Control system and control method for planning possible trajectories of a motor vehicle based on sampling

Abstract

A control system (10) is suitable for use in an own motor vehicle (12) and is configured and determined to monitor the current driving conditions of the own motor vehicle and another motor vehicle and determine an optimal trajectory for subsequent driving operations of the own motor vehicle (12) based on environmental data of the own motor vehicle (12) provided on a controller, which optimal trajectory the own motor vehicle (12) should follow. The control system is configured and determined to obtain information about the current driving conditions of the own motor vehicle (12) and / or at least one other motor vehicle based on the provided environmental data. The control system (10) is also configured and determined to obtain multiple lateral positions and multiple longitudinal positions and / or speeds based on the information about the current driving conditions of the own motor vehicle and / or another motor vehicle. Finally, the control system (10) is configured and determined to obtain, from the plurality of lateral positions and the plurality of longitudinal positions and / or speeds, a stopping point for an optimal trajectory which the own motor vehicle (12) should follow when performing a driving maneuver, and to determine the optimal trajectory of the own motor vehicle (12) by means of a spline-based interpolation between the obtained stopping points and depending on the lateral positions at the stopping points.

Description

Technical Field

[0001] A control system and a control method are described for determining a trajectory that a host motor vehicle is to follow in order to perform a driving maneuver. The control system and the control method are based in particular on environmental sensors in the host motor vehicle and assist the driver or the autonomously driven motor vehicle. In this case, in semi-autonomous motor vehicles and autonomously controlled motor vehicles, the control system and the control method contribute to increasing the safety and driving comfort of the motor vehicle passengers. Background Art

[0002] The growing interest in the various functions of automated motor vehicles has led to increasing demands for reliable and fast control systems and algorithms for motion planning (trajectory planning) of at least semi-autonomous motor vehicles. Not only the growing number of traffic participants, but also the cognitive tasks that must be handled in highly complex traffic scenarios with these traffic participants pose a huge challenge. In the architecture of control systems for semi-autonomous or autonomous motor vehicles, in addition to the recognition and control of certain traffic situations, it is mainly decision-making and motion planning that have a significant impact on the overall performance of the respective motor vehicle.

[0003] The control systems and algorithms for trajectory planning of semi-(autonomous) motor vehicles, which have been continuously developed in recent years, demonstrate the complexity of automated driving. In this regard, the challenge of motion planning is generally to obtain a comfortable and otherwise practicable and collision-free trajectory within a limited time in order to thereby ensure the real-time performance of the semi-(autonomous) system of the motor vehicle in a dynamically changing environment.

[0004] In the past, different approaches for trajectory planning have been developed. These relate, for example, to motion planning algorithms or techniques for trajectory planning based on sampling. The latter are generally used to simplify the motion planning problem in a specific way in order to meet specific requirements, such as limited computing time. Here, the (simplification) goal is to cover a continuous space with an infinite number of samples (sampled values).

[0005] In today's motor vehicles, driver assistance systems (ADAS) provide a number of monitoring and prompting functions to make the driving of motor vehicles safer. Here, the environment of the motor vehicle is monitored based on environmental data about the driving process of the own motor vehicle acquired by one or more environmental sensors located at the motor vehicle.

[0006] Known driver assistance systems, for example, know whether the motor vehicle is located in the lane and whether the driver does not want to drift to the side of the lane or is about to leave the lane. These driver assistance systems generate an "image" of the street and, in particular, the lane from the acquired environmental data. Objects are detected and, during driving, for example, sidewalk edges, lane boundaries, direction arrows, etc. are tracked.

[0007] Today's driver assistance systems also include so-called "blind spot detectors." These blind spot detectors detect, for example, using radar, lidar, video, etc., whether there are other motor vehicles, traffic participants, or objects to the side and / or behind the motor vehicle, so that changing lanes or turning the own motor vehicle may lead to a collision with them.

[0008] In addition, in so-called ACC systems (Adaptive Cruise Control), the speed of the own motor vehicle is automatically adjusted to the speed of the motor vehicle traveling ahead. In this case, a certain distance is always maintained from the motor vehicle traveling ahead. For this purpose, such systems detect the direction of movement and / or the speed of the motor vehicle traveling ahead in order to avoid the own motor vehicle crossing the path of the motor vehicle traveling ahead and causing a dangerous situation. This applies on the one hand to lane changes or turning processes and on the other hand to rear-end collisions.

[0009] In motor vehicles driven by humans, driver assistance systems usually provide a warning function in order to warn the driver before a dangerous situation or a corresponding operation, or to advise the driver on appropriate operation of the motor vehicle. Driver assistance systems can also be used in autonomously controlled motor vehicles in order to provide the autonomous controller with corresponding environmental data.

[0010] Basic Questions

[0011] In street traffic, situations may arise in which a driving maneuver is performed by the driver or a (semi-)autonomous driver assistance system of the own motor vehicle. For example, a curved lane course may already require a corresponding driving maneuver of the own motor vehicle.

[0012] However, the current situation of one's own motor vehicle is not constant, but changes continuously in reality. As a result, other traffic participants may change lanes or speeds intentionally or unintentionally due to an accident, for example. In addition, the current driving situation of one's own motor vehicle has changed due to one's own driving behavior and / or the constantly changing lane direction. Reacting to such changes in the current situation in a reasonable and timely manner poses a huge challenge not only to conventional driver assistance systems but also to human drivers. In addition, the dynamics of many traffic situations and the ever-increasingly dense traffic are reflected in the ever-increasing resource consumption of conventional driver assistance systems.

[0013] The object is therefore to provide a control system and a control method for a motor vehicle which increase the driving safety and driving comfort of the motor vehicle as a function of the current traffic situation in which the motor vehicle finds itself. In addition, resources for the control system are saved in this case by calculating trajectories for subsequent driving maneuver decisions of the motor vehicle in an efficient manner.

[0014] Proposed Solution

[0015] This object is achieved by a control system having the features of claim 1 and a control method having the features of claim 12 .

[0016] Preferred embodiments can be seen in the dependent claims 2 to 11 and 13 and in the following description.

[0017] One aspect relates to a control system configured and determined to be used in a own motor vehicle. Such a control system identifies lanes, road boundaries, road signs, other motor vehicles and / or other different objects based on environmental data acquired by at least one environmental sensor assigned to the own motor vehicle. The at least one environmental sensor is configured to provide environmental data to an electronic controller of the control system that reproduces the area in front of, beside and / or behind the own motor vehicle. The control system is at least configured and determined to be used to obtain information about the current driving condition of the own motor vehicle and at least one other motor vehicle based on the provided environmental data. The control system is also at least configured and determined to be used to obtain multiple lateral positions based on the information about the current driving condition of the own motor vehicle and / or the other motor vehicle, and to obtain multiple longitudinal positions and / or speeds based on the information about the current driving condition of the own motor vehicle and / or the other motor vehicle. Finally, the control system is also configured and determined to: obtain stopping points for a trajectory of the own motor vehicle from the multiple lateral positions and the multiple longitudinal positions and / or speeds, the own motor vehicle should follow the trajectory when performing a driving maneuver, and determine the trajectory of the own motor vehicle by means of spline-based interpolation between the obtained stopping points and depending on the lateral positions at the stopping points.

[0018] When obtaining lanes, road boundaries, road signs, other motor vehicles and / or other different objects, the control system can be configured and determined to, for example, obtain the lane in which the own motor vehicle is located and other lanes in which other motor vehicles are located in the area in front of, beside and / or behind the own motor vehicle, and determine the other motor vehicles in the area in front of, beside and / or behind the own motor vehicle as other motor vehicles to be detected. The other lanes can correspond to the own lane or be adjacent to the own lane. This ensures that not only motor vehicles traveling on the same lane as the own motor vehicle but also other motor vehicles traveling on lanes adjacent to the lane of the own motor vehicle can be determined as other motor vehicles to be detected.

[0019] In addition, in particular lateral movements of (the longitudinal axis of) the own motor vehicle and / or the other motor vehicle in the direction of the lane boundary of the lane in which the own motor vehicle and / or the other motor vehicle is located can be detected. In particular, the other motor vehicle can be identified by means of the control system as a further motor vehicle that is located in front of and / or behind and / or laterally in front of and / or laterally behind the own motor vehicle and that has a minimum distance to the own motor vehicle in the current driving situation (at the respective current point in time).

[0020] The control system can, for example, learn about the own lane and other lanes by means of provided environmental features (e.g., lane boundaries and / or lane markings). The control system can also recognize these environmental features based on environmental data acquired by means of the at least one environmental sensor. In addition, the control system can also acquire environmental features from an external system (e.g., from a GPS system) for use.

[0021] The other motor vehicle can be detected over a predetermined time span or continuously by means of the at least one environmental sensor of the own motor vehicle, and the corresponding data can be provided to the control system of the own motor vehicle. Here, the control system is also configured and determined to: obtain a change in distance between the longitudinal axis of the other motor vehicle and the center line, at least one lane boundary or at least one lane marking of the relevant lane in which the other motor vehicle is located during a predetermined time span or continuously. The center line and lane boundary of the relevant lane can here be a virtual, current center line or current lane boundary of the relevant lane determined by the control system. The control system can also obtain a change in distance between the longitudinal axis of the other motor vehicle and a virtual or actual lane marking or lane boundary in which the own motor vehicle is located during a predetermined time span or continuously.

[0022] The control system can be configured and determined to be used to: obtain the distance between the own motor vehicle and other motor vehicles located in front of and / or to the side and / or behind the own motor vehicle, as well as the speed difference and / or acceleration difference between the own motor vehicle and the other motor vehicles in order to obtain the current traffic conditions, wherein the own motor vehicle and the other motor vehicles can be in the same lane or in adjacent lanes.

[0023] The control system can combine the determined distances, speed differences and / or acceleration differences between the own motor vehicle and the other motor vehicle in any desired manner and link them to one another.

[0024] It is understood that the control system can also be configured and determined to detect other motor vehicles participating in the traffic by means of the at least one environmental sensor over a predetermined time span or continuously in order to obtain the current traffic situation. Thus, the control system can also obtain the previously described changes in relative distances, speed differences and / or acceleration differences over a predetermined time span or continuously, specifically relative to all other motor vehicles or other objects in the environment close to the own motor vehicle in the current traffic situation.

[0025] In certain embodiments, the control system may also be configured and determined to: obtain a basic driving operation based on the provided environmental data, the motor vehicle itself should perform the basic driving operation, and determine at least one of the multiple lateral positions based on the obtained basic operation and / or assign a numerical value to at least one of the lateral positions, the numerical value representing a basic operation type, wherein the basic operation type is included in a basic operation type set, the basic operation type set being composed of at least basic operations for changing lanes to the left, changing lanes to the right, and keeping lanes.

[0026] Alternatively or additionally, the basic operation type set may include other basic operations of keeping lane left and / or keeping lane right.

[0027] The control system may also be configured and adapted to determine the plurality of longitudinal positions based on a longitudinal acceleration of the own motor vehicle and / or of the further motor vehicle.

[0028] This may be, for example, the maximum possible acceleration of the own motor vehicle based on the engine power of the own motor vehicle. Alternatively or additionally, this may be a maximum acceleration value of the own motor vehicle and / or another motor vehicle derived for safety and comfort reasons, which may be predetermined based on the current driving situation or in another suitable manner.

[0029] In certain embodiments, the control system may also be configured and designed to determine the plurality of longitudinal positions by means of an adaptive discretization, wherein accelerations of the own motor vehicle and / or the further motor vehicle are used as control parameters for the adaptive discretization.

[0030] This may be, for example, the longitudinal and / or lateral acceleration described above or respectively a corresponding maximum value thereof.

[0031] Furthermore, this may particularly involve the acceleration of the own motor vehicle in the direction of travel and / or the acceleration of the other motor vehicle in the direction of travel which is present at the respective point in time when the current driving situation of the own motor vehicle and / or the other motor vehicle is determined.

[0032] By means of an adaptive discretization, a non-linear mapping of the linearly sampled longitudinal positions and / or velocities of the own motor vehicle and / or of the further motor vehicles can be established.

[0033] Furthermore, by means of the adaptive discretization, in particular when evaluating a highway and / or freeway scenario, a changing longitudinal acceleration of the own motor vehicle and / or other motor vehicles can be taken into account when determining the trajectory of the own motor vehicle.

[0034] The control system may also be configured and designed to determine the trajectory of the own motor vehicle taking into account a static crash test based on a plurality of lateral positions of the own motor vehicle and / or the further motor vehicle.

[0035] The control system can further be configured and designed to determine the trajectory of the own motor vehicle taking into account a dynamic crash test, wherein the dynamic crash test is based on the multiple lateral positions and / or on the multiple longitudinal positions and / or speeds of the own motor vehicle and / or the other motor vehicle.

[0036] In this case, the control system can also be configured and designed to carry out the dynamic crash test taking into account a maximum acceleration of the own motor vehicle and / or of the further motor vehicle.

[0037] Here, the control system may also be configured and designed to carry out the static collision test and / or the dynamic collision test using a separation set.

[0038] The control system can also be configured and designed to check the trajectory of the own motor vehicle with the aid of an objective function that is based on a cost function.

[0039] The target function can be assigned a target state and the deviation of the target function from the target state can be used as the target of the target function. Alternatively or additionally, a further target of the target function in the form of a lateral and / or longitudinal acceleration of the own motor vehicle and / or in the form of a lateral and / or longitudinal jerk of the own motor vehicle can be used as the target state of the target function.

[0040] A comprehensive objective function can then be obtained, the individual items of which respectively model the avoidance of dynamic collisions, the avoidance of static collisions, feasibility and driving comfort for the current traffic situation of the own motor vehicle 12, so as to select the trajectory of the own motor vehicle for each current driving situation and for each operation type.

[0041] The control system can also be configured and determined to: determine the multiple lateral positions and / or multiple longitudinal positions and / or speeds of the own motor vehicle and / or the other motor vehicle in curvilinear coordinates with respect to a reference lane provided to the control system of the own motor vehicle via the environmental data, and to verify the trajectory of the own motor vehicle in orthogonal coordinates.

[0042] In this case, interpolation in orthogonal coordinates can also be performed.

[0043] As a result, trajectory planning for driving on straight and curved sections can be performed independently of the street geometry.

[0044] In contrast to the evaluation of candidate trajectories in curvilinear coordinates, a time-intensive point-by-point transformation of the trajectories obtained for obstacles (e.g., other motor vehicles) is not necessary. In addition, the transformation of only the stop points together with the interpolation and evaluation in orthogonal coordinates (motor vehicle coordinates) makes it possible to avoid the need to transform each individual trajectory point. For reasons of accuracy, the resolution of the stop points can be selected as a function of the street curvature, thereby minimizing deviations from a reference lane (reference trajectory) due to the interpolation.

[0045] The control system may also be configured and designed to determine the trajectory by means of an objective function based on a cost function.

[0046] The control system may also be configured and determined to perform a static collision test when determining the trajectory of the own motor vehicle. The static collision test is based on the plurality of first sample values.

[0047] The control system may also be configured and determined to perform a dynamic collision test when determining the trajectory of the own motor vehicle. The dynamic collision test is based on the plurality of second sample values.

[0048] The control system may also be configured and adapted to carry out the dynamic crash test taking into account a maximum acceleration of the own motor vehicle.

[0049] Here, the control system can also be configured and designed to carry out the static crash test and / or the dynamic crash test using a separate set.

[0050] Another aspect relates to a control method for identifying road boundaries, road signs, other motor vehicles and / or objects in an area in front of, beside and / or behind the own motor vehicle based on environmental data acquired by at least one environmental sensor assigned to the own motor vehicle, wherein the control method is implemented in particular with the aid of the control system described above. The control method comprises at least the following steps:

[0051] - obtaining information about the current driving situation of the own motor vehicle and of at least one further motor vehicle based on the provided surroundings data,

[0052] - obtaining a plurality of lateral positions based on the information about the current driving situation of the own motor vehicle and / or the further motor vehicle,

[0053] - obtaining a plurality of longitudinal positions and / or speeds based on the information about the current driving situation of the own motor vehicle and / or the further motor vehicle,

[0054] - obtaining a stopping point for a trajectory of the own motor vehicle, which the own motor vehicle should follow when carrying out the driving maneuver, and

[0055] The trajectory of the own motor vehicle is determined by means of a spline-based interpolation between the determined stopping points and as a function of the lateral positions at the stopping points.

[0056] Yet another aspect relates to a motor vehicle comprising the control system described above.

[0057] Compared with conventional driver assistance systems, the solution proposed here improves the correct prediction and correct identification of the current driving conditions of the own motor vehicle and other motor vehicles. In addition, a two-stage trajectory candidate planning and evaluation scheme implemented in real time is provided, which can accurately and efficiently obtain the best possible (optimal) trajectory for the future driving operation of the own motor vehicle.

[0058] This best possible trajectory can therefore be determined as a reasonable response to the current traffic situation that the own motor vehicle finds itself in. The environmental data acquired by means of the at least one environmental sensor here continuously change depending on the actual traffic situation and driving situation and can be updated periodically.

[0059] If the best possible trajectory for the future driving maneuvers of the own motor vehicle is used, this improves the driving comfort and driving safety of the own motor vehicle, since not only static and dynamic crash tests are carried out, but also the driving dynamics of the own motor vehicle are taken into account when determining the trajectory. The control system can also ensure the driving dynamics achievability of the best possible trajectory, which additionally improves driving safety. In addition, in other embodiments of the control system, the driving behavior of the own motor vehicle is optimized by taking into account the driving dynamics of the own motor vehicle when tracking the trajectory. Driving dynamics are to be understood here, for example, as the longitudinal acceleration and the lateral acceleration of the own motor vehicle.

[0060] It will be appreciated by those skilled in the art that the previously described aspects and features may be arbitrarily combined in a control system and / or a control method. Although some of the features described above are described with respect to a control system, it will be appreciated that these features may also be applicable to a control method. The features described above with reference to a control method may also be applicable to a control system in a corresponding manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Further objects, features, advantages and application possibilities emerge from the following description of exemplary embodiments, which are not to be understood as limiting, with reference to the drawings. All features described and / or illustrated in the drawings represent the subject matter disclosed herein individually or in any combination. The dimensions and proportions of the components shown in the drawings are not to scale. Identical or equivalent components are provided with the same reference numerals.

[0062] Figure 1 A motor vehicle is schematically shown according to a particular embodiment, the motor vehicle having a control system and at least one environmental sensor.

[0063] Figure 2 The architecture of a two-stage approach for planning and evaluating candidate trajectories according to certain embodiments is schematically illustrated.

[0064] Figure 3 The schematic diagram shows sideways reference paths (reference trajectories) for five different sideways operation types according to certain embodiments.

[0065] Figure 4 A nonlinear diagram for going from a linear discretization to an adaptive discretization using different exemplary acceleration values ​​as control parameters according to certain embodiments is schematically illustrated.

[0066] Figure 5 Graph schematically illustrating the relative cost of adaptive discretization compared to linear discretization for different amounts of longitudinal acceleration in accordance with certain embodiments.

[0067] Figure 6 The transformation from curve coordinates K to vehicle coordinates F according to a specific embodiment is schematically shown.

[0068] Figure 7 A set of candidate trajectories obtained for a straight road segment (upper figure) and a curved road segment (lower figure) according to a specific embodiment are schematically shown.

[0069] Figure 8 Schematically illustrating the best possible (optimal) trajectory determined by a control system from a set of candidate trajectories obtained for a straight section (upper diagram) and a curved section (lower diagram) according to a particular embodiment.

[0070] Fig. 9 The schematic shows a separation set applied according to a specific embodiment.

[0071] Fig.10 The average runtime for a dynamic crash test according to certain embodiments is schematically illustrated.

[0072] Fig.11 A schematic diagram illustrating dynamic collision testing using disjoint collections according to certain embodiments.

[0073] Fig.12 Schematic diagram showing a two-stage trajectory planning and evaluation scheme according to certain embodiments, with a first stage (left side) for sampling-based deriving of candidate trajectories and a second stage (right side) for collision testing.

[0074] Fig.13 The first five illustrations (from top to bottom) schematically illustrate a scenario of entering a highway from an off-ramp using a two-stage trajectory planning and evaluation approach according to certain embodiments.

[0075] Fig.14 The next five illustrations (from top to bottom) schematically illustrate a scenario of entering a highway from an off-ramp using a two-stage trajectory planning and evaluation approach according to certain embodiments.

[0076] Fig.15 Schematically shown in Fig.13 and 14 The speed of the motor vehicle during the on-ramp onto freeway scenario shown in . DETAILED DESCRIPTION

[0077] In the following disclosure, specific aspects are mainly described with respect to the control system. However, these aspects are of course also valid within the scope of the disclosed control method, which can be implemented, for example, by a central control unit (ECU) of a motor vehicle. This can be done with appropriate read and write access to a memory assigned to the motor vehicle. The control method can be implemented in the motor vehicle as hardware, as software or as a combination of hardware and software. This also includes digital signal processors, application specific integrated circuits, field programmable gate arrays and other suitable switching components and computing components.

[0078] Figure 1 A motor vehicle 12 (own motor vehicle) is schematically shown, which includes a control system 10. The control system 10 is connected to at least one environmental sensor 14, 16, 18 located at the motor vehicle 12 in order to obtain environmental data from the at least one sensor 14, 16, 18. The control system 10 may include an electronic controller ECU (Electronic Control Unit, not shown in the figure). For example, the control system 10 in this case is configured and determined by means of the ECU and / or other electronic control systems to obtain the trajectory of the motor vehicle 12 and / or obtain the trajectory of another motor vehicle, which is located in the environment of the own motor vehicle in the current driving state of the own motor vehicle 12. For example, the ECU receives signals from the environmental sensors 14, 16, 18, processes these signals and the related environmental data and generates control signals and / or output signals.

[0079] exist Figure 1 1 shows three environment sensors 14, 16, 18, which send corresponding signals to the control system 10 or the electronic control unit ECU. In particular, at least one environment sensor 14 pointing forward in the driving direction of the motor vehicle 12 is arranged on the motor vehicle 12, which detects an area 22 in front of the motor vehicle 12. The at least one environment sensor 14 can be arranged, for example, in the area of ​​the front bumper, the headlights and / or the front radiator grille of the motor vehicle 12. The environment sensor 14 thus detects the area 22 directly in front of the motor vehicle 12.

[0080] In the area of ​​the front window of the motor vehicle 12, at least one additional or alternative environment sensor 16 is shown, which is also directed forward in the direction of travel of the motor vehicle 12. For example, this environment sensor 16 can be arranged between an interior rearview mirror and the front window of the motor vehicle 12. Such an environment sensor 16 detects an area 24 in front of the motor vehicle 12, wherein, depending on the configuration of the motor vehicle 12, the area 24 directly in front of the motor vehicle 12 cannot be detected due to the front section of the motor vehicle 12 (or its geometry).

[0081] Furthermore, at least one environment sensor 18 is arranged laterally on the motor vehicle 12 and / or at the rear of the motor vehicle. This optional environment sensor 18 detects an area 26 located laterally on the motor vehicle 12 and / or behind the motor vehicle 12 in the direction of travel of the motor vehicle. For example, the data or signals of the at least one environment sensor 18 can be used to verify the information detected by the further environment sensors 14, 16 and / or to determine the curvature of the lane in which the motor vehicle 12 is traveling.

[0082] The at least one environmental sensor 14, 16, 18 can be implemented arbitrarily and include a front camera, a rear camera, a side camera, a radar sensor, a lidar sensor, an ultrasonic sensor and / or an inertial sensor. For example, the environmental sensor 14 can be implemented in the form of a front camera, a radar sensor, a lidar sensor or an ultrasonic sensor. The front camera is particularly suitable for the environmental sensor 16 positioned at a higher position, while the environmental sensor 18 arranged in the rear of the motor vehicle 12 can be implemented in the form of a rear camera, a radar sensor, a lidar sensor or an ultrasonic sensor.

[0083] The electronic control unit ECU processes environmental data acquired by environmental sensors 14 , 16 , 18 located at the motor vehicle 12 in order to obtain information about the static environment of the motor vehicle 12 (immovable environmental objects, such as road boundaries) and the dynamic environment (movable environmental objects, such as other motor vehicles or traffic participants).

[0084] The electronic control unit ECU then processes the environmental data acquired by the environmental sensors 14, 16, 18 located at the motor vehicle 12 in order to detect the lane in which the motor vehicle 12 is traveling with a first lateral lane boundary and a second lateral lane boundary in front of the motor vehicle 12. In addition, the electronic control unit ECU processes the environmental data acquired by the environmental sensors 14, 16, 18 located at the motor vehicle 12 in order to detect lanes occupied by objects (the lanes are adjacent to the lane in which the own vehicle is traveling, wherein adjacent means that there may also be one or more other lanes between the adjacent lanes) and lateral lane boundaries of the lanes in front of, beside and / or behind the motor vehicle 12. The other objects are here another motor vehicle moving along a lane adjacent to the lane of the own motor vehicle or any other possible obstacle on the lane in front of this other motor vehicle.

[0085] To this end, the environmental sensors 14, 16, 18 of the electronic control unit ECU provide environmental data that reproduce the area in front of, beside and / or behind the vehicle. To this end, the control system 10 communicates via at least one data channel or bus (in Figure 1The data channel or bus can be implemented by means of cables or wirelessly.

[0086] Alternatively or additionally, the control system 10 or its electronic control unit ECU can also obtain data from one or more other auxiliary systems 20 or other controllers 20 of the motor vehicle 12, which data indicate or can be derived from these data the lanes and lateral lane boundaries in which the own motor vehicle 12, other motor vehicles and other motor vehicles are traveling. The data and information obtained by the other systems can thus already be used by the control system 10.

[0087] In addition, the control system 10 or its electronic control unit ECU uses environmental sensors (that is, based on environmental data acquired by means of the at least one environmental sensor 14, 16, 18) to determine the driving situation. Alternatively or additionally, an already existing assistance system 20 or electronic control unit 20 can also transmit data and / or information here, which define the driving situation or from which the driving situation can be quickly derived. Depending on the driving situation obtained, at least one possible trajectory is then determined, which the own motor vehicle 12 is to follow in the further course of driving.

[0088] The driver assistance system 20 or the electronic control unit 20 can further be configured and designed for semi-(autonomous) control of the motor vehicle. The control system 10 is configured and designed in this case to output data to the driver assistance system 20 or the electronic control unit 20 for autonomous driving. The control system 10 (or its ECU) can in particular output the following data to the component 20: the data specifies the curve of a specific trajectory that the own motor vehicle 12 should follow in the further course (after the current traffic situation). The data can also be transmitted in a wired manner or in a wireless manner via a data channel or bus.

[0089] The information acquired from the environmental data and provided to the control system 10 includes, for example, the position and / or speed and / or acceleration of the own motor vehicle 12 and / or the other motor vehicle in the lateral direction, in the longitudinal direction or in a combination of the lateral and longitudinal direction. In addition, this information includes, for example, the relative speed and / or relative acceleration of the own motor vehicle 12 and the other motor vehicle in the respectively mentioned direction.

[0090] Figure 2An overview of the approach presented here, which is based on real-time sampling of environmental data provided to the control system 10, is provided in order to plan one or more trajectories of the own motor vehicle 12 and in order to determine the best possible (optimal) trajectory among the planned trajectories for the further driving process of the own motor vehicle, which driving process is particularly the basis for the control system 10 and the control method presented here. It can be seen that trajectory candidates are first generated by the control system 10. These trajectory candidates are based on lateral (lateral) operation types: left lane change (SWL), lane keeping left (SHL), lane keeping in the middle (SHM), lane keeping right (SHR) and lane change right (SWR). However, the present disclosure is not limited to this; more or fewer or different operation types can be defined instead.

[0091] The control system / control method then selects the best possible trajectory candidate from the trajectory candidates based on the type of operation using an objective function. Target states are taken into account in the objective function, which relate to the dynamic and static environment of the own motor vehicle 12 in the current driving situation and the driving comfort and feasibility of the best possible trajectory candidate. Within the scope of the present disclosure, a trajectory is then considered unachievable, for example, if the driving characteristics of the own motor vehicle 12 do not result in the acceleration required for the trajectory.

[0092] Static and dynamic collision tests are then performed in order to determine the optimal trajectory of the own motor vehicle 12 among the best possible trajectory candidates. The own motor vehicle 12 then follows this optimal trajectory in the further course of the current traffic situation.

[0093] The proposed sampling strategy also enables independent longitudinal (extending in the direction of travel) operating strategies. The desired operating characteristics at a higher level are formed by the target state. However, the decision about the optimal trajectory at the trajectory planning level is made with the help of a two-level candidate selection strategy. The above-mentioned target function forms the first level, which provides the best possible trajectory from the maneuver type. The second level is characterized by static and dynamic collision tests.

[0094] However, before describing the sampling strategy in detail, the mathematical basis applied in trajectory planning will be explained below.

[0095] The control system and control method presented within the scope of the present disclosure are suitable for referring to the corresponding time state t l To generate the example state (or sample state) x of the stopping point (also called support point or node) l In other words, for each time state t lAt least one stopping point is assigned. That is, at least the lateral and longitudinal position values ​​of the own motor vehicle and / or the other motor vehicle are also taken into account in the determination of the stopping point. Within the scope of the following disclosure and the drawings, the geodetic coordinate system, the motor vehicle coordinate system and the curvilinear coordinate system follow the designations E, F and K. The geodetic coordinate system and the motor vehicle coordinate system follow an orthogonal (e.g. Cartesian) frame. The corresponding letters in capital letters before the corresponding values ​​indicate the corresponding coordinate system.

[0096] In order to be able to interpolate between the sampled (sampled) states, the movement F of the self-propulsive vehicle 12 is determined. x (t) and F y (t), where the interpolated x(t) is defined as follows.

[0097]

[0098] For an η-1 spline segment and at t∈{t l , t l+1} holds true, the spline s is defined as follows.

[0099] s l (t) = c v，l (tt l ) v +c v，l (tt l ) v-1 +…+c 1，l (tt l )+c 0，l .

[0100] The properties of the spline require that it be extended through the stop point s l (t l )=x l Another requirement is that the spline is always between the corresponding spline intervals, that is, it satisfies s l (t l+1 )=s l+1 (t l+1 ).

[0101] Using x l =[ F x l , F y l ] T The corresponding start and end conditions and the corresponding derivation can determine the spline coefficient c 0...v，l Each of the l=1...η support points (underlined in the following formula) of the trajectory determined in this way yields:

[0102]

[0103] The above relationship also holds true for N = (v-1) / 2. In order to determine the optimal time for transitioning from one spline interval to the next, time instances (Zeitinstanzen) are used during sampling. t l The formulation based on splines also allows independent of the support point l and with a fixed prediction horizon T P The choice of resolution with respect to time ΔT is made independently. For this reason, by interpolating n=T P / ΔT+1 obtains the trajectory point x k =[ F x k , F y k ] T The trajectory points are then summarized in a trajectory (which forms a candidate trajectory) as follows.

[0104] x 最优 =[x1,t1,x2,t2,...,x k , t k , ..., x n , T P ] T .

[0105] where a constant time interval t∈[t k , t k+1 ].

[0106] The two-stage approach presented here for trajectory planning and evaluation is therefore particularly advantageous over conventional approaches, since the above-described spline-based formulation is combined with trajectory generation using curvilinear coordinate transformation. The lateral and longitudinal positions of the stopping point obtained in curvilinear coordinates are sampled and subsequently transformed into vehicle coordinates. For example, lane boundaries and / or lane markings of the lane currently being traveled are used as references for the transformation into vehicle coordinates. The transformed stopping point is linked to the above-described spline-based interpolation. This results in a set of candidate trajectories, which are then evaluated in vehicle coordinates.

[0107] exist Figure 3 2 shows an exemplary driving situation of the own motor vehicle 12. The own motor vehicle 12 is moving in the middle lane 30 of a three-lane street / road (for example, a highway) 32. There are no obstacles such as other motor vehicles on the lane 32. In addition, Figure 3The lateral (lateral) reference paths (reference trajectories) 28a to 28e are shown, which represent the operations described at the beginning: changing lane to the left (reference trajectory 28a), keeping lane to the left (reference trajectory 28b), keeping lane in the middle (reference trajectory 28c), keeping lane to the right (reference trajectory 28d) and changing lane to the right (reference trajectory 28d).

[0108] The type of sampling of the lateral position of the own motor vehicle 12 is first based on the structure of the current environment of the own motor vehicle 12. The sampled values ​​are obtained from the environmental data. The set of lateral (lateral) operation types used is based on the planning of the operating characteristics of the own motor vehicle 12 at a higher level (for example by the electronic controller 20 or any other suitable electronic controller of the own motor vehicle 12). The mentioned operation types of lane change left (SWL), lane keeping left (SHL), lane keeping in the middle (SHM), lane keeping right (SHR) and lane change right (SWR) form the selected operation types, but the present disclosure is not limited to this. The control system 10 (and also the control method) can also work alternatively with three (for example lane change left, lane keeping and lane change right) selected operation types.

[0109] However, the greater number of included maneuver types results in a greater variability in the maneuvers of the motor vehicle 12 (compared to a system which, for example, classifies the maneuvers only according to SWL, SWR and lane keeping), in particular with respect to the lane 30 in which it is currently travelling. Complex combined maneuvers can then also be classified (for example with the aid of SHM and SHR). Such maneuvers result, for example, in the motor vehicle 12 approaching another lane (to the left or right of the lane in which it is currently travelling) in order to, for example, fill a gap there. The use of the curve coordinate K here sensibly simplifies the sampling of the lateral position, since each of the five maneuver types is directly related to the lane 30 in which the vehicle is currently travelling. K y t In other words, the lateral sampling of the (static) vehicle environment at each sampling time point provides a value at least based on the acquired operation type or a value assigned to this operation type.

[0110] Next reference Figure 4 The sampling in the driving direction (longitudinal sampling) is described. The figure shows a nonlinear mapping from linear discretization to adaptive discretization. Adaptive discretization is used for longitudinal sampling, that is, to obtain the longitudinal position (or position value) and / or speed (or speed value) of the own motor vehicle 12 and / or other motor vehicles.

[0111] Due to limited computing resources in semi-(autonomous) motor vehicles, it is necessary to generate an infinite but still meaningful set of possible trajectories (trajectory candidates). For this reason, within the scope of the present disclosure, a motion-based sampling strategy is used in addition to a state-based sampling strategy for lateral sampling in order to obtain a good estimate of longitudinal parameters such as position. K x t and / or speed K v t The longitudinal acceleration (in the direction of travel of the motor vehicle) of the own motor vehicle and / or another motor vehicle is sampled. K a x (t) is used as a control parameter. A second-order model is used so that the position is generated according to the following relationship with reference to the curvilinear coordinate system K (again capitalized before the corresponding values ​​in the following relationships): K x t and speed K v t longitudinal sample.

[0112]

[0113] Then, for adaptive discretization, we obtain an infinite number of control input sets K a x ∈u. Assuming that the best possible control input for the currently to-be-performed planning step (planning cycle) of trajectory planning is close in time to the previous planning step, a higher resolution is required in the range of the actual longitudinal acceleration. However, the minimum and maximum acceleration values ​​must be taken into account.

[0114] Therefore, if Figure 4 As shown in FIG. 1 , two parabolas f1 and f2 are used in the adaptive (adapted to the situation) discretization strategy derived therefrom in order to determine the final acceleration value Z of the own motor vehicle 12 and / or of the other motor vehicle online (ie in real time or substantially in real time) 当前 To transform the linear sampled values ​​z (obtained from the environmental data). The coefficients required for this are determined from the following conditions:

[0115] f1(z 最小 )=z 最小

[0116] f2(z 最大 ))=z 最大

[0117] in is defined as The adaptive (matched) sample value f(z) is then determined as follows.

[0118]

[0119] exist Figure 4 The diagram shows the motor vehicle 12 and / or other motor vehicles and a 当前 = 0 (for the respectively applicable sampling condition) acceleration value. The horizontally extending grid lines next to each other in the figure show the 当前 = -2 Higher resolution desired for adaptive discretization within the current value range.

[0120] The advantages gained from the adaptive discretization are shown in the evaluation of scenarios when entering a motorway and / or expressway from an on-ramp. Such scenarios involve varying longitudinal accelerations that are required in order to successfully handle dynamic everyday traffic. Figure 4 ] shows a linearly discretized reference trajectory with nearly continuous values, on which the adaptive discretization 4 is recorded. Thus, the best possible solution in terms of the objective function is approximated.

[0121] From the reference solution Φ 优化 The relative cost ΔΦ is calculated as the difference between the open-loop result of the suboptimal (next best possible) solution Φ of the solution with the corresponding longitudinal sampling strategy. Four variants are tested here, which are used to compare linear discretization and adaptive discretization in terms of the sampled longitudinal acceleration ζ=5 and ζ=10. The results are Figure 5 It can be seen that adaptive sampling has an advantage over linear sampling due to its lower relative cost.

[0122] If the adaptive sampling strategy ( Figure 5 In 自适应 =5 and ζ 自适应 = 10) are compared with each other, then in ζ 自适应 = 10 can be explained as follows: For discretization, the best possible acceleration value obtained in the last (previous planning step before the current one) planning step is used in the current planning step.

[0123] It should be noted that the cost function Φ 优化 The reference solution only forms an approximation to the best possible (optimal) solution in order to be able to achieve a relative cost of less than zero. In a dynamic environment, changes occur more slowly than the calculation time of the planning cycle (i.e., a relatively longer time is required), so that the control system or control method can be adapted without any problems and find a suboptimal acceleration depending on the current traffic situation.

[0124] The two-stage planning and evaluation concept as part of the control system 10 is designed for simultaneous planning of longitudinal and lateral movements, since the trajectory to be planned for the host motor vehicle 12 is to follow any desired curve course and simultaneously take into account the traffic situation directly in front of, to the side and behind the host motor vehicle 12. Therefore, the lateral and longitudinal sampled states are transformed and combined in order to generate trajectory candidates.

[0125] Then refer to Figure 6 To transform the curve coordinates into vehicle coordinates, the control system 10 or its electronic controller selects a reference lane marking (see, for example, Figure 3 The reference lane marking 32 is defined as a square polygon f 参考 To calculate the arc length, the infinite integral is approximated and the trapezoidal rule is applied. In this way, a lookup table is generated for queries with the same spacing within the region of the identified lane markings. F x i ∈[x s , x e ]. It follows that K x to the initial solution of the transformation F The starting value (initial value) of the lateral part is given by F y0=(f F x0) is given. Using the orthogonal slope m⊥=-1 / f′ 参考 ( F x) to obtain the transformed position in the x and y directions of the vehicle coordinates:

[0126]

[0127]

[0128] Accordingly, Figure 6 It shows how to transform from the curvilinear coordinate system K to the vehicle coordinate system F. K Evaluation of the lookup table for x = L provides the position F x0 (the position in the x direction in curvilinear coordinates). The arc length is then calculated by numerical integration of the reference polygon. The position is thus obtained F y0 (the position in the y direction in curvilinear coordinates). K When y=N, the desired coordinates in the vehicle coordinate system F are obtained from trigonometric considerations.

[0129] After the position has been transformed into vehicle coordinates, the stopping points (also called support points or knot points) are determined by so-called downsampling and the set of trajectory candidates is calculated by means of the spline interpolation described above. Figure 7 Such a generated set of trajectory candidates is shown in FIG. 1 within the scope of an exemplary traffic scenario (also referred to as a driving situation or traffic situation). (reference number 36 ). In each case, one can see a host motor vehicle 12 , which is following another motor vehicle 28 on the right-hand lane 34 of a two-lane road 32 ′. Figure 7 The upper image of shows a scenario with another motor vehicle 28 as a (static or dynamic) obstacle for a straight section. Figure 7 The corresponding scenario for a curved road section is described in the lower image of .

[0130] In other words, Figure 7 A plurality of combined candidate trajectories 36 (also referred to as sets) of the own motor vehicle 12 are shown, that is to say candidate trajectories that are composed of or derived from lateral and longitudinal samples (position and / or speed). The individual trajectories that make up the set of candidate trajectories 36 are each shown as gray lines and are connected by stop points that are shown on the line in the form of dots.

[0131] An important aspect of trajectory planning based on sampling is to obtain candidate trajectories 36 accurately and quickly. Therefore, the two-stage planning and evaluation scheme mentioned at the beginning is used within the scope of the present disclosure. Thus, the comfortable driving for humans is combined with the aspect of accurately avoiding collisions. This should reflect human driving behavior on the one hand in the context of candidate trajectories 36 (and the optimal trajectory ultimately obtained from the candidate trajectories mainly for the current traffic conditions of one's own motor vehicle 12), but on the other hand, safe and accurate operation must also be guaranteed. Therefore, the candidate trajectories 36 of each operation type are evaluated (tested) by a comprehensive objective function, which models the physically affected safety distance for avoiding dynamic collisions, and in addition takes into account the avoidance of static collisions, feasibility targets (feasibility targets) and comfort targets for the current traffic conditions of one's own motor vehicle 12. Thus, the optimal (best possible) trajectory is selected for each current driving condition and for each operation type.

[0132] For safety reasons, each selected optimal trajectory is searched using an accurate geometric image of the participating vehicles. If the search fails, an alternative strategy can be switched to, which is not described in detail here.

[0133] In order to determine the (globally) optimal trajectory from the trajectory candidates 36, the best possible trajectories of the individual maneuvering types are compared with one another. This comparison is performed with the aid of the described cost function Φ, so that the optimal trajectory simultaneously represents the best maneuvering type for the current driving situation of the own motor vehicle 12. The following expression represents the optimal trajectory for determining the optimal trajectory: comparison.

[0134]

[0135] Here, Φ represents the cost of the objective function. d and b s The following definition:

[0136]

[0137]

[0138] In addition, the expression ω d and ω s Forms weighting parameters for accurate collision control. Figure 8 Show that for Figure 7 The global optimal trajectory among the trajectory candidates 36 for the presented road segment or driving condition Here, the optimal trajectory The other trajectories used in the trajectory candidates 36 are shown as thicker and darker dots.

[0139] Next, the calculation of the objective function will be described, which is defined as a comprehensive objective function for evaluating (determining) the best possible trajectory for the own motor vehicle 12 Therefore, the various terms of the objective function take into account the avoidance of dynamic collisions with the help of the dynamic environment (field), the avoidance of static collisions with the help of the static environment (field), and the consideration of following the optimal trajectory. The feasibility / implementability of the best possible trajectory and the driving comfort of the own motor vehicle 12. Figure 8 In the figure above (where the optimal trajectory for a straight section is evaluated ), and in Figure 8 In the figure below (where the optimal trajectory for the curved section is evaluated ), the evaluation results in a left lane change operation, so that when the own motor vehicle 12 follows this obtained optimal trajectory in the further progress of the current traffic situation , the own motor vehicle can overtake the other motor vehicle 28, for example because the other motor vehicle is traveling slower than the own motor vehicle 12 and / or the other motor vehicle must perform a braking process (which forces the own motor vehicle 12 to perform an evasive maneuver).

[0140] The cost function factored into the objective function is defined as follows:

[0141] Φ=e T Ωe.

[0142] Here, Ω forms the weighting matrix The expression e represents the error vector, which is determined as follows:

[0143] e=[o1,o2,...,o z ,χ(h1),χ(h2),...,χ(h n-1 )] T

[0144] Expression o i Here, the objective function is formed. k )=max{[0,h k ]} Unequal conditions are taken into account. In this way, the conditions of the current driving situation imposed on the own motor vehicle 12 by the static and dynamic environment are also included. In addition, conditions and incomplete vehicle characteristics are detected regarding the vehicle dynamic characteristics of the own motor vehicle 12 (such as its maximum acceleration, which can be defined by the resultant force circle (Kammschen Kreis)). Deviations from the desired target state can also be formulated as the target of the objective function. This target state is defined by a higher-level planning module and is not described in detail here. In order to meet the goals regarding driving comfort (o i ), so that in following the optimal trajectory The lateral (lateral) acceleration in the lateral direction and the longitudinal direction of the own motor vehicle 12 as well as the (longitudinal) acceleration occurring in the direction of travel and the jerk (the derivative of the acceleration with respect to time) are minimized.

[0145] Next reference Fig. 9 The collision test basis used here is explained. In order to ensure that the candidate trajectory 36 for the own motor vehicle 12 avoids collisions, an accurate geometric model of the own motor vehicle 12 and the other motor vehicle 28 in the current situation (i.e., in general, a model of the motor vehicle that could potentially collide) is important. Within the scope of the present disclosure, bounding boxes are used to form the shape of the own motor vehicle 12 and the other motor vehicle 28 and / or other motor vehicles involved in the current traffic situation. This can be done, for example, from Fig. 9 Rectangular boxes are particularly suitable for mapping convex shapes (as are often used in motor vehicle bodies) with high accuracy.

[0146] In the scope of dynamic collision test, the high accuracy of the model also leads to a higher computational workload of the control system (in the present case, the control system 10) implementing the collision test. In order to avoid this, for example, hierarchical pruning of possible collisions can be performed, but this in turn causes a reduction in the number of accurate collision tests. In contrast, the multiple interference test realizes the test of each individual candidate trajectory point by point for all obstacle trajectories (i.e., the current driving condition of the motor vehicle 12 of the own motor vehicle 12 and / or other motor vehicles 28 based on its current driving behavior and the motor vehicle 12 of the own motor vehicle 12 and / or other motor vehicles 28). This point-by-point test corresponds to the most serious situation in terms of the computing resources of the control system being implemented, but its calculation must always be taken into account in the feasible time for safety considerations. In order to reasonably evaluate the dynamic environment, collision tests are performed in space and time. A separation set (Eidelheit set) is used here to form a temporal overlap between two rectangles.

[0147] Fig. 9 It can be seen that the own motor vehicle 12 and the other motor vehicle 28 (both of which are formed as rectangles) do not collide if there is a straight line onto which the two rectangles (i.e., objects) can be projected and the objects do not overlap on the straight line. Fig. 9 This is the case in the example shown in . Within the scope of the dynamic collision test, for each current time point t k , it is tested whether all trajectory candidates 36 of the own motor vehicle 12 overlap with the trajectory data of the further motor vehicle 28 obtained by the own motor vehicle 12 .

[0148] The runtime of the dynamic collision test is Fig.10 . In the present exemplary embodiment, for an accurate dynamic collision test using bounding boxes, the average run time obtained for 10,000 test runs with a trajectory consisting of 31 trajectory points is used. It can be clearly seen that the average run time depends on the number of obstacles (e.g. other motor vehicles) in the environment of the own motor vehicle 12. For a larger number of obstacles (e.g. seven), an average run time of approximately 100 μs results.

[0149] Fig.11 An example of an accurate dynamic collision test in the space-time domain using a separation set is shown. The own motor vehicle 12 and the other motor vehicle 28 are modeled as rectangles. In this case, each individual rectangle 12 shown (in Fig.11The rectangles 28 in the figure, whose positions change over time in the x direction, represent the positions of the own motor vehicle 12 in the space-time domain. The stacked rectangles 28 each form the positions of the other motor vehicles 28 in the space-time domain. At each time point t k The optimal trajectory is tested against one or more trajectories obtained for the other motor vehicle 28 (for which the control system 10 of the own motor vehicle 12 is configured and determined) Overlap or not. Detected overlaps are displayed here with possible collisions.

[0150] In addition, a static collision test is performed, for example to detect an impending collision of the own motor vehicle 12 with a street boundary. This is achieved by calculating the minimum distance between the bounding box of the own motor vehicle 12 and the bounding box of the corresponding lane marking.

[0151] exist Fig.12 For the sake of clarity, the two-stage or two-layer trajectory planning and evaluation scheme or its two planning levels are shown again. Here, the left-hand picture with a "1" in the circle shows the first planning stage. In which trajectory candidates 36 are determined from the objective function (see also Figure 2 and related instructions). The second planning step involves the collision tests just described (static and dynamic) and Fig.12 is shown on the right side.

[0152] refer to Fig.13 and 14 , the mode of operation of the control system 10 or the two-stage trajectory planning and evaluation concept will now be demonstrated using a real traffic scenario. Fig.13 and 14 The traffic scenario involves entering a highway from a ramp, where multiple other motor vehicles (in the ramp section) Fig.13 and 14 All motor vehicles not marked with the number 12 in FIG. 1 ), for example, another motor vehicle 28. Fig.10 and 11 There are 2 seconds between each of the five graphs in .

[0153] For a specific example, Fig.13 and 14The own motor vehicle 12 shown in FIG. 1 is equipped with radar sensors at its front, rear and sides and additionally has a camera that monitors the area in front of the own motor vehicle 12. Static and dynamic environmental information is then acquired in the described manner from the environmental data provided to the control system 10. In the current driving situation, the control system 10 of the own motor vehicle 12 selects a reference lane for the transformation from the curvilinear coordinates K to the orthogonal vehicle coordinates F based on the camera data. In this case, the following lane marking is selected as the reference lane marking in the current driving situation or sampling situation: compared with other existing lane markings, the camera of the own motor vehicle 12 can detect the largest area from this lane marking. The accuracy and therefore the quality of the result are increased by the further inclusion of map data, which are provided to the control system 10, for example, by a navigation system present in the own motor vehicle 12 or in any other suitable manner, wirelessly or by wire.

[0154] The control system 10 operates, for example, with model predictive control (English: receding horizon control), since the environmental data used are updated cyclically for trajectory planning and collision avoidance.

[0155] For a non-limiting example of what is considered herein as the functionality of the control system 10, for a set of one hundred trajectories, an average runtime of 40 ms was used for a single core processor with a clock frequency of 3.30 GHz and a 6 MB cache. For the spline-based interpolation used to generate the candidate trajectories 36, an order of v = 5 was determined, and for the prediction horizon T P Three seconds was selected.

[0156] The trajectory of the basic maneuver in the middle lane (SM) is modeled as a spline with η = 3 support points, with time t between the splines. l＝2 =1,5s. For the trajectory of other types of operation, the number of support points is η=2.

[0157] exist Fig.13 and 14 In the context of the ramp-on-motorway maneuver shown in FIG. 1 , the vehicle identified in the figure by ID: 7 is first selected by the control system 10 as the further motor vehicle 28. This selection is based on the relevance of all further motor vehicles for the current driving situation of the own motor vehicle 12, which is obtained by the control system 10. The own motor vehicle 12 is in the situation shown ( Fig.131 (the first figure from the top) is traveling behind another motor vehicle 28 in the same lane 34 and currently has a higher driving speed than the other motor vehicle 28. The lane 34 is a curved ramp to the highway. Then the own motor vehicle 12 and the other motor vehicle 28 change from the lane 34 to the right lane of the highway.

[0158] At this point in time, the driver of own motor vehicle 12 or driver assistance system 20 wishes to increase the longitudinal speed of own motor vehicle 12 to 90 km / h. At the beginning of this driving scenario, an overtaking process by own motor vehicle 12 is not possible, even though other motor vehicle 28 is currently traveling more slowly, for example at 70 km / h. The trajectory involved in a lane change to the left could with a high probability lead to a collision with other motor vehicle 36 that is itself overtaking. Later in the process (from Fig.13 ), the overtaking maneuver of the own motor vehicle 12 is prevented by the other motor vehicle 38. Therefore, the control system 10 selects a trajectory for lane keeping (SHM) and the own motor vehicle 12 is braked (for example by the driver assistance system 20) in order to maintain a constant distance from the other motor vehicle 28 traveling ahead (see Fig.13 2 to 5 from the top).

[0159] exist Fig.14 As can be seen in the top figure, the control system 10 examines the overtaking process for the current situation. That is, the control system 10 selects a track that keeps the lane to the left (SHL, see also Figure 3 The reference path 28b) in is used as the optimal trajectory That is, own motor vehicle 12 carries out a driving maneuver in which the own motor vehicle approaches a lane located to the left of its current lane in order to initiate an overtaking process.

[0160] Then in Fig.14 As can be seen in the second figure from the top, control system 12 again selects a trajectory that maintains the lane in the middle from trajectory candidates 36. That is, the overtaking maneuver (e.g., by control system 10 and / or driver assistance system 20) is canceled and own motor vehicle 12 is again essentially in the middle of its currently traveling lane.

[0161] The reason can be found in Fig.14 A motor vehicle 40, which is traveling at a higher speed than the own motor vehicle 12, approaches the own motor vehicle 12 from behind in the overtaking lane. Here, there is again a risk of collision, so that an overtaking maneuver of the own motor vehicle 12 is not possible.

[0162] exist Fig.14As can be seen from the fourth figure from the top, the faster-moving motor vehicle 40 is also moving in the longitudinal direction ( Fig.13 and 14 E in x ) has produced a relatively large distance with respect to its own motor vehicle 12, so that the overtaking operation of its own motor vehicle 12 is now possible. At this point in time, no other motor vehicle that is also traveling faster approaches from the rear on the overtaking lane.

[0163] Since the own motor vehicle 12 is to accelerate to an exemplary speed of 90 km / h and the other motor vehicle 28 is travelling more slowly, the other motor vehicle 28 no longer constitutes the most relevant traffic participant for trajectory planning as soon as an overtaking manoeuvre is possible. The control system 10 therefore selects the other motor vehicle 40 as a reference (traffic participant) for trajectory planning and calculates the optimal trajectory from the manoeuvre type of left lane change as described above. (See Fig.14 The own motor vehicle 12 then uses the optimal trajectory from the maneuver type in the middle lane to The other motor vehicle 40 is followed in the passing lane and the slower-moving further motor vehicle 28 is passed.

[0164] Fig.15 The motor vehicle 12 shown in FIG. Fig.13 and 14 The speed in each condition (each figure). Fig.13 and 14 The overtaking process shown in Fig.15 It starts at time t = 8s and Fig.15 As mentioned, at time t = 26s. Fig.13 and 14 There are two seconds between each graph. Fig.15 As can be seen in FIG. 1 , the own motor vehicle can first be accelerated shortly before reaching t=26 s in order to achieve the desired speed (for example, predetermined by the driver assistance system 10 ) of 90 km / h. This means that Fig.14 The last two figures (counting from the top) show an overtaking process performed by own motor vehicle 12 .

[0165] The present disclosure is directed to a control system (10) for planning candidate trajectories based on sampling. Suitable trajectory candidates 36 are generated and evaluated in real time. The sampling of the lateral is based on the basic operation type of the lateral. The discrete structure of the lane is thus used for semi-autonomous driving. In contrast to this lateral sampling of lateral values ​​in state space, longitudinal samples (i.e. sampled values ​​in the driving direction) are generated from the (dynamic) action space, wherein the longitudinal acceleration of the own motor vehicle 12 is used as a control parameter. The efficiency during sampling is increased by adaptive discretization.

[0166] The spline-based formulation of the trajectory and its use in curvilinear coordinates offers the advantage that the control system 10 can (essentially) be implemented and used both for curved and straight sections.

[0167] The two-stage approach for planning and evaluating trajectory candidates 36 is used to avoid collisions with trajectories of other road users who are located in the immediate driving situation in the vicinity of own motor vehicle 12. These trajectories are also respectively acquired by control system 10. In addition, the required driving comfort and the feasibility of the possible optimal trajectory are taken into account in the evaluation.

[0168] The lateral and longitudinal driving behavior of the own motor vehicle 12 is not predetermined by a higher control level, since the higher control level only provides the target state. Accordingly, in addition to trajectory planning for lateral maneuvers (such as lane changes), the presented sampling and evaluation strategy is inherently also capable of executing further driving maneuvers (such as following the vehicle in front or merging into a lane when the number of lanes is reduced).

[0169] It should be understood that the exemplary embodiments described above are not exhaustive and do not limit the subject matter disclosed herein. It will be apparent to those skilled in the art that the features of the various embodiments may be combined with one another and / or the different features of the embodiments may be omitted without departing from the subject matter disclosed herein.

Claims

1. A control system (10) configured and intended for use in a own motor vehicle (12), the control system identifying lanes, road boundaries, road signs and / or obstacles on the lane in an area (22, 24, 26) in front of, beside and / or behind the own motor vehicle based on environmental data acquired by at least one environmental sensor (14, 16, 18) assigned to the own motor vehicle, wherein the at least one environmental sensor is configured to provide an electronic controller of the control system (10) with environmental data reproducing the area in front of, beside and / or behind the own motor vehicle, and wherein the control system is at least configured and intended for: - obtaining information about the current driving situation of the own motor vehicle (12) and at least one further motor vehicle based on the provided environmental data, - obtaining a plurality of lateral positions based on the information about the current driving situation of the own motor vehicle (12) and / or the further motor vehicle, - obtaining a plurality of longitudinal positions and / or speeds based on the information about the current driving situation of the own motor vehicle (12) and / or the further motor vehicle, - determining the plurality of lateral positions and / or the plurality of longitudinal positions and / or the speeds of the own motor vehicle (12) and / or the other motor vehicle in curvilinear coordinates with respect to a reference lane provided to the control system (10) of the own motor vehicle (12) by means of the environmental data, and deriving from the plurality of lateral positions and the plurality of longitudinal positions and / or the speeds a stopping point for a trajectory of the own motor vehicle (12) which the own motor vehicle (12) is to follow in order to carry out a driving maneuver, and - determining the trajectory of the own motor vehicle (12) by means of spline-based interpolation in orthogonal coordinates between the determined stop points and as a function of the lateral position at the stop points.

2. The control system (10) according to claim 1, further configured and determined to: - based on the provided environmental data, a basic driving maneuver is obtained, which the own motor vehicle (12) is to carry out, and -determining at least one lateral position of the plurality of lateral positions based on the obtained basic operations and / or assigning a numerical value to at least one lateral position of the lateral positions, the numerical value representing a basic operation type, wherein the basic operation type is included in a basic operation type set, the basic operation type set being composed of at least basic operations for changing lanes to the left, changing lanes to the right and keeping lane.

3. The control system (10) according to claim 1 or 2, wherein the control system is further configured and determined to: - the plurality of longitudinal positions is determined based on a longitudinal acceleration of the own motor vehicle (12) and / or of the further vehicle.

4. The control system (10) according to claim 1 or 2 is also configured and determined to be used to: determine the multiple longitudinal positions by means of adaptive discretization, wherein the acceleration of the own motor vehicle (12) and / or the other motor vehicle is used as a control parameter for the adaptive discretization.

5. A control system (10) according to claim 1 or 2, wherein the control system is also configured and determined to be used to: determine the trajectory of the own motor vehicle (12) taking into account a static collision test, wherein the static collision test is based on multiple lateral positions of the own motor vehicle and / or the other motor vehicle.

6. The control system (10) according to claim 1 or 2 is also configured and determined to be used to: determine the trajectory of the own motor vehicle (12) taking into account a dynamic collision test, wherein the dynamic collision test is based on the multiple lateral positions and / or on the multiple longitudinal positions and / or speeds.

7. The control system (10) according to claim 6, further being configured and designed for carrying out the dynamic collision test taking into account the maximum acceleration of the own motor vehicle (12) and / or the further motor vehicle.

8. The control system (10) of claim 5, further configured and determined for: performing the static collision test using a separation set.

9. The control system (10) of claim 6, further configured and determined to perform the dynamic collision test using a separation set.

10. The control system (10) according to claim 5 is also configured and determined to: determine the trajectory of the own motor vehicle (12) taking into account a dynamic collision test, wherein the dynamic collision test is based on the multiple lateral positions and / or on the multiple longitudinal positions and / or speeds, and to perform the static collision test and the dynamic collision test using separate sets.

11. The control system (10) according to claim 1 or 2, further configured and designed to check the trajectory of the own motor vehicle (12) by means of an objective function based on a cost function.

12. The control system (10) according to claim 11, wherein a target state is assigned to the objective function and wherein a deviation of the objective function from the target state is used as a target for the objective function.

13. A control system (10) according to claim 12, wherein a further target of the target function in the form of lateral and / or longitudinal acceleration of the own motor vehicle (12) and / or in the form of lateral and / or longitudinal jerk of the own motor vehicle is used as the target state of the target function.

14. A control method, in a own motor vehicle (12), based on environmental data acquired by at least one environmental sensor (14, 16, 18) assigned to the own motor vehicle (12), identifying lanes, road boundaries, road signs and / or obstacles on the lane in the area in front of, beside and / or behind the own motor vehicle (12), wherein the control method comprises at least the following steps: - obtaining information about the current driving situation of the own motor vehicle (12) and at least one further motor vehicle based on the provided environmental data, - obtaining a plurality of lateral positions based on the information about the current driving situation of the own motor vehicle (12) and / or the further motor vehicle, - obtaining a plurality of longitudinal positions and / or speeds based on the information about the current driving situation of the own motor vehicle (12) and / or the further motor vehicle, - determining the plurality of lateral positions and / or the plurality of longitudinal positions and / or the speed of the own motor vehicle (12) and / or the further motor vehicle in curvilinear coordinates with respect to a reference lane provided to the control system (10) of the own motor vehicle (12) by means of the environmental data, obtaining from the plurality of lateral positions and the plurality of longitudinal positions and / or the speeds a stopping point for a trajectory of the own motor vehicle which the own motor vehicle is to follow when carrying out a driving maneuver, and - determining the trajectory of the own motor vehicle (12) by means of spline-based interpolation in orthogonal coordinates between the determined stop points and as a function of the lateral position at the stop points.

15. The control method according to claim 14, wherein the control method is implemented by means of a control system (10) according to any one of claims 1 to 13.

16. A motor vehicle comprising a control system according to any one of claims 1 to 13.

Images