Maneuvering assistance procedure and maneuvering assistant

Abstract

Driving assistants are known in the art, but their ability to maneuver in confined spaces, such as parking garages, is limited. Therefore, the objective is to improve maneuverability and reduce the risk of collision. This objective is achieved, in particular, by a method, particularly a computer-implemented one, which is designed to monitor, plan, and / or execute such a driving maneuver. The method is characterized by comprising the steps of: - Creating and / or maintaining an environmental model designed to depict obstacles in the environmental model; - Analysis of a driving situation, designed to predict a trajectory ahead; and - Planning a set of trajectories taking into account the environmental model and / or the analysis of the driving situation; and - Identifying at least one advantageous trajectory from the set of trajectories.

Description

The invention relates to a method, in particular a computer-implemented method, a driver assistance apparatus, a control device, a vehicle and a computer program product.Driving assistants are known in the prior art. In common road traffic, these serve a driver to assist in parking more safely, changing a lane and / or as a cruise control in order to maintain or adapt a speed.For driving assistants, manoeuvring in narrow places or below a certain speed, in particular in the region of the step speed, is difficult. Thus, it is difficult, in particular in parking garages, but also in narrow roads, for example in central European interior towns with their historically grown gate structure, to rely on the driving assistants or to use them at all.Thus, the problem exists in the prior art that known driving assistants must be improved in order to be able to use them even in bottleneck locations for driving maneuvers in which there is an increased risk of collision due to bottleneck locations.It is therefore an object of the invention to improve the usability of driving assistants, in particular in the region of bottleneck areas, by improving their maneuverability in order to reduce a risk of collision and thus to increase the safety of a driving maneuver.The object is achieved in particular by a method according to claim 1, a driver assistance apparatus according to claim 12, a control device according to claim 13, a vehicle according to claim 14 or a computer program product according to claim 15. In particular, the independent claims of one claim category can also be developed analogously to the dependent claims of another claim category.According to one aspect, the object is achieved in particular by a method, in particular implemented as a computer-implemented method, wherein the method is configured to monitor, plan and / or perform a driving maneuver of an ego vehicle. The method has the steps of creating and / or preparing an environment model in order to map obstacles in the environment model; analyzing a driving situation, in such a way as to predict a trajectory lying ahead; planning a trajectory group with the inclusion of the environment model and / or the analysis of the driving situation; and identifying at least one advantageous trajectory from the trajectory group. As a result, maneuverability can be increased, whereby the risk of collision decreases, which in turn increases the safety of a driving maneuver at low speeds and in narrow places.In this context, a driving maneuver may be, in particular, the actual operative travel of an ego vehicle, which may be planned and / or carried out. In this context, the ego vehicle may be referred to as one's own vehicle, while other vehicles are the vehicles of other road users. Other vehicles may be moving, but may also be parked.The creation of an environment model can ensure that objects (also called obstacles) are mapped in the environment model. In this case, above all, static and solid obstacles are mapped with respect to free spaces in the environment model. The mapping can be understood here in particular as a mathematical operation which makes it possible to create a digital twin of at least part of an environment and to keep it available for the further steps of the method. For example, it is difficult for a driver to recognize certain pillars, for example in a parking garage, and / or to re-resemble them after parking. It is thus important that these obstacles are mapped in the environment model such that their distance from the ego vehicle and / or their position in space for identifying an advantageous trajectory is known to the method.This applies in particular to those static objects which can be hidden in their "blind spots" for a given vehicle type. The blind spot may be an angular range that is not visible by the driver, since the view into this angle is blocked, for example, by the A, B or C pillar of the ego vehicle. However, the static objects may also be too low, as a result of which they do not project into a field of view. This also makes it possible to take a place in particular on an area next to the vehicle, i.e. in the lateral direction, since these areas can quickly lead to the ego vehicle touching these static objects.In this case, a reservation can be understood in particular in such a way that the environment model is kept ready in a memory for carrying out the method. In particular, the environment model can therefore already be present at least partially and a corresponding creation is not necessary in all situations.In this case, a driving situation can be analyzed in such a way as to predict a trajectory lying ahead. The currently expected trajectory can thus be predicted.A driving situation can relate to the totality of the parameters and / or conditions which are present at a point in time of the trip. The parameters and / or circumstances can in this case comprise in particular at least one of the following parameters, alone or in any conceivable combination:a position, a relative position, a speed, an angular speed, a rotation, an acceleration and / or an angular acceleration.The analysis of the driving situation allows the dynamic monitoring of the entirety of the driving situation, that is to say, for example, to what extent the position of the ego vehicle changes over time. As a result, a trajectory can be pre-planned and / or predicted, which allows the route path lying ahead to be connected to the position of the ego vehicle therein. This can be referred to as a trajectory lying ahead.In other words, the term "trajectory lying ahead" can thus be understood here in particular as the further route profile to which the ego vehicle is exposed when the journey continues. This can consequently be the further route and / or the further movement profile which is in the future of the journey if the journey is continued unchanged.A trajectory group can be planned with the inclusion of the environment model and / or the analysis of the driving situation. Thus, the model can "play through" various situations.In this case, a trajectory group can be planned on the basis of a current location of the ego vehicle. A trajectory group can represent the set of possible paths of the ego vehicle. The course of adjacent trajectories can differ, for example, by a predefined difference of possible different steering angles (the term steering angle can be used interchangeably here and elsewhere, which can describe the position of a steering system, for example in relation to a central position for driving straight ahead, but which can also be used to describe the angle between the tangent of the previous trajectory to a newly driven trajectory by the steering) of the own motor vehicle. Other parameters may also be included, such as variance from the instantaneous speed. In particular, the method can be configured such that the trajectory lying ahead is included in the trajectory group and is therefore available as an option for the further travel course. In other words, the trajectory ahead remains an option to continue the trip.In this case, the environment model and / or the analysis of the driving situation can / can be included in particular.At least one advantageous trajectory from the trajectory family can be identified. An advantageous trajectory can be in particular such a trajectory whose driving avoids a collision. Furthermore, it may be advantageous if a trajectory does not require a reset, for example in order to pass around a narrow point with a sharp curve or in order to pass around a static object. Such trajectories, without resetting and / or without intervention and / or without intervention change, can be referred to as smooth trajectories, since there is no interruption of the forward travel. In other words, a smooth trajectory can be one which allows the constriction to be traversed by means of initial parameters which have been chosen once.The method can be embodied as a computer-implemented method or can be a computer-implemented method.According to a further aspect, it can be provided that the method further comprises the step of weighing an intervention decision, designed in such a way as to make a decision about an intervention or an intervention change. This decision can be based on the prediction of the trajectory ahead, the planned trajectory group and / or the identified advantageous trajectory. As a result, a decision can be made by means of the method as to whether an intervention is necessary or not. This can improve the safety of driving in narrow places.In this case, the weighing can be understood to mean that the method, in particular computer-implemented, involves the prediction of the trajectory ahead, the planned trajectory group and / or the identified advantageous trajectory in an analysis which outputs the probability of travelling (better or best) which trajectory is suitable, for example in order to prevent a collision. In this case, a trajectory which is being traveled on can be continued even if it has been identified as a collision-free trajectory, even if it does not represent the optimum trajectory. In this case, an optimum trajectory can also be delimited, for example, by other parameters with respect to the "purely advantageous" trajectory, for example because a smaller number of steering movements, a smaller number of control commands by the driver and / or corresponding interventions (included steering interventions) are necessary. Alternatively, the optimum trajectory may be preferred. For example, the optimum trajectory can be a smooth trajectory.An intervention may represent any form of intervention in the control of a vehicle, such as braking, accelerating, clutching and / or shifting. An intervention change may thus refer to a change in a current intervention, such as when an intervention in the control of a vehicle is re-adjusted while currently still running an intervention. This serves in particular to iteratively control to a preferred trajectory and / or to change a corresponding trajectory again, for example because the oncoming traffic reacts in turn and the oncoming foreign vehicle can thus change its trajectory. This can change the predicted collision point.A steering intervention can in this case represent, in particular, an intervention in the steering, as a result of which the direction of the vehicle can be changed. Furthermore, however, a steering intervention change can also be provided, which changes a steering intervention while a steering intervention is still running. Intervention, change in intervention, steering intervention and / or change in steering intervention can therefore also be, in particular, such that an action of a driver is counteracted, for example by counter-steering, braking instead of accelerating, accelerating instead of braking.According to a further aspect, it can be provided that the method is designed to carry out at least one intervention and / or at least one intervention change, based on the intervention decision. As a result, the method can directly perform a control and control function. The definitions listed above with respect to the intervention and change in intervention apply accordingly.The intervention decision can also be designed in such a way as to avoid a disadvantageous trajectory. The avoidance of a disadvantageous trajectory can relate in particular to leaving the disadvantageous trajectory. This serves in particular for collision avoidance. In this case, a collision can have the highest priority in the course of travel. The predicted and / or planned passing trajectory can lead to a collision in particular when moving units (other vehicles, persons on the roadway, autonomous robot vehicles) cross the trajectory, which could not be foreseeable in the initial prediction.Furthermore, it can be provided that the method is designed to perform a driving maneuver semi-autonomously or autonomously. Parking and / or unparking by a vehicle according to the invention can therefore also be carried out semi-autonomously or autonomously, in particular by means of a driving assistance device according to the invention. This increases the safety of a driving maneuver. Here, the terms semi-autonomous and autonomous are described elsewhere and repetition is omitted here.According to a further aspect, it can be provided that the method is designed in such a way that the environment model is created based on at least one piece of map information and / or is created on at least one piece of information from an experiential sensor. As a result, real-time data can be introduced into the method and / or a prediction of the trajectory lying ahead, the planning of the trajectory group and / or thus the identification of an advantageous trajectory can be improved. This increases the safety of a trip and reduces the risk of a collision in bottleneck areas.The map information can be a geographical map information. This can transmit position information of the vehicle into the environment model and / or locate it relative to a location of the environment to be considered a hazard zone. A hazard zone can be, in particular, a zone in which the dimensioning of the vehicle relative to the environment represents a narrow zone for the trip. This can be provided in that either the dimensioning of the free spaces is only at a small distance from the vehicle. Alternatively or additionally, a hazard zone can also be present, however, if the angles (for example the sharpness of curves and / or turns, in particular in parking garages) are designed such that a driving maneuver is difficult to carry out. The map information can thus be the basis that allows the ego vehicle to be located in the environment model.Alternatively or additionally, however, the map information can also be a construction plan, for example of a parking garage or a parking garage, which has corresponding information about travel paths. These are often not present in geographical maps. This also allows geographical maps and / or construction plans to be supplemented. This is possible, for example, in the area of entry and exit from a parking garage in a city with narrow gate boxes.The localization of the ego vehicle can relate in particular to the determination of the position of the ego vehicle, for example by means of a global navigation satellite system ("Global Navigation Satellite System"), which can be fed into the method and / or can be matched to the map information using a GNSS. Additionally or alternatively, at least one piece of sensor information can be used by means of an experieroceptive sensor and / or sensor system in order to support the local position. In this case, a dynamic component with respect to other vehicles and / or passers-bys, which move, for example, in the area of bottleneck locations in cities and / or parking garages, can also be included. The experiential sensors are described elsewhere in detail and reference is made to this description for compactness and readability.The most important GNSS at the time of registration is the Global Positioning System (GPS) of the USA, officially called NAVSTAR GPS (Engl.: "Navigational Satellite Timing and Ranging--Global Positioning System"). Alternative systems are "Galileo" from EU, and "GLONASS" from Russian Federation, "Beidou" from People's Republic of China (a network available worldwide since 2004 for the Asian sector is in the setup), the Indian Regional Navigation Satellite System in the setup and limited to India and the quasi-zenith satellite system in the setup and designed for regional coverage, from Japan.At least one exterior-receptive sensor can be provided, wherein said sensor can generate, in particular, information with respect to at least one other vehicle and / or an object and / or the ego vehicle. The at least one exuteroreceptive sensor can be selected from one of the following sensors and sensor systems:an inclination sensor,a laser source with a reflection sensor,a lidar system with a lidar sensor,an ultrasonic system with an ultrasonic sensor,a radar / radar sensora camera system.Inclination sensors can measure inclination angles on one or two axes, for example (for example) on the x-axis and / or the y-axis of the ego vehicle. These products can be based in particular on the robust MEMS technology, in which capacitance differences can be converted into analog voltage in a micromechanical sensor chip. This analog voltage may be proportional to the angle to which the sensor is exposed. Modular concepts make it possible to easily adapt the sensors to specific desires.An inclination sensor can be used in particular in situations where accurate position determination and / or constant monitoring of the angle with respect to gravity is required. An inclination sensor can measure the angle in particular with respect to a horizontal position, wherein the imaginary line to the center of the earth serves as a reference. Inclination sensors can have a very broad field of application in particular, since they can be mounted everywhere and / or offer a high degree of freedom of mechanical design. The sensor output can be composed in particular of two components. These can be, in particular, a static (inclination) component and a dynamic (acceleration) component. The sensor output can in this case be in particular a combination of these two components.A laser source can be a continuous wave radiation laser source and / or can detect the reflected laser radiation by means of a reflection sensor, which has changed in frequency, for example on the basis of the speed.Alternatively or additionally, a LIDAR (from Engl.: "Light Detection and Ranging" accordingly in the German "Light Detection and Distance Measurement") can be used. Scanning LIDAR sensors can operate in particular according to the scanner principle: the laser beam from a light source can be directed onto the environment to be observed line by line and point by point by rotating mirrors and the reflected light can be directed to the sensor on the reverse path. Solid-state LIDAR sensors make do without mechanical components in particular. The transmitting and receiving units, which are part credit card-sized, can each accommodate 128 transmitting points in 80 lines, which transmitting points are simultaneously receivers of the light pulse reflected by the detected object. The laser can in particular pulse at a frequency of 25 Hz; each line is in particular scanned several hundred times per scan. In the computer unit belonging to the system, the information from all LIDAR sensors of the vehicle can be merged into a 360° position image.Further, ultrasound may be used. Ultrasonic sensors can be used in particular in vehicles in order to monitor the immediate surroundings of the vehicle and / or to measure distances to obstacles. Ultrasonic sensors can be beam-based sensors: they can transmit and receive sound waves whose frequencies are above the range perceptible by human hearing (above 16 kHz). These sound waves (usually transmitted from so-called piezo actuators via a membrane) propagate in the surrounding air and are reflected by obstacles. The echo signals can be registered by the sensors and evaluated by a central control unit. From the propagation time, i.e. the time required for the echo signal to arrive at the transmitter, the distance of the obstacle is calculated.Ultrasonic sensors can thus function in particular as a basis for driver assistance systems. Ultrasonic sensors can be integrated in the vehicle front and / or in the vehicle rear. The determined distance to obstacles can be indicated to the driver acoustically and / or visually depending on the manufacturer and system. Ultrasonic sensors can be used in particular in autonomous / automated driving, especially for environment recognition in the near range up to six meters and / or at low speeds. Ultrasonic sensors can be used in areas up to 5.5 meters apart. Due to the low frequency and a relatively large wavelength, the ultrasound technology is less demanding and thus relatively cost-effective.The range thus also moves in particular in dimensions which are typical for the dimensioning of people cars at the time of registration. Any constrictions can be understood here in particular in such a way that they have about 15% more width and / or height and / or length than the maneuvering ego vehicle. Furthermore, bottleneck areas can have only 10% more or only 5% more width and / or height and / or length than the maneuvering ego vehicle.Radar may be used for advanced driver assistance systems that are capable of immediately measuring distance, angle, and speed and / or generating detailed images of the environment. This technology enables safety-critical functions such as emergency braking. Radar sensors are used in particular in vehicles in order to monitor the environment of the vehicle by measuring distances to obstacles and their relative speeds.Radar (German "Radio Detection and Ranging") is based on radar sensors, beam-based sensors, which can be used to detect objects, for example other vehicles and / or pedestrians and / or landmarks, and / or to measure their distance from the ego vehicle and their relative speeds. For this purpose, electromagnetic waves can be emitted which are reflected by the objects. These reflected electromagnetic waves can be received and / or evaluated: the measured values can be converted in particular into electrical signals which can be evaluated in a control device configured for this purpose.Radar sensors may operate at radar frequencies between 76 and 77 GHz. In addition, there are designs which use a frequency range of 24 GHz. These are used in particular at distances of approximately up to 160 meters in front and up to 100 meters in rear. In a range from about 5 meters, short-range radar sensors can be used. Mid-range radar sensors can operate in particular with an operating frequency of 24 GHz or 77 GHz. In the far range up to about 250 meters, in particular far range radar sensors with a frequency of 77 GHz can be used. Radar sensors at a distance of up to 250 meters as one of several sensor principles may also provide important 360-degree environmental information for an automated / autonomous driving vehicle, as will be described in detail elsewhere. Compared to LIDAR sensors, radar sensors are less sensitive to weather influences such as rain, snowfall or fog.A camera system comprising one or more cameras can also be provided, which is connected to a computing unit in such a way that object recognition is implemented. As a result, the camera system can analyze the images for patterns and / or recognize objects, which it can then make available to a modeling of an environment model.The sensor information can form the basis for the function of numerous safety systems which are intended to avoid accidents with corresponding warnings and vehicle interventions. These include cruise control systems, collision warning and avoidance systems. Here, a combination of the assistance functions can be provided, for example, in order to monitor and / or initiate a journey along a selected trajectory, in particular after it turns out that the new simulated trajectory would be advantageous when checking whether a trajectory is advantageous compared to an alternative, simulated trajectory.According to a further aspect, it can be provided that the method is configured to plan the trajectory group on the basis of the trajectory ahead predicted from the driving situation, wherein a minimum variance is used as the basis for the predicted trajectory ahead. This makes it possible to make the possibly necessary intervention as small as possible. Abrupt driving maneuvers can thus also be prevented.In this case, the variance can relate directly to the course of the trajectories. These can therefore be varied, wherein a deviation from the trajectory lying ahead can be limited. From the variance of the trajectory, it is then possible to infer the parameters which would be necessary to travel on the individual trajectories. Alternatively or additionally, however, it may also be possible to configure the initial parameters such that they are varied with a certain variance, wherein trajectories can be determined therefrom. The latter can then be examined and / or evaluated for their deviation from the trajectory ahead and / or examined for their property of being an advantageous trajectory.Based on the properties as an advantageous trajectory, with the lowest variance, the decision can then be made as to which trajectory is selected in order to continue the journey. In this case, it is also possible to take into account the extent or how much intervention has to be made by a control device according to the invention and / or a driving assistance device according to the invention in order to travel on the corresponding trajectories, if appropriate. In this case, as a rule, trajectories in which the intervention is as minimal as possible can be preferred in order to make the driver, insofar meaningful and possible, control over the control of the vehicle.According to a further aspect, it can be provided that the method is designed in such a way that the at least one advantageous trajectory is determined on the basis of the prediction of possible collision probabilities for the trajectories of the trajectory family, and of the trajectory lying ahead (also referred to as "currently traveled"). A driving maneuver can thus be made more secure.The collision probability can be determined based on the current position. The probability can change in particular when mobile units, such as foreign vehicles, are also included and / or their trajectories are predicted, for example based on information from experiential sensors and associated extrapolations of movement patterns of the mobile units. The collision probability can also change if the driver and / or the ego vehicle itself performs changes in his driving situation. It may also be possible to weight the trajectories after the time of the occurrence of a collision, wherein a previous collision occurrence is evaluated as being more disadvantageous than a later one. This can be based on the assumption that there is still time for an alternative advantageous trajectories to be determined in the event of a later collision occurring.According to a further aspect, it can be provided that the method is designed such that the method is carried out at a speed of less than 30 km / h, further less than 25 km / h, further less than 10 km / h, further less than step speed, in particular such that the method is carried out continuously. It is thus possible to carry out the method in speed ranges, as is present in parking garages and in narrow roads and gasses of cities. A continuous execution of the method can be understood here in particular to mean that the method runs continuously, for example if a set maximum speed is undershot. The method and in particular a driving assistance device according to the invention can thus be automatically switched on in order to assist the driver. The driver therefore no longer has to think of activating the system when he enters a potential hazard zone.According to a further aspect, it can be provided that the method is designed in such a way that a decision is made based on the environmental model, a speed of the ego vehicle is limited to a maximum speed, for a case in which the maximum speed is exceeded and / or the maximum speed is set as an upper speed limit, for the case in which the speed is below the maximum speed, wherein for each case the maximum speed is set in particular depending on the environmental model. This makes it possible to prevent a speed from being recorded which increases a risk of collision in the hazard zone, such as a bottleneck. Thus, safety of a trip can be improved. In this case, the above-defined maximum speeds can be used in particular, and repetition is omitted here for reasons of compactness and readability. Here, too, a deceleration, in particular a braking, can be initiated automatically, solely on the basis of the environmental model, without a collision imminent. This can also allow for example that in parking garages and in town gasses, often unpredictable passengers will run into the roadway. These events, which are in some cases very difficult or even unpredictable, can thus be "taken into account overall".The environmental model can also comprise so-called "virtual bumper bars" and / or "virtual road sleepers" and / or "virtual hoots". Virtual bumper bars" are possible regions in which the environmental model places artificial "virtual objects" that are not present in reality in order to block a journey through these regions. Real objects can also receive a "virtual halo" in order to enlarge them in the environment model in such a way that the method starts from a larger object in the further steps than is actually the case. This allows earlier adaptation of the journey and / or earlier trajectory change to be carried out. Other trajectories may also be advantageous compared to a simulation without a "virtual halo". "virtual road thresholds" may in this context be regions in which a speed reduction is imposed since the environmental model simulates or artificially specifies an obstacle (here too the obstacle is not present in reality), which although it allows a journey across it, seems to assume a corresponding speed for a safe passage (apparent is awakened by the pure virtuality of the obstacle and interpreted by the method for real).According to a further aspect, it can be provided that the method is designed in such a way that only stationary objects are included in the environmental model and / or designed to include objects that are movable in such a way, in particular if the speed falls below a maximum speed threshold and / or the environmental model specifies a maximum speed. As a result, the method can be adapted to the field of application in which a driving maneuver is to be made safer. Static objects at rest can be parked cars, columns in a parking garage, objects on the road such as flower plugs, lamp posts, road signs, signboards, pylons and / or posts. Movable objects can be movable units as already described at another point.According to a further aspect, it can be provided that the method is designed to identify a closed vehicle platoon as a stationary object. This makes it possible to simplify the prediction of the trajectory and the ascertainment of advantageous trajectories, since the individual vehicles of a closed vehicle platoon no longer have to be taken into account individually. Rather, the vehicle platoon can be "interpreted" as a virtual wall and processed accordingly. The platoon of vehicles may be, in particular, a platoon of parked vehicles parked in the parking spaces, for example, at the edge of the way or in a parking garage.According to a further aspect, it can be provided that the method is designed to give the driver a recommendation for action and / or a warning, in particular simultaneously with an intervention. The driver can thus be informed about an action to be taken and / or the driver can be warned, including that an intervention will take place, in order, for example, to prevent the driver from being startled during the intervention. The maneuver and also the intervention are thus made more secure.A recommendation for action here may be any recommendation regarding an action performed by a driver in controlling a vehicle to control a vehicle, in particular one of steering, clutching, braking, acceleratoring and / or shifting.According to a further independent aspect, the object can be achieved in particular by a driver assistance device, wherein the driver assistance device is configured in particular in such a way as to carry out a method according to the invention. Furthermore, the driver assistance device can be designed to create an environment model in order to map obstacles in the environment model; to analyze a driving situation in such a way as to predict a trajectory lying ahead; to plan a trajectory group with the inclusion of the environment model and the analysis of the driving situation; and to identify at least one advantageous trajectory from the trajectory group.The driver assistance device can be specified in particular by the features and / or functions of the method according to the invention and / or of the control device according to the invention.A driver assistance device may consist of a plurality of devices and / or may be used in a motor vehicle. For example, the host vehicle and / or its driver assistance device can access and / or comprise a control unit according to the invention or a system comprising a plurality of control units according to the invention.The driver assistance device can also be further designed and configured to play at least one piece of information about the further maneuver profile, in particular with the display of at least one trajectory, and / or at least one warning and / or at least one action instruction to a driver, in particular via a visual field display and / or via an augmented reality interface. The driver can thus adapt a corresponding driving profile in real time before a collision occurs. The safety of the driving maneuver is thus improved.According to a further independent aspect, a control device or vehicle control device can be provided, designed and / or set up in such a way as to carry out a method according to the invention. The control device can in this case in particular also be specified by the features of the method. The control device can also be connected in particular to exterior-receptive sensors, i.e. sensors which sense the surroundings of a movable unit (here the ego vehicle), in order to be able to make the corresponding information (landmark positions, for example parking space markings in the parking garage or lane markings) available to a method according to the invention. In addition, the control device can be designed to control a highly automated movable unit. The control device can in this case comprise a storage medium (flash storage medium, magnetic storage medium, hard disk, USB stick, etc.) and a CPU.A control unit or a system comprising a plurality of control units can be used in a motor vehicle. For example, the host vehicle and / or its driver assistance system may include the control unit or the system of a plurality of control units.The control unit or the system can be set up and determined for use in a motor vehicle. The control device or the system can have an electronic controller. The control unit or the system can be or have an electronic control unit (ECU). A plurality of control units can be provided. The plurality of control units can be connected via a bus system, for example a "controller area network" (CAN), and / or can exchange data with one another. The electronic controller and / or the control device or system can have a microcomputer and / or processor. The control unit or system can comprise one or more sensors and / or be connected to these. The control device or system can comprise the computer program product described above and / or below. The control device or system can have a memory. The computer program product may be stored in the memory. The control unit or system can be designed to carry out the method described above and / or below.According to a further independent aspect, a vehicle or semi-autonomous vehicle or autonomous vehicle can be provided, which comprises a control device according to the invention and / or a driving assistance device according to the invention. The vehicle may be a motor vehicle, for example an ego vehicle. The vehicle may be configured and / or intended to carry out the method described above and / or below. The vehicle may include the computer program product, control device or system of a plurality of control devices described above and / or below. The vehicle can also comprise a driver assistance device which is in communication connection in particular with the control device according to the invention in order to carry out a method according to the invention.In this case, a vehicle of level 1 to level 4 can be understood in particular as a semi-autonomous vehicle, wherein a vehicle of level 5 can be provided as an autonomous vehicle. All vehicles of these categories can be highly automated vehicles in particular. The levels can be defined as follows, in particular according to the German federal legislation at the application time (or at the priority date):Level 1: Assisted Driving, wherein drivers constantly control their vehicle.Level 2: Partially automated driving, wherein drivers constantly control their vehicle.Level 3: Highly automated driving, wherein drivers may temporarily deviate from driving task and traffic.Level 4: Fully automated driving, wherein drivers can provide vehicle guidance over longer distances and in different traffic situations.Level 5: Autonomous driving, wherein there are only passengers without a driving task.The vehicle can be specified in particular by the features and / or functions of the method according to the invention and / or of the control device according to the invention and / or of the driving assistance device according to the invention.According to a further independent aspect, a computer program product can be designed in such a way that it can be executed on a control device according to the invention in order to carry out a method according to the invention. The computer program product can be designed in particular such that it can be stored (=stored) on the storage medium of the control unit. In particular, the computer program product can be configured such that it can be executed on the CPU of the control unit in order to carry out the method according to the invention. The computer program product can cause a device, such as an, for example, electronic, controller and / or controller and / or arithmetic unit / device, a control system, a driver assistance system, a processor or a computer, to execute the method described above and / or below. For this purpose, the computer program product can have corresponding data sets and / or program code means and / or the computer program and / or a storage medium for storing the data sets or the program. The computer program product may comprise program code means for, when the computer program product is executed on a processor, carrying out the method described above and / or below.The method can be stored as a computer program at least partially on a computer, microcomputer, in an electronic control and / or computing unit, in a control system, on a storage medium or on a machine-readable carrier and / or implemented there. The computer program can be distributed in software technology onto one or more storage media, control and / or computing units, such as electronic control units (ECUs) or computers, etc., in particular in the own vehicle (ego vehicle). The storage medium may be a semiconductor memory, a hard disk memory, or an optical memory.Developments and advantageous embodiments are evident from the dependent claims.The invention is described in more detail by the figures. Shown therein are: FIG. 1A shows a driving situation of an ego vehicle in the presence of objects in blind spots; FIG. 1B shows a driving situation of an ego vehicle when approaching a curb edge; FIG. 2 shows a driving situation of an ego vehicle by series of parked cars; and FIG. 3 shows a schematic illustration of a method according to the invention.The same reference numerals are used for identical and identically functioning features.FIG. 1A shows a driving situation 50 of an ego vehicle 1 in the presence of objects 9 in blind spots 10. These support elements, also called columns, are objects 9 in the sense of the method 60 according to the invention, as is described in detail in FIG. 3. These objects 9 are in this case in the way of free driving through the parking garage. These objects 9 are often even arranged directly around the parking space, which is why there is a high risk of collision with them during parking and parking. In particular, even minimal trajectory changes can lead to touching of the objects 9, which are often arranged laterally with respect to the ego vehicle 1. This touching or any other form of collision is to be avoided in this case. The ego vehicle 1 can accelerate from rest (starting) or carry out a journey (straight-line movement, but also accelerated movement). In both cases, this results in a trajectory 5 lying ahead, which is based on the current steering angle and the selected speed v and acceleration a.Here, for orientation, a coordinate system is schematically displayed when viewed from the bird's eye view, wherein the x direction represents the longitudinal direction of movement of the ego vehicle, the y direction represents the lateral direction of movement, and w represents the rotation of the moving vehicle, for example during cornering.The aim of the method 60 according to the invention is to make it possible to safely carry out a driving maneuver in a corresponding driving situation 50. For this purpose, a trajectory group 13 is planned which proposes different trajectories on the basis of a variance of the initial parameters of the journey. These are evaluated (weighted) accordingly in order to be able to decide whether the trajectory 5 lying ahead should be traveled on further or whether it is more advantageous to choose another trajectory, a so-called advantageous trajectory 4.An advantageous trajectory 4 is in particular such a trajectory that can be traveled without a collision 12. A disadvantageous trajectory 6 would here direct the ego vehicle 1 to a direct collision 12. Alternatively, however, the trajectory 5 lying ahead can also be selected if this itself represents an advantageous trajectory 4 or at least no disadvantageous trajectory 6. In other words, the currently running driving maneuver can therefore continue to run without intervention if there is no risk of collision. The method can nevertheless continue to run continuously in order to analyze the driving situation continuously.In order to ascertain the trajectory to be traveled from the trajectory family, an environment model 100 is created, which contains the objects 9 and also a virtual representation 11 of the ego vehicle 1. In this case, a virtual collision 12 acan be ascertained if a negative trajectory 6 were followed. This is schematically represented here by the shifted virtual representation 11 a, which, in the event of an excessively early steering angle during unparking, leads to a collision 12 with one of the two pillars next to the ego vehicle 1. This disadvantageous trajectory 6 is thus to be avoided.Here, the objects 9 are provided with a virtual halo 9 a. The ego vehicle 1 or its virtual representation 11 must respect this virtual halo 9 awhen the method 60 is carried out. As schematically shown here, the virtual station 9 a artificially enlarges the actual object 9 in the environment model in order to take into account a safety distance between the ego vehicle 1 and the object 9. In this case, a virtual collision 12 acan be assumed, although physically no collision 12 would yet take place, when driving on a corresponding negative trajectory 6. Thus, firstly security is increased, but secondly the method is also implemented in a resource-saving manner.The method also makes it possible to identify objects 9 which are arranged in a so-called blind spot 10 of the ego vehicle, as a result of which the object 9 cannot be seen by the driver. In order to detect the objects nevertheless, a whole series of experiential sensors are used, the position and arrangement of which is only shown schematically here and which can depend on the vehicle type.The ego vehicle 1 can have electronic systems in order to include the environment in planning a driving profile and in particular in carrying out a parking process or parking process. A control device 35 according to the invention (also as a control unit, control module, control device) comprises a computer program product 35 a, likewise according to the invention. The control device 35 can be connected to an inclination sensor 37, a laser source with a reflection sensor, a LIDAR system with a LIDAR sensor 38, an ultrasonic system with an ultrasonic sensor 39 and / or a radar 41, that is to say experiential sensors, in order to record real-time information which is determined via the sensors.The information recorded by the experieroceptive sensors of the ego vehicle 1 can relate to at least one other vehicle 2. This can be a moving other vehicle 2, but also parked other vehicles 2. For moving other vehicles 2, a relative speed, an absolute speed, a braking behavior, an acceleration behavior and / or an approach, in particular from the side during the braking process, can additionally be determined. Alternatively or additionally, this information can relate individually or in all conceivable combinations to a moving other vehicle 2. As a result, its trajectory can also be predicted.However, the experieroceptive sensors can also measure information about the environment itself in order to enable the driver assistance system 36 according to the invention (also referred to as a maneuver assistant, maneuver driving assistance device or maneuver driving assistance system) to play this information to a driver of the ego vehicle 1. This information can also be used to prepare an emergency operation in the event of a direct risk of collision in order to determine and carry out a steering intervention. Furthermore, this environmental information can serve to create and / or update an environmental model 100.Furthermore, the experieroceptive sensors can record information about the positions and the distribution of objects 12 in the environment in order to feed them into a map of the environment, in order to use this map as a basis for a method 60 according to the invention in order to create an environment model 100.Map information is also stored in memory 45 of a control device 35 according to the invention, on which ego vehicle 1 is localized by means of a GNSS module 40. The position and also the dimensioning of the ego vehicle 1 are thus included in the environment model 100. The previously described digital (or virtual) representation 11 of the ego vehicle 1 is created.FIG. 1B shows a driving situation of an ego vehicle when approaching a curb 8, wherein the ego vehicle is equipped with exterior-receptive sensors in a manner corresponding to that described in FIG. 1A and is referred back to it. The curb 8 is treated here like a wall and is created as a virtual obstacle which indicates virtual collisions 12 ain the case that a trajectory leads to a displaced virtual representation 11 awhich collide with the virtual wall. This makes it possible to prevent an ego vehicle or its driver from getting too sharp a curve (intersecting the curve), as a result of which damage to the tire and / or the rim and / or the underbody may possibly arise. In such a case, the disadvantageous trajectory 6 is likewise discarded. An advantageous trajectory 4 will be chosen to continue the trip.FIG. 2 shows a driving situation 50 of an ego vehicle 1 by series of parked other vehicles 2. an other vehicle 3 standing on the roadway is also in the travel path of the ego vehicle 1. In this case, three obstacles are present in this driving situation, the rows of parked other vehicles 2 are lateral (left and right) of the ego vehicle 1 and the other vehicle 3 in the path. If an environment model 100 of the garage or of the parking garage is already reserved, the environment model 100 is updated with the new (and temporarily present) obstacles. In order to facilitate a calculation and the prediction of the advantageous trajectory 4, the rows of parked foreign vehicles 2 are treated as singular, static objects 7. As a result, it is no longer necessary to image each individual parked other vehicle 2 of the row of parked other vehicles 2. In the present example, an advantageous trajectory 4 is selected via the trajectory 5 lying ahead in order to travel over this advantageous trajectory, since this trajectory allows it to travel around the other vehicle 3 standing in the path, but also since this avoids a virtual collision 12 awith the singular, static object 7. This also prevents a real collision 12 with the parked other vehicles 2. The disadvantageous trajectory 6 would lead here to a frontal collision with the parked, in-path other vehicle 3. Therefore, this is discarded here. However, touching an obstacle would also be disadvantageous. The collisions 12 and the virtual collisions 12 a, which would result from the driving of the trajectory 5 ahead or the driving of the disadvantageous trajectory 6, are not shown for reasons of clarity.FIG. 3 shows a schematic representation of a method 60 according to the invention. the information of the experieroceptive sensors, as described in detail, is introduced into the method 60 and used together with map information for the creation of an environment model. The ego vehicle 1 is located on the map and thus in the environment model by means of a GNSS module 40. Thus, environment modelling 61 is performed. The resulting environmental model 100 is then transmitted 62 to a driving behavior analysis 63. The driving behavior analysis 63 is then matched 64 with the environmental model from the environmental modelling 61 in such a way as to enable trajectory planning 66. In the trajectory planning 66, both a trajectory 5 lying ahead and a trajectory group 13 resulting therefrom are determined. This resulting trajectory group 13 is fed to an analysis for the property as an advantageous trajectory 4 or disadvantageous trajectory 6, wherein a value is determined for each trajectory of the trajectory group 13, which value specifies a measure for the trajectory, whereby a decision 68 about a control intervention 69 can be made. In this case, a trajectory to be traveled is transmitted 65 to the intervention decision, and the trajectory to be traveled is matched to the trajectory 5 lying ahead. In the event that there is a difference between these two trajectories, the intervention decision 68 would be positive and a control intervention command would be output 67 in such a way as to perform a control intervention 69. However, information can also be fed back 42 from each method step to the environment modelling 61 in order to update the environment model 100, the driving behavior analysis 63, the trajectory planning 66 and / or the intervention decision 68 and, if appropriate, to adapt it to new circumstances. The driver is also provided with a recommendation 75 for action and a warning 76, in the case of control intervention 69.As a result, it is possible to implement a method 60 that makes a driving maneuver 50 safer in towns, for example in parking garages and / or in narrow roads and lanes of cities.The method is not limited to the embodiments shown here, but rather can also cover the various implementations which are evident to the person skilled in the art from the description, figures and the claims presented here.Optional features of the invention are designated in particular by "can". Accordingly, there are also developments and / or exemplary embodiments of the invention which additionally or alternatively have the respective feature or the respective features.Isolated features can also be extracted from the combinations of features disclosed here, if necessary, and used in combination with other features to delimit the subject matter of the claim, resolving a structural and / or functional relationship optionally existing between the features. The order and / or number of all steps of the method can be varied.Reference numerals denote reference numerals1 Ego vehicle / host vehicle 2 Other vehicle / other vehicle 4 advantageous trajectory 5 Trajectory 6 lying ahead Trajectory / collision trajectory 7 Planned trajectory of the ego vehicle 8 Curb 9 Object / pillar 9 aof a virtual farm of an object / virtual farm of a pillar 10 11 Blind angle Vehicle Approximation / virtual representation of the ego vehicle 11 aVehicle Approximation at Predicted Position / Displaced Virtual Representation of the ego vehicle 12 Collision in the physical space 12 a Virtuelle 13 Trajectory Group 35 Control device 35 a Computer program product installed Storage medium of the control device and executed on control device CPU 36 Driving assistance device 37 Inclination sensor 38 LIDAR sensor 39 Ultrasonic sensor 40 GNSS module 41 Radar sensor 42 recursion information flow 45 memory 50 driving situation 60 method 61 environment modelling 63 driving behavior analysis 64 comparison 66 trajectory planning 67 output control command 68 intervention decision 69 control intervention 75 action recommendation 76 warning 100 environment model v ego vehicle speed w angle of rotation y y dimension of the movement from bird's eye view / lateral direction x x dimension of the movement (bird's eye view) / longitudinal directionReferences included in the specificationThis list of documents cited by the applicant has been produced in an automated manner and is only included for the better information of the reader. The list is not part of the German patent application or utility model application. The DPMA does not take any adhesion for any faults or omissions.Cited Non-Patent LiteratureGalleo" from EU, and "GLONASS" from Russian Federation, "Beidou" from Peoples republic of China (since 2004

[0032]

Claims

Method (60), in particular computer-implemented, designed to monitor, plan and / or perform a driving maneuver of an ego vehicle (1), the method (60) being characterized in that it has the steps of: - creating (61) and / or holding an environment model (100) designed to image objects (9) in the environment model; - analyzing (63) a driving situation (50) designed to predict a trajectory (5) lying ahead; and - planning (66) a trajectory group (13) by including the environment model (100) and / or the analysis of the driving situation (50); and - identifying at least one advantageous trajectory (4) from the trajectory group (13).Method (60) according to Claim 1, characterized in that the method (60) further has the step of weighing an intervention decision (68), designed in such a way as to make a decision about an intervention, based on the prediction of the trajectory (5) lying ahead, the planned trajectory group (13) and / or the identified advantageous trajectory (4).Method (60) according to either of Claims 1 and 2, characterized in that the method (60) is designed in such a way as to carry out at least one intervention and / or at least one intervention change, based on the intervention decision (68).Method (60) according to one of the preceding claims, characterized in that the method (60) is furthermore designed in such a way that the environment model (100) is produced on the basis of at least one map information item and / or is produced on the basis of at least one information item of an experiential sensor.Method (60) according to one of the preceding claims, characterized in that the method (60) is furthermore designed in such a way that the trajectory group (13) is planned on the basis of the trajectory (5) ahead predicted from the driving situation (50), wherein a minimum variance is used as the basis for the predicted trajectory (5) ahead.Method (60) according to one of the preceding claims, characterized in that the method (60) is furthermore designed in such a way that the at least one advantageous trajectory (4) is determined on the basis of the prediction of possible collision probabilities for the trajectories of the trajectory group (13) and of the currently traveled-on, preceding trajectory (5).Method (60) according to one of the preceding claims, characterized in that the method (60) is further designed such that the method (60) is carried out at a speed of less than 30 km / h, further less than 25 km / h, further less than 10 km / h, further less than step speed, in particular such that the method (60) is carried out continuously.Method (60) according to one of the preceding claims, in particular according to Claim 7, wherein the method (60) is designed to make a decision based on the environment model (100) to throttle a speed (v) of the ego vehicle (1) to a maximum speed, for a case in which the speed (v) of the ego vehicle (1) exceeds the maximum speed and / or to set the maximum speed as an upper speed limit, for the case in which the speed (v) of the ego vehicle (1) is below the maximum speed, wherein, for each case, the maximum speed is set in particular as a function of the environment model (100).Method (60) according to one of the preceding claims, characterized in that the method (60) is furthermore designed in such a way as to pick up only stationary objects (9) in the environmental model (100) and / or designed in such a way as to include movable objects (9), in particular if the speed (v) falls below a maximum speed threshold and / or the environmental model (100) specifies a maximum speed.Method (60) according to one of the preceding claims, characterized in that the method (60) is furthermore designed in such a way as to identify a closed vehicle platoon as a stationary object (7).Method (60) according to one of the preceding claims, characterized in that the method (60) is furthermore designed in such a way as to give the driver a recommendation for action (75) and / or a warning (76), in particular simultaneously with an intervention.Driving assistance device (36), characterized in that the driving assistance device (36), in particular configured in such a way as to carry out a method (60) according to one of the preceding claims, is configured in such a way as to create an environment model (100) in order to image objects (9) in the environment model (100); to analyze a driving situation (50) in such a way as to predict a trajectory (5) lying ahead; to plan a trajectory group (13) by including the environment model (100) and the analysis of the driving situation (50); and to identify at least one advantageous trajectory (4) from the trajectory group (13).Control device or vehicle control device (35), designed and / or configured in such a way as to carry out a method (60) according to one of the preceding Claims 1 to 11.Vehicle (1) or semi-autonomous vehicle or autonomous vehicle, comprising a driving assistance device (36) according to claim 12 and / or a controller (35) according to claim 13.Computer program product (35a) designed to be executed on a driver assistance apparatus (36) according to Claim 12 and / or on a control device (35) according to Claim 13 in order to carry out a method (60) according to one of Claims 1 to 11.

Images