of the method according to the invention. The storage medium can be provided at least partially as a non-volatile data memory (e.g. as a flash memory and / or as an SSD - solid state drive) and / or at least partially as a volatile data memory (e.g. as a RAM - random access memory). The storage medium can be arranged in the computer or computer network. However, the storage medium can also be operated on the Internet, for example, as a so-called app store server and / or cloud server. The computer or computer network can provide a processor circuit with, for example, at least one microprocessor.The program code may be provided as binary code and / or as assembly code and / or as source code of a programming language (e.g. C) and / or as a program script (e.g. Python).

[0034] The invention also encompasses combinations of the features of the described embodiments. The invention therefore also encompasses implementations that each comprise a combination of the features of several of the described embodiments, unless the embodiments are described as mutually exclusive.

[0035] Exemplary embodiments of the invention are described below. Shown are: Fig. 1 a schematically illustrated motor vehicle according to an exemplary embodiment; Fig. 2 is a schematic process diagram according to an exemplary embodiment; Fig. 3 a configuration space with configuration samples for an exemplary motion planning; Fig. 4 an exemplary representation of provided movement trajectories for each parking option.

[0036] The exemplary embodiments explained below are preferred embodiments of the invention. In the exemplary embodiments, the described components of the embodiments each represent individual features of the invention that can be considered independently of one another, each of which also develops the invention independently of one another. Therefore, the disclosure is intended to encompass combinations of the features of the embodiments other than those shown. Furthermore, the described embodiments can also be supplemented by further features of the invention already described.

[0037] In the figures, the same reference symbols designate elements with the same function.

[0038] In Fig. 1 shows a schematically illustrated motor vehicle 10, which may be configured as an autonomous motor vehicle. In particular, the motor vehicle 10 may be configured to perform motion planning for a parking maneuver and to park autonomously based on the motion planning. For this purpose, the motor vehicle 10 may have a sensor device 12, a computing device 14, and a control device 16.

[0039] The sensor device 12 can comprise one or more sensors that can record data from the surroundings of the motor vehicle 10 and / or data from the motor vehicle 10 itself. For example, the sensor device 12 can comprise optical sensors, in particular camera sensors and / or laser sensors, radar sensors and / or ultrasonic sensors, with which the surroundings of the motor vehicle 10 can be recorded. Furthermore, the sensor device 12 can comprise sensors that can determine a current steering angle, a speed and / or an orientation of the motor vehicle 10 in the surroundings. Furthermore, parking options can also be detected by means of the sensor device 12, wherein the parking options can be determined from the recorded data, for example, using an image recognition algorithm.

[0040] The computing device 14 and the control device 16 can be provided as a vehicle computer of the motor vehicle 10, wherein they can be provided as a unit in a common device or as a common vehicle computer, or as separate devices. The computing device 14 can at least be designed to carry out movement planning to one or more predetermined parking positions in the surroundings of the motor vehicle 10 determined by the sensor device 12, wherein the movement planning can be based on a sampling-based algorithm in which a plurality of sample values ​​can be determined which, as a sequence, can together result in a geometric path to the parking positions. In particular, the computing device 14 can comprise a graphics processing unit 15, by means of which fast parallel processing of multiple sample values ​​is enabled.

[0041] The control device 16 can be configured to control the motor vehicle 10 using the motion planning provided by the computing device 14, in particular using a provided motion trajectory, such that the motor vehicle 10 moves autonomously into the parking position. In this case, the motion trajectory can take into account, in particular, multiple steering maneuvers with changes in direction, including small or large steering angles, as well as steering angles when the motor vehicle 10 is stationary.

[0042] In Fig. 2 shows a schematic process diagram for the movement planning of an autonomous motor vehicle 10 for a parking operation according to an exemplary embodiment, wherein the movement planning can be carried out by means of the sensor device 12, the computing device 14 and the control device 16.

[0043] In a step S1, an initial state of the motor vehicle 10 and the surroundings of the motor vehicle 10 can be determined by the sensor device 12 of the motor vehicle 10. The surroundings can include static and dynamic objects in the surroundings of the motor vehicle 10. The initial state can provide an initial configuration or an initial state of the motor vehicle 10, based on which the motion planning can be carried out. For example, the initial state can include a current speed, a current steering angle, and / or a current orientation in space or in the surroundings.

[0044] In a step S2, parking spaces can be determined from the determined surroundings of the motor vehicle 10 by the computing device 14. For each parking space, a target state that the motor vehicle 10 should assume for parking in the respective parking space can also be determined. The determination of the parking space can be carried out based on the data recorded by the sensor device 12, wherein the data can in particular comprise image data from a camera device (not shown), which can be determined using an image recognition algorithm operable on the computing device 14. Preferably, only those parking spaces that match a predetermined dimension of the motor vehicle can be recognized and provided.

[0045] In a step S3, a virtual configuration space 18 can be generated by the computing device 14 from the environment determined by the sensor device 12, which virtual configuration space represents the environment of the motor vehicle 10. Such a configuration space 18 is shown by way of example and very schematically in Fig. 3. The computing device 14 can then calculate configuration sample values ​​20, 20', 20" in the configuration space 18 using a sampling-based algorithm, preferably using at least one graphics processing unit 15, which include possible positions of the motor vehicle 10 with a respective associated state of the motor vehicle 10. The configuration sample values ​​20, 20', 20" have, as a state, at least one respective steering angle required for the position and an orientation of the motor vehicle 10 at this position. In particular, the configuration sample values ​​20, 20', 20" can be generated in several sampling steps that build on one another.For example, in a first sampling step, the configuration samples 20 can be determined, in a subsequent second step the configuration samples 20', which are based on the first configuration samples 20, in particular on the state provided in the configuration samples 20, and correspondingly in a third sampling step, further configuration samples 20". The number of sampling steps is shown only as an example, and a plurality of further sampling steps can be carried out up to a final sampling step in which the target state is reached in each parking space.

[0046] Particularly preferably, it is provided that the configuration sample values ​​20, 20', 20" are determined simultaneously up to each target state of each parking option. Simultaneous means that a calculation is not carried out individually for each parking option, but rather the entire configuration space 18 for each parking option is spanned in a single sequence of sampling steps. In particular, all possible movements that are possible by the motor vehicle 10 in the configuration space 18 can be calculated using the configuration sample values ​​20, 20', 20".

[0047] In order to limit the number of configuration samples 20, 20', 20" in the respective sampling step and thus provide real-time capability for motion planning, it can further be provided that the configuration samples 20, 20', 20" are generated in the configuration space 18 as a function of a motion model of the motor vehicle 10, thus excluding impermissible movements of the motor vehicle 10. In particular, the motion model can be provided so that positions and / or states of the motor vehicle 10 are not generated in the configuration space 18 for which a movement of the motor vehicle 10 is excluded, wherein all other possible movements of the motor vehicle are generated, preferably also assuming a steering movement when the motor vehicle is stationary.A single-track model can preferably be used as the motion model, allowing steering angles and the resulting movement and orientation of the motor vehicle 10 in space to be easily determined. In the single-track model, it can preferably be provided that a front-axle steering of the motor vehicle 10 is assumed and movements in the forward and / or reverse direction can be generated. Thus, for example, multiple trains with changes of direction at any location can also be planned.

[0048] From sequences of the generated configuration samples 20, 20', 20", geometric paths 22, 22' of the motor vehicle 10 can be determined, which preferably lead to the respective target state of the respective parking possibility after the last sampling step.

[0049] In a step S4, the possible paths 22, 22' can be evaluated depending on a path criterion, wherein the path criterion evaluates the possible paths 22, 22' based on a distance to respective objects in the environment and / or based on a number of required steering maneuvers on the respective path and / or based on a number of required changes of direction and / or based on a path length and / or path duration of the respective path and / or based on steering maneuvers while the motor vehicle is stationary. This evaluation can preferably be performed after each scanning step, and preferably only collision-free paths can be pursued.

[0050] This means that steps S3 and S4, in particular, can be performed iteratively. Thus, after each scanning step, the generated paths can be evaluated, and for the subsequent scanning step, paths that should not be pursued can be excluded based on the evaluation in step S4. Thus, only collision-free paths that lead to the parking position in a final scanning step can be generated.

[0051] In a step S5, the computing device 14 can select a path to each parking space from all possible paths, in particular depending on the evaluation of the respective path. Preferably, the best-rated path 22' can be selected for each parking space, wherein a respective movement trajectory of the motor vehicle 10 can be generated for the respectively selected paths 22', which additionally includes a speed profile. Thus, a respective movement trajectory of the motor vehicle 10 to the respective parking space can be generated from the geometric paths.

[0052] In a step S6, the generated movement trajectories can then be evaluated by the computing device 14 using a maneuver criterion, and a ranking of the movement trajectories 24 can be provided depending on the evaluation. Using the maneuver criterion, the movement trajectories 24 can be evaluated, for example, based on a target state or a distance of the motor vehicle from other objects in the parking position and / or based on an estimated fuel consumption for the respective parking position and / or based on a duration of the parking process for the respective parking position.For example, movement trajectories 24 in which the distance to other objects, such as other vehicles, is small in the parking position may be rated worse than those with a large distance, high fuel consumption may result in a worse rating of the movement trajectory than for movement trajectories 24 with low fuel consumption and / or a short duration to the parking possibility may be rated better than a long duration.

[0053] An exemplary representation of generated movement trajectories 24 is shown in Fig. 4. In this case, several parking options or parking positions can be present in the surroundings, which can be reached by the movement trajectories 24. After an evaluation of the movement trajectories 24 based on the maneuver criterion, a ranking list can be provided, which can be displayed, for example, to a user of the motor vehicle 10. In this case, the parking options 1, 2, 3, 4, 5 can be displayed according to their placement in the ranking list, from which the user can, for example, select the movement trajectory 24 and thus the parking option via a user interface (not shown). The user interface can, for example, be a display device of the motor vehicle 10, wherein the ranking list is displayed as an overview of the surroundings, for example in the form of the Fig. 4 shown overview.

[0054] In a step S7, a motion trajectory 24 can then be selected from the ranking list, for example, via a user input via the user interface. Alternatively, the motion trajectory can also be selected automatically by the computing device 14, in which case, in particular, the best-rated motion trajectory can be used.

[0055] Finally, in a step S8, the selected movement trajectory can be provided to the control device 16 of the motor vehicle 10, wherein the control device 16 can park the motor vehicle 10 autonomously by means of the selected movement trajectory 24.

[0056] Should the surroundings change during the parking maneuver, which can be verified using the sensor device 12, a new movement trajectory 24 can be selected from the ranked list of movement trajectories 24, which replaces the previously applicable movement trajectory 24 and is not restricted by the changed surroundings. Thus, the parking maneuver can be continued using the new movement trajectory 24 without recalculating all movements of the motor vehicle 10.

[0057] Overall, the examples demonstrate how a strategy for simultaneously computing multiple paths can be provided. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] WO 2022 / 268323 A1

[0009]

Claims

[1] Method for planning the movement of an autonomous motor vehicle (10) for a parking operation, the method comprising the following steps: - determining (S1) an initial state of the motor vehicle (10) and an environment of the motor vehicle (10) by a sensor device (12) of the motor vehicle (10); - determining (S2) parking possibilities from the determined surroundings of the motor vehicle (10) by a computing device (14) of the motor vehicle (10), wherein a target state is determined for the respective parking possibility; - Determining (S3) configuration sample values ​​(20, 20', 20") in a configuration space (18) by the computing device (14), wherein the configuration sample values ​​(20, 20', 20") are generated as a function of a movement model of the motor vehicle (10) and thus impermissible movements of the motor vehicle (10) are excluded, wherein the configuration space (18) represents the surroundings of the motor vehicle (10) and the configuration sample values ​​(20, 20', 20") comprise possible positions of the motor vehicle (10) with a respective associated state of the motor vehicle (10), wherein the configuration sample values ​​(20, 20', 20") are determined using a sampling-based algorithm in predefined sampling steps, wherein in each sampling step, a plurality of configuration sample values ​​(20, 20', 20") are processed in parallel using at least one graphics processing unit (15). the computing device (14),wherein the configuration samples (20, 20', 20") generated in each sampling step are based on the state and position of the configuration samples (20, 20', 20") of the previous sampling step, wherein respective sequences of configuration samples (20, 20', 20") define possible paths (22, 22') of the motor vehicle (10) through the configuration space (18) to the respective parking options for reaching the respective target state, wherein the possible paths (22, 22') for the respective parking options are determined simultaneously by the computing device (14);, - evaluating (S4) the possible paths (22, 22') of the motor vehicle (10) to the respective parking possibility after each scanning step by the computing device (14) as a function of a path criterion; - selecting (S5) a respective path (22') to the respective parking possibility from the possible paths (22, 22') on the basis of the evaluation by the computing device (14), wherein a respective movement trajectory (24) of the motor vehicle (10), comprising a speed profile of the motor vehicle (10) to the respective parking possibility, is generated from the respectively selected paths (22'); - evaluating (S6) the generated movement trajectories (24) on the basis of a maneuver criterion by the computing device (14) and providing a ranking list of the movement trajectories (24) depending on the evaluation; - selecting (S7) a movement trajectory (24) from the ranking list; - Providing (S8) the selected movement trajectory (24) to a control device (16) of the motor vehicle (10) for moving the autonomous motor vehicle (10) to the target state in the parking space. [2] Method according to claim 1, wherein the respective state of the motor vehicle (10) comprises at least an orientation of the motor vehicle (10) in the environment and a steering angle, wherein configuration samples (20, 20', 20") which are excluded by the orientation of the motor vehicle (10) and the steering angle are not generated by the movement model. [3] Method according to one of the preceding claims, wherein the motion model is a single-track model. [4] Method according to one of the preceding claims, wherein the parking possibilities in the determined surroundings of the motor vehicle (10) are determined by an image recognition algorithm, wherein only parking possibilities are provided which match a predetermined dimension of the motor vehicle (10). [5] Method according to one of the preceding claims, wherein the evaluation of the possible paths (22, 22') by the path criterion is carried out on the basis of a distance of the respective path (22, 22') to objects in the environment and / or on the basis of a number of required steering maneuvers on the respective path (22, 22') and / or on the basis of a path length of the respective path (22, 22'). [6] Method according to one of the preceding claims, wherein the generated movement trajectories (24) are evaluated by the maneuver criterion on the basis of the respective target state and / or on the basis of an estimated fuel consumption and / or on the basis of a duration of the parking process. [7] Method according to one of claims 1 to 6, wherein the ranking list with the movement trajectories (24) is provided to a user interface of the motor vehicle (10), wherein the selection of the movement trajectory (24) from the ranking list is carried out by means of the user interface. [8] Method according to one of claims 1 to 6, wherein the best-rated movement trajectory (24) is automatically selected from the ranking list by the computing device (14). [9] Method according to one of the preceding claims, wherein if a change in the environment of the motor vehicle (10) is detected by the sensor device (12), a new movement trajectory (24) is selected from the ranked list of movement trajectories (24) which matches the changed environment. [10] Motor vehicle (10) comprising a sensor device (12), a computing device (14) and a control device (16), - wherein the sensor device (12) is designed to determine an initial state of the motor vehicle (10) and an environment of the motor vehicle (10); - wherein the computing device (14) is designed to • to determine parking options from the determined surroundings of the motor vehicle (10) comprising a target state for the respective parking option; • to determine configuration samples (20, 20', 20") in a configuration space (18) using a sampling-based algorithm in predefined sampling steps and to generate the configuration samples (20, 20', 20") depending on a movement model of the motor vehicle (10) and thus to exclude inadmissible movements of the motor vehicle (10), wherein the configuration space (18) represents the surroundings of the motor vehicle (10) and the configuration samples (20, 20', 20") comprise possible positions of the motor vehicle (10) with a respective associated state of the motor vehicle (10), wherein in each sampling step a plurality of configuration samples (20, 20', 20") can be generated in parallel using a graphics processing unit (15) of the computing device (14), wherein the configuration samples (20, 20', 20") on the state and position of the configuration samples (20, 20',20") of the previous scanning step, wherein respective sequences of configuration samples (20, 20', 20") define possible paths (22, 22') of the motor vehicle (10) to the respective parking possibilities for reaching the respective target state;, • to determine the possible paths (22, 22') for the respective parking options simultaneously, • to evaluate the possible paths (22, 22') of the motor vehicle (10) to the respective parking possibility depending on a path criterion; • to select a respective path (22') to the respective parking possibility from the possible paths (22, 22') on the basis of the evaluation and to generate a respective movement trajectory (24) of the motor vehicle (10), comprising a speed profile of the motor vehicle (10), to the respective parking possibility from the respectively selected paths (22'); • to evaluate the generated movement trajectories (24) based on a maneuver criterion and to provide a ranking list of the movement trajectories (24) depending on the evaluation; and • provide a movement trajectory (24) selected from the ranking list to the control device (16); - wherein the control device (16) is designed to control the motor vehicle (10) autonomously to the target state in the parking space based on the provided movement trajectory (24).

Images